Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Response of deep-ocean ostracodes to climate extrema of the Paleogene: Ecological, morphological, and geochemical data from the Eocene -Oligocene transition and late Paleocene thermal maximum
(USC Thesis Other)
Response of deep-ocean ostracodes to climate extrema of the Paleogene: Ecological, morphological, and geochemical data from the Eocene -Oligocene transition and late Paleocene thermal maximum
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. ProQuest Information and Learning 300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA 800-521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESPONSE OF DEEP-OCEAN OSTRACODES TO CLIMATE EXTREMA OF THE PALEOGENE: ECOLOGICAL, MORPHOLOGICAL, AND GEOCHEMICAL DATA FROM THE EOCENE-OUGOCENE TRANSITION AND LATE PALEOCENE THERMAL MAXIMUM by Stephen Allen Schellenberg A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA in Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (EARTH SCIENCES) May 2000 Copyright 2000 Stephen Allen Schellenberg Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3054898 Copyright 2000 by Schellenberg, Stephen Allen All rights reserved. _ _ ® UMI UMI Microform 3054898 Copyright 2002 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA 90007 This dissertation, written by Stephen A. Schellenberg under the direction of /l.A?...... Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School in partial fulfillment of re quirements for the degree of DOCTOR OF PHILOSOPHY Dean of Graduate Studies Date DISSERTATION COMMITTEE Chairperson Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stephen Allen Schellenberg David J. Bottjer Response of Deep-Ocean Ostracodes to Climate Extrema of the Paleogene: Ecological, Morphological, and Geochemical Data from the Eocene-Oligocene Transition and Late Paleocene Thermal Maximum The Eocene-Oligocene transition is marked by a pronounced -1.5 % o decrease in foraminiferal 5,80 termed Oi-1, reflecting ice-volume expansion and bottom-water cooling. To assess temperature change at ODP Site 744A (Kerguelen Plateau), the temperature-dependent Mg/Ca ratios of Krithe ostracodes were determined by ICP-MS. Krithe Mg/Ca shows no net trend through Oi-1, indicating that the 51 flO shift largely reflects major Antarctic ice-sheet development. Hoever, major deep-ocean ostracode faunal abundance, richness, and evenness changes coincide with Oi-1 at ODP Sites 744A and 689B (Maud Rise). Following Oi-1, Site 744A faunal composition shifted slightly back towards initial compositions, whereas Site 689B faunas remained relatively distinct. The Oi-1-coincident faunal changes include both uniform (e.g., disappearance of Cytherella and Bairdioppilata at both sites) and reversed (e.g., Site 689B disappearance and Site 744A appearance of Trachyleberis and Actinocythereis). Together, these faunal patterns imply a common response to increased Oi-1 surface productivity and a disparate response to either subtle productivity intensity differences or circulation changes. At Site 744A, morphometric examination of seven Algulhasina quadrata instar stages show a strong positive correlation with biogenic opal, indicating that increased organic carbon Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. delivery after Oi-1 allowed increased instar size through sequestering of greater metabolic reserves. The faunal response of deep-ocean ostracodes to the Late Paleocene Thermal Maximum (LPTM) was also examined at ODP Sites 690B (Maud Rise) and 738C (Kerguelen Plateau). Abundances drop precipitously at the major 5'3 C negative excursion and then recovers to roughly two-thirds to fully pre-event levels. The Site 690B fauna was overall taxonomically richer than the Site 738C fauna. Site 738C richness recovered after the LPTM, whereas Site 690B richness remained low, but less variable overall, perhaps reflect lingering deleterious conditions. Both sites contain three distinct faunas bounded by the carbon isotope spike and later recovery to relatively stable isotope values. Prolonged LPTM bottom-water dysoxia is not supported by the stable proportion of dysoxia-tolerant filter-feeding ostracode taxa. The restriction of specific taxa to particular stratigraphic intervals within the event provides provisional paleoecological preferences for testing at other LPTM sites and other intervals. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In memory of Fay Gross (22 September 1922 — 9 December 1999) “ What is laid down, ordered, factual, is never enough to embrace the whole truth: life always spills over the rim of every cup.” — Boris Pasternak “Complicated? It’s alright.” — Poi Dog Pondering “. . . and it really did happen.” — Paul Thomas Anderson Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements To my parents I truly owe it all. Jo and Karl’s love and support were unfailingly present — the same for my brother John and sister-in-law Kara. I would dedicate this work to my family, but I cannot imagine a worse compliment. Fellow young scientists at USC were a source of encouragement, insight, advice, and support, particularly Kathleen Campbell, Andrew Meigs, Whitey Hagadorn, Nicole Fraser, Heather Moffatt, and Becky Robinson. The USC Earth Science staff provided an incredibly supportive work environment, particularly John McRaney and Cindy Waite, who kept me in the green and on track, respectively. Students in my USC and CSU- Dominquez Hills courses inspired me to truly understand through articulation many concepts that I took for granted. Numerous mentors have influenced me in ways they may or may not have imagined, or perhaps wish acknowledged. Warren Allmon (then USF, now PRI) bears original blame for his investment of time and energy in a young Florida public university student with more attitude than ability. A Smithsonian NMNH RTP internship with Doug Erwin made it clear to me that nothing happens after the Paleozoic, and I’m here to report that I find nothing more fascinating. Terry Quinn (USF) provided the opportunity to get my feet into aqueous solutions through an M.S. degree in geochemistry. For the Ph.D. degree this tome represents, I traded south Florida humidity for southern California smog, coming to USC to work with Dave Bottjer, who provided the freedom to wander to many dead-ends but never over cliffs, before I (finally) settled on a dissertation project, only to have it torpedoed by a Canadian. Thus, I followed my heart on this project, originally my minor proposal, and I thank Dave for all his support, feedback, and patience through the years, as well as the freedom to chase research endeavors north (UC-Santa Cruz) and east (Smithsonian NMNH) towards the end of my Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i v Ph.D. journey. Conversations with Ai Fischer were always enlightening and too few and far between, largely due to his inspiring globe-trotting. Many other USC faculty were of great help during my studies, particularly Greg Davis, Bob Douglas, Don Gorsline, Doug Hammond, Steve Lund, Loren Smith, and Lowell Stott. Special thanks to those poor souls who served (rather long) sentences with no parole on my Ph.D. committee. At the Smithsonian NMNH, Richard Benson provided a world view fundamentally different from that of my academic upbringing - simultaneously confusing my paradigm, focusing my questions, and broadening my perspective. In addition, Dick provided the financial support that allowed me to extend my productive stay at the NMNH. Special thanks to Carlita Sanford for giving up her place in the sun (literally) and all the help in making the science go forth. Ralph Chapman, master of myriad morphometries, was a touchstone of pragmatism, statistical insight, and most excellent music. Brian Huber graciously allowed use of his image analysis system and picking of his brain regarding Cenozoic paleoceanography. Peter Wilf provided excellent programming assistance and patience. Amanda Ash worked her magic upon digital plates and largely kept me sane whilst in an institution — if only I could have reciprocated by keeping Hyde Park squirrels better at bay. Lunchtime discussions with the “paleochlorophyll crew” (Scott Wing, Amanda Ash, and Peter Wilf) taught me that plants are only slightly more interesting that I’d previously thought, constantly reminded me that I had yet to finish my Ph.D., and collectively clarified many analytical procedures upon the trusty lunchtable blackboard. Walt Brown, now enjoying the good, wet, retired life in the Pacific Northwest, and Susanne Braden allowed my unintentional, but exhaustive, search of the NMNH SEM for every minor software and hardware glitch. Enriqetta Barrera (U. of Akron), Lisolette Deister-Haass (U. ties Saarlandes), Gerta Keller (Princeton), Jim Kennett (UC-Santa Barbara), Lowell Stott (USC), Ellen Thomas (Yale/Wesleyan), Jim Zachos (UC-Santa Cruz), ard Rainer Zahn (Geomar) are Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V thanked for their time, ODP samples, and paleoenvironmental data; the latter provided the crucial independent framework for this study. On the geochemical front, thanks to Father Bill Cain (Loyolla Marymount University) for liberal access to and experience gained on his ICP-AES, and to Dan Sampson and Terry Quinn for help in my “ commando ICP-MS missions” to UC-Santa Cruz. On the ostracode front, Tom Cronin (USGS), Paul Steineck (SUNY-Purchase), and the Honorable Chap Robin Whatley (U. of Wales, Aberystwyth) are thanked for their thoughts and comparative materials. Finally, immense gratitude to all of my friends for keeping me sane, carrying me through some tough times, and getting me out of the lab (sometimes not a challenge). In addition to the above young scientists, these include, on the left coast, Tom Lazaroff, Suzanne Link, Susan Nyman, Mike/Jim Reynolds, Tim “ Tasty” Ross, and Amber and Daniel Defabio — the latter are thanked for providing the simplest explanation of my job for those in the LA entertainment industry: “He is like Ross on Friends.” On the right coast, thanks to Allison Alcott, Mikial Carroll, Gretchen Cook, David Ellenberger, Mark Friaili, Norissa Giangola, Heather Goguen, Gayle Levy, Peg Sewall, John/Jack Vaughan, and Jenn Young. Most of them will never read the following and for that they are likely grateful. Funding and support for this study was provided by the JOI/USSAC Ocean Drilling Fellowship Program, Smithsonian Predoctoral Fellowship Program, Smithsonian Institution, Amoco Corporation, USC Department of Geology, Wrigley Institute of Marine Science, Geological Society of America, and, as with many endeavours (and foibles) throughout the years, HKS Enterprises. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vi Table of Contents Dedication......................................................................................................................................... ii Acknowledgments........................................................................................................................... iii 1. Introduction......................................................................................................................................1 1.1. Scope of Dissertation...............................................................................................................1 1.2. Deep-ocean ostracode biology and ecology..........................................................................6 1.3. Modem ocean-atmosphere circulation in the Southern Ocean...................................1 8 1.4. Transient climate extrema................................................................................................. 2 4 1.5. Geologic, stratigraphic, and geochronologic setting......................................................2 7 1.5.1. Eocene-Oligocene Transition....................................................................................... 2 7 1.5.2. Late Paleocene Thermal Maximum.............................................................................2 8 2. Krithe Mg/Ca paleothermometry during the Eocene-Oligocene transition: An independent assessment of 81 8 0 -based paleoceanographic reconstructions 3 0 2.1. Introduction 3 0 2.1.1. Scope 3 0 2.1.2. The importance and difficulty of reconstructing paleotemperature 3 3 2.1.3. Mg/Ca paleothermometry: Physiology and thermodynamics 3 8 2.2. Materials and Methods 4 3 2.2.1. Stratigraphic and isotopic context of sample population 4 3 2.2.2. Krithe sampling, cleaning, and image archiving 4 4 2.2.3. Diagenetic screening through cathodoluminescence and SEM 4 4 2.2.4. Elemental determination via ICP-MS 4 6 2.3. Results 4 6 2.3.1. Phenotypic, environmental, and elemental relationships 4 6 2.3.2. Interpretation of stratigraphic elemental variations 6 3 2.4. Discussion............................................................................................................................... 6 8 2.5. Conclusion................................................................................................................................7 0 2.6. Future directions and suggestions.....................................................................................7 1 3. Deep-ocean ostracode faunal response to the Eocene-Oligocene transition: High-resolution (104 yr) records from the Southern Ocean (ODP Sites 744A and 689B )..................................................................................................................................... 7 5 3.1. Introduction............................................................................................................................7 5 3.1.1. Scope..................................................................................................................................7 5 3.1.2. Eocene-Oligocene transition.......................................................................................7 8 3.1.3. Global ostracode response............................................................................................8 8 3.2. Stratigraphic and paleoenvironmental context..............................................................9 3 3.2.1. Site descriptions............................................................................................................9 3 3.2.2. Paleoenvironmental framework................................................................................ 9 4 3.3. Materials and methods..........................................................................................................9 8 3.4. Results...................................................................................................................................1 0 0 3.4.1. Ostracode abundance and richness........................................................................... 1 00 3.4.2. Richness assessment through rarefaction.............................................................108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. v i i 3.4.3. Faunal distributions...................................................................................................11 4 3.4.4. Diversity analysis...................................................................................................... 1 28 3.4.5. Faunal Turnover..........................................................................................................1 48 3.4.6. Cluster analyses — stratigraphic and taxonomic................................................ 1 4 9 3.4.7. Principal components analysis — stratigraphic and taxonomic........................1 55 3.5. Discussion............................................................................................................................1 82 3.5.1. Overview......................................................................................................................1 8 2 3.5.2. Ostracode abundance, richness, and diversity.....................................................1 82 3.5.3. Krithe, Algulhasina, and potential gradual temperature change...................... 1 8 7 3.5.4. Pre-Oi-1 taxa — the platycopid response............................................................1 88 3.5.5. Reverse-response taxa............................................................................................. 1 92 3.5.6. Other taxa.....................................................................................................................1 93 3.5.7. Faunal patterns within Oi-1....................................................................................... 1 9 4 3.6. Conclusions......................................................................................................................... 1 95 3.7. Future Directions..............................................................................................................1 9 6 4. Linkage of surface productivity and benthic ostracode size during the Eocene- Oligocene transition: Algulhasina quadratica ontogeny at ODP Site 744A (Kerguelen Plateau, Southern Ocean)..................................................................................1 97 4.1. Introduction........................................................................................................................ 1 9 7 4.2. Paleoenvironmental Setting............................................................................................201 4.3. Materials, and Methods..................................................................................................... 2 0 4 4.4. Results..................................................................................................................................2 0 7 4.4.1. Stratigraphic abundance...........................................................................................2 0 7 4.4.2. Ontogenetic size analysis of Algulhasina quadratica.......................................... 2 0 8 4.5. Discussion.............................................................................................................................2 2 6 4.6. Conclusions.......................................................................................................................... 2 3 4 4.7. Future Directions...............................................................................................................2 3 5 5. Deep-ocean ostracode faunal response to the Late Paleocene Thermal Maximum: High-resolution (104 yr) records from the Southern Ocean (ODP Sites 690B and 738C)....................................................................................................................................2 3 6 5.1. Introduction......................................................................................................................... 2 3 6 5.1.1. Scope.............................................................................................................................. 2 3 6 5.1.2. Deep-ocean ostracode response............................................................................... 2 4 0 5.2. Materials and Methods...................................................................................................... 2 4 6 5.2.1. Site descriptions......................................................................................................... 2 4 6 5.2.2. Age models.....................................................................................................................2 4 6 5.2.3. Paleoenvironmental framework..............................................................................2 4 7 5.2.4. Faunal identification and tallying........................................................................... 2 5 2 5.3. Results..................................................................................................................................2 5 3 5.3.1. Primary data................................................................................................................ 2 5 3 5.3.2. Sample size and richness relationships: Rarefaction assessment...................261 5.3.3. Relation of faunal abundance and richness to paleoenvironment......................271 5.3.4. Diversity analysis.......................................................................................................271 5.3.5. Faunal Turnover.......................................................................................................... 2 7 7 5.3.6. Cluster analyses — stratigraphic and taxonomic.................................................2 8 2 5.3.7. Principal components analysis — stratigraphic and taxonomic.................... 2 9 3 5.4. Discussion............................................................................................................................31 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vi i i 5.4.1. Overview...................................................................................................................... 3 1 3 5.4.2. Ostracode abundance, richness, and diversity.....................................................3 2 0 5.4.3. Whole-fauna patterns through the LPTM.............................................................3 2 2 5.4.4. Podocopids, platycopids, and benthic foraminifera: constraints on paleooxygenation levels.............................................................................................32 4 5.4.5. LPTM taxa: implications for paleoecology and dispersal..................................3 2 9 5.4.6. Comparison of LPTM records on Maud Rise: Sampling constraints and environmental comparison....................................................................................... 331 5.5. Conclusions..........................................................................................................................3 3 2 5.6. Future Directions.............................................................................................................. 3 3 3 6. Summary of major conclusions........................................................................................... 3 3 5 6.1. No evidence for temperature change in Kerguelen Plateau bottom-waters through the latest Eocene to earliest Oligocene...........................................................3 3 5 6.2. Common and reversed faunal changes at Kerguelen Plateau and Maud Rise coincide with the onset of O i-1...................................................................................... 3 3 6 6.3. Algulhasina quadratica shows significant ontogenetic size increase in response to early Eocene surface productivity increase......................................... 3 3 7 6.4. Common faunal changes at Kerguelen Plateau and Maud Rise support temperature increase, but not prolonged dysoxia, during LPTM .......................... 33 7 Bibliography................................................................................................................................... 3 4 0 Appendix 1. Age model data........................................................................................................... 3 6 3 Appendix 2. Ostracode faunal data...............................................................................................3 6 5 Appendix 3. Rarefaction Code.......................................................................................................39 7 Appendix 4. Bootstrap Description, Code, and Results........................................................... 40 0 Appendix 5. Algulhasina Morphometries Macro..................................................................... 4 0 6 Appendix 6. Algulhasina Morphometric Data...........................................................................41 0 Appendix 7. Taxonomic Plates..................................................................................................... 3 7 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. j X Table of Figures Figure 1.1. Global composite record of benthic foraminiferal stable isotope values for the early Cenozoic................................................................................................2 Figure 1.2. Southern hemisphere paleogeography and locations of Kerguelen Plateau and Maud Rise during the Late Eocene (-40 Ma)................................................4 Figure 1.3. Assumption versus testing of taxonomic uniformitarianism......................... 7 Figure 1.4. Generalized podocopid body plan as exemplified by Bicornucythere bisanensis..................................................................................................................... 9 Figure 1.5. Summary of modern bathymetric distribution of nine “key” ostracode taxa with respect to water masses in the Atlantic Ocean................................ 1 3 Figure 1.6. Summary of modern bathymetric distribution of seven “key” ostracode taxa with respect to water masses in the Southwest Pacific and Southern Ocean.................................................................................... 1 5 Figure 1.7. Simplified meridional cross-section of the Atlantic Ocean.............................1 9 Figure 1.8. Simplified block diagram of horizontal and vertical water mass movement in the Atlantic Ocean from -40° to 70° S...................................... 21 Figure 1.9. Simplified model of a climate extreme produced by the crossing of some threshold limit of stability......................................................................... 2 5 Figure 2.1. Global early Cenozoic composite record of benthic foraminiferal stable isotope values and paleogeographic location of Site 744A during the Latest Eocene............................................................................................................. 31 Figure 2.2. Timing of Paleogene glaciomarine sedimentation in the Southern Ocean | g and 5 O-based estimates of ice-volume and BWT based on the global 1 8 5 Ob f composite.........................................................................................................3 6 Figure 2.3. Core-top variations in Krithe spp. ostracode Mg/Ca versus observed bottom water temperatures from the North Atlantic and Arctic..................41 Figure 2.4. Elemental concentrations of internal reference standards for determination of Mg/Ca and Sr/Ca ratios via ICP-MS................................... 4 7 Figure 2.5. Krithe valve measurement and elemental ratios for Site 744A....................4 9 Figure 2.6. Relationships between Krithe valve size, mass, and clarity..........................5 2 Figure 2.7. Relationships of Krithe valve size, mass, and clarity (luminance) to determined elemental ratios of Mg/Ca and Sr/Ca.............................................5 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. X Figure 2.8. Relationships of Krithe Mg/Ca to biogenic opal and dissolution indices of % fragmented planktic foraminifera and % benthic foraminifera 57 Figure 2.9. Relationships of Krithe Sr/Ca to biogenic opal and dissolution indices of % fragmented planktic foraminifera and % benthic foraminifera....... 5 9 Figure 2.10. Relationships of Krithe valve size, mass, and clarity to determined elemental ratios of Mg/Ca and Sr/Ca within the two most sampled intervals.................................................................................................................. 6 1 Figure 2.11. Stratigraphic distribution of Krithe valve mass, area, and luminance for the chemically analyzed sample population.............................................. 6 4 Figure 2.12. Stratigraphic variations in Mg/Ca and Sr/Ca in the Krithe ostracodes through the Eocene-Oligocene event................................................................... 6 6 Figure 2.13. Relationship between BWT, Krithe valve mass, and Krithe Mg/Ca in the modern Atlantic Ocean..................................................................................... 7 3 Figure 3.1 Southern hemisphere paleogeography and ODP site locations during the Late Eocene (-40 Ma).....................................................................................7 6 Figure 3.2 Global composite record of benthic foraminiferal stable isotope values for the early Cenozoic.............................................................................................8 0 Figure 3.3 Cenozoic evolution of thermohaline circulation proposed by Kennett and Stott (1990).................................................................................................... 81 Figure 3.4 Stable isotope contours through time and paleodepth for ODP Sites 689, 690, 699, and 703......................................................................................8 5 Figure 3.5 Global summary of ostracode faunas from 1,600 50 cc samples from 155 DSDP Sites spanning the Cretaceous and Cenozoic..................................8 9 Figure 3.6 Rho-group similarity analysis of ostracode faunas along an east-west transect of the South Atlantic..............................................................................91 Figure 3.7 Paleoenvironmental framework for Site 744A and 689B.......................... 9 6 Figure 3.8 Ostracode abundance and richness at Sites 744A and 689B..................... 101 Figure 3.9. Ostracode abundance and dissolution indices comparison..........................1 03 Figure 3.10. Correlation of dissolution indices and ostracode valve abundance and richness at Site 689B..........................................................................................1 06 Figure 3.11. Ostracode abundance versus richness for Sites 744A and 689B...............1 10 Figure 3.12. Rarefaction curves for Site 744A and 689B................................................112 Figure 3.13. Absolute and relative abundances of taxa at Site 744A and 689B...........11 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.14. Pelecocythere foramen abundance versus total abundance and its relative abundance versus 81 8 0 ......................................................................... 11 9 Figure 3.15. Bradleya dictyon abundance versus total abundance and its relative abundance versus 81 8 0 at Site 744A.................................................................121 Figure 3.16. Indeterminate Genus E abundance versus total abundance and its relative abundance versus 81 8 0 at Site 744A................................................. 1 23 Figure 3.17. Pennyella abundance versus total abundance and its relative abundance versus 81 0 O at Site 744A.................................................................1 25 Figure 3.18. Cytherella spp. abundance versus total abundance and its relative abundance versus 61 8 0 at Sites 744A and 689B............................................ 1 2 9 Figure 3.19. Bairdoppilata abundance versus total abundance and its relative abundance versus 81 s O at Site 744A................................................................1 31 Figure 3.20. Abundance of key taxa with distinct stratigraphic distributions versus the dissolution index of % fragmentation of planktic forams for Site 689B........................................................................................................133 Figure 3.21. Abundance of key taxa with distinct stratigraphic distributions versus the dissolution index of % benthic forams for Site 689B............. 1 35 Figure 3.22. Actinocythereis abundance versus total abundance and its relative abundance versus 81 8 0 at Sites 744A and 689B........................................... 1 37 Figure 3.23. Trachyleberis abundance versus total abundance and its relative abundance versus 81 8 0 at Sites 744A and 689B........................................... 1 39 Figure 3.24. Bradleya sp. abundance versus total abundance and its relative abundance versus 81 8 0 at Site 744A................................................................ 1 41 Figure 3.25. Henryhowella asperima abundance versus total abundance and its relative abundance versus 81 8 0 at Sites 744A and 689B........................... 1 43 Figure 3.26. Stratigraphic variations in faunal evenness (E), Shannon information function (H), and its maximum potential value for a given sample (Hm a x )...................................................................................146 Figure 3.27. Intersample ostracode appearances and disappearances, turnover, and turnover rates for Sites 744A and 689B............................................... 1 50 Figure 3.28. Complete-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 744A......................................................... 153 Figure 3.29. Complete-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 689B......................................................... 156 Figure 3.30. Q-mode PCA eigenvalues for Site 744A faunal matrices........................... 1 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. x ii Figure 3.31. Q-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of Site 744A log-transformed abundances..............................................161 Figure 3.32. Varimax-rotated Q-mode PCA factor score plots for Site 744A log- transformed abundances.......................................................................................163 Figure 3.33. Q-mode PCA eigenvalues for Site 689B faunal matrices............................166 Figure 3.34. Q-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of Site 689B log-transformed abundances..............................................168 Figure 3.35. Varimax-rotated Q-mode PCA factor score plots for Site 689B log- transformed abundances.......................................................................................1 70 Figure 3.36. R-mode PCA eigenvalues for Site 744A faunal matrices............................1 74 Figure 3.37. Site 744A R-mode eigenvalue coefficients of taxa for varimax- rotated PCA of absolute abundance, log-transformed, and relative (%) abundance matrices......................................................................1 76 Figure 3.38. R-mode PCA eigenvalues for Site 689B faunal matrices............................ 1 78 Figure 3.39. Site 689B R-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of absolute, log-transformed, and relative (%) abundance matrices......................................................................1 80 Figure 4.1. Late Eocene paleogeography of Southern Ocean and ODP Site 744A on Kerguelen Plateau.................................................................................................. 1 99 Figure 4.2. Environmental context and Algulhasina quadratica abundance at ODP Site 744A........................................................................................................2 0 2 Figure 4.3. Orientation and measurements for Algulhasina quadratica......................... 2 0 5 Figure 4.4. Linear growth series (length versus height) for all complete Algulhasina quadratica specimens..................................................................... 2 0 9 Figure 4.5. Algulhasina quadrata length versus circularity and area............................21 1 Figure 4.6 Stratigraphic distribution of valve lengths of Algulhasina quadratica instars...................................................................................................................... 21 3 Figure 4.7. Normal distribution test of valve length for complete, pre-Oi-1, and post-Oi-1 intervals...................................................................................... 21 5 Figure 4.8. t-test (a = 0.05) of mean valve length differences between each pre-Oi-1 and post-Oi-1 instar population....................................................21 7 Figure 4.9. Linear correlation of each lengths within each instar to % biogenic opal and 51 8 0 for the entire study interval...................................................... 2 2 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. xiii Figure 4.10. Linear correlation of length within each instar to 51 3 C and % sand for the entire study interval.............................................................................. 2 2 2 Figure 4.11. Statistical significance of correlations between instar length and paleoenvironmental parameters for entire study interval........................ 2 2 4 Figure 4.12. Linear correlation of each lengths within each instar to % biogenic opal and 81 8 0 for post-Oi-1 interval........................................................................ 22 7 Figure 4.13. Linear correlation of each lengths within each instar to 51 3 C and % sand for post-Oi-1 interval..........................................................................2 2 9 Figure 4.14. Statistical significance of correlations between instar length and paleoenvironmental parameters for post-Oi-1 interval........................... 231 Figure 5.1. Late Paleocene (-55 Ma) paleogeography of the Southern hemisphere and locations of ODP Site 690B and 738C.......................................................2 3 7 Figure 5.2. Global summary of ostracode faunas from 1,600 50 cc samples from 155 DSDP Sites spanning the Cretaceous and Cenozoic.....................241 Figure 5.3. Response of deep-ocean ostracodes at Site 689B, Maud Rise.....................2 4 4 Figure 5.4. Paleoenvironmental framework for Site 690B.............................................2 4 8 Figure 5.5. Paleoenvironmental framework for Site 738C............................................. 2 5 0 Figure 5.6. Absolute and relative abundances of taxa at Site 690B and 738C............ 2 5 4 Figure 5.7. Ostracode abundance and richness at Sites 690B and 738C.......................25 9 Figure 5.8. Ostracode abundance versus richness for Sites 690B and 738C..............2 6 2 Figure 5.9. Rarefaction curves for entire sites and for LPTM divisions within each site..................................................................................................... 2 6 5 Figure 5.10. Rarefaction curves for individual Site 690B samples................................ 2 6 7 Figure 5.11. Rarefaction curves for individual Site 738C samples.................................2 6 9 Figure 5.12. Stable isotope values versus ostracode abundance and richness for Sites 690B and 738C.................................................................................... 2 72 Figure 5.13. Site 690B and 738C stratigraphic variations in faunal evenness...........2 7 5 Figure 5.14. Site 690B intersample ostracode appearances, disappearances, turnover, and turnover rate...............................................................................2 7 8 Figure 5.15. Site 738C intersample ostracode appearances, disappearances, turnover, and turnover rate...............................................................................2 8 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. xiv Figure 5.16. Average-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 690B......................................................... 2 8 4 Figure 5.17. Average-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 738C......................................................... 2 8 7 Figure 5.18. Average-linkage dendrogram of Jaccard coefficient matrix for taxonomic samples of Site 690B.......................................................................2 8 9 Figure 5.19. Average-linkage dendrogram of Jaccard coefficient matrix for taxonomic samples of Site 738C....................................................................... 291 Figure 5.20. Q-mode PCA eigenvalues for Site 690B faunal matrices...........................2 9 5 Figure 5.21. Q-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of Site 690B log-transformed abundances............................................2 9 8 Figure 5.22. Varimax-rotated Q-mode PCA factor score plots for Site 690B log- transformed abundances..................................................................................... 3 0 0 Figure 5.23. Q-mode PCA eigenvalues for Site 738C faunal matrices...3 0 4 Figure 5.24. Q-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of Site 738C log-transformed abundances............................................ 3 0 6 Figure 5.25. Varimax-rotated Q-mode PCA factor score plots for Site 738C log-transformed abundances............................................................................. 3 0 8 Figure 5.26. R-mode PCA eigenvalues for Site 690B faunal matrices...31 1 Figure 5.27. Site 690B R-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of absolute, log-transformed, and relative (%) abundance matrices.....................................................................3 1 4 Figure 5.28. R-mode PCA eigenvalues for Site 738C faunal matrices...31 6 Figure 5.29. Site 738C R-mode eigenvalue coefficients of taxa for non- and varimax-rotated PCA of absolute, log-transformed, and relative (%) abundance matrices....................................................................31 8 Figure 5.30. Stratigraphic abundance of ostracode valves and percent abundance of filter-feeding podocopid taxa for Sites 690B and 738C........................ 3 2 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 1. Introduction 1.1. Scope of Dissertation Paleobiologists face an apparent paradox in reconstructing the history of life. On one hand, we embrace taxonomic uniformitarianism, commonly using the ecological and environmental distribution of living individuals to reconstruct the paleoenvironments of their fossil counter-parts (e.g., Murray, 1995). On the other hand, we appreciate that taxa evolve, with genomic, ecological and environmental shifts not necessarily accompanied by morphologic (i.e., taxonomic) change (e.g., Schopf et al., 1975). Resolution of this paradox is neither straightforward, unique, nor inevitable — it is case-by-case and based on a mix of past experience, salient data, and provisional assumption, considered together within the operational framework of historical science sensu Simpson (1963). The following study is no exception. This dissertation presents detailed faunal, morphological, and geochemical data on the response of deep-ocean ostracodes to two climate extrema of the Paleogene: the Late Paleocene Thermal Maximum and the Eocene-Oligocene “greenhouse-icehouse” transition (Figure 1.1). ODP study sites for each time interval are located on the Kerguelen Plateau and Maud Rise in the Southern Ocean (Figure 1.2), an important region in the history of global ocean-atmosphere circulation, particularly thermohaline deep-water production. Interpretation of these data are based on consideration of both the modern ecologies of living representatives and independently-derived paleoenvironmental data. Although modem ecologies may not be directly applicable to Paleogene individuals given intervening ecological changes via evolution, it is assumed that paleoecological limits of specific taxa did not change significantly during the geologically short intervals of study. Thus, discussions are generally grounded in an amalgam of independent paleoenvironmental data, modern ecological data, and ancient Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 Figure 1.1. Global composite record of benthic foraminiferal stable isotope values for the early Cenozoic. Horizontal line indicates the major changes in 81 8 0 and 51 3 C associated with the Eocene-Oligocene transition (“Oi-1”) and the Late Paleocene Thermal Maximum (“LPTM”) (after Zachos, pers. comm., 1998) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 813C 5180 3 — I I 1 , 1 I I I I, 1 I ^ C l -L— I------1 ------> I I----- 1 ■ 1 J - L . ' I ----- 1 J 1----- I - Q|PP!tAl| Apeg aie-ilAjjeg -| eiPPiyy lApeg a^ei|A|jeg 8 U 0 O O !i^ o6;io 0 9 |B d > 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 Figure 1.2. Southern hemisphere paleogeography and locations of Kerguelen Plateau and Maud Rise during the Late Eocene (-40 Ma). LPTM sites consist of Site 738C (Kerguelen Plateau) and 690B (Maud Rise). Eocene-Oligocene sites consist of Site 744A (Kerguelen Plateau) and Site 689B (Maud Rise) (after Lawver et a/., 1992). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 689/690 738$<W Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 faunal data — the strong signals recorded in the first provide an opportunity to test the applicability of the second to the third (Figure 1.3). Four chapters form the core of the dissertation. Each is written as an independent manuscript for subsequent rumination, refinement, and publication. Chapters two, three, and four examine the geochemical, faunal, and morphological response, respectively, of deep-ocean ostracodes to the Eocene-Oligocene transition from “greenhouse" to “icehouse" state. This transition was marked by the onset of significant circum-polar surface-water circulation, high southern latitude bottom- water production, and Antarctic ice-sheet development. Chapter 5 examines the faunal response of deep-ocean ostracodes to the Late Paleocene Thermal Maximum, a geologically short-lived, but extreme warming event driven by massive sublimation of shelf and slope methane hydrates that led to significantly increased pC02, associated global warming, and a pronounced geochemical and biotic signature. Given this brief outline of contents, the remainder of this introductory chapter provides a broad context for these studies through brief overviews of (1) deep-ocean ostracode biology and ecology, (2) modern ocean-atmosphere circulation in the Southern Ocean, (3) transient climate extrema, and (4) general geologic, stratigraphic, and geochronologic setting of the study sites. 1.2. Deep-ocean ostracode biology and ecology The Ostracoda are a class of minute crustaceans with an exceptional fossil record of 40,000-50,000 species ranging from the Cambrian to Recent. The class is characterized by trunk segment reduction, organ system loss, lateral body compression, and five to seven pairs of specialized appendages (Figure 1.4). The body and appendages are enclosed within a bivaived carapace, a cuticular fold derived embryonically from the cephalic segment (Brusca and Brusca, 1990). The external surface of the carapace is Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 Figure 1.3. Assumption versus testing of taxonomic uniformitarianism. If one assumes no ecological shifts between modern and ancient representatives of a given taxon, then the paleoenvironment of the ancient representative may be interpreted. Conversely, if one compares the modern ecosystem to a robust, independently-derived paleoenvironment, the validity of taxonomic uniformitarianism may be tested. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Modern Ecosystem Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9 Figure 1.4. Generalized podocopid body plan as exemplified by Bicornucythere bisanensis. A. Lateral view with left valve removed. B. Transect view through plane one in A. C. Transverse view through plane two in A. Schematic cross-section through carapace. Abbreviations: a1 = first antennae; a2 = second antenna; am = adductor muscle; an = anus; ca = carapace; col = calcified outer lamella; cprol = calcified procuticle of outer lamella; do = domociliar cavity; dt = digestive tract; ec = epidermal cells; epil = epicuticle of inner lamella; epol = epicuticle of outer lamella; evp(4) = epipodial ventilatory plate on fourth limb; hi = hinge; hs = haemocoelic space; il = inner lamella; li = ligament; Iv = left valve; mb = membrane; ol = outer lamella; me = medial eye; pb = posterior wall of body; pril = procuticle of inner lamella; ps = posterior spine; re = reticulated ornament; rv = right valve; sdc = subdermal cell; 5-7 = fifth to seventh pair of appendages, (from Vannier and Abe, 1995) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 impregnated with low-Mg calcite, while the uncalcified interior surface acts as a medium for gas exchange; housed within the carapace are cellular compounds and functional systems (Vannier and Abe, 1995). The outer carapace surface (generally termed valves) is heavily calcified and morphologically complex within higher taxa such as the Subclass Podocopida, reflecting various environmental conditions, structural solutions, and functional properties (Benson, 1981). Such groups have an exceptional fossil record and are employed for paleoenvironmental reconstruction (e.g., Kornicker and Wise, 1960; Cronin, 1979; Van Harten and Van Hinte, 1984; Whatley, 1991; Cronin et a i, 1999), biostratigraphic correlation (e.g., Sheppard, 1978; Cronin and Hazel, 1980; Keen 1982; Hazel, 1988), and evolutionary studies (e.g., Benson, 1982, 1983; Cronin, 1985; Whatley et al., 1986; Schweitzer and Lohmann, 1990). The Subclass Podocopida exhibits the greatest carapace calcification and dominates ostracode abundance and diversity in the post-Paleozoic benthic marine realm, particularly in bathyal and abyssal depths (Benson et ai, 1961; Dalingwater and Mutvei, 1990; Maddocks, 1992). Podocopids have a ventral gape through which thoracic appendages extend to crawl upon or through sediment and may feed upon any organic material, living or detrital. Respiration organs have been secondarily lost; gas exchange and osmoregulation occurs directly across the inner carapace and body surface, which constrains the subclass to a relatively small size (McMahon and Wilkens, 1983; Keyser, 1990; Vannier and Abe, 1995). The majority of biological and ecological studies of podocopids have focused on shallow-water faunas given their higher abundances and greater accessibility (e.g., Benson, 1959; Kaesler, 1966; Heip, 1976; Herman and Heip, 1982; Herman et a i, 1983; Martens, 1985; Cohen and Morin, 1990; Bodergat, 1993; Vannier and Abe, 1995). In recent years, however, a number of studies have examined the modern deep- ocean distribution of podocopid ostracodes within the context of bottom-water Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 2 circulation and surface-water productivity (e.g., Rosenfeld and Bein, 1978; Benson et al., 1983; Steineck et al., 1988; Dingle et al., 1989; Dingle and Lord, 1990; Correge and DeDeckker, 1997; Ikeya and Cronin, 1993; Cronin et al., 1994; Ayress et al., 1997). Together, these two processes strongly control local environmental parameters of bottom-water temperature, salinity, dissolved oxygen, C 0 2 -related corrosivity, sediment composition, and organic carbon form and abundance. The resulting data suggest that deep-ocean ostracodes are sensitive to benthic environmental conditions, paralleling similar data on modem benthic foraminifera (e.g., Lohmann, 1978; Douglas and Woodruff, 1981; Bremer and Lohmann, 1982; Lutze and Coulburn, 1984; Thomas, 1992; Gooday, 1993; Miao and Thunnell, 1993; Rathburn and Corliss, 1994; Thomas and Gooday, 1996; see Schnitker, 1994 for review). These ostracode and benthic foram studies, however, have yielded few unique taxon-environment correlations that hold on global scales (v. Thomas and Gooday, 1996; Schnitker, 1994). For example, based on a compilation of six regional ostracode studies in the Atlantic Ocean, Dingle and Lord (1990) demonstrated that the upper depth-limits of many taxa generally coincide with water mass boundaries (Figure 1.5). However, Ayress et al. (1997) found the Southern Ocean distribution of some predominant deep- ocean taxa to differ significantly from their Atlantic distribution with respect to water mass boundaries (e.g., Poseidonamicus has an upper depth-limit in lower North Atlantic Deep Water within the Atlantic, versus its predominate restriction to shallower Antarctic Intermediate Water in the South Pacific) (Figure 1.6). In the search for such organism-environment relations, it is important to employ a “ total evidence” approach in characterizing the ambient environment of a given fauna. Most modern ostracode studies, including the two regional syntheses above, have concentrated on more easily characterized conservative water-mass properties (i.e., temperature, salinity) over non-conservative properties related to local surface Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.5. Summary of modern bathymetric distribution of nine “ key” ostracode taxa with respect to water masses in the Atlantic Ocean. A. Study locations: A = Cape Basin of southwest Africa (Dingle et al., 1990), B = Angola Basin (Peypouquet and Benson, 1980), C = northwest Africa (Rosenfeld and Bein, 1978), D = DSDP Leg 94 sites (Whatley and Coles, 1987), E = Newfoundland (Benson et al., 1983), F = southeast USA (Cronin, 1983). B. Summary of major water masses and their relation to study sites: Vertical scale in kilometers; horizontal scale in degrees of latitude. Vertical bars indicate depth range covered at study sites. Arrows indicate general flow of water masses. Water masses: AABW = Antarctic Bottom Water, AAIW = Antarctic Intermediate Water, LSW = Labrador Sea Water, NADW = North Atlantic Deep Water, ML = Mixed Layer, MW = Mediterranean Water. C. Bathymetric relations of common ostracode taxa to water masses. Solid lines are known, dashed lines are uncertain, vertical scale in kilometers. Taxa: 1 = Henryhowella, 2 = Krithe, 3 = Echinocythereis, 4 = Rugocythereis, 5 = Ligitimocythere, 6 = Buntonia, 7 = Poseidonamicus, 8 = Bosquetina, 9 = Dutoitella. (from Dingle and Lord, 1990) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14 A 30 ATLANTIC OCEAN CAST WEST LSW 2 0 40 30 SO r SW Africa i ? . i 4 s « r a i AAIW NAOW AABW Angola NW Africa MW MW Leg 94 i i ] 4 ) ir if SE USA l ? 3 4 » « r « 9 Newfind 1 234541 1 9 . LSW Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 5 Figure 1.6. Summary of modern bathymetric distribution of seven “key” ostracode taxa with respect to water masses in the Southwest Pacific and Southern Ocean. A. Study locations. B. Summary of major water masses and their relation to study locations. Water masses: WSPCW = West South Pacific Central Water, APS = Australasian Subantarctic Water, AAIW = Antarctic Intermediate Water, NADW = North Atlantic Deep Water, CPDW = Central Pacific Deep Water, DW = Deep Water, AABW = Antarctic Bottom Water. Arrows indicate main flow directions: N = northwards, W = westwards, E = eastwards. C. Bathymetric relations of common ostracode taxa to water masses. Horizontal scale in meters, sample depths marked by vertical marks at bottom margin of each plot, finely dashed vertical lines indicate oxygen minimum zone (OMZ), heavily dashed vertical lines indicate major water mass boundaries, (from Ayress et al., 1997) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 6 Kttf ««NOH P » 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 7 productivity and water-mass age (i.e., carbon:carbonate flux, dissolved 0 2, C 0 2, Si, P 0 4, N 0 3, and alkalinity). Studies of foraminifera have better incorporated such non conservative properties into their more comprehensive faunal-environment databases, and these appear to be as important or even more so than conservative water properties in controlling benthic foram biogeography (e.g., Schnitker, 1980; Douglas and Woodruff, 1981; Schnitker, 1994). In particular, the supply of organic carbon may strongly control benthic foram abundance (Herguera and Berger, 1991), but no comparable modern or ancient data exists for ostracode abundance. In addition to these modern ostracode-environment studies, recent high- resolution late Neogene studies provide insight upon environmental preferences of taxa via their respective “ waxing and waning" through glacial-interglacial cycles. These studies provide a certain advantage, in a sense, over modern “snapshot” studies given their intrasample time-averaging on the order of thousands of years. Cronin and Raymo (1997) documented pronounced 41-kyr cycles in Pliocene ostracode faunas in the North Atlantic, with faunal changes more likely related to surface productivity than water- mass shifts. Cronin et al. (1999) found a similar faunal correspondence to Late Quaternary 100-kyr cycles in the North Atlantic, where Krithe dominated glacials, Propontocypris and Cytheropteron characterized deglacials, and numerous trachyleberid genera (Poseidonamicus, Echinocythereis, Henryhowella, Oxycythereis) were most abundant during interglacials. These 41-kyr and 100-kyr periodicities had a similar impact on the ostracode fauna, although the temperature amplitude of the Pliocene obliquity cycles was one-half to one-third that of the Late Quaternary precessional cycles. Even given a fairly robust and global taxon-environment correlation, application of these relationships through taxonomic uniformitarianism may be a strong assumption for the commonly long-lived ostracode taxa. This is particularly true for the earlier Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 8 Paleogene, before the deep-ocean fauna experienced pronounced ecological and evolutionary changes in response to progressive thermal isolation from shallow-water environments (Benson, 1975a, 1984, 1990; Benson et al., 1984, 1985). Thus, through consideration of modern ecological and paleoenvironmental data, the most parsimonious interpretations of the ostracode distributions, and their implications for paleoceanography, are presented. These interpretations are considered hypotheses requiring further testing in other regions, particularly lower latitudes. 1.3. Modern ocean-atmosphere circulation in the Southern Ocean The Southern Ocean plays a central role in modern ocean-atmosphere circulation by transferring heat from deep- and surface-waters to the atmosphere, where the extremely low water-vapor concentrations allows most of this long-wave radiation to escape to space (Figure 1.7). An estimated -3.0 x 102 2 cal/yr flux into the region from sub-surface currents, largely North Atlantic Deep Water (NADW), whereas only -0 .5 x 102 2 cal/yr flux out of the region as Antarctic Bottom Water (AABW) and Antarctic Intermediate Water (AIW) (Kort, 1962). This order-of-magnitude difference relates to heat exchange at the ocean surface, which receives roughly - 1.0 x 102 2 cal/yr from solar irradiance, but transfers over -3.3 x 102 2 cal/yr into the atmosphere as long wave radiation, much of which escapes into space (Kort, 1962). The majority of sensible and latent heat transferred to the high-southern latitudes is thereby exported from the planet, thus maintaining an approximate radiative steady-state through time. Although separated by -90° of longitude, the Kerguelen Plateau and Maud Rise study sites share a similar modern oceanographic setting within the Southern Ocean (Figure 1.8). Both are located south of the Antarctic Polar Front Zone within the vigorous, eastward-flowing Antarctic Circumpolar Current (ACC) that is driven by the southern hemisphere westerlies. Regionally, the upper water column is strongly Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 9 Figure 1.7. Simplified meridional cross-section of the Atlantic Ocean. Arrows indicate direction of major water masses of North Atlantic Deep Water (NADW), Antarctic Intermediate Water (AAIW), and Antarctic Bottom Water (AABW). Waters to the left of the 34.8 % o isohaline are more saline. Inflow of warm, saline Mediterranean Water is indicated by “M” at -35° N. 0 2 max and min are controlled by the relative age of deep waters and carbon export from the surface leading to respiration at depth. Heat is transported by both surface and deep waters to the Southern Ocean, where a net heat loss occurs through the atmosphere (modified from Brown et al., 1989) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 P > 00 o a. g * 1 - o N o C O o eg O n GO C M e j \ o I CM O ? i = 9 © CO LO C M 00 U J > | Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.8. Simplified block diagram of horizontal and vertical water mass movement in the Atlantic Ocean from -40° to 70° S. NADW becomes entrained into the ACW and rises to the surface at the Antarctic Divergence, which is driven by latitudinal differences in westerlies intensity. From the divergence, northward flowing surface waters cool sufficiently to sink as AAIW at the Antarctic Polar Front Zone and southward flowing surface waters contribute to AABW formation. Isotherms from 0 to 10 are in °C. (modified from Brown et al., 1989; flux values from Kort, 1962) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 k . C % £ ° - 3 O 05 co Q cm co in co o Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 affected by seasonal upwelling and primary productivity pulses of the Antarctic Divergence immediately to the south, driven by latitudinal differences in westerlies intensity and the interaction of the eastward-flowing ACW and westward-flowing coastal Polar Current. Both study sites are bathed in upper bathyal Antarctic Circumpolar Water (ACW), which accounts for an estimated two-thirds of the total ACC flux. Broecker et al. (1985), using potential temperature, salinity, and initial phosphate concentrations, estimated ACW in the Indian Ocean sector (i.e., Kerguelen Plateau region) to consist of 45% Weddell Sea water, 30% Pacific/Indian Ocean intermediate water, and 25% North Atlantic Deep Water (NADW), while ACW in the Atlantic Ocean sector (i.e., Maud Rise region) consists predominately of NADW. Underlying these intermediate waters and the study sites at lower bathyal to abyssal depths is the deepest water mass affected by the ACC, Antarctic Bottom Water (AABW), which forms predominately in the Weddell and Ross Seas through supercooling and the development of ice-shelves and polynyas (Foldvik and Gammelsrfd, 1988). Thus, the modem Southern Ocean plays a central role in mixing waters from major ocean basins, maintaining the Antarctic ice-sheet, and modulating the global carbon cycle through seasonal upwelling and primary productivity. For the Late Paleocene study interval, ocean-atmosphere circulation was much different that that of today, with perhaps unique paleoceanographic conditions present during the LPTM. The later Eocene-Oligocene study interval likely coincides with a major developmental step towards the modern ocean-atmosphere circulation regime. These critical intervals of Paleogene climate are examined from the perspective of deep-ocean ostracodes from ODP sites on the Kerguelen Plateau and Maud Rise, sites selected for their prime Southern Ocean locations, high sedimentation rates, stratigraphic completeness, and independently-derived, high-resolution paleoenvironmental frameworks (e.g., stable isotope and sedimentological data). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24 1.4. Transient climate extrema Climate change is nothing new. Evidence abounds for local to global climate changes over various time-scales in the familiar parameters of temperature, precipitation, evaporation, and salinity. Such changes reflect the net response of the climate system to changes in insolation distribution (e.g., Milankovitch forcing), atmospheric composition (e.g., cloudiness, COz, H2 0, and other greenhouse gas concentrations), continental configuration (e.g., albedo, oceanic gateways, mountain ranges, etc.), and related factors. Exactly how these myriad factors control climate through interactions and feedbacks within the lithosphere, hydrosphere, atmosphere, cryosphere, and biosphere remains the major goal of paleoclimate research. For most of Earth's history, climate change appears relatively gradual, with varying lag-times in response to forcing factors producing a quasi-steady-state climate regime. In some cases, however, such gradually changing forcing factors produced abrupt transient climate extremes attributed to the crossing of some critical threshold that separated distinct climate states of relative equilibrium (Figure 1.9; Manabe and Stouffer, 1988; Crowley and North, 1988). As a given climate parameter approached such a threshold, only a slight further gradual forcing produced a transient non equilibrium climate extrema driven by physical and chemical feed-backs. Positive feed-backs within the climate system drive this initial “run-away” climate trajectory, which is subsequently ameliorated and reversed by negative feed-backs (Broecker et al., 1985; Zachos et al., 1993). The impact of two such climate extrema, the Late Paleocene Thermal Maximum and the Eocene-Oligocene transition, on the ecology, evolution, and geochemistry of deep-ocean ostracodes is the focus of this dissertation. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 Figure 1.9. Simplified model of a climate extreme produced by the crossing of some threshold limit of stability. Within climate state one, progressive change in the forcing factor(s) approaches an equilibrium threshold whose crossing produces an extreme climate through “run-away" positive feedbacks. This resultant climate extreme attenuates over time through negative feed-backs within he system and subsequently returns to tracking of the forcing factor(s) within the “new” climate state. The “ tracking” of the forcing factor(s) by climate may be exaggerated in this simplified model; climate may remain relatively stable between shifts from one climate equilibrium state to another. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. < D E r/1 Climate State 2 y m - Climate Extreme \ Critical Threshold v .* ’ 'v 3 i) Climate State 1 Forcing Factor "Climate" Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 1.5. Geologic, stratigraphic, and geochronologic setting 1.5.1. Eocene-Oligocene Transition ODP Hole 744A was drilled on the southern slope of Kerguelen Plateau (61°34.66'S, 82°47.25'E) at a water depth of 2,307 m (Figure 3.1). Advanced piston coring produced effectively 100% recovery of a Late Eocene-Early Oligocene sequence of predominately coccolith carbonate ooze, with a significant increase in biogenic opal from <2 to as much as 18 weight % in the earliest Oligocene (Barron et al., 1989a; Salamy and Zachos, 1999). Due to the plateau’s distance from Antarctica and elevated topography since the Late Cretaceous (>80 Ma), terrigenous sediments are rare except for intervals of ice-rafted debris (IRD; e.g., quartz, feldspars, and lithic fragments), particularly at the isotope excursion Oi-1 (Barron et al., 1989a; Lawver et al., 1992; Zachos et al., 1996). Estimated Late Eocene-Early Oligocene paleodepths are estimated at -1,800 meters, shallower than the estimated paleosurface of the carbonate compensation depth (CCD) (Barrera and Huber, 1993). ODP Hole 689B was drilled on the northern crest of Maud Rise (64°31.01'S, 03°06.00'E) at a water depth of 2,080 meters (Figure 3.1). Advanced piston coring produced effectively 100% recovery of a Late Eocene-Early Oligocene sequence of coccolithophore carbonate ooze, where % opal varies from -10-90 % (Barker et al., 1988a; Diester-Haass, 1993). Maud Rise had also developed significant topographic relief by the Late Cretaceous and was paleogeographically isolated from Antarctica, restricting most terrigenous sediment to that delivered as IRD (Barker et al., 1988a; Lawver et al., 1992). Late Eocene-Early Oligocene paleodepths are estimated at 1,400- 1,650 meters, also shallower than the estimated paleosurface of the CCD (Kennett and Stott, 1990). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 Age models for Site 744A and 689B are based on calibrating their magnetostratigraphy (Spiess, 1990; Keating and Sakai, 1991) to the Geomagnetic Polarity Time Scale (GPTS) of Cande and Kent (1995), and assume constant sedimentation between chron boundaries. Paleomagnetic reversal depths for each site and GPTS chron ages are provided in Appendix 1. All published data of previous studies were re-calibrated to this uniform time-scale using MBSF data. 1.5.2. Late Paleocene Thermal Maximum ODP Hole 690B was drilled on the southwestern slope of Maud Rise (65°9.63'S, 1°12.30'E) at a water depth of 2,914 m. Advanced piston coring (APC) produced effectively 100% recovery of a Late Paleocene sequence of almost exclusively calcareous biogenic sediment (Barker et al., 1988b). Late Paleocene paleodepths are estimated at - 2,000 meters, shallower than the estimated average paleosurface of the carbonate compensation depth (CCD) (Thomas et al., 1990; Kennett and Stott, 1990; Van Andel, 1975). ODP Hole 738C was drilled on the southern slope of the Kerguelen Plateau (64°42.54'S, 82.° 47.25'E) at a water depth of 2,252 meters. A rotary core barrel (RBC) produced an 11R section with less than 50% recovery of the Late Paleocene sequence of calcareous chalk with chert nodules and fragments (Barron et al., 1989b). Late Paleocene paleodepths are estimated at -1,300 meters, also shallower than the estimated paleosurface of the CCD (Barrera and Huber, 1991) and roughly equivalent in depth to Steineck and Thomas’s (1996) study of Site 689B. The age model for Site 690B is based on three datums and assumes constant inter-datum sedimentation rates. The oldest datum is the C24r/C25n boundary, located at a mean depth of 185.47 mbsf (bounding paleomagnetic sample intervals of 185.25 and 185.70 mbsf; Spiess, 1990), with a Geomagnetic Polarity Time Scale age of 55.904 Ma (Berggren et al., 1995; Cande and Kent, 1995) The middle age-model datum is the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29 abrupt benthic foram extinction horizon, taken at 170.48 mbsf, to which Aubry et al. (1996) assign an age of 55.5 Ma. The youngest datum is the NP10/NP9 boundary at a mean depth of 148.90 mbsf (bounding nannoplankton sample intervals of 148.40 and 149.40 mbsf) and assigned an age of 55.0 Ma (Aubry et al., 1996). However, disagreement exists regarding the magnetobiochronology of the Late Paleocene-Early Eocene interval at Site 690B; the most recent reviews and arguments by selected workers are provided in Aubry et al. (1996) and Berggren and Aubry (1999). Age datums for the Site 738C study interval are rare given its position within a long coring interval of extremely poor recovery. In addition, ship-board paleomagnetic determinations for this and many other Site 738C intervals were subsequently deemed unreliable due to magnetometer malfunctions. Lu and Keller (1993) provide ages for samples examined in this study, however, their biochronological basis is not explicitly outlined. Therefore, Site 738 samples are examined by depth interval (mbsf) alone, with no attempt to assign absolute ages or to correlate temporally to Site 690B. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 2. Krithe Mg/Ca paleothermometry during the Eocene-Oligocene transition: An independent assessment of 81 0 O -based paleoceanographic reconstructions 2.1. Introduction 2.1.1. Scope During the late Eocene to early Oligocene, the relatively ice-free “greenhouse” of the earlier Paleogene shifted towards the present “icehouse” climate (Fischer, 1981). Proposed forcing mechanisms for this global change include atmospheric C 0 2 draw-down through carbon cycle shifts (Freeman and Hayes, 1992) and progressive Antarctic “refrigeration” as the circum-polar oceanway widened (Kennett and Shackelton, 1975; 1 8 Bartek et al., 1992). These events led to progressive benthic foraminiferal 8 O 1 8 (hereafter 8 O b l) punctuated by several positive relative extrema, representing rapid bottom water temperature (BWT) cooling and/or glacial expansion (Figure 2.1). The most pronounced and complex of these, termed Oi-1 by Miller et al. (1981), occurs just above the epoch boundary with a benthic foraminiferal S1 8 0 increase of ~1.5%o. Two -100 kyr sub-cycles occur within the -350 kyr event, probably representing major ice-volume fluctuations (Zachos et al., 1996). Oi-1 is interpreted as a non-linear climate “overshoot” from crossing the critical threshold dividing “ greenhouse” and “icehouse” states (Berger, 1982). This overshoot likely lowered atmospheric C 0 2 through positive feedbacks (i.e., increased C 0 2 solubility, Co r g production/burial, shelf carbonate weathering) and subsequently attenuated to a new equilibrium through negative feedbacks (i.e., global nutrient sequestering, ocean alkalinity shifts) (Zachos et al., 1996). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 Figure 2.1. Global early Cenozoic composite record of benthic foraminiferal stable isotope values and paleogeographic location of Site 744A during the Latest Eocene. Horizontal line indicates the major shift in 5,aO and 5’3 C associated with the Eocene- Oligocene transition (“Oi-1”) (after Zachos, pers. comm., 1999). Late Eocene-Early Oligocene paleodepths at ODP Site 744A are estimated at -1,800 m (after Lawver et at., 1 9 9 2 ). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T - 1 a ip p i^ j A | J B 3 aieilApea -I a | P P ! i A l A | J B 3 3 ) B ~ | A p B 3 8* B T e u e o o jy v ■ o 6||0 0U 0O O 3 O 0|B d >1 O C O o o LO o C O C O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33 18 Reconstructions of Oi-1 are based largely on the 5 Ob l records, but resolving the relative contributions of BWT and ice-volume to this signal remains problematic (e.g., Kennett and Stott, 1990; Zachos et al., 1996). To provide a new geochemical perspective on the thermal history of Oi-1, the recently developed Mg/Ca paleothermometer for Krithe ostracodes was used to examine thermal, and thereby 1 8 cryospheric, history during this rapid -1 .5 % o shift in 8 Ob ( at the Eocene-Oligocene boundary. (Figure 2.1). Elemental results provide scant evidence for net BWT change through the Oi-1 event, but must be tempered due to currently poorly understood aspects of ostracode physiology, ecology, and preservation. If observed Krithe Mg/Ca is a true recorder of BWT, then current 5 8 Ob ( -based estimates of -3-4 ° C BWT cooling and cryosphere growth to at least 40 % modern volume during Oi-1 are greatly over- and under-estimated, respectively, based on local conditions on the Kerguelen Plateau. 2.1.2. The importance and difficulty of reconstructing paleotemperature Paleotemperature data represent the spatial distribution of planetary heat through time and are therefore fundamental in reconstructing global climate history and dynamics. The global ocean is a major medium for redistributing this heat through linked mixed layer and thermohaline circulation, with a net poleward transport of low- latitude “surplus” heat. This ocean circulation, coupled with atmospheric circulation, produces thermal quasi-steady-states within a global system defined by primary solar heat influx (e.g., Milankovitch forcing), atmospheric composition (e.g., cloudiness, C 0 2, H2 0, and other greenhouse gas concentrations), continental configuration (e.g., albedo, oceanic gateways), and related factors. The primary seawater flux from surface to thermohaline circulation occurs where surface water heat content and salinity produce densities equal to or exceeding that of an underlying water-mass (c.f., other minor fluxes, such as pressure-driven downwelling Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34 in gyre centers). Under these conditions, surface waters will spontaneously sink (conserving their temperature and salinity parameters and precluding further atmospheric exchange) and enter the thermohaline system, which is driven by such buoyancy fluxes, the coriolis effect, and upwelling processes within the context of density-stratified water masses and bathymetric barriers {e.g., ridges and sills). This geostrophic system affects global climate by further re-distributing heat and dissolved solids as well as modulating atmospheric C 0 2 (Schmitz, 1995). In addition, thermohaline production and circulation strongly influence the vast oceanic benthic ecosystem: production sets ambient temperature and salinity, while changes in the tempo and mode of circulation affect initial dissolved oxygen, carbonate equilibria, etc. on local to global scales. Thus, the history of deep-ocean circulation is a critical component for understanding global climate dynamics and their forcing of biosphere evolution and extinction. The primary means for reconstructing the thermal history of water masses 1 8 bathing the ancient deep-ocean floor is the benthic foraminifera, whose 5 O values archive past BWT, salinity, and ice-volume (Epstein et at., 1953; Emiliani, 1955; Craig and Gordon, 1965; Shackleton, 1967). However, partitioning 6,8 Ob t among these three contributing paleoclimate parameters is difficult (Equation 1). 5,8 O b ( [PDB] = S'8 0 s e a w a te r [SMOW] + (2.78 x 106 x BWT 2 [K] - 33.3557 (McCorkle et al., 1990) [Equation 1] 18 S Os e a w a le r values are controlled locally by phase-change fractionations of evaporation- precipitation at the surface (highly correlated with salinity; Craig and Gordon, 1965) and globally by cryospheric sequestering of 1 6 0-rich precipitation. This primary Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35 18 8 O s e a w a ie r value is further altered by a temperature-dependent fractionation of - 0.22 % o per C ° during biomineralization. Biological “ vital effects" within and among species may introduce additional variance and offsets. Thus, non-unique solutions exist for 1 6 ancient 5 Ob t values unless two contributing parameters are independently known or constrained. Different assumptions and approaches may improve paleoclimatic 1 8 interpretations of 5 O b t records, but unique solutions remain difficult. Attempts to solve this quandary include assumption of constant sea surface temperatures in the open-ocean equatorial Pacific, thereby providing a planktic foram 18 8 O-based proxy of global ice-volume to calculate local SST and BWT in other regions (e.g., Matthews and Poore, 1981; Prentice and Matthews, 1988). However, the assumption of relative stability in equatorial SST remains controversial (CLIMAP, 1981; Beck eta/., 1992; Guilderson et a/., 1994; Billups and Spero, 1996) and is highly susceptible to selective planktic foram dissolution, which would overestimate calculated ice-volume. A similar approach employs an informed assumption of limited 18 to constant BWT and interpretation of 8 O values exceeding these values as lower (but not upper)limits on ice-volume (Miller et at., 1987; Zachos et a/., 1993, 1996). Across the Oi-1 event at Site 744A, Zachos et at. (1996) assumed minimum BWT to be equal to the modern BWT of 1° C. Based on this minimum BWT, the observed -1.3%o shift at Oi-1 represents rapid cryospheric development to at least 40% of modern ice-volume concomitant with -3-4 C° BWT decline (Figure 2.2). If BWT did not decline by this amount or minimum BWT did not reach 1° C, then ice-volume would be commensurably greater (up to a maximum of -125% assuming no BWT change). This rapid cryosphere growth is consistent with coeval ice-rafted debris and clay mineralogy data from numerous circumpolar locations (Figure 2.2; Wise, 1991; Ehrmann and Mackensen, 1992; Diester- Haass eta/., 1993; 1996). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36 Figure 2.2. Timing of Paleogene glaciomarine sedimentation in the Southern Ocean and 18 18 5 O-based estimates of ice-volume and BWT based on the global 5 Ob t composite. Glaciomarine data from Wise et al. (1991). Benthic foram values adjusted for vital 1 8 effect disequilibrium (+0.6%o) and mean cryosphere 5 O assumed constant (-45%o SMOW). Black fill represents minimum ice-volume assuming a minimum BWT of 1° C and stippled region indicates possible range of ice-volume assuming BWT between 1 and 4 0 C. Numerical ages based on timescale of Berggren et al. (1985). (from Zachos et al., 1 993) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. llogion Rost Soa Marie Oytdland Wedded See Prydi Bay Kemuelen Plateau Sle DSO P 270 USSTS 1cnos 1 Ml PaliM OOP 893 OOP 739 O PO 742 OOP 738 OOP 744 O OP 748 LaMuda 77-S 7S-S 77*S 76-S 71*3 67*S 67“S 63 "S 6TS sa-s PERCENT PRESENT ICE VOLUME 0 S O 100 ISO 200 Ma Age Calculated Ice Volume Glacial/Glaclomaflne Wed Documented D lsconlorm ity I Gladai/Glack)marine Less wed Documented Hyatodasitite Calculated Temperature (Ice-Free) o 5 to 15 BOTTOM WATER TEMPERATURE C M 38 The consistency of these isotopic and glaciomarine records provides a well-grounded hypothesis of ice-volume and BWT dynamics that may be tested through Mg/Ca paleothermometry. 2.1.3. Mg/Ca paleothermometry: Physiology and thermodynamics While benthic foraminifera and stable isotopes are of clear utility in reconstructing deep-ocean history, they are not the only potential means for reconstructing paleotemperatures. The Ostracoda, a class of minute bivalved crustaceans, are common deep-ocean metazoans that provide an abundant fossil record of low-Mg calcite valves. Modern studies demonstrate that BWT strongly controls Mg2 * incorporation into the low-Mg calcite valve of Krithe, a cosmopolitan Cenozoic ostracode genus (Correge, 1993; Dwyer et al., 1995; Cronin et al., 1996). Equation 2 illustrates the modern relationship between Mg/Ca ratios in Krithe (hereafter M g /C a ^ J and ambient BWT in the North Atlantic. Using this Mg/CaK n W ie paleothermometer, fossil ostracodes may provide a 8,aO-independent estimate of ancient BWT. BWT [K] = (584 x M g /C a ^ , [mols]) + 276.93 (Dwyer et al., 1995) [Equation 2] The above positive relationship between temperature and Mg/Ca is common in many biogenic carbonates (e.g., foraminifera, mollusks, corals, brachiopods, etc.) and is controlled by both thermodynamics and physiology. For the chemical system C aC 03 + M g 2+ <-» MgCO + C a 2+, the thermodynamic coefficient for Mg in calcite ( kc ) is determined by: Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. where k c and k m are the solubility products of pure calcite and magnesite, kC aC 03 and ^Mgco 3 are the activity coefficients of CaC03 and MgCOa in the solid precipitate, and yca 2+ and y^ 2+ are the activity coefficients of aqueous Ca2 * and Mg2 * in seawater. K Temperature-induced changes in are controlled by ratio changes in ------ (assuming K negligible temperature effects on solid and aqueous activity coefficients) by the relationship / \ * t 2 A H 0 ' 1 1 ' Kr., V 1 1 J 2 3R J l T 2 , where a h 0 is the reaction enthalpy for the initial temperature, R is the gas constant, and T is temperature in Kelvin. These thermodynamic aspects are reviewed in detail by Morse and Bender (1990) and Mucci and Morse (1990). Using these relations, Rosenthal et al. (1997) calculated that a 25° to 5° C temperature change would produce a relative distribution coefficient decrease (K5 »c/K2 5 c) of 0.59, which compares favorably to experimental calcite precipitation values of 0.69, 0.70, and 0.53 (Katz, 1973; Mucci, 1987; Oomori et al., 1987, respectively). The calcitic valves of ostracodes, however, are not produced in a thermodynamic system — significant metabolic energy is expended during physiological precipitation of a highly calcified valve pair over the course of hours. Therefore, thermodynamic principles are difficult to apply directly with any faith. Instead, “real-time” studies of modern ambient environmental conditions and resultant elemental ratios in biogenic carbonates generally provide empirical relationships that are then applied to fossil materials. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40 Given this predominance of physiology over thermodynamics, a brief overview of salient ostracode biology is helpful. As do all members of the Arthropoda, the class periodically molts their exoskeleton, including their highly calcified valves, eight times until reaching sexual maturity. Turpen and Angell (1971) determined new valve calcite to be derived directly from ambient seawater with no resorbtion of the preceding valve calcite. During valve biomineralization by a constellation of epidermal cells, some amount of Mg2 * substitutes for Ca2 * in the C aC 03 lattice. The degree of substitution is strongly correlated with temperature, which appears to control major metabolic and physiologic aspects of biomineralization through kinetic and microenvironmental effects. Under uniform conditions, ostracode Mg/Ca ratios differ between sub-families, decrease through successive molts, and vary across individual valves (Cadot et al., 1972; Cadot and Kaesler, 1977; Chivas et al., 1983). In the core-top-based paleothermometry equation below and in the present study, Mg/Ca variability due to such factors is minimized through exclusive analysis of whole, adult valves of the deep- ocean ostracode genus Krithe. Three recent core-top studies document a strong positive relationship between M g /C a^ g and ambient BWT in the modern deep-ocean (Figure 2.3; Correge, 1993; Dwyer et al., 1995; Cronin et al., 1996). Imperfect correlation within and between the three linear regression is attributed to measured BWT uncertainties, analyses of potentially sub-modern specimens, and undocumented non-BWT-related effects. Improved calibration through culturing experiments and more precise core-top data are currently underway or planned (Cronin et al., 1996; Dwyer, pers. comm., 1998; Schellenberg, in prep.). For single Mg/CaK n (/)e values, the North Atlantic calibration (Equation 2) produced a BWT estimate with an error of ±1.3 C° at a 95% confidence interval. Application of this technique to Pliocene and Quaternary intervals produced eustatic sea-level estimates consistent with stratigraphic evidence. These findings Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41 Figure 2.3. Core-top variations in Krithe spp. ostracode Mg/Ca versus observed bottom water temperatures from the North Atlantic and Arctic. Atlantic equation is BWT [°C] = (584 x M g/Ca^.^ (molsj) - 3.78 (r = 0.87) and Arctic equation is BWT [°C] = (375 x Mg/CaJ fn fh e [mols]) - 3.53 (r = 0.77). (from Cronin et al., 1996) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42 0.035 0.025 IS o 0.020: o 0.015: 0.0101 . . 0.005 Water Temperature (:C) i_:ne ' Atlantic t = 3 5£= _.n e 2. Atlantic a r c A :::-: * = C-.3"~C. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43 indicate Mg: C a^.^ is sensitive to orbitally-forced shifts in dominance of different bottom-water masses in the North Atlantic (Dwyer et al., 1995; Cronin et al., 1999). In this study, the technique is applied to significantly older Paleogene materials to attempt to place constraints on BWT and ice-volume changes associated with the Eocene- Oligocene Oi-1 event. 2.2. Materials and Methods 2.2.1. Stratigraphic and isotopic context of sample population ODP Hole 744A was drilled on the southern slope of Kerguelen Plateau (61°34.66'S, 82°47.25'E) at a water depth of 2,307 m (Figure 2.1). Advanced piston coring produced exceptional recovery (>98%) of a Late Eocene-Early Oligocene sequence (130-147 mbsf; -31.7-33.6 Ma) of predominately coccolith carbonate ooze (Barron et al., 1989a). Estimated Late Eocene-Early Oligocene paleodepths are estimated at -1,800 meters, shallower than the estimated paleosurface of the carbonate compensation depth (CCD) (Barrera and Huber, 1993). The age model for Site 744A is based on calibrating the core magnetostratigraphy of Keating and Sakai (1991) to the Geomagnetic Polarity Time Scale of Berggren et al. (1995) and assumes constant sedimentation between chron boundaries. From 137.47- 157.01 mbsf, 10 cc volumes were sampled at -1 0 cm intervals, yielding a 10-12 kyr/sample resolution. Zachos et al. (1996) determined for each sample the average 81 8 0 and 81 3 C of 1-15 individuals of Cibicidoides spp. from the >150 pm size-fraction. Salamy and Zachos (1999) determined for each sample the weight percent of the >63 pm (sand) size-fraction and bulk weight percent composition of C aC03, opal, and terrigenous sediments. Relative dissolution intensity was assessed through percent planktic foram fragmentation (fragments/whole tests) and percent benthic forams (benthic/planktic forams) (Diester-Haass, 1996; Salamy and Zachos, 1999). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44 2.2.2. Krithe sampling, cleaning, and image archiving Whole adult female Krithe valves were picked from the >150 pm size-fraction of approximately every other 744A sample (n = 101, mean sampling interval = 20±12 cm). Adult Krithe differ from juvenile instars in their pronounced vestibule and radial pore canal system. Adult male Krithe valves, distinguished by their more elongate form, were quite rare and not analyzed. Significant morphological variation exists within the genus, with moderate success in construction of a species-level taxonomy (see Coles et al., 1994). Variation in radial pore canal and vestibular structure was pronounced, but largely continuous, within the Site 744A study population. Based on quantitative comparison, these structures do not correlate with any of the preservational or geochemical parameters below. Therefore, the sample population was treated as Krithe spp., but additional systematic work on this material is merited. Selected specimens were cleaned by brush and multiple short (1-2 second) sonications in ethanol and deionized water, with a final deionized water rinse and air drying on filter paper. Under transmitted light of constant intensity, each specimen was recorded photographically (Leica 35 mm w/ valve-centered autoexposure metering) and digitally (CCD camera and PC framegraber board). Valve area and mean luminance gray value (as an inverse measure of valve opacity) was measured using the Bioscan Optimas 5.5 image analysis program. Individual valve masses were calculated as the average of three to seven separate pg-scale measurements. 2.2.3. Diagenetic screening through cathodoluminescence and SEM Cathodoluminescence (CL) is the emission of visible light from a material in response to bombardment by cathode rays. The principal CL excitor in carbonates is manganese (Mn), which may substitute for calcium within the crystal lattice (Dromgoole and Walter, 1990). The net intensity of CL emission depends upon the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45 elemental concentrations of “ activators”, such as Mn, and “quenchers”, principally iron (Fe) (Orgel, 1955; Medlin, 1968; Hemming et al., 1989). Unaltered marine carbonates, both abiogenic and biogenic, do not generally exhibit CL because of Mn’s relatively low concentration in seawater (5 x 10 9 moles/kg; Bruland, 1983) and its low partition coefficient into calcite (3.8-16; Dromgoole and Walter, 1990). This observation has been parlayed into a general proxy for retention of primary carbonate chemistry in ancient carbonates (e.g., Popp et al., 1986). In contrast, diagenetic carbonate often exhibits CL because of increased Mn concentrations in porewaters (in turn related to reduced Eh) and relatively slow precipitation rates, both of which increase Mn’s partition coefficient into calcite (Lorens, 1981; Mucci, 1988; Pingitore et al., 1988; Dromgoole and Walter, 1990; Major and Wilber, 1991). The relatively rapid rates of ostracode biomineralization and limited CL examination of modern ostracode valve calcite (Schellenberg, personal observation) are consistent with a presumption of no CL in pristine, unaltered low-Mg valve calcite. The selected Krithe valve population, as well as a small population of poorly preserved valves (based on visual characteristics), were individually screened for CL, and thereby presumed presence of diagenetic calcite, using a Technosyn cold cathode luminescence scope at the common settings of 20 kV and 0.5 Ma. Examined in complete darkness, the selected Krithe valves generally showed no visually discernible CL to a rare, extremely weak homogenous CL. In contrast, most of the poorly preserved valves displayed intense CL ranging from homogenous to heterogeneous across the valve surface, commonly with concentrations along edges and fractures. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46 2.2.4. Elemental determination via ICP-MS From the total Krithe sample population, a sub-population was selected for elemental analysis to maximize stratigraphic coverage. Individual valves were dissolved in 1 ml 1% H N 03 within trace-clean microcentrifuge tubes with ~5 minutes of sonication bath to ensure complete reaction. This solution was then micro-pipetted to a trace-clean scintillation vial, into which was micro-pipetted 60 pi internal standard solution (Figure 2.4) followed by an additional 5 ml 1% H N 03. Isotopic concentrations of the nuclides 4 3 Ca (0.135%), 4 8 Ca (0.187%), 2 4 Mg (78.99), and 8 6 Sr (9.86%) were determined using a Finnegan Element magnetic-sector ICP-MS equipped with a Meinhard nebulizer. These nuclides were selected to minimize required dynamic range and interference effects (e.g., use of rarer Ca nuclides in relatively Ca-rich solution). Examination of repeated intersample determinations of standard solution concentrations (Figure 2.4) and internal Sc standards produced Mg/Ca and Sr/Ca uncertainties of always less than 5% and generally less than 2%. 2.3. Results 2.3.1. Phenotypic, environmental, and elemental relationships Resultant data on Krithe valve area, mass, opacity, and determined elemental ratios are presented in Figure 2.5. Valve mass and area are highly significantly correlated with one another, but neither show a strong correlation to valve luminance, a relative measure of valve clarity (Figure 2.6). In pristine modern Krithe, valves are highly translucent and tend to become progressively more opaque with post-mortem dissolution of the valve surface. Luminance is weakly positively correlated with valve area (Figure 2.6b), likely due to smaller valves having relatively greater proportion of highly curved margins of greater apparent thickness. Conversely, luminance is weakly negatively correlated with mass (Figure 2.6c), being particularly affected by a small Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 7 Figure 2.4. Elemental concentrations of internal reference standards for determination of Mg/Ca and Sr/Ca ratios via ICP-MS. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 8 Internal SI andard Concentration Element Cone. Units Sc Yr Bi Ce 102.3 107.1 0.5 0 54 0.3 5 88 PPb PPb PPb PPb Reference Standard Solution Concentrations Element S-1 S-2 S-3 S-4 Units Mg 8 .627 1.718 0.5914 0 .1 7 27 PPb Ca 3028.1 599.91 206.496 60.286 PPb Sr 5 5.85 11.065 3.809 1.112 PPb Ur 15.876 3 .1 4 5 1.083 0.316 ppt Ba 23.1 4 .5 7 7 1.576 0.46 ppt Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49 Figure 2.5. Krithe valve measurement and elemental ratios for Site 744A. Ages based on age model described in methods section. Stable isotope values from Zachos et al. (1996). Elemental ratios in mmol/mol. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50 as O ® • r v 05 in ^p in TP C M pv C O in 05 in C O in o in C O C O C M in C O in C O in in C O o C O C M co Pv C O C M © © © © © © o © © © © © © Pv © o © Pv © w 0 ) $ C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M © © © © © © © © © C O O n V CO C O 05 p*. C O in 05 05 pv in K rv C O ^p C O C O tv in in C O O ^p C M 05 in C O C O in C O C O 05 C M r v |V O o o O O © © O © © © © © © © © © © © © o o © © s C O C O C O in co C O C O rv C O fv C O tv C O C O rv C O C O C O © © © © © © Pv © © rv rv pv rv © Pv co O n V rv ao o 05 C O o ao C M 05 05 C M C O C O C O C O o rv C M C O o 03 o m C O C O C O C O C O in C O 05 © © o C M © © * r Pv © pv * • © © o © o © M’ o © © © 05 2 3 C O rv C O C O C O C O C O rv 05 |v C O rv C O C O pv 05 C O C O © © © Pv © © rv pv © © rv rv pv © Pv CL > © D 5 5 44 o - C M - - o o C M - C M v - o o - - - © C M o © o o - - © o -H 44 44 44 44 44 44 44 44 44 44 44 44 44 44 44 44 44 44 44 1 44 44 44 44 44 44 44 44 44 44 44 44 X C T a. 1 8 2 4 W “ o C M 2 4 4 3 1 2 5 2 0 1 6 1 6 3 3 3 0 00 C M 2 3 ___2 1 C M 2 1 2 2 © 2 0 © 2 5 I 6 4 2 4 I 3 7 © 2 2 © © © © > 0) C D C O 05 C O C O o C O C O O C O C O C O pv C M C O C O o Tp C M © 05 C O © © C M C O C O C M © O © Pv © C M © o rv © o © © © T - © © rv © © © Q 5 3 o C M 05 r v in C O ▼ - 05 C D C M C M o C M pv C M C M rv 05 T - o C M 05 TT C M C O C M © C M pv © C M © © © © © Pv © © © O © X © 44 44 44 44 44 44 +4 44 44 44 44 44 44 44 I 44 44 44 44 44 1 44 44 44 44 44 44 44 44 44 44 44 i o c C O c C O 05 C M O 05 in T — C O O O co O rv O C O O in C O Pv © C O in C O C O 05 C M C M C O O pv C M © C M © © © © rv © © © © r v © © E 3 _J (O O m G O in o in 00 ^P in in rv in rv in o in O C O 05 in 05 in C O in C O in 05 in O C O © © Pv © © r - © © © © © pv © © © © © © © © C O C O o C O T f o in C M C O co C O C O 05 rv 05 C M 05 C M C O 05 in o © C M C M C O C O C M 05 pv © © © © © © rv © rv © © © © © © © © © < C D C O C M C O i n C M in C M rv in rv in C M 05 rv C M 05 C O C M iv tv C M rv C M C O C M C M T in C M in C M C M C O o C O C O 05 C M I in C M C O in C M in I V C M C O © C M © © C M 1 © © C M © © C M © C M ^ r pv © © © © © © © © © © © © © © © © © i Spec. # iv 00 o - C M - C M C O V“ C M C O C M C O * — C O - C M C O C M C M C M © - © - © - © - © © O C D © 05 in 05 in 05 in 05 in 05 C O C M C O C M ao C M C O C O C O fv O rv o rv o o C M o C M in in © V O © © C M © C M © © © o © o © © O C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M © © © © © © © © © o n o o o o o C M C M C M tv C O Pv C O rv C O •M - 05 C O 05 C O C O Tf C O © © © © C M O © © © o © o © pv Tf pv © M’ © tp © © < -D *“ V— * - *“ V — T“ * - *“ r - *“ *“ 5 C O pv C O C O pv © C O pv C O C O pv C O C O pv C O in co 05 in C O 05 in C O 05 05 o 05 o 05 o in in o in in o in in © C O o Tf C O o C O 05 o C O 05 O C O Oi © © © © C M © C M Pv C M pv © © © © © © © © © © © © © pv © Pv © rv © © © © © a < C M C O C M C O C M C O C M C O C M C O C M C O C M C O C M C O C O C O C O C O C O C O C O C O C O C O C O CO C O C O C O C O C O C O C O C O C O C O © © © © © © © © © © © © © © © © © © © © © © © © © © © © In C O o a ) o C O o C O o C O o C O co C D C O C O C O C M C M TP C M C O C O C O C O C O C O C O i r co TT 05 © 05 co 05 © Oi O © © C M Pv C M Pv C M © © © © © pv © rv © T"“ rv © a 2 CO C O C O C O ® CO C O C O 00 C O C O C O 00 C O C O C O 0 5 C O 05 C O 05 C O 0 5 C O 05 C O 05 C O 05 C O 05 C O 05 C O 05 C O 05 C O o o C M © © © © © M- o 0 5 o 05 o 0 5 o 05 o 05 o in o in o in C O in C O in C O in in C O in co in C O ao rv ao rv C M C M C M © © © « © © © © © © © © © © © © © © C O C O © C O ao C O C O C O C O C O © 00 C O in in in C O C O C O C O C O C O C O rv C O rv 05 05 Oi 05 © © C M C M © © © © rv © Pv © © © © pv C O - - -- C M C M C M C M C M C M C M C M C M C M C M C M © © © © © © © Core 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H Hole II7 4 4 A 7 4 4 A 7 4 4 A II7 4 4 A II7 4 4 A || 7 4 4 A || 7 4 4 A II7 4 4 A | | 7 4 4 A II7 4 4 A < N || 7 4 4 A 7 4 4 A II7 4 4 A || 7 4 4 A < M ’ pv || 7 4 4 A 7 4 4 A II7 4 4 A | | 7 4 4 A | | 7 4 4 A | 7 4 4 A II7 4 4 A | | 7 4 4 A < *r p v | 7 4 4 A < Pv | | 7 4 4 A II7 4 4 A || 7 4 4 A | | 7 4 4 A f| 7 4 4 A | | 7 4 4 A Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 (0 O C O ▼ C M o r v C M o C M C O in C O C M in C M in C M C O C M C M C O m in in C O C O C O C O C M in C O C O C M C O fv C M O i C O C M C M f v f v ^T O i CO fv O i fv in o CO C O in CO in CO S C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M <0 O n V 0 0 0 0 CO C M o> C M C O CO 00 C O O i C M CO CO 00 0 0 CO C M f v C O 00 O i C M C O C O C O O O C O o CO O (O C M in 00 f v O i in in C O CO O i C M CM O ^ r O i in a 2 CO r v CO C O in C O o> r v C O C O CO - CO C O C O f v C O f v C O f v f v f v 0 0 (0 f v C O C O C O C O IV f v <5 O w CO o m C O in C M CO |V C M C O | v C O CO o C M O i O i c o C O o co in O i TT C O C O f v O O i r v in C M o C M C M r v C M O in f v CO O i O i 00 f v C O m 00 C O CO f v f v O 2 C M f v C O r v C O in CO O i IV C O C O C O f v C O C O IV fv f v (O f v f v rv C O <o f v CO C O C O f v f v r v £L > 0 ) - - - T“ o O - O - O o C M - - T“ - o o - o o C M o o o o o - CO 41 1 + 1 41 + i 4 1 •H 41 4 1 41 4 1 4 f 4 1 4 1 4 1 4 T 4 T 4 1 4 1 4 1 4 T 4 1 4 1 4 1 41 41 4 1 4 1 4 1 4 1 41 41 4 1 41 X CT a 0 ) C M C M C M o © O i O in C O | v C M f v C O IV O I C M o C M C O C M C M *r C M C M C M C M IV in C M o CO O in O C M O C M 00 C M C O C M C M C M o C M O i in C M > C M O C O C M C O o C O IV C O C O in T " C M C M O CO CO o m f v o C M f v V * in f v O i in IV C O o 00 o C O C M CD O C M C M t f C O O i C M f v CO C M in o <o in o o 5 C M C M C M C O f v o C M O i O i O i C O o> C M C M C O 00 C M C O C O o i C O O i C O CO C O O i CO in 00 O i CO X 0) 1 ■ H 41 4 1 ■ H 41 4 1 4 1 4 1 1 4 1 4 1 41 4 1 4 T 4 1 4 T 4 1 4 1 4 T 41 4 1 41 4 1 4 1 41 4 1 41 4 1 1 41 4 1 41 o c <0 c r v o C O in C O C M C O CO C O C O in O i IV C M C O r v CO 00 in f v C O O C O o C O O i C O in r v C O C M CO CO C O C O C M in r v CO CO C O r v C O in CO o m 0 0 in O C O C M CO E 3 - J | v in o in O i C O f v O i C O in O i C O C M CO r v in in in in t r C M C O rv T f f v o in CD C O C O in C M o in in 00 <0 TT C M G O C O r v O i O i in o o CO C O C M TT O i C M **• 00 f v o C M CO C M f v O i C O O i m f v o C O in O i C O < f co C O f v C O C O m o O i C O in C O C M O i CO < 1 C M C O C M in C M 00 C M O i C O ® C O C M C O C M in C M C M o C O C M 1 00 C O CO m C M o m rv C M C O in C M O fv C M |v C M m r v C M in in C M IV in C M f v in C M C M fv C M C M C M f v C M c o CO C M o in C M O i C M in C M 00 C O r» 1 00 CO C M C M CO C M O i in C M S p e c . # C M - C M C O T“ - C M C M C O - C M - C M C M C O C M C O - C M - C O - C M C O C M C O r f C M CO o C O n o C O C O CO CO C O C O C O C O C O C O C O C O *3“ C O C O o o CO o C O in m rv o IV O f v o C O o C O o O i O i Oi Oi in m in in in in in in ▼- o O **• o ro C M C M C M C M C M C M C M — — * - — - * - o O *- ■ — T “ * “ *“ *“ o n o |v in h * in | v in in in C O C O 00 CO C O C O C O o C O C M ® C M C O 0 01 0 01 in 00 in C O in C O C O f v C O f v 0 01 0 01 O i O i in C O in C O in oo in C O in ao in ao T f o o o * - * - o o o o o o o o o o o O o o o o o o o o fl> 2 C M O i CO C M C O * • C M CO C M C O in C O IV O i |v O i f v | v C O IV |v C O IV N C O 00 f v f v IV f v r v IV C O r v C O rv O i C O f v O ) C O r v O i ao r v o i f v O i o o r v o i o f v O i O O i O i O i O i O i O i O i C M o> C M O i C M CO C M CO C O C M CO CO C M CO < D 09 < CO CO C O C O CO C O CO C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O CO CO C O CO C O C O C O C O C O C O C O C O CO C O C O C O C O C O C O C O C O C O C O CO C O •M* CO C O C O CO CO CO t f l iv C O 0 01 CO O i 00 O i N in O i in o i | v t v f v o IV C M f v C M IV o r v o r v f v f v f v C O f v 00 f v o f v o O i CO O i 00 O i C O O i 00 O i 00 O i C O 0 0 CD 0 0 ao 0 0 CO .O 2 ^ r in in ^ r in f v V * f v f v G O CO CD O i O i O i O i n O i O i Tf* T“ O i o in o in T“ in in in in in in in T“ in CM m C M in C M in O i o C O o CO o CO O i o r v C M |V C M C O C O C O * * C O O i CO O i CO O i O i O i C O O i C O O i CO O i f v O i f v CO CO CO CO O i T f O i i f C O C O C O C O C O T“ C O T“ o 0 0 o G O o 00 f v C O C M CO C M C O C M N o in C M m C M f v CO | v C O f v N r v C O f v C O f v C O IV f v f v r v 55 C O f v T f f v O i C M r — O i C M O i C M O i C M O i C M O i C M r “ o o f v a o f v 00 f v CO in <0 C O C O C O C O C O V - - - C M C M C M C M C M C M C M CO C O C O CO O O C O C O C O C ore 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H 1 6 H X 00 X 00 X C O 1 8 H 1 8 H X 00 X C O 1 8 H 1 8 H 1 8 H 1 8 H X C O 18H X C O 18H 18H X ao 18H 18H 18H 18H X 00 18H 18H 18H 18H I Hole II7 4 4 A < f v 7 4 4 A || 7 4 4 A ||7 4 4 A 17 4 4 A II7 4 4 A II7 4 4 A 7 4 4 A < n fv < f v | | 7 4 4 A II7 4 4 A ||7 4 4 A < fv | | 7 4 4 A 7 4 4 A II7 4 4 A | | 7 4 4 A < fv |7 4 4 A II7 4 4 A | | 7 4 4 A H7 4 4 A | | 7 4 4 A 7 4 4 A < f v | | 7 4 4 A 7 4 4 A < f v II7 4 4 A 7 4 4 A | | 7 4 4 A Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52 Figure 2.6. Relationships between Krithe valve size, mass, and clarity (luminance). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 o > to to 70 Mass = 0.1457 (Area) - 13.278 r = 0.71 60 50 40 30 20 0 0 300 100 250 350 150 200 Relative Area 70 60 o .1 50 E 40 Luminance = 0.0119 (Area) + 47.398 r = 0.09 30 100 150 200 250 Relative Area 300 350 70 60 © o c (0 .£ 50 E 40 30 Luminance = -0.05 (Mass) + 51.565 r= 0.08 O 10 20 30 40 Mass (ng) 50 60 70 Critical value (a=0.05; d.f.=66,70) for r significance = 0.20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54 number of relatively massive larger valves (upper left of Figure 2.6a), where greater thickness would tend to reduce light transmittance. Determined Mg/Ca and Sr/Ca ratios were regressed against each of these valve- based measures for possible phenotypic effects upon observed ratios (Figure 2.7). Neither elemental ratio shows a significant relationship to luminance (Figure 2.7a). Mg/Ca and Sr/Ca show insignificant, but slightly negative and positive, respectively, slopes with respect to area (Figure 2.7b). This pattern is also present with respect to valve mass and slightly more robust (i.e., higher r values), with a statistically significant relationship between valve Sr/Ca and mass at the 95% confidence level. The potential physiological and early diagenetic effects of organic carbon export from the surface and possible increases in carbonate undersaturation on observed elemental ratios were examined via biogenic opal and the dissolution indices of % fragmented planktic forams and % benthic forams (Diester-Haass, 1996; Salamy and Zachos, 1999; Figures 2.8 and 2.9). The later two proxies were drawn from Diester- Haass’s (1996) offset sampling of the Site 744A core and therefore plotted values represent averages of stratigraphically adjacent sample pairs. None of these proxies show statistically significant correlations at the 95% confidence level, but both Mg/Ca and Sr/Ca values tend to decrease as these indices increase. While tempting to make interpretations of the above phenotypic and environmental factors on the observed elemental ratios, each of these correlations assume that other factors are constant (e.g., BWT, alkalinity, food availability, etc.), which is clearly not the case for a population sampled across the Oi-1 event. To isolate potential phenotypic effects from major environmental trends through time, the largest samples from relatively low productivity (34.199 Ma, n=6, biogenic opal = 0.88) and relatively high productivity (32.876 Ma, n=; % biogenic opal = 4.34) intervals were compared to size, mass, and translucence (Figure 2.10). These data show generally Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55 Figure 2.7. Relationships of Krithe valve size, mass, and clarity (luminance) to determined elemental ratios of Mg/Ca and Sr/Ca. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mmol/mol mmol/mol mmol/mol 56 12 10 8 6 4 i 2 I A -A- O Q 24Mg/48Ca = 0.0081 (L) + 6.7654 r = 0.05 00 ° ° o <5 ° O 6Sr/48Ca = 0.0004(L) + 2.4667 r = 0.02 30 1 2 r i 10 8 6 4 2 40 o oJo 50 Luminance 60 70 24Mg/48Ca = -0.0023(A) + 7.7555 r = 0.11 0— 0 ; o° 6 O c cp < 5 6 A A A A ________A A A / I I'fW V AA A B 6 Sr/48Ca = 0.0004(A) + 2.3948 r = 0.12 100 12 10 8 6 4 2 0 150 200 250 300 350 Area Mg/48Ca = -0.0155(M) + 7.5573 r = 0.14 86Sr/48Ca = 0.0033(M) + 2.408 r = 0.21 10 20 30 40 Mass (|ag) 50 60 70 Critical value (a=0.05; d.f.=66,70) for r significance = 0.20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57 Figure 2.8. Relationships of Krithe Mg/Ca to biogenic opal and dissolution indices of % fragmented planktic foraminifera and % benthic foraminifera. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58 Fragmentation Planktic Foraminifera o CO O in o o CO o C M CD CD U - M 0 ° o < P C M o > 0 0 (O in in c o o c o in C M © C M in in |edo 0!ue6o;g pue ejdjiuiiuejo^ omiueg % Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mg/4 8 Ca 59 Figure 2.9. Relationships of Krithe Sr/Ca to biogenic opal and dissolution indices of % fragmented planktic foraminifera and % benthic foraminifera. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 0 % Fragmentation Planktic Foraminifera o r « . o c o o in o o c o o C M (0 O m m u- □ c m in c o o c o in C M o CM in in |edo o!ue6o;g pue ejajiuiuiejoj omtuag % in C M c o o o c o in C M O in C M in C M c v i o o c v i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 06Sr/48Ca 61 Figure 2.10. Relationships of Krithe valve size, mass, and clarity (luminance) to determined elemental ratios of Mg/Ca and Sr/Ca within the two most sampled intervals. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62 12 10 8 6 4 2 •Mq«: 'C:1 = 0 0 6 4 ,L; * 4 0236 i = o • : Mg/ Ca = 0.0147(L) + 5.7487 • " j r = 0.20 8 06Sr/48Ca = -0.003x + 2.628 r = 0.19 ’ Sr ‘ 'Co = -0 0 2 3x - 3 7 j 81 - A r = 0 72 30 40 50 Luminance 60 70 12 10 8 6 4 2 ' * ^ /lg/48Ca = 0.0028(A) + 5.949 r = 0.54 ' ? ■ “ Cn = 0 ! - 0 4 f; A A * - - - - - 86Sr/48Ca = 0.0007(A) + 2.335 100 150 200 250 Area 300 350 12 10 8 6 4 2 0 10 " M g ' C;- - 0 '43-: .! 843 7 r - 0 61 24Mg/48Ca = 0.0386(M) + 5.8805 r = 0.74 C i - 0 311V> 86Sr/48Ca = 0.0093(M) + 2.3174 r = 0.80 20 30 40 Mass (pg) 50 60 70 Open symbols = 32.876 Ma, Gray Symbols = 34.199 Ma Critical value (a=0.05; d.f.=3,4) for r significance = 0.81 (32.876 Ma),0.73 (34.199 Ma)) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63 positive and some statistically significant relationships between elemental ratios and phenotype. Given that each analyzed Krithe population for each horizon was not randomly selected but comprised of the best visually preserved individuals, it is difficult to fully evaluate these data within or between stratigraphic horizons. For completeness, each valve measurement for each specimen, and the mean value per horizon, is plotted stratigraphically in Figure 2.11. These data should not be interpreted as representative of the entire sample population. 2.3.2. Interpretation of stratigraphic elemental variations Stratigraphic variations of elemental ratios in individual Krithe, and the mean ratio per horizon, are presented in Figure 2.12. Stratigraphic samples containing multiple Mg/Ca determinations had an average relative standard deviation of 8%, slightly greater than the 6% observed by Dwyer et al. (1995) for late Quaternary and late Pliocene intervals. The mean sample Mg/Ca ratio was 7.13 ± 0.73 mmol/mol and ranged from 5.44 to 9.37. Using the slope of Equation 2, this 3.9 mmol/mol range equals a relative temperature range of 2.29 C°. For the entire interval, estimated BWT increased by 0.16 C°/Myr, a weak but statistically significant relationship (r = 0.58; a = 0.05; d.f. = 69; r(crit) =0.20). Assuming an Eocene-Oligocene Mg/Caseawater similar to today (i.e., straightforward application of Equation 2 slope and intercept), average BWT would have been 0.38 ± 2.29 °C through the interval. Using the paleothermometry equation for the Arctic bottom waters (BWT = 0.375 (Mg/Ca [mmol/mol]) -3.53; Cronin et al., 1996), average BWT decreases to -1.11 °C and is less variable with a total range of 1.47 °C. However, as discussed below, such a straight-forward application is unrealistic given likely changes in Mg/Caseawater over time (Hardie, 1996). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64 Figure 2.11. Stratigraphic distribution of Krithe valve mass, area, and luminance for the chemically analyzed sample population. Note that these data should not be interpreted to reflect “real” shifts in the parent population from which samples were selectively drawn. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65 a e i o 9U03O6j|O A|JB3 0 U0 OO3 0 JB“| Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66 Figure 2.12. Stratigraphic variations in Mg/Ca and Sr/Ca in the Krithe ostracodes through the Eocene-Oligocene event. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67 ■ • ; c r « M c o © 33.0-' c ( 1 ) O O .O ) I o 1 2 3 3 .5 - * ( 3 4 .0 - ■ CO O . , Eocene 3 4 .5 -■ < 2 ■ > ; tow m m m r 1 1 1 0 9 8 - i — i— i— i— H 1 ----1 --- I--- H 24Mg/48Ca (mmol/mol) 6 5 4 3 2f — I 1 ----1 ---- 1 ----t •* I » I -t- J _ • V i T O V » - rt'fcV H 1 ---- 1 ---- F H — I---- h -2 -1 0 +1 +2 24Mg/48Ca-based Relative Temperature Change (C°) 2.2 2.4 2.6 2.8 “ S r / ^ C a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68 2.4. Discussion 2.4.1. Assumptions for relative BWT reconstruction from fossil Krithe Mg/Ca ratios The above relative BWT estimates are highly speculative as they involve three critical assumptions: ( 1) Mg/Caseawater is unknown, but effectively constant, through Oi-1, (2) the Mg/Ca paleothermometer is insignificantly affected by infaunal life-mode of Krithe, and (3) diagenetic alteration of valves is insignificant. These are discussed below. The first is deemed justified from first principles, but the later two require explicit testing through modern studies. Assuming that these three assumptions are robust, the implications of this study are discussed. Mg/CaS e a w a ie r is unknown. but effectively constant, through Oi-1: The residence times of Mg2 ’ and Ca2 ', and therefore Mg/Ca , are controlled by their fluxes into and ^ ® seawater * out of the global ocean system (Broecker and Peng, 1982). In the modern ocean, Ca2 ' residence time is 1.1 x 106yr, Mg2 ' residence time is 1.5 x 107 yr, and Mg/Caja a w a ie r is 5.14 (Bruland, 1983). Many Late Neogene and Quaternary Mg/Ca paleothermometry studies have employed these modern values for absolute temperature calculations. During Paleogene intervals, however, secular flux variations may have produced Mg/Ca , values significantly different (~±10%) from today (Lasaga et al., 1985). S 6 3 W 3 l6 r * For example, the Paleocene peak in ocean-crust production and associated hydrothermal exchange of Mg2 ' for Ca2 ' likely strongly decreased Mg/Cajo a )(a te r (Larson, 1991). For the -1 0 s yr events addressed here, Mg/Ca is assumed constant, but of unknown absolute ' seawater value. This assumption restricts reconstruction of climate parameters to a relative nature (e.g., 4 C° change versus 2 °C to 6 °C change over X kyr), but appears to be robust: Kasting and Richardson (1985), using the BLAG geochemical cycle model, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69 demonstrated that an instantaneous, four-fold increase in global hydrothermal fluid flux would require >105 years to significantly alter global Mg/CaM iw a ie (. Mg/Ca paleothermometer is insignificantly affected by infaunal life-mode of Krithe: In the upper few centimeters of sediment where ostracodes live and biomineralize, carbonate dissolution may significantly alter the ambient porewater chemistry by increasing Ca2* relative to Mg2*. Majoran and Agrenius (pers. comm.) cultured deep-ocean Krithe under natural conditions (except pressure) and found individuals living as deep as 20 mm and always deeper than 3 mm. Because individuals may biomineralize at these different sediment depths, constraints must be placed on how porewater chemistry may change with depth. Factors affecting carbonate dissolution include Co r g /C aC 03, saturation state of bottom waters, depth profile of Co r g remineralization rate, and the rate constant of C aC 03 dissolution (both aragonite and calcite) (Martin and Sayles, 1996). A primary control on carbonate dissolution, however, appears to be oxygen influx, which limits organic decay and thereby production of C 0 2. Maximum change in Mg/Ca p o r e w a te r compared to overlying Mg:Ca s e a w a ie r may be estimated using Fick’s first laws and “worst-case” natural limits of variation. By assuming a maximum dissolved 0 2 concentration in bottom water (-300 pmols) and maximum 0 2 flux into sediments (1 0 1 1 mol c m ' s ’), it can be shown that at 5 cm depth the maximum decrease in Mg/Cap o r e w a te r compared to overlying Mg/Cas e a w a le r would be less than 5% (Hammond, pers. comm., 1999). Diagenetic alteration of valves is insignificant This final assumption represent a major concern for the application of Mg/Ca paleothermometry to ancient materials. Although a much greater amount o.‘ research effort has been exerted on elemental diagenesis in foraminifera (e.g., Delaney et al., 1985; McCorkle et al., 1995; Nurnberg et al., 1996; McIntyre et al., 1997; Rosenthal et al., 1997; Hastings et al., 1998; Stoll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70 et al., 1999), controls upon primary values and subsequent susceptibility to diagenesis remain a major issue for development of the field. In contrast, little similar study has been conducted on ostracodes, making evaluation of possible diagenesis difficult. Cathodoluminescence provided no evidence for recrystallization, but SEM analyses of coeval specimens does indicate that minor dissolution may have occurred. Observed Oi-1 Sr/Ca ratios averaging around 2.5 mmol/mol are slightly lower than the -3 mmol/mol values present in the modern Krithe (Dwyer et al., 1995). These lower values may reflect partial diagenetic recrystalization given differences in Sr partition coefficients between inorganic and biogenic calcite (Baker et al., 1982; Delaney, 1989). However, complete recrystallization would produce an -4-fold decrease in observed ratios, much greater than that observed for Site 744A compared to modern Krithe. Conversely, the higher inorganic than biogenic Mg partition coefficients would increase observed Mg/Ca (Delaney, 1989). Selective dissolution of more susceptible valve regions (i.e., higher organic matrix, Mg/Ca values, smaller calcite crystals, etc.) could affect the mean Mg/Ca of the remaining valve material, but additional studies are required to evaluate this and other possible controls. 2.5. Conclusion The Eocene-Oligocene Oi-1 event is marked by a pronounced benthic foraminiferal 8,80 decrease of -1.5 % o which reflects some combination of development of major Antarctic ice-sheets and thermal cooling of bottom-waters. However, partitioning the two controls is difficult without independent constraints on ice-volume, seawater 51 8 0 , or bottom water temperatures. To address this uncertainty, the recently documented temperature-dependence of Mg/Ca in the low-Mg calcitic valves of Krithe ostracodes was applied to material from the Kerguelen Plateau using inductively coupled plasma mass-spectrometry (ICP-MS). Comparison of valve area, mass, and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71 translucence, as well as environmental and dissolution indices, show some correlations with elemental ratios, but no clearly resolvable or predominant relationships. Diagenetic screening, principally by cathodoluminescence (CL) and limited SEM study, show no positive evidence for diagenetic alteration except for minor dissolution. Individual Krithe Mg/Ca shows relatively high intersample variability and no net trends through the oxygen isotope excursion. Simplistic interpretation of this pattern supports the hypothesis that the 51 s O decrease largely reflects major Antarctic ice-sheet development in response to thermal isolation of the continent following its isolation by oceanic gateways. This absence of a major temperature factor at Kerguelen Plateau is consistent with on-going benthic foram-based Mg/Ca studies at Maud Rise (Elderfield, pers. comm, 1999). The “signal failure” in ostracode and foram Mg/Ca-based records from two different sites has important implication for either the nature of the event or the applicability of such proxies in deep-time. Thus, the field of elemental paleothermometry remains in its infancy with major questions on the primary (i.e., physiological) and secondary (i.e., diagenetic) controls upon modem and ancient biological carbonate. Thus, these data are considered a small experimental exercise rather than robust evidence for the nature of climatic dynamics during the Eocene- Oligocene transition. 2.6. Future directions and suggestions As with many scientific endeavors, this initial study of Mg/Ca ratios in the ostracode genus Krithe through the Eocene-Oligocene transition has produced more questions instead of a unique answer. For example, does the absence of a strong Mg/Ca- based paleotemperature change accurately reflect historical events or a subsequent overprint of diagenetic alteration? To what degree does ecophenotypy not directly related to temperature effects (e.g., food availability) affect observed ratios through Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 2 biological (biomineralization) or physicochemical (dissolution) means? In addition, re-evaluation of existing data dictates that the biological and ecological context of biomineralization be better appreciated. For example, Krithe Mg/Ca values decrease with increasing valve mass in the modern Atlantic (Figure 2.13). What is the ecophysiological basis of this relationship with that of temperature? In science, as in religion, the most recent convert often becomes the most fervent disciple, but this application of Mg/Ca paleothermometry to Paleogene materials encourages skepticism. The patterns and processes of Mg/Ca incorporation and alteration in both benthic foraminifera and ostracodes are the focus of future postdoctoral research to be conducted at UC-Santa Cruz. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73 Figure 2.13. Relationship between BWT, Krithe valve mass, and Krithe Mg/Ca in the modern Atlantic Ocean. Note the logarithmic correlation between decreasing Mg/Ca and increasing valve mass in lower plot. Data courtesy of Cronin. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110 Mg/Ca O M as s r(c rit) = 0 .1 4 ( a = 0 .0 5 ; d .f.= 1 4 5 ) Mg/Ca = 0.82 (BWT) + 8.93 r = 0.69 O Mass = -0.69 (BWT) + 23.44 r = 0.19 ' 20 BWT (°C) 25 20 j e E _ 1 5 Mg/Ca = -2.0837Ln(Mass) + 16.304 r = .54 r(c rit) = 0 .1 4 (a = 0 .0 5 ; d .f.= 1 4 5 ) \ oi I 10 j i 0 10 20 30 40 50 60 70 80 90 100 Valve Mass (ng) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Valve M ass (Mg) 75 3. Deep-ocean ostracode faunal response to the Eocene-Oligocene transition: High- resolution (104 yr) records from the Southern Ocean (ODP Sites 744A and 6 8 9 B ) 3.1. Introduction 3.1.1. Scope Ostracodes are the only commonly preserved benthic metazoa in the deep-ocean and afford one of the most continuous records of Cenozoic multicellular life. Thus, the order provides insight on paleoecological and paleoenvironmental changes in the extensive bathyal and abyssal realms. Furthermore, ostracodes provide an independent means to test environmental reconstructions based on more commonly studied unicellular benthic foraminifera. During the Eocene-Oligocene transition, the relatively ice-free “greenhouse" climate state of the earlier Paleogene shifted towards an “icehouse” climate state more similar to today (Fischer, 1981, 1984; Prothero and Bergrren, 1992). Benson et al. (1984; 1985) identified this transition as an important Cenozoic milestone for deep-ocean ostracode faunas based on a global DSDP census of approximately one million year sampling intervals. This study documents the response of deep-ocean (bathyal) ostracodes at much higher resolution (104 yr), focusing on ODP Sites 744A and 689B in the Southern Ocean, a pivotal region for global ocean circulation through thermohaline bottom-water production as well as carbon cycling through significant primary productivity (Figure 3.1). Ostracode faunal data are examined within a robust paleoenvironmental framework of stable isotopic and sedimentological data, and compared to previous paleoceanographic reconstructions of the Eocene-Oligocene Southern Ocean region. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76 Figure 3.1 Southern hemisphere paleogeography and ODP site locations during the Late Eocene (-40 Ma). (after Lawver et al., 1992) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78 At both sites, the major changes in ostracode abundance and faunal composition coincide with the major 5'8 0 step increase known as Oi-1. These faunal changes likely represent allogenic replacement in response to environmental changes, versus evolution, extinction, or autogenic succession (c.f., Rollins et al., 1979; Gould and Calloway, 1980). This assertion is based on the global range-through of all formally described taxa and limited by the supra-ecological temporal scale of the observed changes. Although the study’s temporal resolution approaches the lower limit for deep- ocean records (i.e., tens of thousands of years; Schindel, 1980), the observed patterns imply processes operating at time-scales much longer than that of known ecological processes. The ostracode response reflects significant changes in surface productivity and ambient water mass conditions, although changes in the latter have not previously been documented through sedimentological or benthic foraminiferal studies. The stratigraphic distribution of taxa with known environmental preferences/tolerances (e.g., platycopids) provide insight on the nature of these environmental changes, while existing paleoenvironmental data provides provisional constraints on taxa of uncertain paleoecology that require further testing. 3.1.2. Eocene-Oligocene transition Proposed forcing mechanisms for the Eocene-Oligocene transition from a “greenhouse” to “icehouse” state include COz draw-down through carbon cycle shifts (Lasaga et al., 1985; Freeman and Hayes, 1992) and progressive Antarctic “refrigeration” as a circum-polar current developed through the growth of oceanic gateways (Kennett and Shackleton, 1976, Bartek et al., 1992; Lawver et al., 1992; Bice, unpublished data). These events drove significant cryospheric growth and progressive deep-ocean cooling through increased bottom-water production at high southern latitudes (Kennett and Barker, 1990; Miller, 1992). The global benthic Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79 foram 8,0O record of this transition is punctuated by several positive relative extrema, representing some combination of cryosphere expansion and bottom-water cooling (Figure 3.2). The most pronounced and complex of these extrema, termed Oi-1 by Miller et al. (1981), occurs just above the epoch boundary with a 5,0O increase of ~1.5%o. High- resolution isotope analyses have revealed two -100 kyr sub-cycles (Oi-1 a and Oi-1 b) within the -350 kyr event, representing major ice-volume fluctuations or pulses of colder bottom-water production (Zachos et al., 1996). Oi-1 is interpreted as a non linear climate “overshoot” from crossing a critical threshold dividing “greenhouse” and “icehouse” states (Berger, 1982). This overshoot presumably lowered atmospheric C 0 2 through positive feedbacks (e.g., increased C 0 2 solubility, Co r g production and burial, shelf carbonate weathering), which promoted lower bottom water temperatures and greater ice-volume (Gibbs and Kump, 1994; Derry and France- Lanord, 1996). The resultant climate extrema subsequently attenuated to a new equilibrium through negative feedbacks (e.g., global nutrient sequestering, ocean alkalinity shifts) (Zachos et al., 1996). Similar scenarios have been proposed for the Miocene (Vincent and Berger, 1985) and the last glacial maximum (Broecker, 1982). In a broader Cenozoic perspective, the Eocene-Oligocene transition may be placed within Kennett and Stott’s (1990) paleoceanographic model based on stable isotope evidence from Maud Rise (Leg 113; Sites 689B [-1,500 m paleodepth at 35 Ma] and 690B [-2,300 m paleodepth at 35 Ma]) (Figure 3.3). During the early Paleogene, a thermospheric “Proteus Ocean" was halothermally driven by relatively warm, saline deep water masses (WSDW) produced in low- to mid-latitude Tethyan or North Atlantic regions (v. Corfield and Norris, 1996). This WSDW was overlain by less-dense (i.e., cooler and/or less saline) water masses derived from high southern latitudes. Perhaps as early as the late Paleocene, but certainly by the Eocene-Oligocene transition and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 Figure 3.2 Global composite record of benthic foraminiferal stable isotope values for the early Cenozoic. Horizontal line indicates the major positive step increase in 8’8 0 and, to a lesser extent, in 5,3 C, associated with the Eocene-Oligocene transition, (from Zachos, pers. comm.) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C O o ® I P P ! I / M A | J B 3 a te -) A | j b 3 e ie -| | 9 I P P I I / M A ( J B 3 A |J B 3 9 J B 1 auaoo(i/\| auaoo6no auaoog auaooeiHd > 1 o CM o CO O o m o co o Is - Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 Figure 3.3 Cenozoic evolution of thermohaline circulation proposed by Kennett and Stott (1990). The three panels summarize the transition from halothermal- to thermohaline-driven circulation of deep and intermediate waters. AABW = Antarctic Bottom Water; AAIW = Antarctic Intermediate Water; NADW = North Atlantic Deep Water; WSDW = Warm Saline Deep Water; MED. = Mediterranean Outflow Water. (from Kennett and Stott, 1990) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83 ANT. ARCTIC 80° 70 60 S O 40 30 20 10 0 10 20 30 40 50 60 70s S N Latitude OCEANUS: Modem, psychrospheric (Thermohaline circulation) km WATER WSDW 80 70 60 50 40 30 20 10 10 20 30 40 50 60 70 N PROTO-OCEANUS: Oligocene (Mixed halothermai & thermohaline circulation) km WATER WSDW 80° 70 60 50 40 30 20 10 0 10 20 30 40 50 60 70* PROTEUS: Eocene (Halothermai circulation) N Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 84 into the early Oligocene, Antarctic circum-polar surface-water cooling and regional sea-ice production produced two thermohaline-driven water masses: Antarctic Bottom Water (AABW), sufficiently cold and saline to displace WSDW upward and progressively flood the deep ocean, and Antarctic Intermediate Water (AAIW), a less dense, cool water- mass overlying the WSDW mass. These new water-masses shifted global circulation towards a psychrospheric “Proto-Oceanus” mode. In the Latest Oligocene, increased thermohaline production, perhaps in concert with decreased WSBW production, led to the disappearance of WSDW from the Maud Rise region and initiated the “ Oceanus” stage, similar to modern conditions. Mead et al. (1993) paleobathymetrically extended the Kennett and Stott (1990) model with stable isotope data from ODP Sites 703 (Meteor Rise; -1,000 m paleodepth at 35 Ma) and 699 (abyssal plain of the South Atlantic; -3,400 m paleodepth at 35 Ma). These data confirm the consistent 8,aO maximum at Site 689B depths during the Paleogene, with a near constant 5,80 vertical gradient pattern before and after the Oi-1 event (Figure 3.4a). The 81 3 C data show no strong vertical gradient at any given time, but a transient overall enrichment during the Eocene-Oligocene transition (-35 Ma), followed by generally lower 51 3 C overall (Figure 3.4b). This 8’3C pattern implies that the Southern Ocean was well-mixed or supplied by a single deep-water source region. These two implications, however, are confounded by the consistent 81 8 0 gradient through time. Other regions show generally similar 8'3 C values and little vertical gradation during this transition, perhaps reflecting global oligotrophy and/or multiple bottom- water sources that precluded pronounced global and bathymetric S1 3 C variations (Miller and Fairbanks, 1985; Miller, 1992; Mead ef al., 1993; Wright and Miller, 1993). During Oi-1, a global increase in deep-water 8,3C values relative to surface values, together with abundant unconformities (Miller et al., 1987), suggests increased circulation fluxes at all depths (Zachos et al., 1993). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85 Figure 3.4 Stable isotope contours through time and paleodepth for ODP Sites 689, 690, 699, and 703. Note the (a) consistent 5,80 maximum at Site 689B throughout the Paleogene and (b) general absence of strong vertical 8,3C gradients, (from Mead et al., 1 9 9 3 ) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 6 ^ •4 0 Q. *0 727 ° * 71 3.524 ? 315 a.s»; • t 013 UJ041 Q A 7 % £ < > ■ * °4 7 ' « 7 0 2 (0 0 5 5 ] 1 040 t 626 1 .7 5 1 42 40 38 28 36 26 30 32 34 Age (Ma) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Paleodepth (km) 87 • 30 o # o f c e > ’ • 'x . t44 • j « j o r2% ° * 9 t 0C20 0 f H O.S3t 0 S 1 ? / * G* i oii ( 0 ,0 4 ) 1 1 6 0 * M 2 y 0 S«s O M I 0 MOOj 0 -5 2 8 3 r t t tc *i 26 28 30 32 34 38 40 36 42 Age (Ma) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 These southern high-latitude paleoceanographic reconstructions are generally consistent with the Zachos et al. (1993) global compilation of surface- and deep-water temperatures for five Paleogene intervals (i.e., late Paleocene, early Eocene, middle early Eocene, late Eocene, and early Oligocene). Following the early Eocene thermal maximum, high-latitude surface temperatures declined 8-9 C° by the late Eocene, and declined another 2 C° from the late Eocene to early Oligocene. However, low-latitude surface temperature for these intervals remains questionable due to upwelling-biased site distributions, preferential planktic foram dissolution, and ice-volume uncertainties. Global deep-water temperatures parallel those of high-latitude surface temperatures within an offset of <2 C° until the late Eocene, when deep-waters north of 45° S warmed by up to 3 C° relative to high-latitude surface- and deep-waters. Numerous caveats exist for the general paleoceanographic model for the Eocene- Oligocene Southern Ocean . Restriction of paleobathymetric data to shallower than -3,500 m does not allow comment on the development, presence, or intensity of proto- AABW at deeper depths. In addition, the similarity of benthic foram faunas at Sites 689B and 690B throughout the interval provide no positive evidence for different water masses (Thomas, 1990, 1992). Finally, recent higher-resolution stable isotope analyses of Sites 689 and 690 (R. Zahn, unpublished data) fail to replicate Kennett and Stott’s ('990) lower-resolution 8’8 0 inversion between the sites. Additional details on, and deviations from, this general pattern at Sites 744A and 689B are addressed in the discussion. 3.1.3. Global ostracode response Benson et al. (1984; 1985) and Benson (1990) documented the Eocene- Oligocene as a major event in deep-ocean ostracode history, with globally decreased abundance, increased diversity, and decreased provinciality (Figure 3.5 and 3.6). These Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89 Figure 3.5 Global summary of ostracode faunas from 1,600 50 cc samples from 155 DSDP Sites spanning the Cretaceous and Cenozoic. Letters A-E are A, K/T boundary; B, Eocene-Oligocene origin of the psychrosphere; C, Middle Miocene “event”; D, Messinian Salinity Crisis; and E, the 3.5 Ma event, (from Benson, 1990) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. % samples wflti ostracods atandng generic d versly 90 8 8 8 8 8 8 8 v — • j eueoofs ! ou®oofl|o eueoog i m r -» tO O C D IO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 - W dveraRy *•»«•» titnover 91 Figure 3.6 Rho-group similarity analysis of ostracode faunas along an east-west transect of the South Atlantic. Each asterisk represents one or more samples of a given age from a given site; connecting lines indicate samples sharing a 0.76 or greater level of similarity, (from Benson et al., 1985) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92 _ cm < n to to n U i co to co io c o 1 0 0 9 5 i i i r 1 r t ~ r r r * r — i • ■ • • ■ 1 • • 1 I P -*-* * * CD O J m o > CM IO N . CM in £ to o CM CM CM in c o m (ei/\|) a iu ij. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R h o -g ro u p s 0 .7 6 93 changes were interpreted to reflect an increasingly homogenized global deep-ocean environment as bottom-water masses shifted from warm, slow, and salinity-driven (thermospheric) to cold, fast, and temperature-driven (psychrospheric), with increased nutrient and oxygen fluxes. In contrast, some higher-resolution regional studies in the northern hemisphere report relatively little faunal change across the epoch boundary (e.g., Coles, 1990; Coles et al., 1990). This disparity may result largely from differences in spatio-temporal scales of study; its resolution will require sub-sampling of the global record to allow direct comparison to these more regional studies. In southern high-latitudes, ostracode studies have focused largely on Quaternary and living deep-ocean faunas (e.g., Dingle et al., 1989, 1990; Ayress et al., 1997), with only two published pre-Neogene studies for latitudes higher than 35° S (i.e., Maastrichtian of the Atlantic sector of the Southern Ocean by Majoran et al., 1997, 1998; Late Paleocene Thermal Maximum at Site 689B on Maud Rise by Steineck and Thomas, 1996). Additional low-resolution [106 '7 yr] data exist in unpublished dissertations conducted at the University of Wales, Aberystwyth, under the advisement of R. C. Whatley (e.g., Millson, 1987; Balman, 1997). Thus, the response of southern high-latitude deep-ocean ostracodes to late Paleogene cooling, and the Oi-1 event in particular, is poorly known. This study provides detailed ostracode faunal data during the Eocene-Oligocene “greenhouse-icehouse” transition as recorded on Kerguelen Plateau (ODP Site 744A) and Maud Rise (ODP Site 689B). 3.2. Stratigraphic and paleoenvironmental context 3.2.1. Site descriptions ODP Hole 744A was drilled on the southern slope of Kerguelen Plateau (61°34.66‘S, 82°47.25'E) at a water depth of 2,307 m (Figure 3.1). Advanced piston Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94 coring produced effectively 100% recovery of a Late Eocene-Early Oligocene sequence of predominately coccolith carbonate ooze, with a significant increase in biogenic opal from <2 to as much as 18 weight % in the earliest Oligocene (Barron et al., 1989a; Salamy and Zachos, 1999). Due to the plateau’s distance from Antarctica and elevated topography since the Late Cretaceous (>80 Ma), terrigenous sediments are rare except for intervals of ice-rafted debris (IRD; e.g., quartz, feldspars, and lithic fragments) (Barron et al., 1989a; Lawver et al., 1992). Estimated Late Eocene-Early Oligocene paleodepths are estimated at -1,800 meters, shallower than the estimated paleosurface of the carbonate compensation depth (CCD) (Barrera and Huber, 1993). ODP Hole 689B was drilled on the northern crest of Maud Rise (64°31.01'S, 03°06.00'E) at a water depth of 2,080 meters (Figure 3.1). Advanced piston coring produced effectively 100% recovery of a Late Eocene-Early Oligocene sequence of coccolithophore carbonate ooze, where % opal varies from -10-90 % (Barker et al., 1988a; Diester-Haass, 1993). Maud Rise had also developed significant topographic relief by the Late Cretaceous and was paleogeographically isolated from Antarctica, restricting most terrigenous sediment to that delivered as IRD (Barker et al., 1988a; Lawver et al., 1992). Late Eocene-Early Oligocene paleodepths are estimated at 1,400- 1,650 meters, also shallower than the estimated paleosurface of the CCD (Kennett and Stott, 1990). 3.2.2. Paleoenvironmental framework Previous studies of the Eocene-Oligocene transition at Sites 744A and 689B provide an independent paleoenvironmental framework in which to examine the faunal response of deep-ocean ostracodes. Age models for each site are based on calibrating their magnetostratigraphy (Spiess, 1990; Keating and Sakai, 1991) to the Geomagnetic Polarity Time Scale (GPTS) of Cande and Kent (1995), and assume constant Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95 sedimentation between chron boundaries. Paleomagnetic reversal depths for each site and GPTS chron ages are provided in Appendix 1. All published data of previous studies were re-calibrated to this uniform time-scale using MBSF data. Stable-isotope records from both sites document the pronounced step increase in 5,80 known as Oi-1 (Figure 3.7; Isotope Interval A/B boundary). For Site 744A, Zachos et al. (1996) determined average 81 8 0 and 8’3 C for 1-15 individuals of Cibicidoides spp. from the >150 pm size-fraction. For Site 689B, Diester-Haass and Zahn (1996) determined average 5,80 and 8’3 C for 1-10 individuals of C. mundulus and C. praemundulus from the >150 pm size-fraction. The higher-resolution stable isotope stratigraphy of Site 744A clearly reveals the two relative maxima, Oi-1 a and Oi-1b, within the Oi-1 event (i.e., Isotope Intervals B and D). Analogous relative extrema are present at Site 689B, but are poorly defined in its lower-resolution sampling. Therefore, Site 689B was not as finely sub-divided as Site 744A for most analyses (i.e., only Isotope Intervals A and B as defined by the Oi-1 event were employed). Irregardless, the excellent agreement between the two records reinforces the global nature of the shift as well as the robustness of the age model employed. In addition to stable-isotope data indicating some combination of ambient bottom- water cooling and cryospheric expansion, sedimentological studies at each site provide a first-order record of associated primary productivity changes and dissolution intensity (Figure 3.7). For both sites, Diester-Haass (1995; 1996) analyzed 800 counts (if present) in the 63-125, 125-250, 250-500 pm size-fractions to determine weight percent radiolaria and diatoms. For Site 744A, these coarse-fraction % biogenic opal values are in general agreement with coeval chemically-determined bulk % biogenic opal values of Salamy and Zachos (1999). However, magnitudes differ strongly between the methods: % biogenic opal prior to Oi-1 averages -10% optically versus -1-2% spectrophotometrically, while the peaks within Oi-1 average 65-75% optically and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96 Figure 3.7 Paleoenvironmental framework for Site 744A and 689B. Ages based on core magnetostratigraphy (Spiess, 1990; Keating and Sakai, 1991) calibrated to the GPTS of Cande and Kent (1995). Stable isotope data from Zachos et al. (1992) and Diester- Haass and Zahn (1996). Sedimentological data from Diester-Haass (1993, 1995). Relative % opal values are based on standardizing original values for each core and method to its highest and lowest values. White and gray triangles represent stratigraphic position of 10 cc samples examined for ostracodes at Sites 744A and 689B, respectively. Intervals A through E are defined by major inflection points of the stable isotope record at Site 744A, where intervals B and D contain relative maxima within the Oi-1 event (B-E). At Site 689B, the Isotope Interval boundary between A and B is well-defined, but less so for subsequent boundaries. Thus, all later analyses of Site 689B only employ the Isotope Interval A/B boundary. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97 O co AAAAA A A A M A& AAAJLAA A A AAA AA A. A AAA A A . . . . ■ . . m i o U Z I D NSIO m i o 8U0OODHO A|JB3 8 U0 OO3 0 je -| Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98 14-18% spectrophotometrically. Thus, point-count data show much higher overall values as the <63 pm size-fraction is predominately carbonate. Regardless, both methods show a comparable range of 8-9 times their respective base values. In comparison, the Site 689B % biogenic opal record does not show a major step increase in productivity, instead absolute point-count values vary widely and generally between 30-60%, with exceptions of a relative low at 34.2 Ma and slight peaks coincident with those at Site 744A. Both records support increased surface productivity related to enhanced seasonal upwelling, likely driven by polar front shifts/intensification, with maximum productivity coinciding with 5'80 “glacial-condition” maxima during Oi-1 a and Oi-1b. The productivity increase was extreme at 744A, but relatively weak at 689B, where biogenic opal content was already high, but variable (Figure 3.7). Diester-Haass (1995; 1996) also calculated dissolution indices of percent planktic foram fragmentation (i.e., % FPF = fragments /(fragments+whole tests)*100) and percent benthic forams (i.e., % BF = benthic forams/(benthic+planktic forams)*100) for each sample. These dissolution indices generally increase in more corrosive bottom waters because planktic forams are more susceptible to carbonate dissolution, thereby increasing their test fragmentation and decreasing their abundance relative to bethic forams (Berger, 1973; Thunell, 1976; Peterson and Prell, 1985). The relation of surface productivity history to the dissolution indices are discussed in detail below as they appear to influence observed ostracode faunal abundance. 3.3. Materials and methods All ostracode valves and valve fragments were picked from the >150 urn size- fraction of 102 10 cc samples from Site 744A (mean sampling interval = 20±11 cm; 20±14 kyr) and 67 10 cc samples from Site 689B (mean sampling interval = 22±8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99 cm; 53±21 kyr). The total time-interval examined for Site 744A ranged from 32.78- 34.82 Ma (137.28-157.01 mbsf) and for Site 689B ranged from 32.72-36.16 Ma (115.14-129.43 mbsf). Note that only the common interval of 32.5-35.0 is shown in most site-to-site comparisons, although all Site 689B samples were considered for all statistical analyses. Examination of numerous >63 pm size-fractions of samples from each site revealed rarely identifiable fragments and extremely rarely complete individuals. Therefore, this larger volume, finer-grained size-fraction was not examined further and all subsequent discussion is based on the >150 pm size-fraction. Valves were assigned to genus, and where confident, species, through reference to primary literature, unpublished works (e.g., Millson, 1987; Balman, 1997), and systematic collections in the Smithsonian NMNH. Remaining unidentifiable fragments were retained, but not considered further. Given the paucity of comparable high- southern latitude Paleogene ostracode systematic studies, brevity of the geologic interval examined, and rareness of many taxa, formal systematic description was not done. Eight morphologically cohesive groups of individuals not confidently assignable to described taxa were defined as operational taxonomic units using the letters A-l. Deep-ocean ostracode genera are often represented by a single species at a given locale; exceptions in this study include Bradleya, Henryhowella, Pelecocythere, Cytherella, Krithe, and Cytheropteron. Two species each are present in the first three genera, while the total species number in the latter three is uncertain, and therefore all individuals are considered together under their collective genus. Individuals in each taxon were classified into four valve types: left adult, right adult, left juvenile, right juvenile; sexes were not differentiated. Valve fragments exceeding -50% completeness were tallied as complete individuals; less complete, but still identifiable valve fragments were tallied by equating two 25-50% complete fragments (F1 in Appendix 2) or three <25% complete fragments (F2 in Appendix 2) to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 one complete valve. Total and taxon-specific lett/right and adult/juvenile ratios showed no obvious temporal trends, with the possible exception of Algulhasina (discussed below). Therefore, to maximize specimen abundance and information content, all subsequent faunal analyses are based on the summation of each taxon’s left adult, right adult, left juvenile, and right juvenile valves, together with fragment-based “individuals” of the F1 and F2 categories rounded up to the next whole number. Although this summation approach likely increases the relative abundance of taxa with abundant juveniles, it ensures that rarer taxa are not discounted. 3.4. Results 3.4.1. Ostracode abundance and richness Total valve abundance for each site is presented in Figure 3.8. At Site 744A, abundance is greatest at -34.6 Ma, decreases to a relatively stable level from 34.5 to 33.9 Ma, oscillates to a minimum by the onset of Oi-1, remains low through most of the event, and then increases after the Oi-1b maximum. In contrast, ostracode abundance at Site 6896 is lower overall, with two broad maxima prior to Oi-1 and a maximum after Oi-1 a. To examine the possible effects of seawater corrosivity and mass accumulation rates (MAR) on these patterns, ostracode abundance data were compared to % FPF and % BF dissolution indices (Diester-Haass, 1993, 1995) and bulk MAR data (Barker et al., 1988a; Barron et al., 1989a). Bulk MAR were calculated from the linear sedimentation rate multiplied by dry bulk density; specific sediment component MARs may be calculated by multiplying the bulk MAR by their respective weight %. Ostracode abundance appears inversely correlated to foram dissolution indices at both sites (Figure 3.9). At Site 744A, three major pulses in both % BF and % FPF coincide with low abundances at the base of the record, just prior to Oi-1 (-33.75 Ma), and at the onset of Oi-1. Compared to Site 744A, Site 689B % BF values are higher and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101 Figure 3.8 Ostracode abundance and richness at Sites 744A and 689B. White and gray triangles represent stratigraphic position of 10 cc samples for Sites 744A and 689B, respectively. Isotope Intervals B and D contain the two relative 8iaO maxima of the Oi-1 event. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 8 0 7 4 4 A R ich n ess 6 8 9 B Richness 102 O - - C M W . . O . . O T - - I V N j v a / v / • L O 0 o c C O - -c < cc o CO . _ L O 0 o r; 0 T D C ' < < . o i r > F - o UJ O O Q A M AAA/W \W A“ M » a a a a a a a a m a A A AAAAAA A A AAA AA A A AAA A A A /W/WVWMra«l^ /X V W \flV W W W aM ttY Y Y Y X ^^ I f ) 'CM o 'CM I f ) i n C O C O i n CO C O o 3 i n 3 NCI.0 ye io NSIO eueooB jio A jjeg aueoog e)e~ | Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 Figure 3.9. Ostracode abundance and dissolution indices comparison. Ostracode abundance (shaded background) and dissolution indices of benthicrplanktic forams (“% BP; solid line) and fragment:whole planktic forams (% FPF; dashed line) show a weak relationship between abundance and dissolution intensity (Site 744A indices from Diester-Haass, 1996; Site 689 indices from Diester-Haass, 1993). Black triangles to right of each site indicate coinciding intervals of intense dissolution and low valve abundance. Also provided is Site 744A mean MAR data for carbonate, opal, and terrigenous sediments, showing increased sedimentation overall during Chron 13N (Barron et al., 1989a; Salamy and Zachos, 1999). Site 689B mean MAR does not change strongly through the study interval at Site 689B (Barker et al., 1988a) and is therefore not shown. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Site 744A A b u n d a n c e Site 744A M A R ( g / c r n ^ / k y r ) Site 689B Abundance 05 1 .0 1 .5 2.0 2.5 % 0 744A : 34.5- V 6 H 9 B Opal Terrigenous I I t I 0 2 0 40 6 0 8 0 0 0.05 0.10 0 2 0 40 6 0 80 1 0 0 % Benthics and Fragmentation MAR (g/cm2/kyr % Benthics and Fragmentation 104 105 more variable, while % FPF are higher and less variable, with a broad % BF pulse centered at 34.6 Ma and two pronounced relative maxima within the peak of a broader maximum centered on the Oi-1 event. Because Site 689B faunal data and dissolution indices are drawn from the same sample series, their relationship may be directly compared (Figure 3.10). The two dissolution indices are strongly correlated with one another at lower percentages, but highly fragmented samples range in 50-90+ in their % BF values, and show a significant correlation coefficient (r= 0.58 at a=0.05, d.f.=45). Linear regression of Site 689B valve abundance and richness against these dissolution indices ocue correlation coefficients of between 0.53-0.65 — all are significant at a=0.05 (Figure 3.10). MAR data provides insight on possible “ dilution” of ostracode valves through surface productivity. Site 744A valve abundance is broadly inversely related to MAR, with the two- to three-fold increase in carbonate, opal, and terrigenous sediment calculated for Chron 13N coinciding with low (<30) valve abundance (Salamy and Zachos, 1999; Barron et al., 1989a) (Figure 3.9). Little rate variation (-<20% of the mean) exists within each MAR calculation interval, with the exception of high opal MAR variation in Chron 13N. In contrast, Site 689B MAR (not plotted) is relatively low overall, decreasing stepwise from only 0.75 to 0.52 g/cm2 /kyr through the study interval (Barker et al., 1988a; Diester-Haass and Zahn, 1996). Thus, accumulation rates and seawater corrosivity may affect the observed ostracode valve abundances at Site 744A, particularly just prior to and during the Oi-1 event. At Site 689, the relatively constant MAR, combined with less variable, higher relative % opal and % FPF values, implies that dilution was likely less important than seawater corrosivity in producing the observed valve abundances. Richness, the simple tally of the number of taxa per sample (Peet, 1974), generally varies with abundance at both sites (Figure 3.8). At Site 744A, richness Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106 Figure 3.10. Correlation of dissolution indices and ostracode valve abundance and richness at Site 689B. % FPF and % BF covary and each is weakly correlated to various degrees with valve abundance and richness for coeval samples at Site 689B. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Site 689 Dissolution Indices versus Abundance and Richness 100 & 90 ■ o © c 80 © I 70 © w u. s? 60 50 r = 0.58 10 20 30 40 50 60 % Benthic Forams 70 80 90 100 100 • Abundance O Richness 80 r=0.53 60 00 40 •O per® 20 0 0 0 20 40 60 80 % Benthic Forams [BF/(BF+PF)*100] • Abundance O Richness r=0.58 «A0 O oo o o 50 60 70 80 90 % Planktic Foram Fragmentation [FPF/(FPF+WPF)*100] 100 r(crit) = 0.29 at a =0.05 and d.f.=45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Richness Richness 108 decreases toward the onset of Oi-1, but remains high (-1 0 taxa) relative to the very low valve abundance, and then returns to pre-Oi-1 values after the Oi-1b maximum. At Site 689B, overall richness is roughly half that at Site 744 and generally covaries with abundance. 3.4.2. Richness assessment through rarefaction Comparison of the above ostracode abundance and richness records at each site demonstrate a persistent phenomenon in ecological sampling: sample size can strongly affect observed richness (Figure 3.11; Connor and McCoy, 1979). This “species-area effect” precludes robust, straightforward comparison of richness and other diversity indices for samples of differing abundance or incomplete census. Rarefaction estimates how many taxa would be “ captured” if n specimens were randomly sub-sampled from a sample population of N abundance (Sanders, 1968; Hurlbert, 1971). Plotting the “rarefied” richness from n = 1 to each sample’s N produces a set of hypergeometric curves that allow robust richness comparisons at any common n. Rarefaction values are purely deterministic, being a direct function of number of taxa per sample and their relative frequency {i.e., evenness). The method is the only diversity measure sensitive to rare taxa and unbiased by sample size (Smith and Grassle, 1977) and is applicable here given the sample's similar taxonomic composition, mode of collection, and paleoenvironmental context (Raup, 1975; Tipper, 1979). The analytical expression for rarefaction (E(S„)) is: i=i i - N - N i n N n (Hurlbert, 1971), where: Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109 5 is the number of taxa in the sample population (richness), N is the number of individuals in the sample population, N, is the number of individuals in the ith taxon, and n is the number of individuals in the rarefied sub-sample population. In the E(Sn ) equation, the binomial coefficient w ~ Ni represents the number of ways of choosing n specimens from N - N, specimens (i.e., all specimens not in the /th species), whereas the binomial coefficient ^ represents the number of ways of choosing n specimens from all specimens N. The entire pre-summation term (in brackets) represents the probability, from zero to one, of choosing a given taxon ; when sub sampling n specimens. Summing these individual probabilities from / = 1 to S provides a deterministic estimate of richness if n specimens were sub-sampled from the original population N. £(S„)values were calculated using the computer program Mathematica; program code and results are provided in Appendix 3. For each site, rarefaction curves were calculated for pooled sample populations for the entire site and pooled sample populations defined by pre- and post-Oi-1 intervals (Isotope Intervals A and B-E, respectively) (Figure 3.12a,b). In addition, the Site 744A record was divided into five pooled sample populations (Figure 3.12c). Curves for pooled sample populations are based on a calculation series of n = 10, 20, 30, . . . N. Rarefaction-based richness of Site 744A exceeds that of Site 689B at all common n sub-sample sizes (Figure 3.12a). At both sites, pre-Oi-1 richness exceeds that of post-Oi-1, although the trajectory of the 689 post-Oi-1 curve implies that its richness could exceed the pre-Oi-1 interval given additional samples (n = 23; roughly half that of the 689B pre-Oi-1 interval) (Figure 3.12b). At Site 744A, rarefaction curves for Isotope Intervals A through E indicate richness consistently increases from Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 10 Figure 3.11. Ostracode abundance versus richness for Sites 744A and 689B. Site 689B samples tend to be less diverse at higher abundances relative to Site 744A. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Richness 111 25 A 20 A A A A A A A A A A A A A A A A A A A A AA A A 1 5 A A A A A AAAJti • A A A A 4 A A A A A A A A * A A* A A *4AA • • 1 0 A A A 4 A A AA A A • • A O L 9 • • • A A • € & • • 5 A M ft • m m • i a m % A A A A 0 A - a 744A • 689B 25 50 75 1 0 0 1 2 5 15 0 1 7 5 Valve Abundance i 200 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 1 2 Figure 3.12. Rarefaction curves for Site 744A and 689B. A. Pooled sample populations for entire sites. B. Pooled samples defined by the onset of Oi-1 (Isotope Intervals A and B-E). C. Pooled samples defined by Isotope Intervals A through E for Site 744A. Curves for pooled sample populations are based on a calculation series of n = 10, 20, 30, . . . N. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R i c h n e s s R i c h n e s s R i c h n e s s 113 40 7 4 4 A (n=101) 3 0 20 6 8 9 B (n=66) 10 Rarefaction of Sites 0 0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0 7 0 0 0 8 0 0 0 n 4 0 3 0 20 10 0 0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0 7 0 0 0 8 0 0 0 n 4 0 3 0 20 10 0 0 5 0 0 1 0 0 0 1 5 0 0 2 0 0 0 2 5 0 0 3 0 0 0 3 5 0 0 4 0 0 0 4 5 0 0 5 0 0 0 n E (n=6) (n=23) ► C (n=10) B(n=11) Rarefaction of Site 744A by Isotope Intervals A-E Rarefaction of Sites by Oi-1 Boundary Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 14 Isotope Interval B through E following the richness drop at the onset of Oi-1 (Figure 3.12c). The species-area effect was also examined by a bootstrapping approach (Appendix 4), which produced results consistent with those from rarefaction. Note that the rarefaction method does not support extrapolation beyond sample population richness to estimate actual parent population richness (Tipper, 1979; Gotelli and Graves, 1996). Such extrapolations are generally based on “ taxon-effort” curves, which plot new taxa sampled relative to the exerted search “effort” (e.g., in entomological studies: time of searching, number of synchronized bush beatings, volume of vegetation gassed, etc.; v. Colwell and Coddington, 1994). This approach is inappropriate for paleoecological studies as sampling effort is a function of both the sampler and myriad, often unquantifiable taphonomic factors (e.g., transport, time- averaging, habitat-averaging, selective destruction, etc.). 3.4.3. Faunal distributions Faunal matrices of ostracode abundance distributions (i.e., left adult, right adult, left juvenile, right juvenile, fragment size class 1, fragment size class 2) at Sites 744A and 689B are presented in Appendix 2. Taxa in which adults and juveniles were difficult to differentiate due to low abundance, high size variation, etc., were tallied only as left and right “ adults” and are identified by their shaded appendix categories. Absolute (total number of individuals) and relative (% of sample) abundances of each taxon are plotted in Figure 3.13. From this mosaic distribution of taxa through time and between sites, four basic patterns may be defined: 1. Low-abundance. wide-distribution — Less abundant taxa tend to be more stratigraphically patchy in their distribution, likely reflecting the “species-area” effect (discussed below), and thereby preclude robust comments on their paleocologies. At both sites, Bosquetina, Cytheropteron spp., Propontocypris, Pelecocythere foramen, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115 Figure 3.13. Absolute and relative abundances of taxa at Site 744A and 689B. Absolute values in black; relative (% of sample) in gray. For each site, the upper-axis hash- marks indicates 10 individuals and lower-axis hash-marks indicate 10% of the sample. Note that absolute hash-mark distance vary between taxa, but are constant between sites for each taxon. The 8’8 0 , relative percent opal, and isotope-based time divisions are the same as in Figure 3.8. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116 £ . - . v . A i . w — J L l< ■ s . a: « o I rVT^UaJU..---.. i . . ► -CiiMjijLj o > 0 0 C O O (O o o to C O CO in o o in i n o o in o Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 117 CD f t , / G O OQ , o s O ) CO CO £ (0 ,0 , i n Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 1 8 Aversovalva sp. A, Aversovalva sp. B, and the indeterminate genera A, B, D, and Fare relatively rare, but widely distributed, with abundances generally greater at Site 744A. Less abundant, widely distributed taxa restricted to Site 744A consist of Bradleya dictyon, Henryhowella philofelicula, Pelecocythere sp., Bairdia, Aversovalva sp. B, Pennyella, Munseyella, and the indeterminate genera C, E, G, H, and I. These rarer taxa are generally restricted to isolated samples within the pre-Oi-1 interval, but this distribution may relate to the interval's higher average ostracode abundance (i.e., a sampling artifact). Although rare, the distribution pattern of four of these taxa may reflect biological and ecological conditions rather that purely sampling probability. The first is P. foramen, which is most common at Site 744A immediately following the 8’8 0 and paleoproductivity maxima of Oi-1a and Oi-1-b (Figure 3.14). The second is Bradleya dictyon, which is consistently present prior to 33.8 Ma and then largely restricted to the later portions of relative 5’8 0 maxima within Oi-1 (Figure 3.15). The third is Indeterminate Genus E, restricted to the post-Oi-1 interval and largely to relative 51 8 0 minima in Oi-1 (Figure 3.16). The fourth is Pennyella, which is most consistently present prior to -33.8 Ma and during relative 81 s O minima within Oi-1 (Figure 3.17). 2. High-abundance. wide-distribution — Krithe spp. and Poseidonamicus are abundant and widely distributed at both sites, but each show abundance changes through time. At Site 744A, Krithe spp. absolute abundance is roughly sinusoidal through the study interval, with relative maxima around 34.25 Ma and the top of the study interval. Krithe spp. relative abundance increases steadily in the early portion of the record and, with the exception of two major drops just prior to (-33.75 Ma) and coincident with onset of Oi-1, averages -30% for the remainder of the record Krithe spp. absolute and relative abundance at Site 689B varies within the same range, but at a much higher stratigraphic frequency. In addition, Krithe spp. is absent during a short interval prior Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 1 9 Figure 3.14. Pelecocythere foramen abundance versus total abundance and its relative abundance versus 81 8 0 . The taxon is rare overall, but tends to be most abundant after the 5,80 step at Site 744A (top) during 81 8 0 minima (bottom). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pelecocythere foramen APre-Oi-1 (744A) AP°St-0i-1 (744A) • Pre-Oi-1 (689B) O Post-Oi-1 (689B) ,_ _ _ A M A I I I I I — O O D T C D O CO C D em m m m m a m • • • 50 100 150 200 Total Abundance 0.12 r A Post-Oi-1 (744A) A Pre-Oi-1 (744A) * 0.10 j • Post-Oi-1 (689B) QPre-Oi-1 (689B) A I ® O c ( Q ■ o c 3 £ i < 0.08 0.06 ■i 0.04 o c C O « 0.02 0.00 OO o W ° A A a l A dA A _____ O O IB H D D D O O O * • • 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 6180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121 Figure 3.15. Bradleya dictyon abundance versus total abundance and its relative abundance versus 6,80 at Site 744A. Although the taxon is present before and after the 81 8 0 step increase (top), it tends to occur during 81 8 0 minima after the onset of Oi-1 (bottom ). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bradleya dictyon a > 0 1 3 3 X I < c 2 o * x to y- 1 0 ▲ A A Pre-0i-1 (744A) O Post-Oi-1 (744A) A O A A A O A O A A O A A O A O A O tO A A A A A A A O A A O A O A IA A A A A D A A A A A A O O 9 OOB O O U P O 0 0 O O A A A A A A A 0 50 100 150 200 Total Abundance 0.14 0.12 < D S 0.10 < Q * 0 C E 0.08 < © I 0.06 t o a ) tr c 0.04 o x £ K 0.02 0.00 A A A i W 4 * A Pre-Oi-1 (744A) O Post-Oi-1 (744A) ° O ° 0^ 0° I 0.75 1.00 1.25 1.50 A A a p o o t n g o a p 1.75 2.00 2.25 2.50 8180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123 Figure 3.16. Indeterminate Genus E abundance versus total abundance and its relative abundance versus 8,0O at Site 744A. The taxon is restricted to the post-Oi-1 interval, generally during 51 8 0 minima. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Indeterminate Genus E < Q 1 2 2 ■O < C o X < 0 O O ODOOO O O A Pre-Oi-1 (744A) O Post-Oi-1 (744A) O aODOaOBCDD 000 A M A A A A 4 50 0D A A 100 Total Abundance 150 c ( 0 •o c 2 > m ® D C c o X ( 0 I” 0.10 0.08 0.06 0.04 0.02 0.00 A Pre-Oi-1 (744A) O Post-Oi-1 (744A) 0.75 1.00 1.25 1.50 o 8 O o °o % 1.75 5180 O O A QD 2.00 2.25 2.50 200 I 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125 Figure 3.17. Pennyella abundance versus total abundance and its relative abundance versus 81 8 0 at Site 744A. Although the taxon is present before and after the 8’8 0 step increase (top), it tends to occur during 81 0 O minima after the onset of Oi-1 (bottom). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pennyella O 0 1 2 c 3 £ } < i 1 < 0 A Pre-Oi-1 (744A) O Post-Oi-1 (744A) O H A A i i A M A O A O A A 0 ( 0 ) 0 A JL A A t 0 * O O A A 000 O 00 O A A A M A A t A I A A M 50 100 150 Total Abundance 0.10 0.08 ® O t j 0.06 c 3 n < > 0.04 T o « C C § 0.02 0.00 A Pre-Oi-1 (744A) 0 Post-Oi-1 (744A) A A A IS- A A A o o Q > o A A 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 5180 200 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 127 to Oi-1 (-33.8-33.6 Ma), but total faunal abundance is also quite low. At Site 744A, Poseidonamicus shows four progressively less-pronounced abundance peaks prior to Oi- 1, a relative abundance maximum coincident with Oi-1 a, and increased abundance following Oi-1b. At Site 689B, Poseidonamicus shows two broad abundance peaks prior to Oi-1 and is relatively rare thereafter. Eucythere is less abundant than either Krithe spp. or Poseidonamicus, but is widely distributed through each site, with barren intervals corresponding with low total abundance at Site 689B. In contrast, Algulhasina predominates through most of Site 744A, but is absent altogether at Site 689B. This genus shows a steady decline in relative abundance throughout the interval from over 60% to less than 30%. 3. P re -O i-1 — Cytherella spp., Bairdoppilata, and Cytherelloidea are largely restricted to pre-Oi-1 intervals, with very rare occurrences in the Oligocene. Cytherella spp. is present at both sites until Oi-1 and then absent except for rare occurrences in the latest portion of each record (-32.8 Ma). Bairdoppilata has a low but consistent presence at Site 744A until the onset of Oi-1, after which it is only present immediately following Oi-1 a and at -32.8 Ma. At Site 689B, Bairdoppilata is occasionally present until -34.35 Ma. Although extremely rare, Cytherelloidea is also restricted to the pre-Oi-1 interval at Site 744, with its last spotty appearance at -33.7 Ma. Of the entire fauna, these taxa include the only members of the Order Platycopina and exhibit a fundamentally different life-mode from all other taxa of the Order Podocopida. Possible explanations for this pattern and its implication for paleoenvironmental conditions at both sites are presented in the discussion. Irregardless, the loss of these taxa is not attributable to simple sampling artifact of decreased total abundance in the post-Oi-1 interval: Cytherella spp. and Bairdoppilata are absent from nearly all post-Oi-1 samples, although many post-Oi-1 samples have Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128 total abundances equal to most pre-Oi-samples (Figure 3.18 and 3.19). In addition, their relative abundances show significant correlation with either dissolution index, supporting a biological, not preservational, control upon distribution (Figure 3.20 and 3 .2 1 ). 4. Reverse-response — This pattern is defined by taxa present within both sites, but with reversed or highly disparate distributions between sites. These taxa include Actinocythereis, Trachyleberis, Bradleya sp., Henryhowella asperima, and Dutoitella. Both Actinocythereis and Trachyleberis are completely restricted to the pre-Oi-1 interval at Site 689B and predominately restricted to the post-Oi-1 interval at Site 744, with sparse and rare occurrences prior to Oi-1. As with the “pre-Oi-1” faunal pattern discussed previously, these distributions are not easily attributed to sampling artifact (Figure 3.22 and 3.23). Bradleya sp. is present throughout Site 744A (except for the relative 51 8 0 minima within Oi-1), but does not appear until just prior to Oi-1 at Site 689B, after which it predominates the ostracode fauna (Figure 3.24). In contrast, H. asperima is common throughout Site 689B, but does not become a consistent and relatively common constituent of the Site 744A fauna until the onset of Oi-1 (Figure 3.25). Finally, Dutoitella is included in this group because it is a common Site 689B constituent until the onset of Oi-1, after which it only occurs only during the first relative 81 8 0 minimum, while its Site 744A distribution is rare and spotty overall. 3.4.4. Diversity analysis In all of ecology, the term “diversity” itself probably contains the greatest diversity of potential meanings beyond that of simple taxonomic richness sensu strictu. A plethora of denotations and connotations has grown from the general concept’s application in both taxonomy (i.e., number of taxa; e.g., “richness") and ecology (i.e., resource partitioning; e.g., “evenness”). In both fields, the term may be invoked at Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 129 Figure 3.18. Cytherella spp. abundance versus total abundance and its relative abundance versus 8'8 0 at Sites 744A and 689B. The taxon is common in all but the lowest abundance samples prior to Oi-1 (shaded symbols), then absent after (open symbols) with three exceptions in the latest portion of the study interval , in spite of only a slight decrease in the range of total abundances. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cytherella spp. 12 10 8 a > u c " O c 3 . a < c o x (0 • A • • • A A * A A • * 0 A A • A • A A A A • • A A A A A A 2 . % m % • M A A A A A A A A M M M • • A M A A A A A A 0 A / w v w w m A f l A -----------------A A A M A i K * . .. _ * * _ A Post-Oi-1 (744A) A Pre-Oi-1 (744A) t m i ) CD 0 O CD M M * • O Post-Oi-1 (689B) • Pre-Oi-1 (689B) 0 50 100 150 200 Total Abundance a ) o c a T J c 3 ■O < 0 C 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 A Post-Oi-1 (744A) A Pre-Oi-1 (744A) O Post-Oi-1 (689B) • Pre-Oi-1 (689B) A A A O O 0.75 1.00 1.25 1.50 1.75 5180 2.00 2.25 2.50 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 131 Figure 3.19. Bairdoppilata abundance versus total abundance and its relative abundance versus 51 8 0 at Site 744A. The taxon is common prior to Oi-1, then absent afterwards (with three exceptions in the latest portion of the study interval), in spite of only a slight decrease in the range of total abundances. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bairdoppilata 6 r 5 ! APre-Oi-1 (744A) OPost-Oi-1 (744A) 4 A A A © 0 1 ! 1 3 A A A I ! § 2 A O A A A A A A 1 OA M A JiA A A A O A o ffw n — oo o oo c d o o ^AA A A A A A A J f e A A A J I A A 0 50 100 150 Total Abundance 0.10 0.08 < u u c ■ S 0.06 I 3 s s © c c g 0.02 < 0 0.00 APre-Oi-1 (744A) OPost-Oi-1 (744A) A A QD < 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 5180 200 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 133 Figure 3.20. Abundance of key taxa with distinct stratigraphic distributions versus the dissolution index of % fragmentation of planktic forams for Site 689B. Note the general absence of significant correlation of both absolute or relative abundances to dissolution intensity. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Selected Site 689B Taxa vs Dissolution Intensity 30 r 25 a > “ 20 a T J c 3 -O < C o x < 0 I- + C y th e re lla A B a ird o p p ila ta □ A c tin o c y th e re is O T ra c h y le b e ris O B ra d le y a 15 10 | 50 _ O * t » Q 0 ^ n o o o 60 70 80 90 % Planktic Foram Fragmentation 100 ® O c m T 3 c 3 -O < c o 0 ) C C + C y th e re lla A B a ird o p p ila ta □ A c tin o c y th e re is O T ra c h y le b e n s O B ra d le y a O 0 * 0 9 70 80 % Planktic Foram Fragmentation 100 Correlation Coefficient (r) Taxon Absolute Abundance Relative Abundance Cytherella spp. 0.24 0.22 Bairdippilata 0.13 0.05 Actinocythereis 0 .3 6 0.21 Trachyeberis 0.20 0.16 Bradleya sp. 0.06 ■'^.7 ■ r(crit) = 0.29 at a =0.05 and d.f.=45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 135 Figure 3.21. Abundance of key taxa with distinct stratigraphic distributions versus the dissolution index of % benthic forams for Site 689B. Note that the absolute abundances of three of the taxa show a significant correlation with % benthic forams, while only the relativel abundance of Bradleya show a statistically significant (and positive) correlation with % benthic forams. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 136 Selected Site 689B Taxa vs Dissolution Intensity 30 r c 1 0 T J C 3 •O < C o + Cytherella A Bairdoppilata Actinocythereis o Trachyteberis O Bradleya 40 60 % Benthic Forams 100 c < 0 ■ o < c o 0 1 c c A Cytherella A Bairdoppilata □ Actinocythereis O Trachyteberis O Bradleya O O 40 60 % Benthic Forams 100 Correlation Coefficient (r) Taxon Absolute Abundance Relative Abundance Cytherella spp. 0 .3 6 0.03 Bairdippilata 0.23 0 .14 Actinocythereis 0 .3 7 0 .16 Trachyeberis 0.30 >m-- 0 .1 6 Bradleya sp. 0.10 r(crit) = 0.29 at a =0.05 and d.f.=45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137 Figure 3.22. Actinocythereis abundance versus total abundance and its relative abundance versus 8,aO at Sites 744A and 689B. The taxon is present predominately prior to Oi-1 at Site 689B and after Oi-1 at Site 744A (shaded symbols). There are no occurrences in post-Oi-1 samples at Site 689B and occurrences in only some of the most abundant pre-Oi-1 samples at Site 744A, where it constitutes less than 5% of the total fauna (bottom). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Actinocythereis APost-Oi-1 (744A) A P re-O i-l (744A) OPost-Oi-1 (689B) • Pre-Oi-1 (689B) A • c 4 < o H T3 I 3 c g 2 8 7 6 5 « • • A A • • O • • A O O • A A • A M A A A A A A m • Mk J A A A A A A A A A A A ▲ A A A m m - A A A A A A A A \ A A /W /W W S A A A A M M a A CC03D CD O OO CD L ____ 0 50 100 150 200 Total Abundance c C O T 3 c 3 A < C o 8 j 7 I i I 6 { | 5 ! i 4 i I 3 I 2 1 0 L 0.75 A • • i 'i f i T i f l — A • A/MWMV A A A • • A Post-Oi-1 (744A) APre-Oi-1 (744A) OPost-Oi-1 (689B) • Pre-Oi-1 (689B) A A A A O aaDCODO o o 1.00 1.25 1.50 1.75 2.00 2.25 2.50 5180 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 39 Figure 3.23. Trachyteberis abundance versus total abundance and its relative abundance versus 5,80 at Sites 744A and 689B. The taxon is common prior to Oi-1 at Site 689B and after Oi-1 at Site 744A (shaded symbols). There is only one occurrences in post- Oi-1 samples at Site 689B and occurrences in only some of the most abundant pre-Oi-1 samples at Site 744A, where it constitutes less than 5% of the total fauna (bottom). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Trachyteberis APost-Oi-1 (744A) APre-Oi-1 (744A) OPost-Oi-1 (689B) •Pre-Oi-1 (689B) A ftA /V*A A A 50 100 Total Abundance 150 200 ® o c < a T3 c 3 A < J 2 © < r c o 0.35 - 0.30 ! 0.25 j 0.20 * I o .i5 j 0.10 i I I 0.05 ! 0.00 -0.05 • • A Post-Oi-1 (744A) ^Pre-Oi-1 (744A) OPost-Oi-1 (689B) •Pre-O i-1 (689B) 0.75 1.00 AA A A O O B Q B O O O • • m • • • 1.25 1.50 1.75 2.00 2.25 2.50 8180 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141 Figure 3.24. Bradleya sp. abundance versus total abundance and its relative abundance versus 5,80 at Site 744A. After the onset of Oi-1, the taxon is less frequently present at Site 744, but first appears and largely predominates the Site 689B fauna, even though average sample size range does not change. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bradleya sp. 30 25 20 U c ^ A Post-Oi-1 (744A) APre-Oi-1 (744A) 0 0 Post-Oi-1 (689B) O Pre-Oi-1 (689B) I 15 • A < # A A A A . A A A o 1 J • “ * \ “a MA A a A 2 L “ A X &A COD O O O O 0 0 < 3 D 50 100 150 200 Total Abundance 1.00 r © o c < 0 * o c 3 A < f f i G C c o X < 0 0.80 0.60 0.40 0.20 ‘ 0.00 APost-Oi-1 (744A) APre-Oi-1 (744A) 0 Post-Oi-1 (689B) o Pre-Oi-1 (689B) . M i 0.75 < n > o < 1.00 ^Aa 1.25 O O 1.50 • • • * • A * A O 1.75 5180 • • 2.00 2.25 2.50 2.75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 143 Figure 3.25. Henryhowella asperima abundance versus total abundance and its relative abundance versus 8iaO at Sites 744A and 689B. The taxon is common throughout the study interval at Site 689B, but rarely present at Site 744A prior to Oi-1. After the onset of Oi-1, H. asperima becomes a major constituent of the Site 744A fauna, being more common in lower-abundance post-Oi-1 samples than in higher-abundance pre- Oi-1 samples at the site. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Henryhowella asperima 20 18 16 14 12 0 ) 0 1 10 § 8 I 6 * 4 2 A Post-Oi-1 (744A) APre-Oi-1 (744A) • Post-Oi-1 (689B) O Pre-Oi-1 (689B) O A • A O A O A A O A O A A A O M A A A A A A « L * A O J f t A A A O A Ok O A A CXMMJkO A A A A A A A crxm oaacA q a o a a a a a oik A — A A A /AA/ftJVTWJfW A £ \ & ◦ O O CD O O 0 50 100 150 200 Total Abundance 0.50 0.40 A Post-Oi-1 (744A) APre-Oi-1 (744A) ® Post-Oi-1 (689B) O Pre-Oi-1 (689B) is u c (0 ■ o c 3 ■O < a > c c 0.30 0.20 9 0.10 0.00 _ O < % o ° As o • ^ i 1* A A A • • • A M 0 S & AA A O C D dTD O O 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 5180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 45 three hierarchical levels (alpha, beta, gamma), each addressing different spatial scales and concepts. Although diversity indices are a powerful means to standardize and summarize large amounts of ecological information, care must be taken to select indices appropriate to the question at hand. For a single sample (i.e., community), alpha diversity indices exist for richness (i.e., Margalef’s index, Menhinik’s index, log-log index, Fisher’s alpha), evenness (i.e., Simpson’s index, Berger-Parker index, McIntosh index), and a combination of the two (i.e., Shannon index, Brillouin index). Each index’s appropriate use, as well as strengths and weaknesses are discussed at length by Margalef (1958), Magurran (1988), and Hayek and Buzas (1997). A problem common to all indices is that they are sampled from unknown distributions, which precludes assignment of probability values. Consequently, there is no basis for evaluating the statistical significance of diversity index differences between any two samples. Note also that these indices are only sensitive to net changes in relative proportions — between two samples, a complete faunal turnover yielding a similar proportional distribution would not produce a change in index values. To examine ostracode community structure during the Eocene-Oligocene transition, the abbreviated Shannon information function (H), its maximum value (Hma*)< and evenness (E) were calculated for each sample using s h = - £ Pi infc>;) (Shannon, 1948), i = 1 Hm ax = inS) (Shannon, 1948), eH e = — (Buzas and Gibson, 1969), s where S is richness and p, is the proportion of the /th species (Figure 3.26). The Shannon information function was selected to account for both the number of taxa and their relative proportions. For any given sample, the maximum value of H occurs when Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 146 Figure 3.26. Stratigraphic variations in faunal evenness (£), Shannon information function (H), and its maximum potential value for a given sample (Hm a x ). Indices indicate increased evenness during Oi-1 at Site 744A, but no comparable trends at Site 689B . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 147 co C O o o C O C O ; V v , I C M v \ O O Q o C M i n co O A M A M A M M M M Ift AAMA AAAA>A A A ,________________________ A A /fflW VW VM M M W M M M ftAiM BM PQ M M BV /W Y V V X yyX yV V ^R < X W Y Y V X W X Y Y ^M X yV Y W l(V V X \ A X X Y ft AAAAAA A A AAA AA A A AAA A i n co co i n o c o C O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 148 all taxa are equally distributed (i.e., evenness is maximum). Samples of different richness, but the same evenness, will have different H values (i.e., a taxonomically richer sample will be higher than a taxonomically poorer sample). Therefore, H values are plotted with their maximum potential value Hm a x to compare actual and potential H values — greater convergence between the two measures indicates greater evenness. For Site 744A, evenness increases from a mean of -0.45 to -0.75 between -33.9-33.8 Ma, oscillates about this new average until the end of Oi-1 b, followed by a slight decreasing trend for the remainder of the record. The H— Hm a x values also reflect this pattern through their convergence approaching Oi-1 a and subsequent divergence following Oi-1b. These two patterns both support increased evenness during the Oi-1 interval. For Site 689B, evenness shows greater variance, but no temporal trend. 3.4.5. Faunal Turnover Turnover, or change in the constituents of a community through time, is another metric of potential use for examining ecological response to environmental change. Turnover measures are generally calculated from data collected by one of two distinct sampling strategies: pairwise, where censuses taken at the beginning and end of an otherwise unsampled interval are compared, and cumulative, where all appearances and disappearances are tallied within an interval and adjacent intervals are compared. Pairwise turnover was calculated for each site using the following indices: A p p e a ra n c e s (A = * B ) + D isappearances (A => B ) titersam p e T u rn o v e r = -------------------- — --------------------------------------------------------------------------------------- , X C Taxa p re s e n t i i A and B ) and A pp earan ces (A => B ) + D isappearances [A => B ) T u rn o v e r ra te = -------------------------------------------------------------------------------------------------------- , X C Taxa p re s e n t in A and B ) x Census In te rv a l where A = older stratigraphic sample, andS = younger stratigraphic sample. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 149 For each site, the number of taxonomic appearances and disappearances from each older sample to the stratigraphically adjacent younger sample show no distinct trends through the study interval (Figure 3.27a). Unsurprisingly, the directly related intersample turnover also shows no distinct trend (Figure 3.27b). Calculation of turnover as a rate, by inclusion in the denominator of the time interval between two censuses, shows a pronounced increase at, and oscillations within, the Oi-1 interval at Site 744A (Figure 3.27c). However, this pattern is an artifact of the dense sampling in these intervals, together with the high probability of rarer taxa not occurring in every adjacent sample in these low abundance samples. 3.4.6. Cluster analyses — stratigraphic and taxonomic Similarity coefficients and clustering algorithms were used to examine correlations between each stratigraphic sample pair (i.e., taxa as variables) at each site. To minimize effects due to sample size, rank abundance, etc., variable values were transformed into presence/absence form (i.e., 1/0) prior to similarity coefficient calculation (Cheetham and Hazel, 1969). Taxa occurring in a single sample and stratigraphic samples containing no individuals were culled from presence/absence matrices. Aimilarity matrices were calculated using the simple matching equation: Simple matching: sS M = —a + c— a + b + c where: a = number of variable pairs with common occurrence (i.e., positive match), b = number of variable pairs with only one occurrence, c = number of variable pairs with no occurrence (i.e., negative match). The simple matching coefficient includes negative matches between two compared samples in both the numerator and denominator and is effective for examining stratigraphic (community) structure through time (Kaesler, 1966; Maddocks, 1966; Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 50 Figure 3.27. Intersample ostracode appearances and disappearances, turnover, and turnover rates for Sites 744A and 689B. A. Values indicate the number of taxa that appear or disappear from the previous stratigraphic interval. B. Values represent the number of appearances and disappearances from the previous sample divided by the total number of taxa in the previous and current sample. C. Values represent the number of appearances and disappearances from the previous sample divided by the total number of taxa in the previous and current sample times the census interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 151 O O Q AA AA A AAA A A A A A A A A A A A AAA A A A A AAA A /flim yiC flflM infll' I’l I’lMBTIMTi /WYyvxYYW WMWW YWxy^^ I U Z I O N e i o y e i d N S IO 9 U 9 3 0 D H 0 Ajjeq 9 U 9 0O3 9 )9 H Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 0 5 1 0 0 5 1 0 1 5 0 2 4 6 8 0 0 .5 1 . 0 1 . 5 2.0 % o Intei sample Disappearances Intersample Turnover Tumover/10 kyr 152 Mello and Buzas, 1968). Similarity matrices calculated from presence/absence matrices for Site 744A and 689B stratigraphy and fauna are presented in Appendix 5; all matrix calculations were conducted through Systat 8.0 (PC-version). Clustering algorithms allow graphic examination of similarity matrices through hierarchical grouping of more similar variables into dendrograms. The method is largely exploratory, lacking the probabilistic underpinnings of approaches such as discriminant analysis, where the number and members of groups are known a priori. Often a similarity level is chosen above which clusters are considered “significant” (e.g., “> 0.5”), but these levels are arbitrary. Stratigraphic and faunal simple-matching similarity matrices for Sites 744A and 689B were clustered using complete-linkage algorithms through Systat 8.0 (PC- version). Complete-linkage iteratively agglomerated individual samples into larger, more inclusive groups in order of their most distant constituent sample's similarity values (i.e., “ furthest neighbor") until all are contained within a single group. For Site 744A, most Isotope Inteval A stratigraphic samples clustered together in a middle “Pre- Oi-1” cluster, except for a few exceptions (A44, A45, A47, A52) nested within the “lntra-Oi-1” cluster (Figures 3.28). The third cluster (“Post-Oi-1") consists of Isotope Interval D and E samples. This dendrogram topology indicates that sufficient faunal cohesiveness exists within sample populations before and after Oi-1 to agglomerate them as two distinct clusters. Furthermore, faunas return somwhat to pre- Oi-1 compositions in the post- Oi-1 interval , with Isotope Intervals D and E samples (“Post-Oi-1") clustering proximal to those of Isotope Interval A (“Pre-Oi-1"). For Site 689B, complete-linkage of simple matching coefficients for stratigraphic samples produced a “Pre-Oi-1” (Isotope Interval A) cluster and a “Pre- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 153 Figure 3.28. Complete-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 744A. Sample numbering consists of Isotope Interval letter followed by relative stratigraphic position from base of given Isotope Interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P o s t - O i - 1 P r e - O i - 1 O i - 1 154 I Site 744A SM Coefficient Complete Linkage 0 c D ; D D ; D j d ; 0.0 0.2 0.4 0.6 0.8 1.0 Distance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 155 Oi-1 to Post-Oi-1” cluster (Figure 3.29). No stratigraphic ordering is apparent within or among these sub-clusters. Although pre- and post-Oi-1 sample populations cluster separately, there is no indication that later faunal compositions became more similar to pre-Oi-1 compositions. 3 .4.7. Principal components analysis — stratigraphic and taxonomic Principle components analysis (PCA) is a multivariate statistical technique for reducing a matrix of potentially correlated variables into an uncorrelated (i.e., orthogonal in multi-dimensional space) set of indices termed principal components (PCs). For a given matrix, the first PC is oriented in the dimension of maximum multivariate variance, the second PC is oriented in the dimension of maximum remaining multivariate variance orthogonal to the first, and so on. The relative amount of total variance accounted for by each PC is represented by its eigenvalue. The contribution of each variable to each PC is represented by its eigenvector coefficient. The linear position of each case along each PC axis is represented by its factor score. Standard PCA produces a mathematically unique solution based on maximized orthogonal variance rankings. However, this solution is but one of an infinite number that can describe the original data with equal precision, but not parsimony with respect to maximum orthogonal variance ranking. To facilitate PC interpretation as a composite variable for multiple variables, PC indices are often orthogonally rotated post-hoc to align more closely with clusterings or arrays of eigenvector coefficients. PCA parameters are then recalculated, maximizing specific eigenvector coefficients on specific PCs and improving PCA explanatory power, but no longer from a perspective of maximum orthogonal variance ranking. Numerous rotation methods exist, with Kaiser’s (1958) varimax method being the most common. Manly (1994) provides additional discussion of principle components analysis and rotation methods. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 156 Figure 3.29. Complete-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 689B. Sample numbering consists of Isotope Interval letter followed by relative stratigraphic position from base of given Isotope Interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P r e - O i - 1 P r e - O i - 1 ► O i - 1 157 B23----- B22----- B5— I B6— f B14----- B18----- B8----- B7----- B11— L B21— f B9----- B2----- Site 689B S M C o e f f i c i e n t C o m p l e t e L i n k a g e B4 B19 B17 BIO B12 B20 B3 I-------------------1 -------------------1 ------------------1 -------------------1 ------------------ 1 0.0 0.2 0.4 0.6 0.8 1.0 Distance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 158 Q-mode PCA was performed on absolute, log-transformed, and standardized (%) matrices for Site 744A and 689B, using stratigraphic samples as cases and taxa as variables; all calculations were performed through Systat 8.0 (PC-version). Log- transformation used the natural log of each variable value (plus one) to minimize variance due to extremely abundant taxa (e.g., Algulhasina at Site 744A). Standardization converted each variable’s value into its percent of a given sample (i.e., 5 Xus yus in a sample containing 20 specimens = 0.25). Samples containing no specimens, and taxa occurring in only one sample, were not included. Each matrix was analyzed using non-rotation and varimax rotation PCA. The varimax rotation of the log- transformed matrices are discussed in detail below as they typify the general structure. Eigenvalue structure of all six PCA analyses of Site 744A are presented in Figure 3.30. Absolute and log-transformed eigenvalues have a generally similar structure, with non-rotated eigenvalues more right-skewed than varimax-rotated eigenvalues. PC1 for these four analyses explains 14-22% of the total variance, the second 10- 14%, the third 7-9%, and so on, although the particular loading structure accounting for this variance differs between analyses. Eigenvector coefficients for each taxon in the Site 744A log-transformed matrix are presented in Figure 3.31. The non-rotated and varimax-rotated loading structures are generally congruent in PC-1 to PC-4, with decreasing congruence in PC-5 to PC-10. PC-1 is highly loaded by taxa that are present throughout the study interval (e.g., Bairdia, Pennyella, Krithe spp., Bosquetina) and those that are generally less common (e.g., Algulhasina, Bradleya sp.) to exceptionally rare (e.g., Cytherella spp., Bradleya dictyon, Bairdoppilata) after Oi-1. This loading structure is manifest in PC-1 factor scores by an initially high mean score, followed by a general decrease prior to Oi-1, after which scores increase only after Oi-1b (Figure 3.32). PC-2 is also highly loaded by taxa generally present through most of the study interval, but less common (e.g., Poseidonamicus, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 159 Figure 3.30. Q-mode PCA eigenvalues for Site 744A faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. S ite 7 4 4 A — P C A E ig e n va lu e s — Q -M o d e Principal Absolute Abundances Log-Transformed Abundances Standardized (%) Abundances Component N o Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Axis EioenvaHje % Variance Eiaenvalue % Variance %2I Eiaenvalue % Variance % £ Eiaenvalue % Variance % I Eiaenvalue % Variance Eiaenvalue % Variance 1 6 18 18.71 18.71 481 14 56 1456 6.98 21 15 21.15 5 79 1754 1754 4 79 14.51 14.51 232 703 7.03 2 4.31 13.05 31.76 4.10 1242 26.98 3.92 11.89 33 04 3.37 1022 27.76 2 62 794 22.45 2.07 628 1331 3 2.63 7.96 39.72 2.98 9.02 35 99 2.32 7.02 40.06 262 794 35.70 2 26 6.85 29.30 2.57 7.79 21.10 4 1 96 595 45.67 206 6 23 4222 1 77 5.37 45 43 1.80 545 41.15 1 94 589 35.18 264 7.99 2908 5 1 64 496 50 63 1 82 551 47 73 1 51 4 57 50 01 1.38 4.18 45 33 1 76 535 40 53 2 43 737 36 45 6 1 43 4 3 5 54.97 1 54 4 66 5239 1 45 4 39 54 40 142 431 4964 1 65 4 99 45 52 1.55 4 70 41 15 7 1 37 4 16 59 13 1 48 448 56 86 1 33 4 03 58.43 1 58 478 54 41 1 43 434 4986 1.60 4 85 46 00 8 1 24 3.75 62.88 1.43 432 61.16 1 22 370 62.12 1 68 509 59 50 1.35 4 09 53 94 1 34 4 07 50.07 9 1.09 331 6619 1 47 4 46 65 64 1.12 338 6551 1 42 4 29 63 79 1 31 398 57 92 1 31 396 54 03 10 1 03 313 69 32 121 368 69 32 1 06 321 68 72 162 492 68 71 1 17 354 61 46 1 23 3 74 57 76 11 — — — — — — — — — — — — 111 3 37 64 83 1 21 368 61 44 12 — — — — — — — — — — — — 1 02 3 10 67 93 1 41 427 6572 13 — — — — — — — — — — — — 1.01 3 05 70 98 1.74 526 70.98 □ Absolute Abundances - No Rotation ■ Absolute Abundances - Varimax Rotation ■ Log-Translormed Abundances ■ No Rotation ■ Log-Transformed Abundances - Varimax Rotation □ Standardized (%) Abundances - No Rotation □ Standardized (%) Abundances - Varimax Rotation i i n i M l M l M l M l I 6 7 B Principal Component Axis 10 1 1 12 13 160 161 Figure 3.31. Q-mode eigenvalue coefficients (“loadings") of taxa for non- and varimax-rotated PCA of Site 744A log-transformed abundances. Taxa are sorted and grouped in a step-wise fashion in order of decreasing loadings and increasing PC. Eigenvalues in dark gray cells are 0.75-1.00 and in light gray are 0.50-0.74. Note that little loading structure exists in PC1-PC4, which accounts for most of the total variance. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 744A — Eigenvector Coefficients — Q-Mode Log-Transformed Abundances — No Rotation Log-Transformed Abundances — Varimax Rotation Taxon 1 : 2 3 4 5 6 7 8 9 1 0 1 2 3 4 i 5 6 7 ! 8 j 9 1 0 A lg u lh a s in a B r a d le y a s p . B a ir d i C y th e r e lla s p p . P e n n y e lla K r ith e s p p . B o s q u e tin a B . d lc ty o n B a lr d o p p ila ta P . fo r a m e n In d e l G e n . B T r a c h y le b e r is P o s e id o n a m ic u s P e le c o c y th e r e s p . H . a s p e r im a C v th e r o p te r o n s p p . A v e r s o v a lv a s p . A In d e t . G e n . G In d e l. G e n . C In d e t. G e n . H M u n s e y e lla In d e t . G e n . I A v e r s o v a lv a s p . B H . p h ilo le lic u la In d e t. G e n . A In d e t . G e n . E In d e t. G e n . F D u to ite lla E u c y th e r e C y th e r e llo ld e a A r g illo e c ia In d e t . G e n . D A c tin o c y th e r e is -0.05 0.03 0 . 10! -0.14 0.04 0.07' -0.28 - 0.10 -0.42 0.06 -0.05 -0.06: -0.31 -0.15 0.04 0.24 i 0.13 0.33 - 0.10 -0.05 0.05 -0.06 0.28 -0.29 -0.05 -0.08; -0.03 -0.08 0.06 -0.23 0.73I -0.18 - 0.12 -0.16 0.06 0.111 0.29 -0.14 -0.27 0.04 0.07 -0.04 - 0.01 - 0.10 -0.16 0.45 0.04 -0.05 0.10 0.06 0 09 0.13 -0.03 0.04 0.09 0.02 0.06 0.02 0.70! -0.08 -0.09 0.36 0.19> 0 .3 7 0.05 0.20 -0.03 0.11; 0.05 0.06 -0.19 -0.15 0.08 0.08 -0.09 - 0.01 0.07 -0.06 -0.09 0.02 -0.09 0.25 0.20 •0.50 0.22 0.29 0.18 0.37 -0.36 -0.14 B = ^ 0.75 Eigenvector Coefficient = 0.50-0.74 Eigenvector Coefficient 162 163 Figure 3.32. Varimax-rotated Q-mode PCA factor score plots for Site 744A log- transformed abundances. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S it e 7 4 4 A — P C A — L - T A bundance — Varim ax Rotation 164 o o in c o t - C M C O o o o in C O o in C O in cm M - c o o o co in co co S- « P o in co co in c v i co co o o co co C M co i n C M 0joos jo p ej Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 165 Pelecocythere sp., Cytheropteron spp.) to rare (e.g., Pelecocythere foramen, Indeterminate Genus B, Trachyleberis, H. asperima) prior to the Oi-1 event. These distributional patterns are reflected in factor scores consistently lower than PC-1 for the pre-Oi-1 interval, after which scores display greater variance and relative maxima following Oi-1 a and Oi-1b, increasing strongly in the latest post-Oi-1 samples. The remaining PCs explain progressively less of the total variance, contain little structure in their factor score plots, and generally consist of rare, but widespread, taxa. Eigenvalue structure of the six PCA analyses of Site 689B are presented in Figure 3.33. Eigenvalue structure is similar to that of Site 744A, but with fewer PCs accounting for greater variance. Absolute and log-transformed PCA analyses produced PC-1s explaining between 16-29% of the total variance, PC-2s 13-15%, and PC-3s 8-13%. For the standardized matrices, the variance percentage explained decreases only slightly in subsequent PCs, again reflecting the obscuring of faunal structure by analysis of matrix values based on dividing a given sample’s individual taxon abundance by its total abundance, when little correlation exists between the two values. Eigenvector coefficients for each taxon in the Site 689B log-transformed matrix are presented in Figure 3.34. The non-rotated and varimax-rotated loading structures are generally congruent for PC-1, PC-2, and PC-3, with little agreement in PC-4, PC- 5, and PC-6. PC-1 is highly loaded by taxa predominately restricted to the pre-Oi-1 event (e.g., Actinocythereis, Cytherella spp., Trachyleberis, Dutoitella, Cytheropteron spp.), with lower loadings for the widespread taxa Bosquetina and Krithe spp. Bradleya sp. is also highly loaded within PC-1, but with a negative coeffecient consistent with its predominate presence within the post-Oi-1 fauna. The stratigraphic factor score plot for PC-1 contains three relative maxima at 36.0, 35.0, and 34.25 Ma, followed by minimum values at the Oi-1 event (Figure 3.35). Subsequent PCs are generally loaded with rare taxa, and are therefore of low interpretative value. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 166 Figure 3.33. Q-mode PCA eigenvalues for Site 689B faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 689B — PCA Eigenvalues — Q-Mode Principal Absolute Abundances Loa-Transform »d Abundances Standardized (%) Abundances Component No Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Axis Eiaenvalue % Variance Eioenvalue % Variance Eiaenvalue % Variance Eioenvalue % Variance % I Eioenvalue % Variance Eioenvalue % Variance 1 5.70 27.15 27.15 3.54 16.88 16.88 6.14 29.25 29.25 4.60 21.88 21.88 2.51 11.96 11.96 2.05 9.78 9.78 2 3.00 14.27 41.42 3.04 14.48 31.36 2.95 14.04 43.29 2.79 13.29 35.18 2.31 10.97 22 93 1.31 6.22 16.00 3 1.98 9.44 50.86 2.71 12.92 44.28 1.79 8.53 51.82 1.97 9.38 44.55 1.80 8.55 31.48 1.62 7.71 23.71 4 1.63 7.74 5B.59 2.01 9.56 53.84 1.65 7.84 59.65 2.03 9.69 54.24 1.57 7.46 38.95 1.60 7.62 31.33 5 1.48 7.03 65.63 1.95 9.30 63.14 1.28 6.09 65.74 1.39 6.61 60.65 1.39 6.63 45.58 1.60 7.61 38.94 6 1.24 5.90 71.52 1.61 7.66 70.80 1.18 5.61 71.35 2.21 10.50 71.35 1.31 6.25 51 83 1 42 6.78 45.72 7 1.03 4.89 76.41 1.18 5.61 76.41 — — — — — 1.24 5.89 57.73 1.68 7.99 53.71 e — — — — — — — — — — — — 1.17 5.58 63.30 1.28 6 12 59.82 9 — — — — — — — — — — — — 1.10 5.25 68.55 1.81 8.62 68.44 10 — — — — — — — — — — — — 1 06 5 06 73.61 1.09 5.17 73 61 30 25 * 2 0 Q . X m □ Absolute Abundances • No Rotation ■ Absolute Abundances • Varimax Rotation B Log-Transformed Abundances - No Rotation 0 Log-Transformed Abundances - Varimax Rotation B Standardized (%) Abundances - No Rotation ■ Standardized (%) Abundances • Varimax Rotation f 15 < 0 > o io o 6 7 Principal Components Axis 1 2 167 168 Figure 3.34. Q-mode eigenvalue coefficients (“loadings”) of taxa for non- and varimax-rotated PCA of Site 689B log-transformed abundances. Taxa are sorted and grouped in a step-wise fashion in order of decreasing loadings and increasing PC. Eigenvalues in dark gray cells are 0.75-1.00 and in light gray are 0.50-0.74. Note that little loading structure exists in PC1-PC4, which accounts for most of the total variance. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 689B — Eigenvector Coefficients — Q-Mode Log-Transformed Abundances — No Rotation Log-Transformed Abundances — Varimax Rotation Taxon Actinocythereis Cytherella spp. Trachyleberis sp. A Poseidonamicus Dutoitella Bradleya sp. Bosquetina Cytheropteron spp. Krithe spp. Argilloecia Indet. Gen. A Indet. Gen. D Aversovalva sp. B Indet. Gen. B Bairdoppilata Indet. Gen. F H. asperima Aversovalva sp. A P. foramen Eucvthere 2 3 4 5 6 | -0.40! 0.12 -0.02 0.01! -0.02 j -0-34. 0.04 0.28 -0.13 -0.04 -0.30 -0.06 0.03 0.12 -0.02 | -0.07 0.11 0.00 0.07 0.11 1 -0.21 -0.09 -0.01 0.02 -0.28 i 0.671 0.08 -0.04 -0.20 0.23 0.00 -0.45 0.28 -0.24! -0.45 I 0.01 -0.24 -0.32 -0.28 0.05 I 0.31 -0.03 0.01 0.02 0.07 26 35 0.53 0«73 0.68 ti.57 0.29 -0.04 -0.08, 0.17| 0.07! 0.05 -0.09; -0.46 0.20 -0.051 -0.05 -0.55 43 33 06 -0.36! 0.20 0.47 36 42 -0.07 - 0.261 J3.47 0.59 -0.44 - 0.20 -0.38 0.12 11 05 0.12 -0.16 0.30 0.01 0.71 -0.05 0.34 0.47 0.10! 0.45 -0.22! 0.39 -0.01 0.19 -0.01 0.29! 0.54 0.08 0.36 0.42 0.19 -0.05 0.39 0.10 0.39 0.64] -0.16 0 .0 2 [1 0.65: -0.21 0.13 0.47 -0.19 -0.19 0.20 0.23 0.07 0.03 0.28 0.30 0.22 0.16 1 0.66 0.71 -0.70 0;61 0.49 0.45 0 * 2 0.06 -0.10 0.06 0.01 0.15 0.02 -0.03 0.33 0.13 0.11 0.11 -0.04 0.28 0.15 0.14 0.13 0.34 0.16 0.36 0.08 0.01 -0.04 0.06 -0.22 0.26 -0.34 -0.42 0.01 0.36 -0.32 0.06 0.21 0.01 0.44 -0.21 0.06 0.11 0.15 -0.05 0.16 0.45 -0.05 0.10 0.03 0.01 0.04 -0.11 0.00 6:57! 0.39 -0.01 -0 .03 1 -0.05 0.04! -0.18 -0.20 0.38 -0.16 -0.15 0.56 0.34 0.14 0.53 = > 0.75 Eigenvector Coefficient = 0.50-0.74 Eigenvector Coefficient 169 170 Figure 3.35. Varimax-rotated Q-mode PCA factor score plots for Site 689B log- transformed abundances. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 689B — PCA — L-T Abundance — Varimax Rotation 32.50 32.75 33.00 33.25 33.50 33.75 34.00 34.25 34.50 34.75 35.00 35.25 35.50 35.75 36.00 36.25 7 6 O i - 1 S t e p Increase PC1 PC2 PC3 5 4 3 2 1 0 1 2 8> o o C O o ■s (0 ■Vi Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 689B — PCA — L-T Abundance — Varimax Rotation 32.50 32.75 33.00 33.25 33.50 33.75 34.00 34.25 34.50 34.75 35.00 35.25 35.50 35.75 36.00 36.25 £ o o (0 h- o ts to -1 -2 C B Oi-1 Step Increase -3 173 Each Site 744A and 689B matrix was also subjected to non- and varimax- rotation R-mode PCA analyses, using taxa as cases and stratigraphic samples as variables. In all Site 744A analyses, the first two PC axes accounted for the majority of variance (>70%; Figure 3.36), while one or both of their associated eigenvector coefficients for stratigraphic samples generally exceeded 0.5 (Figures 3.37). As in the cluster analyses of Site 744A, stratigraphic variations in the eigenvector coefficients indicate a general shift in faunal composition during the Oi-1 event: In the varimax- rotation, absolute-abundance analysis, 81% of all pre-Oi-1 (Isotope Interval A) samples have their highest loading on PC-1, whereas 79% of all post-Oi-1 (Isotope Intervals B-E) samples have their highest loading on PC-2. In the varimax-rotation, log-transformed analyses, these values are 100% and 100% respectively. In the varimax-rotation, standardized analyses, these values are 90% and 83%, respectively. Application of R-mode PCA to each Site 689B matrix produced a less skewed and more cohesive set of eigenvalues relative to Site 744A (Figure 3.38). Thus, a larger proportion of stratigraphic samples have higher eigenvalues in the third or higher PC as shown by shaded distributions in Figure 3.39. Considering that the first two PCs account for 53-63% of the total variance, there is a repeat pattern of that at Site 744A: In the varimax-rotation, absolute-abundance analysis, 97% off all pre-Oi-1 (Isotope Interval A) samples have their highest loading on PC-1, whereas 93% of all post-Oi-1 (Isotope Interval B) samples have their highest loading on PC-2. In the varimax- rotation, log-transformed analyses, these values are 91% and 93% respectively. Finally, in the varimax-rotation, standardized analysis, these values are 96% and 98%, respectively. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 174 Figure 3.36. R-mode PCA eigenvalues for Site 744A faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 744A — PCA Eigenvalues — R -M od e Principal Absolute Abundances Loo-Transformed Abundances Standardized (%) Abundances Component N d Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Axis Eiaenvalue % Variance % I Eiaenvalue % Variance Eiaenvalue % Variance Eiaenvalue % Variance Eiaenvalue % Variance % I Eiaenvalue % Variance % E 1 7588 7588 75.88 44.88 44 88 44.88 65.69 65.69 65 69 40.77 40.77 40.77 74.47 74 47 74.47 45 05 45.05 45.05 2 11.66 11.66 87.55 41.76 41.76 86.64 8 78 8.78 74.46 29 38 29.38 70.15 12.47 12.47 8694 41.15 41.15 86.20 3 3.74 3.74 91 28 4.19 4.19 90 63 336 336 77.82 4.77 4 77 74.91 392 392 90 86 4.09 4 0 9 90 29 4 1.86 1.86 9314 2.07 2.07 92.90 2 99 2 99 60.80 3.37 3 37 78.28 226 2 26 9312 260 2 60 92.69 5 1 63 1.63 94 77 1.87 1 87 94 77 2 49 249 83 29 3 07 307 81.35 1.77 1.77 94 89 1.99 1 99 94 89 6 — — — — — — 1 60 1 80 85.09 2.17 2.17 83 52 — _ _ _ _ _ 7 — — — — — — 1.72 1.72 86.81 2 17 2 17 85.69 — _ _ _ _ _ 8 — — — — — — 1 45 145 88 26 1.42 142 87.11 — — _ _ _ _ 9 — — — — — — 1 25 1 25 89.51 1 30 1 30 88.41 _ _ _ _ _ _ 10 — — — — — — 1 08 1 08 90 60 1 31 1 31 8972 _ — — _ _ _ 11 — — — — — — 1 05 1 05 91 64 1.72 1 72 91.44 _ _ _ _ _ _ 12 — — — — — — 1 03 1 03 92 68 1.24 1.24 92 68 - - - - - - 100 (0 70 □ Absolute Abundances - No Rotation ■ Absolute Abundances - Varimax Rotation ■ Log-Trans1ormed Abundances - No Rotation ■ Log-Transformed Abundances - Varimax Rotation □ Standardized (%) Abundances - No Rotation □ Standardized (%) Abundances - Varimax Rotation I — II 6 7 Principal Component Axis 10 11 12 175 176 Figure 3.37. Site 744A R-mode eigenvalue coefficients (“loadings”) of taxa for varimax-rotated PCA of absolute abundance, log-transformed, and relative %) abundance matrices. Samples are presented in stratigraphic order, with eigenvalues in dark gray cells between 0.75-1.00 and in light gray between 0.50-0.74. Note that the majority of varimax-rotated loadings in pre-Oi-1 samples (Isotope Interval A) have their maximum loading on PC1, while majority of post-Oi-1 samples (Isotope Intervals B-E) have their maximum loading on PC2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 177 Ab«oM « Atmndanc— Sit> 7444 — V tn /n n EiQ»w k 1 w C o » fto « w » — R-Mod» L o aT n n D o m M im m cii PfWrnm tetfttU raw 3 4 5 - 1 1 - 04 04 06 • 0 3 05 06 26 0 9 04 - 02 11 - 03 - 04 06 OS 14 • 07 01 21 01 22 09 OS 06 07 12 06 • 02 06 04 • 02 19 . 02 • 0 4 • 07 M i 40 • 35 04 • 0 7 - 04 07 • 01 10 26 2 0 • 09 44 • 03 07 04 -.0 7 - 27 06 39 • 06 32 0 9 • 37 .0 9 • 01 • 07 14 0 0 - 33 03 3 7 \ 24 • 02 10 • 0 6 ' • 26 12 13 ] 45 34 10 05 ^ 02' 12 "_20_] - OS 10 ^ H • 0 6 * 08 r 09 - 03 11J 26 61 17 ’ ’ O .0 6 '2 0 2 l “ 07 J I 2 3 ‘ 03 j To " 04 • 1 5 1 07 07 H I 26 | - 0 3 f - 03 IS 16 * 20 33 h i 09 3 2 * • 02 17 26 ] 05 39 26 ‘ 0 0 * . 00 02 14 * 2 9 16 24 3 9 ! 09 16 15 06 41 40 1 3 0 1 3 “ • 04 10 • 01 * 06 19 • 0 4 0 9 “ 02 00 03 * • 01 21 20 ’ 0 6 ] 03 07 • 0 9 * ■ 04 35 33 23 .3 3 • 17 * 1 7 35 2 0 . • 0 0 * . .0 4 29 03 * 02 42 32 . 16 *• 12 06 •. 12 “ ■ ' 06 23 12 ' • 0 4 *• 14 10 0 2 , • 02 46 32 6 9 ^ 45 • 35 46 32 * 24 39 0 4 * • 06 *•_05 23 • 1 6 * ■ 02 40 00 ' 00 * 11 IS 04 *~ 15 40 2 4 24 *■ 09 16 24 * 22 33 4 4 - 0 3 * 1 1 20 0 0 * 33 43 01 * • 0 7 *'- 20 27 • t a ‘ • 0 6 * » 0 4 * 45 *• 05 Kiaerni E 3 r a n r a 14 - 2 2 00 ‘ .'2 9 ' - 4 S 4 15 • 02 0 0 ■ 0 3 , 02 • 't 0 * 07 ’ 0 4 04 04 I I - 0 9 0 4 0 9 I 0 5 - 0 2 16 I - 0 2 - 0 6 IS I 32 03 01 0.1 3 9 15 0 0 t 03 l ’ • 04 * • 0 6 1 . f t 39 05 09 OS I 2 9 06 - 04 02 1 27 ' 2 5 - 0 3 * 01 f 0 0 • 03 04 0 2 02 .02 31 0 3 02 01 JU lL o r. C l 26 ^ 17 00 ■ 41 - ° 1 . - 03 01 14 1 3 5 11 21 13 ■ H S U • 0 2 * - 06 14 06 | 45 03 ’ 04 07 ■ 3 D 00 * 01 07 06 37 19 * ■ 03 ■ 02 .7 * a t - 06 O S 00 11 26 10 [ 14 04 .n - s M i 0 2 ’ • 03 11 0 6 32 13 * 10 - 1 1 ■ 66 ^ 0 0 ’ - 03 07 0 3 41 04 ' 14 10 ■ 37 * 04 ’ • 02 0 4 - 0 4 27 19 ’ 07 09 ■ 39 * 03 ] 09 • 04 03~ .20 07 * 19 09 ■ .3 0 * 0 4 ] 00 04 - 0 2 45 22 * 14 09 7 » n r 00 * - 01 - 0 2 • 0 6 44 17 ’ 20 03 ■ 27 # 0 5 * • 02 - 0 9 • 0 9 22 25 * 10 07 ■ 10 * 06 05 - 06 01 31 30 ' 07 11 ■ -66 " • 02 0 4 ' 01 0 2 17 34 * 07 11 ■ .6 0 *‘ 0 6 . 04 0 3 0 5 20 ' 39 * 14 ■ 13 ■ 44 ' 00 . o e ' 06' 0 2 4 7 * 29 03 ■ 33 ’ 0 4 ‘ • 0 l ’ 02 0 3 40 E S I. 06 * 27 ■ 3 4 * 02 * 06 02 • 01 ‘ 27 • 04 • 01 ]m 41 * 05 % O S 0 0 0 2 14 A H • 0 9 _ 02 1 45 * 01 * 0 9 03 '• 01 r i* I 12. .1 0 ] ’• 01 ■ 1 0 “ 00 * 02 00 .01 □ l j - 01 . 10 * • 04 ■ 2 2 ’ 05 * 01 02 0 5 1 35 00 33 ’ .14 ■ 17 [ 05 01 05 - 02 | 40 11 ‘ 6 ’ 06 ■ 2 6 [ 02 05 - 01 - 0 3 09 * • 15 * - 0 4 ■ 10 ’ 0 0 . 16 - 03 • 0 5 12 ’ IS * 21 ■ 06 [ 17 01 ^ - 0 5 • 0 3 ' 2 3 n r 14 * 07 ■ 0 6 * 07 ‘ 04 *~• 0 3 • > 0 75 £>g«nv«ctor C o *f1 « w it ■ 0 50-0 74 £«g*nv«etor C o*H < i« oi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 178 Figure 3.38. R-mode PCA eigenvalues for Site 689B faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 689 B — P C A Eigenvalues — R-Mode Principal Absolute Abundances Loo-Transformed Abundances Standardized (%) Abundances Component No Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Axis Eioenvalue % Variance Eioenvalue % Variance Eioenvalue % Variance % I Eioenvalue % Variance Eioenvalue % Variance %L Eioenvalue % Variance % I 1 26.90 41.38 41.38 21.06 32.40 32.40 26.99 41.53 41.53 23.69 36.44 36.44 26.93 41.43 41.43 21.14 32.52 32.52 2 14.44 22.22 63.59 13.94 21.45 53.85 13.44 20.68 62.21 12.73 19.58 56.03 14.44 22.22 63.64 13.94 21.44 53.97 3 5.08 7.82 71.41 6.79 10.44 64.29 4.65 7.16 69.37 5.14 7.91 63.94 5.07 7.81 71.45 6.77 10.41 64.38 4 4.68 7.20 78.61 5.27 8.11 72.40 3.71 5.70 75.07 4.07 6.26 70.20 4.67 7.19 78.64 5 25 8.08 72.46 5 4.17 6.41 85.02 6.23 9.59 61.99 3.37 5.18 80.25 5.24 8.05 78.25 4.14 6.37 85.00 6.22 9.57 82.02 6 2.32 3.57 88.59 3.63 5.59 87.58 2.22 3.42 83.67 2.11 3.24 81.49 2.31 3.56 86.56 3.58 5.51 87.53 7 1.95 2.99 91.58 2.46 3 78 91.36 1.99 3.06 86.74 1.82 2.80 84.29 1.96 3.01 91.57 2.47 3.81 91.34 8 1.35 2.08 93.66 1.31 2.01 93.37 1.71 2.63 89.36 1.80 2.77 87.05 1.34 2.06 93.64 1 32 2.03 93.37 9 1.02 1.57 95.23 1.21 1.86 95.23 1.53 2.36 91.72 1.65 2.54 89.59 1.04 1.60 95.23 1.21 1.87 95.23 10 — — — — — — 1.42 2.18 93.90 1.40 2.16 91.75 — — _ — — — 1 1 — _ — — — — 112 1 73 95.62 2.10 3 23 94.98 — _ — — — 12 — — — — — — 1.04 1.60 97.22 1.46 2.24 97.22 — — — — — — 50 45 40 ■ o © c 35 > o Q. X LU 30 8 c §25 IQ > 2 20 o H o 15 3 9 10 5 0 □ Absolute Abundances • No Rotation ■ Absolute Abundances • Varimax Rotation B Log-Transformed Abundances • No Rotation B Log-Transformed Abundances * Varimax Rotation B Standardized (%) Abundances - No Rotation ■ Standardized (%) Abundances • Varimax Rotation I U : I I W . 6 7 Principal Components Axis 179 180 Figure 3.39. Site 689B R-mode eigenvalue coefficients (“loadings”) of taxa for varimax-rotated PCA of absolute, log-transformed, and relative (%) abundance matrices. Samples are presented in stratigraphic order, with eigenvalues in dark gray cells between 0.75-1.00 and in light gray between 0.50-0.74. Note that the general shift between loadings that coincides with the Oi-1 event. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 181 Varima» Eigenvalue Coefficients — R-Mode Relative (%) Abundances Absolute Loo-Transformed Abundances Abundances .72J .05 26 I .87 .00 36 M i .25 24 | ST 14 -.09 -.05 ; ' -.22 52 I - .13 30 I .5 0 ! .06 28 I .0 6 I -.16 .10 1 .5 5 i .07 r S S 01 K T O i .05 28 B H -.02 20 07 .02 i .02 ■ 0 2 _ .1 1 l6 3 U 0 6 .07 .05 04 01 = a 0.75 eigenvector coefficient = 0.50-0.74 eigenvector coefficient Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 182 3.5. Discussion 3 .5.1. Overview Paleobiologists face an apparent paradox in reconstructing the history of life. On one hand, we embrace uniformitarianism, commonly using the ecological and environmental distribution of living individuals to reconstruct the paleoenvironments of their fossil counter-parts {e.g., Murray, 1995). On the other hand, we appreciate that taxa evolve, with ecological and environmental shifts not necessarily accompanied by morphologic {i.e., taxonomic) change {e.g., Schopf et a/., 1975). Resolution of this paradox is neither straightforward, unique, nor inevitable — it is case-by-case and based on an eclectic mix of past experience, appropriate data, and provisional assumption examined an operational framework of historical science (Simpson, 1963) This study is no exception as it attempts to move beyond documentation of deep- ocean ostracode faunal changes during the Eocene-Oligocene transition to their paleoenvironmental and paleoecological meaning based on the ecology of living representatives and reconstructed paleoecology based on independent paleoenvironmental conditions. Although modern ecological data may not be directly applicable to Paleogene individuals, it is assumed that paleoecological limits of specific taxa did not change significantly during the roughly two million year interval of study. Thus, the discussion is based on an amalgam of independent paleoenvironmental data, modem ecological data, and ancient faunal data — the strong signals recorded in the first provide an opportunity to test the applicability of the second to the third. 3.5.2. Ostracode abundance, richness, and diversity Fossil abundance is the net sum of myriad factors of biology {e.g., reproductive fecundity, success, and rate), ecology {e.g., standing crop, food availability, competition), sedimentology {e.g., “dilution” by other grains), taphonomy {e.g., Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 183 fragmentation, transport, in situ post-mortem dissolution), and post-depositional processes (e.g., compaction, deep-burial diagenesis). Given these myriad factors, constructing unique and testable explanations for observed changes in fossil abundance is often difficult. In this section, mass accumulation rates and dissolution indices are examined as primary controls upon observed ostracode abundances. In general, higher surface productivity increased carbonate and siliceous test flux to the seafloor, potentially diluting ostracode valve abundance. This higher surface productivity would also have increased the organic carbon:carbonate flux to the seafloor, fostering carbonate undersaturation and in situ dissolution through biological oxidation of labile organic carbon. Such data also have implications for changes in ambient water-masses; these implications are discussed later with respect to their fidelity with the major ostracode faunal restructuring that coincided with the onset of 01-1. For Site 744A, Diester-Haass (1996) documented that dissolution indices (i.e., % BF, % FPF) covary with surface productivity (i.e., % biogenic opal) for most of the study interval. Prominent examples of this covariance include the slight increase in surface productivity at 33.8 Ma and the dramatic increase at the onset of 01-1 (inferred from the % biogenic opal record; Figure 3.7), with concomitant peaks in both % BF and % FPF (Figure 3.9). Thus, for much of the interval, organic flux from surface productivity appears to be the primary control upon seafloor carbonate dissolution through bacterial respiration of C 0 2. Diester-Haass (1996) noted two types of exception to this general pattern. The first exception consists of low-productivity intervals where dissolution indices are high, implying the presence of a “young” undersaturated water-mass originating at high-latitudes. As defined by Diester-Haass (1996), such “proto-AABW” occurs before 34.65 Ma and during the intervals of 33.16-33.24 Ma and 32.85-33.0 Ma. The second exception consists of high- productivity intervals where dissolution indices are low, implying the presence of a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 184 warm, carbonate-saturated water-mass that originated at low-latitudes. As defined by Diester-Haass (1996), this “ WSBW” is present at Site 744A from 33.24-33.27 Ma. Increased opal and carbonate mean MARs during Chron 13N are consistent with the major increase in productivity (i.e., % biogenic opal) during the Oi-1 (Figure 3.9). Site 744A ostracode valve abundance may be interpreted in two ways with respect to these dissolution and accumulation data. First, assuming a constant biological “production rate” of ostracode valves, higher dissolution indices may record intervals where fewer valves escaped in situ dissolution. Indeed, valve abundance shows a general inverse relationship to both % BF and % FPF, particularly when these indices are high prior to 34.65 Ma and especially just prior to, and at the onset of, Oi-1. Site 744A valve abundance was not compared to dissolution indices via regression analysis because of their collection from alternating core samples. Although the percent of fragmentation for the abundant, highly reticulate genus Algulhasina appears to weakly covary with dissolution indices, preservation of fine valve features, such as spines and turrets, in many taxa throughout the study interval does not support intense dissolution, except at the major dissolution maxima at the onset of the Oi-1 event. Note that within the Oi-1 interval, biogenic opal weight % and MAR varies considerably, whereas carbonate MAR (i.e., planktic foram production) remains high and stable throughout the event (Figure 3.9). Again, assuming constant valve “production rate”, this high carbonate MAR would have decreased Oi-1 valve abundance through simple dilution. An alternate interpretation of these data discards the rather unlikely assumption that valve production was constant. Instead, times of increased carbonate undersaturation due to high surface productivity may have placed a sufficient metabolic tax upon ostracodes, which likely have the highest ontogenetic carbonaterbiomass ratio of any animal, that population density decreased, even though organic matter flux likely increased overall. Thus, corrosivity likely influenced valve abundance both biologically Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 185 and taphonomically, but deconvolving their relative importance is difficult without further laboratory and field studies (e.g., Swanson, 1995). Finally, there is no apparent correlation of valve abundance with the water-mass history of Diester-Haass (1996) outlined above. Using similar methods, Diester-Haass (1993, 1995) documented % biogenic opal and dissolution indices to generally covary for the Site 689B interval, with corrosivity broadly correlated with surface productivity and no evidence for major changes in bottom-water masses. Site 689B bulk MAR is relatively constant at -0.4- 0.5 g/cm2 /kyr, consistently lower than that for Site 744A, (i.e., -0.7 g/cmz/kyr before and after Chron 13R, >1.75 g/cm2 /kyr within Chron 13R). As at Site 744A, valve abundance is weakly inversely correlated to both dissolution indices (Figure 3.10), but valve abundance is much lower overall and shows greater variance. This low abundance may relate to the much higher dissolution indices at Site 689B, but cannot be attributed to dilution given the relatively constant and low bulk MAR. As organic carbon flux is reflected in % biogenic opal (Berger, 1976), average food supply may have been more abundantly supplied to pre-Oi-1 ostracode populations at Site 689B relative to Site 744A. However, this is not reflected in a higher Site 689B valve abundance per sample, perhaps because bottom water conditions were more corrosive during this interval, keeping observed abundances low either biologically or through dissolution. Richness within a given community is related to numerous factors, including niche partitioning, competitive exclusion, and disturbance. Virtually no data exist on the community structure and dynamics of deep-ocean ostracodes, with recognized trophic strategies consisting of filter versus deposit (i.e., actively foraging) feeding based on soft-part morphology. The class is rarely considered in most ‘‘holistic” benthic studies given their low densities and small size, often easily passing through the finest sampling screen. It is the author’s opinion that most deep-ocean ostracodes are Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 186 “hyper-k-selected” given their: (1) long pre-sexual ontogeny of seven to nine molts over a period of up to two years and possible much longer; and (2) low fecundity given a relatively low body-size:sex-cell ratio (Cohen and Morin, 1991). This is a hypothesis in desperate need of testing. If so, individual ostracodes may rarely “see” one another ecologically, implying little intra-class competition, with abundance and richness controlled by other biological (e.g., predation, meiofaunal competition, etc.) or environmental (e.g., temperature, dissolved oxygen, corrosivity, etc.) factors. Rarefaction analyses demonstrate Site 744A ostracodes to be taxonomically richer than Site 689B at all common sub-samplings. However, rarefaction of pre- and post-Oi-1 intervals produce opposite patterns, with Site 744A more diverse before Oi- 1, and Site 689B presumably more diverse after Oi-1 if one allows “curve extension” due to the low number of post-Oi-1 Site 689 samples. More detailed rarefaction of Site 744A by Isotope Intervals A-E, as defined by 6,80 inflection points among Oi-1 a and Oi- 1b, shows a progressive richness increase following its minimum at the beginning of Oi-1. On a much longer time-scale of the late middle Eocene through earliest Oligocene, benthic foram richness decreases at high-latitudes relative to low-latitudes (e.g., Kaiho, 1994; Thomas, 1992), while ostracode richness globally increased (Benson et al., 1984, 1985; Benson, 1990). Thomas and Gooday (1996) attribute the progressive high-latitude decrease in benthic foram richness to an increasingly “unpredictable” and seasonally fluctuating food supply, with a relative increase in opportunistic phytodetritus-exploiting species such as Epitominella exigua and Alabaminella weddellensis. Relating these low-resolution, long-term trends to the short-term, high-resolution trends at Sites 744A and 689B will require additional studies bridging the different time-scales. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 187 3.5.3. Krithe, Algulhasina, and potential gradual temperature change Two of the most abundant taxa at Site 744A are Algulhasina and Krithe spp. (Figure 3.9). In the pre-Oi-1 interval, the initial predominance of Algulhasina, largely a result of high numbers of juveniles and fragments, gradually shifts to that of Krithe spp. Both decrease in absolute abundance during Oi-1, with Krithe remaining predominant through early Oi-1 and into the latest post-Oi-1 interval. Algulhasina was largely restricted to warm shelf environments off Africa in the Late Cretaceous and early Cenozoic (Dingle, 1981, 1985). The predominance of the genus early in the Site 744A record is largely due to juveniles and valve fragments (see Chapter 4 for complete discussion of genus). The juvenile:adult ratio decreases from a maximum at 34.5-34.6 Ma, to a minimum just before the onset of Oi-1, and remains low until the event’s end at -32.10 Ma. Interpretation of these relations is difficult. Preservation is generally excellent in the robust, heavily calcified adult and juvenile valves, possibly accounting for their abundance compared to the generally weakly calcified instars of other taxa. Dissolution of the relatively thin basal valve layer beneath the strong reticulation is only present during the extreme dissolution maxima around Oi-1. Taphonomic winnowing of Algulhasina juveniles later in the interval is unlikely given that they are typically more massive than the adult valves of many common taxa. However, Algulhasina fragmentation is relatively high and tends to covary with juvenile abundance. The most parsimonious explanation for these observations is that earlier in the interval this taxon had a higher reproductive success, but fewer individuals survived to adulthood. Reproductive success decreased through the interval, as reflected by decreasing juvenile abundances, but roughly the same number of individuals survived to adulthood as at the beginning of the interval. This explanation is highly tentative and does not explain the fragmentation pattern. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 188 In the Late Cretaceous and early Paleogene thermospheric oceans, Krithe was geographically widespread and highly correlated with thermospheric taxa such as Cytherelloidea (Boomer and Whatley, 1995; Dingle, 1981). Majoran et al. (1997) analyzed Maastrichtian faunas of the South Atlantic and found Krithe most common at paleotemperatures of about 15°. This paleoecological regime stands in stark contrast to the modern cryophilic nature of the genus, with maximum temperature tolerance of largely less than 10° C, often predominant in temperatures of <5 °C, and a global distribution largely restricted to the psychrosphere (Whatley and Quanhong, 1993; Coles et al., 1994). Thus, the genus appears to have dramatically evolved in its paleoecological tolerance through the Cenozoic. Krithe’s increased absolute and relative presence through Site 744A is consistent with general cooling proposed for the interval and implies that the genus had developed cryophilic preferences by the Eocene- Oligocene, perhaps thriving in deeper, cooler waters prior to the Oligocene. These data imply that cooling sufficient to re-arrange the rank abundance of these two taxa started prior to Oi-1 at Site 744A, when surface productivity was low and stable. However, the actual temperature change remains uncertain given the poorly know thermal-preference histories of these two genera, as well as 5'8 0-based uncertainties due to combined cryosphere and temperature controls. At Site 689B, Algulhasina is absent and Krithe is relatively common throughout the interval, although overall faunal variance is extremely high and likely related to intensely corrosive seawater as discussed previously. 3.5.4. Pre-Oi-1 taxa — the platycopid response A pronounced faunal change at both sites is the disappearance at the onset of Oi-1 of Cytherella, Bairdoppilata, and Cytherelloidea (although this last genus is extremely rare). Cytherella and Cytherelloidea are both members of the Platycopida, obligate Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 189 filter-feeders that fundamentally differ in respiration, feeding, and reproduction from the generally more abundant deposit-feeding Podocopida (see Whatley, 1991, 1995 for detailed discussion). Nearly all benthic ostracodes lack a circulatory system or specialized respiratory organs, instead molecular gas exchange (as well as osmoregulation) occurs directly across the uncalcified inner lamella. This exchange is assisted by epipodial respiration plates (ERP) that flux seawater through the domociliar cavity, maintaining partial pressure gradients that drive the exchange of seawater dissolved 0 2 and metabolism-derived C 0 2 across the thin epicuticle (McMahon and Wilkens, 1983; Vannier and Abe, 1995). These ERPs are present on the mandibles and maxiilula of all marine benthic ostracodes, but platycopids possess additional plates on the maxilla and 1st thoracic appendages. Although a podocopid, Bairdoppilata bear one additional ERP on their maxilla. This greater number of ERPs in Cytherella, Cytherelloidea, and Bairdioppilata increases the fluid flux past the body, from which the suspended organic matter is sieved through specialized sets of pilate setae. An exaptation of this enhanced circulation for suspension feeding is more efficient gas exchange, allowing platycopids to preferentially survive, and in some cases thrive, over podocopids in reduced oxygen conditions. In addition, all platycopids brood and ventilate their young through multiple early instars, providing a relatively well- oxygenated microenvironment for early development and decided advantage over podocopid brooders as well as non-brooders. Platycopids often constitute the most common taxa within the modern oxygen-minimum zone, with well-documented examples from the southeastern USA (Cronin, 1983) and South Africa (Dingle et al., 1989; Dingle and Giraudeau, 1993). Ancient examples of platycopid-dominant dysoxic intervals include the Eocene (Whatley and Arias, 1993), Cretaceous Oceanic Anoxic Events (OAEs) (Jarvis et al., 1988; Horne et al., 1990; Babinot and Crumiere-Airaud, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 190 1990), Jurassic (Boomer and Whatley, 1992), and Late Devonian (e.g., Lethiers and Whatley, 1994). Given these biological and ecological relationships, the disappearance of low- oxygen-tolerant, generally suspension-feeding taxa at the onset of Oi-1 is somewhat enigmatic. At least five possible explanations exist and some may be tested explicitly with the available data: (1) Disappearance is an artifact of preferential dissolution. Swanson (1995) has argued from the basis of core-top and SEM data that platycopids are highly sensitive to dissolution because of their particular valve microstructure. However, comparison of absolute and relative abundances of these taxa with dissolution indices at Site 689B show no correlation between the two (Figure 3.20 and 3.21), which would be expected if dissolution was a primary control on presence. Furthermore, visual and limited SEM analyses of platycopids indicate that preservation of their more massive, if less dense, valves is comparable to that of most podocopids. (2) Disappearance is an artifact of sampling, where the much lower total abundance during the entire Oi-1 event greatly decreased the probability of “capturing" these taxa within a sample. However, comparison of absolute and relative taxon abundances to total valve abundances demonstrates that both taxa are common prior to Oi-1, then absent afterwards (with the rarest exceptions in the latest portion of the records), even though many post-Oi-1 samples at both sites have total abundances equal to most pre-Oi-1 samples (Figures 3.18 and 3.19). Thus, these taxa are clearly becoming extremely rare with respect to other taxa, implying biologically or ecologically deleterious conditions. (3) Dissolved oxygen levels increased as bottom water production shifted to more proximal locales at Oi-1, fostering proliferation of podocopids at the ecological expense of Cytherella, Bairdoppilata, and Cytherelloidea. However, this proposal is inconsistent Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 191 with the extremely high podocopid:platycopid ratio throughout the pre-Oi-1 interval, which averaged >90%. (4) Bottom-water cooling exceeded the physiological limits of these taxa. Modern and ancient Cytherelloidea are thermophyllic, whereas Cytherella has been reported from abyssal depths in the modern and ancient (Whatley and Coles, 1987). In addition, it is possible that the Bairdoppilata species is more closely aligned with the genus Bythocypris, which Majoran et al.'s (1998) South Atlantic Maastrichtian survey found most abundant at high-latitude Sites 689 and 690 (Northeast Georgia Rise, South Atlantic) at a paleotemperature of 9 °C, the lowest of all sites (range = 9-15 °C). Additional muscle-scar and hinge comparisons are necessary because of poor preservation in SEM specimens examined for this study. Thus, this cooling-based explanation remains viable, but the degree of cooling across Oi-1 remains controversial. Zachos et al. (1996) interpreted the observed ~1.2%o increase in 51 s O at Oi-1 to reflect rapid expansion of the East Antarctic ice sheet coupled with 3-4 °C cooling at Site 744A. However, preliminary Mg/Ca analyses on the genus Krithe, reported in detail in Chapter 1, show no net change in bottom-water temperatures across Oi-1. At Site 689B, benthic foram Mg/Ca paleothermometry analyses also shown no evidence for cooling across the transition (Elderfield, unpublished data). (5) The tempo and mode of surface productivity became increasingly seasonal with the development of stratified shallow waters, delivering a large pulse of highly labile phytodetritus versus a low, relatively steady “rain" of organic material. About the Eocene-Oligocene boundary, benthic forams associated with highly pulsed, seasonal surface productivity appear at many southern high-latitude sites (Thomas and Gooday, 1996). These taxa are interpreted as r-selected opportunists that quickly capitalized on seasonal pulses of highly concentrated phytodetritus, rapidly reproducing and then becoming largely dormant for the remaining annual cycle. Although at first counter- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 192 intuitive, such a change in the tempo and mode of surface productivity may have led to the ecological extinction of filter-feeding ostracodes at both sites. If surface productivity became sufficiently seasonal, obligate filter-feeders with relatively high metabolisms may have been “starved’’ out by the increasingly variable annual cycle. In contrast, deposit-feeding podocopids could actively scavenge the surface and shallow sub-surface, exploiting residual organic matter. An objection to this hypothesis is that % biogenic opal returns to pre-Oi-1 levels just after the Oi-1b peak at Site 689B, but filter- feeding taxa remain absent through the remaining section, except for 1-2 specimens in two of the latest samples. Explanations 1-3 can be largely ruled out based on the counter-evidence presented above. The last two are favored possibilities, but require additional independent data on changes in paleotemperature and the tempo and mode of surface productivity and delivery to the seafloor. 3.5.5. Reverse-response taxa The above taxa show a similar biological response at both sites to some apparently synchronous and therefore widespread environmental shift, in contrast, a number of taxa show a disparate faunal distribution between sites, implying more restricted, regional environmental changes as well. It is unlikely that these faunal appearances and disappearances are random events along the slopes of Kerguelen Plateau and crest of Maud Rise, given the coincidence of nearly all “significant” first and last local appearances around the onset of Oi-1. The primary examples are Actinocythereis and Trachyleberis (Figure 3.22 and 3.23), two common constituents of the pre-Oi-1 Site 689B fauna and post-Oi-1 Site 744A fauna, but not visa-versa. These distributions are not obviously related to either sampling effects (Figure 3.22 and 3.23) or dissolution (Figure 3.20 and 3.21). Unfortunately, little is know about the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 193 paleoecology or ecology of these taxa; most work has been at the alpha taxonomy level (e.g., Whatley and Coles, 1994). Furthermore, there is no clear correlation of these taxa with the independent paleoenvironmental records at the sites. Henryhowella asperima gradually and unevenly increases throughout Site 689B, while its Site 744A abundance increases abruptly at Oi-1. This species predominates in modern AAIW of the Atlantic (Dingle and Lord, 1990; Dingle and Giraudeau, 1993) and may herald the progressive expansion of this cool, moderately saline water-mass from Maud Rise towards Kerguelen Plateau. Dutoitella dwindles in its relative and absolute abundance at Site 689B throughout the pre-Oi-1 interval, and is absent thereafter except for occurrences restricted to the 81 8 0 minimum between Oi-1a and Oi-1b. However, at Site 744A, the genus is rare and spotty throughout the interval, with slightly more consistent presence at and following Oi-1b. In modern oceans, Dutoitella is generally restricted to NADW and AABW, where temperatures are generally less than 2 °C (Dingle and Lord, 1990; Ayress et al., 1997). 3.5.6. Other taxa Some taxa show restrictions or abundance change correlative with specific features of the stable isotope and sedimentological data. For example, little is know regarding the ecological and paleoecological preferences of Penneyella, which also endures an unstable taxonomy. Irregardless, the genus is consistently present in low abundances prior to 33.9 Ma and then most common within 5,80 minima (Figure 3.17). Given this distribution, the taxon is provisionally interpreted to have preferred relatively warmer bottom waters with lower organic supply. Relatedly, Indeterminate Genus E shows a similar preference for 6,0O minima, but is absent prior to Oi-1. This as-yet-undescribed genus is therefore provisionally interpreted to have preferred a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 194 somewhat cooler range of temperatures than Pennyella and/or a somewhat higher, but still restricted range of organic supply. 3.5.7. Faunal patterns within Oi-1 Whereas the above taxa show distinct stratigraphic restrictions and inferred paleoenvironmental preferences, whole-fauna analysis via similarity coefficients, clustering algorithms, and R-mode PCA indicate two general patterns. First, a clear change in faunal composition occurs at the onset of Oi-1. This is illustrated by the cohesive clustering among pre-Oi-1 samples (Isotope Interval A) at both sites. Second, following the Oi-1 event, faunal composition shows different histories at each site. Site 744A faunas tend to transition back to pre-Oi-1 compositions based on the adjacent clustering of Isotope Interval D and E to Isotope Interval A. In contrast, Site 689B post- Oi-1 faunas remain relatively cohesive and distinct in their composition through the end of the study interval. Although not sub-divided into finer isotope intervals given the site's lower-resolution isotope record, examination of Site 689B's Isotope Interval B samples by their stratigraphic position shows no shift towards increased similarity to earlier faunas. The R-mode PCA eigenvector coefficients for both sites also reflect these general faunal differences. The Site 744A pattern is less pronounced in these data, whereas the Site 689B pattern indicates that faunal composition may continue to depart from the earlier post-Oi-1 samples as higher PC show increased loadings in the latest samples. Assuming a tight taxon-environment relationship, these whole-fauna patterns support a slight return to pre-Oi-1 conditions at Site 744A, but lingering post-Oi-1 conditions at Site 689B. This Site 744A pattern may be a partial sampling artifact of the increased richness that accompanies increased abundance following the event, even though this is minimized by the presence/absence nature of the similarity matrix calculations. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 95 3.6. Conclusions Ostracodes are common constituents of Eocene-Oligocene sediments on both Maud Rise (689B) and Kerguelen Plateau (744B), which were separated by roughly 90° paleolongitude, but positioned at similar paleodepths (1,400-1,800 m) and distances from Antarctica. Major changes in the faunal abundance, richness, and evenness coincide with the onset of the major Oi-1 step increase in 81 8 0 marking the beginning of significant Antarctic glaciation and thermohaline cooling. Abundance changes reflect possible dilution effects at Site 744A, and biologically or preservationally deleterious conditions related to seawater corrosivity at both sites, particularly Site 689B, where corrosivity is high throughout the interval. Following the event, Site 744A faunal composition shifted slightly back towards initial compositions, whereas Site 689B faunas remained relatively distinct from earlier faunas. This major increase in high-latitude “refrigeration” likely led to the initiation or increase of thermohaline-driven circulation, and certainly significant increases in primary productivity, likely highly seasonal in nature. The Oi-1-coincident faunal shifts among the sites include changes that are both uniform (e.g., loss of Cytherella and Bairdioppilata at both sites) and reversed (e.g., Site 689B loss and Site 744A appearance of Trachyleberis and Actinocythereis). Together, these faunal patterns imply a common site response to the increase in surface productivity (e.g., Cytherella and Bairdoppilata), as well as a disparate site response to either subtle productivity intensity differences or changing bottom water conditions (e.g., Trachyleberis and Actinocythereis). In contrast, previous benthic foram studies have found little faunal changes at these sites coinciding directly with Oi-1 (Thomas, 1992). Thus, ostracodes appear more sensitive to Oi-1-coincident environmental changes than benthic forams. Ostracodes show no faunal response to the sedimentology-based intervals of proto-AABW and WSBW in the upper and lower portions of the study interval at Site 744A. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 96 3.7. Future Directions One may interpret these findings as the view from “ ground zero” — both sites are very close to Antarctica. Key future directions to improve our understanding of paleoceanographic changes associated with this transition from “greenhouse” to “icehouse” include comparable scale studies at other sites, such as those of recent Leg 177 south of Africa and upcoming Leg 189 in the Tasman Sea. Additional studies at lower-paleolatitudes will also likely remove the strong surface-productivity overprint and allow more straighforward examination of potential bottom-water changes upon ostracode faunas, as well as their diachroneity. This study concentrated on a very short and unique, but climatically important, interval of Paleogene history. Expansion up- and down-section will provide a broader context and examine ostracode relations to other environmental fluctuations. For example, Diester-Haass and Zahn’s (1996) -46-26 Ma study of Site 689 stable isotopes and paleoproductivity proxies indicates that Eocene productivity was greatest during colder periods of increased upwelling, whereas Oligocene productivity was greatest during warmer periods when a proto-polar front moved south of the site. Examination of ostracode faunas from selected cycles within each of these paleoceanographic “modes” will provide additional constraint on paleoecological and paleoenvironmental hypotheses proposed in this study (Schellenberg and Diester-Haass, in prep.). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 97 4. Linkage of surface productivity and benthic ostracode size during the Eocene- Oligocene transition: Algulhasina quadratica ontogeny at ODP Site 744A (Kerguelen Plateau, Southern Ocean) 4.1. Introduction Energetic and material linkage between “separate" ecosystems is an important component of biosphere dynamics. The photic and benthic zones of the pelagic ocean, separated by great microbial-dominated depths, share such a linkage: Energy exported of local surface primary productivity largely controls secondary productivity in the underlying benthos. In addition, surface conditions in often remote regions of deep- water formation set conservative bottom-water properties (i.e., temperature, salinity) as well as initial non-conservative properties (i.e., dissolved oxygen and nutrient concentrations). Conversely, thermohaline cirulation hetergenously redistributes critical nutrients to the photic zone through upwelling. This “planktic-benthic” linkage within the pelagic realm has a complex and dynamic Phanerozoic history as paleoceanographic and biotic components evolved over time (see reviews and examples in Signor and Vermeij, 1994; Fischer and Arthur, 1977). The benthic component of this history is based largely on foraminifera as deep- ocean metazoans are generally not preserved given sparse skeletal material and slow (cm/kyr) pelagic sedimentation. Distinct exceptions to this single-cell bias are the deep-ocean ostracodes, whose robust calcified valves provide an extensive ecological and morphological record of environmental changes from a benthic metazoan perspective (see reviews in Benson, 1988 and Whatley, 1996). Ostracode skeletons contain a complex three-tiered architecture represented by valve reticulation (recording the distribution of valve-precipitating epidermal cells; Okada, 1981, 1982), a sensory pore canal constellation (recording the distribution of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 198 valve-penetrating nervous system setae; Liebau, 1971; Benson, 1972), and their constructional form (including allometric components). These morphological aspects provide a record of evolution in response to mechanical stress, metabolic carbonate budgeting, and environmental conditions (Benson, 1975b, 1981). Given the logistical difficulties of deep-sea ecological studies (see Murray, 1995), most morphometric studies of deep-ocean ostracodes have concentrated on long-term (>107 yr) evolutionary processes recorded within reticulation and pore-canals patterns (e.g., Benson, 1982, 1983). In contrast, the relative ease of sampling shallow-marine, saline-lake, and fresh-water environments has provided numerous studies on the role of temperature, salinity, organic carbon, and other parameters on ecophenotypic morphological changes on ecological time-scales (i.e., seasonal to decadal; e.g., Ishizaki, 1975; Kaesler, 1975; van Harten, 1975; Kaesler and Lohmann, 1976; Semenova, 1979; Foster and Kaesler, 1983; Martens, 1985; Schweitzer and Lohmann, 1990; Kamiya, 1992). This study attempts to bridge this temporal gap between evolutionary and ecological studies through a high-resolution (-1 0 4 yr) ontogenetic size analysis of the ostracode Algulhasina quadratica through the Eocene-Oligocene “ greenhouse-icehouse” transition at ODP Site 744A (Kerguelen Plateau, Southern Ocean; Figure 4.1). Stratigraphic size-variation within seven instars is examined for ecophysiological effects of surface and bottom-water conditions through time. Well-represented intermediate instars exhibit significant size differences before and after the onset of the “ Oi-1” event, which heralded significant Antarctic cryospheric growth and associated regional intensification of surface productivity via shifting/strengthening of polar fronts (Zachos et al., 1996; Salamy and Zachos, 1999). These Algulhasina size differences are most correlated with biogenic opal and 81 8 0 , and are likely driven more by surface productivity changes based on detailed stratigraphic analyses and other data. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 199 Figure 4.1. Late Eocene paleogeography of Southern Ocean and ODP Site 744A on Kerguelen Plateau (after Lawver et al., 1992). Paleodepths through the study interval are estimated at -1,800 meters, shallower than the estimated paleosurface of the carbonate compensation depth (CCD) (Barrera and Huber, 1993). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 0 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 0 1 4.2. Paleoenvironmental Setting The Eocene-Oligocene represents a major step in Cenozoic climate and is marked by a transient “hyperglacial” interval of Antarctic cryospheric development and ocean- atmosphere cooling. This interval, designated as Oi-1 by Miller et al. (1991), is the apparent climatic response to the crossing of a stability threshold as development of the circum-polar ocean led to thermal isolation of Antarctica (Figure 4.1; Kennett and Shackleton, 1976; Lawver et al., 1992). Documented largely from high-resolution stable-isotope ODP studies, the onset of Oi-1 was rapid (<100 kyr) and contained two extrema in 6,aO and 8,3C (Oi-1 a, Oi-b) before the climate system recovered to a new, relatively stable state (Figure 4.2a; Zachos et al., 1994). During Oi-1, ice-sheet development and ocean surface cooling is hypothesized to have increased the flux of weathering-derived micronutrients and wind-stress upwelling to the photic zone of the Southern Ocean (Kennett et al., 1975). This hypothesis is supported through numerous studies of microfossil abundance, stable isotopes, and mass accumulation rates at various southern high-latitude ODP sites (e.g., Baldauf and Barron, 1990; Diester-Haass, 1993, 1995; Diester-Haass et al., 1993, 1996; Salamy and Zachos, 1999). At Site 744A, Diester-Haass (1996) and Salamy and Zachos (1999) have documented a several-fold increase in surface productivity as primary production shifted from calcareous to opaline organisms, likely as an ecological response to seasonally greater divergence driven by polar front intensification (Figure 4.2b and upper c). The potential impact of this photic zone fertilization on the underlying benthos is examined through an analysis of Algulhasina quadratica body size. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 202 Figure 4.2. Environmental context and Algulhasina quadratica abundance at ODP Site 744A. Age model based on linear calibration of core magnetostratigraphy of Keating and Sakai (1991) to the Geomagnetic Polarity Time Scale of Berggren et al. (1995). Open triangles in A indicate stratigraphic position of 10 cc samples of study. A. 51 0 O (gray fill) and 51 3 C (black line) stratigraphy of Zachos et al. (1996) showing pronounced isotopic changes (“Oi-1”) during the Eocene-Oligocene transition. B. Weight % biogenic opal (gray fill) of Salamy and Zachos (1999) showing pronounced increase in diatoms and radiolarians. The dissolution index of % fragmented planktic forams (black line) of Diester-Haass (1996) correlates well with major inflections and extrema in biogenic opal. C. Stratigraphic distribution of adults, juveniles and fragments (F1 = -50-25% complete; F2 = <25% complete) of Algulhasina quadratica (lower axis) and mean mass accumulation rates (MAR = linear sedimentation rate multiplied by dry bulk density of specific sediment component of carbonate, opal, and terrigenous sediments; upper axis; Barron et al., 1989a; Salamy and Zachos, 1999). D. Percentage of adults (black line; adults /adults+juveniles*100) and fragments (gray line; F1+F2/adults+juveniles+F1 +F2*100) of Algulhasina quadratica. Stratigraphic samples containing no specimens are represented by gaps in plot lines. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 0 3 8 snouoQujol s 9 s , °4‘f hiy dZLO NCIO h£lO NSW eud3o6;io AiJeg 0U0OO3 0JB“| Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 204 4.3. Materials, and Methods ODP Hole 744A (southern slope of Kerguelen Plateau, Southern Ocean, 61°34.66'S, 82°47.25'E; 2,307 m) produced a Late Eocene-Early Oligocene sequence of coccolith ooze with an estimated paleodepth of -1,800 meters, shallower than the estimated paleosurface of the carbonate compensation depth (CCD) (Barrera and Huber, 1993). The age model is based on linear calibration of core magnetostratigraphy (Keating and Sakai, 1991) to the Geomagnetic Polarity Time Scale of Berggren et al. (1995). A mean 20±12 cm sampling interval (20±9 kyr/sample) from 137.28- 157.01 mbsf (32.77-34.82 Ma) provided 101 10 cc samples. All valve material identifiable as Algulhasina quadratica, selected for its well- preserved series of seven instars, was removed from the >150 pm size-fraction of each sample and classified as complete (> 50%) adults or juveniles or as incomplete (< 50%) fragments. Final adult instars are distinct from earlier instars by their robust holamphidont hinge and prominent anterior marginal ridge (Figure 4.3). Fragments less than 50% complete were classified into two categories, but their adult/juvenile status was not determined. The F1 category consisted of between 50-25% complete valves and the F2 category consisted of < 25% complete valves. All complete A. quadratica specimens were measured using a CCD-equipped binocular microscope and the Optimas 5.5 digital image analysis system. Specimens were oriented with their hinge chord parallel to the x-axis. Measured x,y coordinates consisted of the anteriodorsal end of this hinge chord and the greatest anterior, ventral, and posterior points. Specimen outlines were then traced clock-wise from the anteriodorsal x,y coordinate to calculate outline area, centroid, circularity, and 64 equally spaced x.y coordinates for future eigenshape analysis. Measurements were exported to Excel using an ALI language macro employing Dynamic Data Exchange (DDE); Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 205 Figure 4.3. Orientation and measurements for Algulhasina quadratica. Maximum valve height and length were calculated from the four cartesian coordinates landmarks. Area, centroid coordinates, circularity (perimeter2 /area), and 64 sequentially and equally spaced coordinates were also collected from a clock-wise outlining of the valve from the anteriodorsal coordinate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 207 macro code is provided in Appendix 7. Maximum valve length and height were calculated by subtracting the posterior x from the anterior x and dorsal y from ventral y, respectively. Circularity (perimetei^/area) is a dimensionless parameter that provides a measure of shape, with a minimum value of 4ji (12.57) for a perfect circle and increasing values towards squares (16) and equilateral triangles (20.78). Measurement error as determined from 30 measurements of a control specimen over the course of study was < 1% for length and height and < 3% for area and circularity. 4.4. Results 4.4.1. Stratigraphic abundance The stratigraphic abundance of Algulhasina quadratica adults, juveniles, and fragment classes is shown in Figure 4.2. These abundances show no strong correlation to either 8,3C or percent sand (not plotted). Through the pre-Oi-1 interval, all categories gradually decrease in abundance, with a slight decrease in the proportion of valve fragments (Figure 4.2c,d). This cross-category trend corresponds to a gradual decrease in the percentage of fragmented planktic forams (% FPF), but no other paleoenvironmental parameters (Figure 4.2a,b). Around 33.76 Ma, juveniles decrease relative to adults, followed by elevated juvenile abundances and % fragments. This faunal pattern corresponds to a minor peak in oxygen isotope, % biogenic opal, and % FPF and the short interval thereafter just prior to the full onset of Oi-1. Within Oi-1, all abundance types are relatively low and % fragments is elevated with peaks corresponding to Oi-1 a and Oi-1b peaks in oxygen, carbon, % biogenic opal, and % FPF. The spikiness of % adults within the interval is largely an artifact of a small and spotty number of relatively complete adults and juveniles. Following the termination of Oi-1 b, abundances in all categories increase, while the % fragments decreases. Note that Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 208 elevated opal, carbonate, and terrigenous mass accumulation rates correspond closely with the interval of decreased valve abundance during the Oi-1 event (Figure 4.2c). 4.4.2. Ontogenetic size analysis of Algulhasina quadratica Measured individuals were assigned post hoc to one of seven instars (Adult through A-6) based on their respective position among seven distinct clusters in the highly correlated (r = 0.99) length-height morphospace (Figure 4.4). In the penultimate and adult instars, male sexual dimorphism (gray symbols) was attributed to individuals of average height but greater length (Whatley and Stephens, 1977; Kamiya, 1988). This male valve dimorphism represents the development of gonads and hemipenes in the penultimate and adult stages (Cohen and Morin, 1990), which commonly account for one-third of the body volume in order to produce and deliver sperm that may exceed body length by an order of magnitude (Winstrand, 1988). The average ratios of size increase between A. quadratica instars range from 1.15 and 1.24, averaging slightly less than 1.26 (the cube root of 2 or the doubling of spherical volume) and within the general range for ostracodes (-1.17-1.35). Valve length (and height) are also highly linearly correlated with valve area and circularity, although these relationship are slightly better described through power function correlations (Figure 4.5). Note that circularity stabilizes in later ontogeny and male sexual dimorphism is also expressed by these measurements. Stratigraphic distribution of the A. quadratica instar group suggests an increase in average valve length of numerous instars within the Oi-1 event, although abundances are generally low (Figure 4.6). To test this hypothesis, each instar population was divided into “pre-Oi-1” and “post-Oi-1" populations at the 33.6 Ma horizon. These normally distributed sub-groupings were subjected to a separate-variance t-test at an alpha level of 0.05% (Figure 4.7, 4.8). The null hypothesis of no significant difference Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 209 Figure 4.4. Linear growth series (length versus height) for all complete Algulhasina quadratica specimens. Values along each axis denote the average proportional increase in each dimension between instar classes. Filled symbols in later instars are interpreted as males given higher valve lengths. Filled A-2 symbols may represent extremely large A-3 or extremely small A-4 specimens based on their relative stratigraphic distribution (see Figure 4.6). Regression excludes male specimens. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 1 0 o C M ^ C O C O <D 3 C cr © _ C O p- k. * 3 | . £ O) C M w c o o + o > c c n © a zL o Oi I' c o _ C M c o o ii g > ® X c o ® C O E co C D T J 3 O X ® c o co C O ® w O ) ® o o o o > o G O ° ! o CO o c n o o CO o CM o 0 0 o o CO CO * ■ ( oitI) I m6|9H o c o o C M Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Length (pm) 21 1 Figure 4.5. Algulhasina quadrata length versus circularity and area. Regressions exclude male specimens. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 1 2 18 1 7 16 ^ i 15 14 Circularity = -0.0254 (Length) + • 16.62 r = 0.79 Circularity = 22.923 (Length) 0 1 0 3 4 r = 0.82 (Regressions exclude males) Adult 13 20 40 60 80 Length (um) 100 120 6000 5000 4000 (0 « : 3000 2000 1000 Area = 66.174 (Length) - 1958.2 r = 0.99 Area = 0.4206 (Length)2 0 4 6 2 r = 1.0 (Regressions exclude males) Adult * • ' / - \ & 20 40 60 80 Length (um) 100 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 213 Figure 4.6 Stratigraphic distribution of valve lengths of Algulhasina quadratica instars. Vertical lines are mean valve length for each instar. Horizontal line is the dividing horizon for pre- and post-Oi-1 populations examined by t-tests. Stratigraphic curve aligned with each mean length shows relative biogenic opal fluctuations throught the interval. Filled A-2 symbols represent possible member of adjacent instars. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 214 6 o ° o 6 0 6 § @ D ? v * * NSI.Pl NCI.P a a o 0U0OODHO A jjeg 0U 0O O q 0| B | Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. L e n g t h ( ^ m ) 215 Figure 4.7. Normal distribution test of valve length for complete, pre-Oi-1, and post- Oi-1 intervals. Linear plotting of each instar indicates normal distribution. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 1 6 c o 3 n *c to O T o E k . 0 z k - £ (0 0) _3 T3 1 0) Q. X L U 3 Post-Oi-1 2 1 0 1 2 C o m p l e t e I n t e r v a l 3 2 1 0 1 2 3 Pre-Oi-1 3 2 1 0 ■ 1 2 3 50 100 Length (*im) 150 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 217 Figure 4.8. t-test (a = 0.05) of mean valve length differences between each pre-Oi*1 and post-Oi-1 instar population. Statistically significant differences are highlighted in gray. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 218 Separate variance t-test of Pre- and Post-Oi-1 valve length of Algulhasina quadratica Instar t d.f. P Diff. Means 95% Cl A (M) -0.73 6.3 0.492 -1.431 -6.167 — 3.306 A 0 .17 8.9 0.872 0.137 -1.741 — 2.016 A-1 (M) — — ----- — — A-1 3.06 17.3 0.007 2.896 0.900 — 4.892 A-2 1.63 32.2 0.113 1.118 -0.281 — 2.518 A-3 5.32 2 4.6 0:000 2.519 1.543 — 3.494 A-4 3.85 17.1 0:001 1.967 0.888 — 3.046 A-5 2.16 14.7 0.047 1.117 0.015 — 2.220 A-6 0.65 21 0.521 0.227 -0.496 — 0.950 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 219 between mean pre- and post-Oi-1 valve length was rejected with 95% confidence level for the instars A-5, A-4, A-3, and A-1. Consideration of stratigraphic distribution and lengths of individuals in the later group of instars suggests that more abundantly and equitably distributed individuals from the post-Oi-1 interval would increase likelihood of rejection of the null hypothesis: A-6 instars are absent through the entire Oi-1 event until the very top of the study interval, when environmental conditions are less extreme than earlier in Oi- 1. A-2 instars show relatively high variance in length throughout the study interval. In addition, the two shortest A-2 specimens, identified by gray infill in Figure 4.6 and based on instar clustering in Figure 4.4, may actually represent the largest of coeval A- 3 instars, to which these specimens plot more closely. Conversely, a similar classification error may be present at the upper size limit of A-2, where five relatively large individuals, again identified by gray infill in Figure 4.6 and based on instar clustering in Figure 4.4, may represent relatively small A-1 instars. Adult specimens are extremely limited to two females and five males in the post-Oi-1 interval, but these largely plot around or below mean length values for each sex. Given these significant size differences, valve lengths for each instar were compared to the four paleoenvironmental parameters of biogenic opal, 51 8 0 , 51 3 C, and percent sand for the entire interval using linear correlation analysis (Figures 4.9 and 4.10). The resultant 36 correlation coefficients were tested for statistical significance at a = 0.05 and n-2 degrees of freedom (Figure 4.11). As with the t-test, the low abundance and restricted distribution of the A-6, A-1 male, and both adult instar sexes may have favored support of the null hypothesis of no significant correlation with any environmental parameter. In contrast, significant correlations exist between valve size and biogenic opal, 51 8 0 , and 51 3 C for instars A-5, A-4, A-3, and A-1, while only biogenic opal was significantly correlated with instar A-2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 2 0 Figure 4.9. Linear correlation of each lengths within each instar to % biogenic opal and 81 8 0 for the entire study interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Length (pm ) Length (urn) 221 120 110 100 90 80 70 60 50 40 30 20 A (M) A-1 A-2 A-3 A-4 A-5 A-6 y = 0.0215x + 99.279 r = 0.29 y = -0.3623x + 109.71 r = 0.03 y = 0.6637X + 83.768 r = 0.45 y = 0.2522X + 67.927 r = 0.22 y = 0.495x + 55.106 r = 0.49 y = 0.4603X + 45.25 r = 0.49 y = 0.2122x + 39.109 r = 0.33 y = 0.0496X + 33.349 r = 0.05 0 1 2 3 4 5 6 7 8 9 101112131415 % Biogenic Opal 120 110 A (M) 100 A 90 80 A-1 70 A-2 60 A-3 50 40 A-4 A-5 30 20 A-6 I y = -1.0259X + 110.2 r = 0.17 y = 0.6958X + 98.481 r = 0.13 y = 2.9717X + 81.113 r = 0.38 y = y = -i- IK W y £ A 1 1 1 * 0 . 9 8 2 1 x + 6 7 . 1 8 3 r = 0 . 1 3 2 . 7 6 8 8 X + 5 2 . 4 7 r = 0 . 5 0 -0.124x + 48.382 r = 0.54 1.4825X + 37.67 r = 0.32 0.3599X + 32.987 r = 0.08 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 5180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 2 2 Figure 4.10. Linear correlation of length within each instar to 51 3 C and % sand for the entire study interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Length (pm) Length (pm) 223 120 110 100 90 80 70 60 50 40 30 20 0 120 110 100 90 80 70 60 50 40 30 20 0 1 2 3 4 5 6 7 8 9 % Sand A (M) & A-3 □ y = 4.1191X + 104.11 r = 0.27 y = 1.2152X + 98.181 r = 0.13 y = 5.67X + 79.133 r = 0.31 y = 1,6702x + 66.798 r = 0.08 y = 4.6171X + 51.321 r = 0.40 y = -0.124x + 48.382 r = 0.48 y = 3.483x + 36.137 r = 0.29 y = -1.7472X + 35.061 r = 0.13 0.5 0.7 0.9 1.1 1.3 1.5 813C A (M) A-1 A-2 A-3 A-4 A-5 A-6 & y = -0.6309x + 112.8 r = 0.34 y = -0.0513X + 99.553 r = 0.06 y = 0.2096x + 83.801 r = 0.12 y = -0.0905X + 68.854 r = 0.06 y = 0.2103x + 54.786 r = 0.16 y = 0.0765X + 45.552 r = 0.07 y = -0.0703X + 39.884 r = 0.07 y = -0.1501X + 34.329 r = 0.18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.11. Statistical significance of correlations between instar length and paleoenvironmental parameters for entire study interval. Significant correlations indicated in bold italics. Regardless of significance, highest values are highlighted darker gray and next highest in lighter gray. Bottom plot shows r values for each parameter for each instar and the r critical value for each sample size. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Correlation Coefficient (r) 225 Correlation of Algulhasina quadratica valve length to environmental parameters for entire study interval r Crit. I Correlation Coefficient (r) I Instar n d.f. a=0.05 I % Opal I §180 513C % Sand A (M) 10 8 0.63 M B U 0.2 7 0 .3 4 I A 29 27 0.37 0.03 0 .0 6 ] A-1 (M) 7 5 0.75 0.20 I 0.02 0.10 A-1 82 80 0.22 ■ ■ 0.38 0.31 0.12 A-2 129 127 0.17 0 1 3 0 .0 8 0.06 A-3 106 104 0.19 0.40 0.16 A-4 85 83 0.21 0.48 0.07 A-5 65 63 0.24 B B B & j F 0.29 0.07 A-6 42 40 0.30 0.05 I 0.08 0 .1 3 Italics = highest correlation coefficient = second highest correlation coefficient = statistically significant at a = 0.05 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 □ t f i 9 »i t i • X • I * I • X ♦ % Opal • d180 A d13C □ % Sand r Crit. % t % 0 \ 0 X 0 % 0 A ♦ , • ‘ .A I » l □ 6 A I 6 □ A A (M) A A-1 A-1 (M) A-2 A-3 A-4 A-5 A-6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 226 An analogous correlation analysis was applied to the post-Oi-1 population (as defined in the t-test) to more closely examine size-environment relationships within Oi-1 (Figures 4.12-14). Unlike the pre-Oi-1 inteval, this interval contains temporal leads and lags between environmental parameters, providing a more dynamic window for identification of potential correlations (Figure 4.2). All instars with relatively widely distributed individuals through the event (i.e., A-3, A-2, A-1; see Figure 4.6) show a strong and significant correlation only with the amount of biogenic opal preserved in the surrounding sediments. 4.5. Discussion The preservation of up to nine “snapshots” of ostracode ontogeny provides both potentials and pitfalls for reconstructing various aspects of ecologic, ontogenetic, and phylogenetic inter-relationships from ecological to evolutionary time-scales. On the potential side, ontogeny represents a major source of variation upon which natural selection may act and therefore heterochrony may represent a major process of evolutionary change (McKinney, 1990; McKinney and McNamara, 1990). On the pitfall side, the discrete molt stages of ostracodes precludes a critical component in such analyses, that of specific developmental timing of particular features. Both of these aspects have been addressed in numerous works, particulary in ostracodes by Kamiya (1988) and Schweitzer and Lohmann (1990). Therefore, the above data are discussed largely in terms of current ostracode growth and development and their relation to environmental conditions. Lifespans of deep-ocean ostracode are estimated at a few years, while each of their eight molt cycles likely transpire over but a few days (Cohen and Morin, 1990). These major physiological events represent only a small proportion of total ontogeny, but appear extremely energetically taxing as active feeding is largely precluded during Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 227 Figure 4.12. Linear correlation of each lengths within each instar to % biogenic opal and 5'80 for post-Oi-1 interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 2 8 120 110 100 90 •g 80 £ 70 O ) c ® _i 60 50 40 30 20 120 110 100 90 v ir JT InD — — □ L-P ° ---------- * H * - ------ ± — --------------------------------------1 - y = -0.3606x + 103.14 r = 1.00 y = 0.9987x + 81.504 r = 0.62 y = -1.6424X + 116.75 r = 0.47 y = 0.5111x + 65.916 r = 0.42 y = 0.2923X + 56.56 r = 0.61 y = 0.0739X + 47.805 r = 0.58 y = 0.277X + 38.565 r = 0.60 y = -0.6605x + 36.765 r = 0.79 2 3 4 5 6 7 8 9 1 0 11 1 2 1 3 1 4 1 5 1 6 % Biogenic Opal E 80 £ 70 o > c ® -I 60 50 40 30 20 A-2 A-3 A-4 A-5 A-6 y = 5.67x + 96.267 r = 0.34 y = -18.555X + 139.13 r = 0.65 y = -4.1499X + 96.041 r = 0.16 y = 2.9378X + 63.206 r = 0.42 -a— -g—n d i^ j g —n a y = i-7796x + 54.578 Q r - A O O x » r = 0.22 y = -0.124x + 48.382 r = 0.40 y = 14.332x + 11.409 r = 0.71 y = -85.87x + 201.34 r = 0.79 1.8 1.9 2.0 2.1 2.2 2.3 2.4 8180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 229 Figure 4.13. Linear correlation of each lengths within each instar to 51 3 C and % sand for post-Oi-1 interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Length (pm ) Length (pm) 230 120 = 4.2756X + 102.95 r = 0.52 0 A (M ) 1.192x + 98.567 r = 0.23 100 -1.636X + 89.477 r = 0.10 90 A-1 80 = 0.8808X + 68.291 r = 0.05 70 A-2 = 2.6207X + 54.96 r = 0.43 □ □ C h m ig--------- X X 60 A-3 -0.124x + 48.382 r = 0.03 50 A-4 10.354X + 28.041 r = 0.76 40 A-5 A-6 = 5.9632X + 27.339 r = 0.79 30 20 0.8 813C 120 110 A (M) 100 A 90 A-1 80 70 A-2 60 A-3 50 A-4 A-5 40 A-6 30 20 5 K -* JK-£ y = -0.4273X + 110.88 ► r = 0.43 y = 1.0793x + 94.802 r = 0.93 y = 0.6674X + 83.036 r = 0.28 y = -0.468X + 72.384 r = 0.25 y = -0.0754X + 58.856 r = 0.10 y = -0.124x + 48.382 r = 0.53 y = -0.7479X + 45.226 r = 0.76 y = -1.9082X + 49.374 r = 0.79 6 7 8 % Sand Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 231 Figure 4.14. Statistical significance of correlations between instar length and paleoenvironmental parameters for post-Oi-1 interval. Significant correlations indicated in bold italics. Regardless of significance, highest values are highlighted in darker gray and next highest in lighter gray. Bottom plot shows r values for each parameter for each instar and the r critical value for each sample size. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Correlation Coefficient (r) 232 Correlation of Algulhasina quadratica valve length to environmental parameters for lntra-Oi-1 interval II Correlation Coefficient (r) r Crit. Instar g=0.05||% Qpal 6180 % Sane 0.34 A-1 (M) Italics = highest correlation coefficient = second highest correlation coefficient = statistically significant at a = 0.05 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 A ♦ □ A (M) a. A A-1 A A-2 □ A-3 □ A-4 □ ♦ A □ A-5 . a % Opal d180 d13C % Sand ■ r Crit. A-6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 233 precipitation of a carbonate mass often exceeding body mass. During intermolt “normal” periods, ostracodes and other crustacean juveniles exceed their basic metabolic requirements for ongoing internal growth and storage of reserves for the subsequent molting cycle (Skinner, 1985). In most crustaceans, this energy reserve increases signifantly prior to proecdysis, with glycogen, fatty acids, and glycerol concentrated in the hepatopancreas and, to a lesser degree, epithelium and subepithelium connective tissue (Travis, 1955; Skinner, 1962). During proecdysis, the epidermis separates from the exoskeleton (apolysis) as epidermal cells enlarge and secrete new pliable epicuticle and exocuticle. These epidermal cell then reduce in size during ecdysis, whereupon the animal emerges from the molt with a “pre-produced” exoskeletal framework that is inflated by hemocoel uptake of water. In ostracodes, this framework is then impregnated with massive amounts of seawater-derived low-Mg calcite by the epidermal cells (Turpen and Angell, 1971). Previous ecological studies have demonstrated that the environment can exert a major effect on ostracode developmental speed and body size (Theisen, 1966; Ishizaki, 1975; Kaesler, 1975; Heip, 1976; Martens, 1985; Latifa, 1987; Kamiya, 1988; Schweitzer and Lohmann, 1990). In general, lower temperatures decrease developmental rate, but increase body size, while increased food availability also favors larger size. Based on these general physiological patterns, the relationship of observed variations in the size of Algulhasina quadratica may be further examined within the Site 744A paleoenvironmental framwork. Through the Oi-1 event, oxygen isotope values reflect some combination of ice- volume and bottom-water cooling, which Zachos eta/. (1996) estimated to be no more than 3-4 °C. Assuming that bottom-waters cooled at the onset of Oi-1, with Oi-1a and Oi-1b largely reflecting an ice-volume effect, the effectiveness of the oxygen isotope record as a bottom-water temperature proxy is further reduced. This estimation Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 234 strengthens the already statistically significant correlation of biogenic opal (i.e., organic flux) as the primary control upon body size. Under higher organic supply, benthic environments become increasingly undersaturated with respect to carbonate as organic material is degraded and XC 02 increases. However, when calcification conditions are bleakest during relative maxima in biogenic opal and fragmentation, valve sizes are the largest. This pattern again indicates a primary role for food availability on the realized body size of Algulhasina quadrata instars during the Eocene-Oligocene transition at Site 744A. 4.6. Conclusions Morphometric examination of seven discrete ontogenetic stages in the ostracode Algulhasina quadrata at Site 744A documented significant size increases in more abundant and evenly distributed instar stages after the onset of the Oi-1 event. The size increase was not accompanied by obvious changes in valve architectural changes or increases in valve-secreting epidermal cells as preserved by reticulation patterns. The significant correlation of body size with % biogenic opal in sediments supports a model where increased food availability allowed increased instar size through sequestering of greater metabolic reserves. However, such hypotheses are difficult to test without independent data on ontogenetic timing of development. This body size:surface productivity linkage is also consistent with the biogenic opal record as a primary signal of surface productivity and not a preservational artifact. Across the Oi-1 transition from relatively low-organic carbon/carbonate- saturated/slightly warmer to relatively high-organic carbon/carbonate- undersaturated/possibly slightly cooler bottom-waters, increased food availability appears to have “ won the battle” over decreased carbonate availability to produce generally larger individuals throughout the later ontogeny of Algulhasina quadrata. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 235 4.7. Future Directions This study documented significant changes in mean body size for a majority of Algulhasina quadratica instars through a major unique perturbation during the Eocene- Oligocene transition. A similar morphometric approach with this and other ostracode taxa could be applied to longer time-scales containing cyclic environmental changes, such as those driven by Milankovitch forcing. Cronin and Raymo (1997) and Cronin et at. (1999) have identified significant ostracode diversity changes at such scales. Therefore, it is not unreasonable to presume that sufficient environmental changes exist to promote intraspecific morphological changes. Examination of such change would allow the degree of ostracode ontogeny-environment coupling to be more explicitly tested. For example, do ontogenetic trajectories track or transcend cyclic paleoenvironmental proxies through time? The realized spectrum of such “lock-step” to “ratchet” patterns has implications for the hierarchy of ecophenotypic variability, natural selection, and genomic changes over ecological to evolutionary timescales. Complementing such historical studies, improved understanding of environmental and physiological controls upon instar size could be achieved through modern ecological studies. Regions of particular interest are those where environmental variability is restricted to a minimum number of parameters. For example, the relatively stable bottom-water mass along the East Pacific Rise is overlain by a major productivity gradient (c.f., benthic foram faunal analysis of the region by Loubere, 1991). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 236 5. Deep-ocean ostracode faunal response to the Late Paleocene Thermal Maximum: High-resolution (104 yr) records from the Southern Ocean (ODP Sites 690B and 738C) 5.1. Introduction 5.1.1. Scope Late Paleocene to Early Eocene global climate was the warmest of the Cenozoic, particularly at high-latitudes (Zachos et at., 1994), and is largely attributed to elevated atmospheric C 0 2 levels from increased ocean-crust production (Owens and Rea, 1985; Rea ef a/., 1990). Given the interval’s relatively low equator-to-pole temperature gradient, varying fluxes of warm saline bottom water (WSBW) are hypothesized to have formed at lower-latitudes through evaporation and strongly affected deep-ocean circulation and benthic environments (Chamberlin, 1906; Brass eta/., 1982; Kennett and Stott, 1990; Corfield and Norris, 1999). This scenario appears to provide the necessary latitudinal heat flux for the observed high-latitude geological and paleoclimatic data (Zachos et al„ 1993; but see Lyle, 1997; Sloan et al., 1998 for alternative scenarios and objections based on climate models). The globally decreasing, then increasing, trend in the early Paleogene S1 8 0 record of benthic forams is consistent with WSBW production, as increased bottom water temperatures (BWT) would dominate the benthic foram 51 8 0 signal over evaporation-related increases in seawater 8,aO. Imbedded within this long-term warming event and just below the Paleocene- Eocene boundary at -55.5 Ma, high-resolution faunal and isotopic studies of benthic foraminifera have identified a transient rapid increase in global temperatures termed the Late Paleocene Thermal Maximum (LPTM; Figure 5.1) (Kennett and Stott, 1991; Stott, 1992; Thomas and Shackleton, 1996). This rapid warming has been attributed to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 237 Figure 5.1 Late Paleocene (-55 Ma) paleogeography of the Southern hemisphere and locations of ODP Site 690B and 738C. (after Lawver et al., 1992) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 239 massive temperature-dependent sublimation of shelf and slope clathrates {i.e., methane hydrates bearing extremely negative 81 3 C values of roughly -32%o; Kvenvolden, 1993; Dickens et al., 1995) as shallow waters warmed in response to increased volcanogenic input of atmospheric C 0 2. These events led to a strong pulse in WSBW production, with high-latitude BWT estimated to have increased by a minimum of 4 C° and possibly up to 15 °C, depending upon evaporation-related seawater 5'8 0 decrease at WSBW production sites (Kennett and Stott, 1991; Zachos et al., 1993; Bralower et al., 1995). The LPTM had a major impact on the biosphere. In the marine realm, planktic organisms responded through major latitudinal shifts, short-term evolutionary events, and increased turnover (Shackleton et al., 1985; Lu and Keller, 1993, 1995a, 1995b; Kelly et al., 1996, 1998; Aubry, 1999; Boersma et al., 1999). The event had a catastrophic impact on benthic organisms, with a global, geologically instantaneous extinction of up to 50% of all benthic foraminifera species (e.g., Kennett and Stott, 1990; Kaiho et al., 1996; Bralower et al., 1995; Thomas and Shackleton, 1996; Thomas, 1998). The effects of the LPTM were not restricted to the oceanic realm. The - 2.5%o 8’3 C excursion is also recorded in terrestrial ecosystems and coincides with the global extinction of numerous archaic mammal orders, the first North American and Eurasian appearances of many new mammal orders, and major perturbations in floral assemblages (e.g., Koch et al., 1995; Wing et al., 1995; Clyde and Gingerich, 1998). Given the biological impact of this event on deep-ocean communities, it is somewhat surprising that the response of deep-ocean ostracodes has not been examined in more detail, an exception being Steineck and Thomas’s (1996) study at Site 689B (Maud Rise). Ostracodes may provide an important metazoan-based test for current LPTM hypotheses and scenarios based largely on benthic foraminifera. In addition, the pronounced environmental perturbations provide a means to examine changing paleoecological preferences of common and long-lived deep-ocean taxa. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 240 This study provides a high-resolution (103 -104 yr) faunal analysis of deep-ocean ostracode response to the LPTM at Sites 690B (Maud Rise; paleodepih -1,000 m deeper than 689B) and 738C (Kerguelen Plateau), both located in the Southern Ocean (Figure 5.1). At each site, most established taxa become extremely rare at the onset of the LPTM, followed by the appearance of new taxa restricted to specific intervals within the stable isotope excursion. Following the event’s end, faunas tend to return towards initial compositions. These rapid and apparently synchronous faunal changes between sites imply some combination of rapid dispersal, nearby refugia, or maintenance of reproductive continuity at extremely low densities. 5.1.2. Deep-ocean ostracode response Very little is known about the global response of ostracodes to the LPTM relative to benthic foraminifera, the only other common deep-ocean benthic fossil group. Most Paleogene ostracode faunal studies have used sampling intervals of 106 years or greater, and therefore likely do not sample or recognize any potential response to the event. Benson et al. (1984; 1985) and Benson (1990) reconstructed the global history of deep-ocean ostracodes at roughly million year intervals from 1,600 50 cc samples from 155 DSDP Sites spanning the Cretaceous and Cenozoic (Figure 5.2), focusing on five major events in the ostracode record (i.e., Cretaceous-Tertiary boundary, Eocene- Oligocene boundary, Middle Miocene, Messinian Salinity Crisis, and a 3.5 Ma event). However, just prior to the Paleocene-Eocene boundary (-60 Ma on Figures 5.2), there is a precipitous drop in average ostracode abundance, richness, and Shannon-Weiner values, as well as the percent of samples with ostracodes. The observed rates of change are much slower than that of the LPTM climate event, but indicate that ostracodes as a whole became globally rarer and less diverse in the latest Paleocene. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 241 Figure 5.2 Global summary of ostracode faunas from 1,600 50 cc samples from 155 DSDP Sites spanning the Cretaceous and Cenozoic. Letters A-E are A, K/T boundary; B, Eocene-Oligocene origin of the psychrosphere; C, Middle Miocene “event”; D, Messinian Salinity Crisis; and E, the 3.5 Ma event, (from Benson, 1990) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. % ca m p les w > h ostracods slanting g e n e ric dversly 242 8 S 8 8 ? 8 8 V v o 10 eueooyv oumoO io I e u e o < > 3 » » T Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S - W d v eraiy I*** tunover 243 Future re-evaluation and expansion of this data set using current age models will test the synchroneity of these changes and their temporal proximity to the LPTM, providing a global framework in which to incorporate additional high-resolution deep-ocean ostracode faunal data. In the southern hemisphere, ostracode studies have focused largely on Quaternary and living deep-ocean faunas (e.g., Dingle et al., 1989, 1990; Ayress et al., 1997), with only two published pre-Neogene studies for latitudes higher than 35° S (i.e., LPTM of Maud Rise; Steineck and Thomas, 1996; Maastrichtian of the Atlantic sector of the Southern Ocean, Majoran et al., 1997, 1998). Additional low-resolution (106'7 yr) data exist as dissertations conducted at the University of Wales, Aberystwyth, under the advisement of R. C. Whatley, but remain largely unpublished (e.g., Millson, 1987; Balman, 1997). Steineck and Thomas (1996) examined deep-ocean ostracode genera across the LPTM event at Maud Rise (Site 689B; mean sampling resolution = 90 kyr) (Figure 5.3). The mid-bathyal site records a faunal turnover coincident with the LPTM’s isotopic extrema. A pre-LPTM assemblage, dominated by heavily calcified, epifaunal taxa, was replaced for -25-40 kyr by a novel assemblage of smaller, less- calcified, ecological generalists. This “disaster fauna” was subsequently replaced by the return of many pre-LPTM taxa, although generally smaller and less calcified. The faunal record is interpreted as a biological response to the initial incursion and lingering effects of low-latitude, 0 2 -depleted, C 0 2-enriched bottom-waters during the LPTM. This study builds upon that of Steineck and Thomas (1996) by expanding ostracode faunal data bathymetrically (i.e., Site 690B; -1,000 m deeper) and geographically (i.e., Kerguelean Plateau; ODP Site 738; -90° longitude to the west). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 244 Figure 5.3. Response of deep-ocean ostracodes at Site 689B, Maud Rise. Note that ostracode abundances increase from right to left. Calcareous nannoplankton and benthic foram zones from Thomas et al. (1990). Age model after Thomas and Shackleton (1996), using stratigraphic interpretation of Aubry et al., 1996. Stable isotopes of benthic foram Lenticulina from Thomas and Shackleton (1996). OTG 5 consists of degraded specimens of Dutoitella, Oertiella, and Pelecocythere. OTG 6 consists of aggraded forms of Abyssocypris, Veenia, Abyssocythere, Dutoitella, Oertiella, and Pelecocythere. Of these, Steineck and Thomas consider all but Abyssocypris to be epifaunal. (from Steineck and Thomas, 1996). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. o c_ 3 2 3 o lo.£ 0 0-18, °/io t 5 . I s> C / 3 < 50 40 30 20 202 1 (55.40) 2 (55.43) 3 (55 46) 4 (55.-: F I 204 OTG 5 206 5 (55.49) 6 (55.50) 208 o o CL - 7(55.67) - 8(55.82) _ 9(55.96) “ 10 (55.98) 8 2 1 0 OTG 6 214 216 -11 (56.26) 218 ' CL -12 (56.37) 220 30 20 10 0 40 30 20 10 0 70 60 50 50 30 50 40 30 2010 0 Percent • C-13,% r o oi 246 5.2. Materials and Methods 5 .2.1. Site descriptions ODP Hole 690B was drilled on the southwestern slope of Maud Rise (65°9.63'S, 1°12.30'E) at a water depth of 2,914 m. Advanced piston coring (APC) produced effectively 100% recovery of a Late Paleocene sequence of almost exclusively calcareous biogenic sediment (Barker et al., 1988b). Late Paleocene paleodepths are estimated at -2,000 meters, shallower than the estimated average paleosurface of the carbonate compensation depth (CCD) (Thomas et al., 1990; Kennett and Stott, 1990; Van Andel, 1975). ODP Hole 738C was drilled on the southern slope of the Kerguelen Plateau (64°42.54'S, 82.° 47.25'E) at a water depth of 2,252 meters. A rotary core barrel (RBC) produced an 11R section with less than 50% recovery of the Late Paleocene sequence of calcareous chalk with chert nodules and fragments (Barron et al., 1989b). Late Paleocene paleodepths are estimated at -1,300 meters, also shallower than the estimated paleosurface of the CCD (Barrera and Huber, 1991) and roughly equivalent in depth to Steineck and Thomas’s (1996) study of Site 689B. 5.2.2. Age models The age model for Site 690B is based on three datums and assumes constant inter-datum sedimentation rates. The oldest datum is the C24r/C25n boundary, located at a mean depth of 185.47 mbsf (bounding paleomagnetic sample intervals of 185.25 and 185.70 mbsf; Spiess, 1990), with a Geomagnetic Polarity Time Scale age of 55.904 Ma (Berggren et al., 1995; Cande and Kent, 1995) The middle age-model datum is the abrupt benthic foram extinction horizon, taken at 170.48 mbsf, to which Aubry et al. (1996) assign an age of 55.5 Ma. The youngest datum is the NP10/NP9 boundary at a mean depth of 148.90 mbsf (bounding nannoplankton sample intervals of 148.40 and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 247 149.40 mbsf) and assigned an age of 55.0 Ma (Aubry et al., 1996). However, disagreement exists regarding the magnetobiochronology of the Late Paleocene-Early Eocene interval at Site 690B; the most recent reviews and arguments by selected workers are provided in Aubry et al. (1996) and Berggren and Aubry (1999). Age datums for the Site 738C study interval are rare given its position within a long coring interval of extremely poor recovery. In addition, ship-board paleomagnetic determinations for this and many other Site 738C intervals were subsequently deemed unreliable due to magnetometer malfunctions. Lu and Keller (1993) provide ages for samples examined in this study, however, their biochronological basis is not explicitly outlined. Therefore, Site 738 samples are examined by depth interval (mbsf) alone, with no attempt to assign absolute ages or to correlate temporally to Site 690B. 5.2.3. Paleoenvironmental framework Previous studies of the LPTM at Sites 690B and 738C provide an independent paleoenvironmental framework in which to examine the faunal response of deep-ocean ostracodes. Kennett and Stott (1990) determined average 51 s O and 81 3 C values for planktic (Subbotina and Acarina) and benthic (Nuttallides truempyi) foraminifera at Site 690B (Figure 5.4). Barrera (unpublished data) determined average 81 s O and 5'3 C values for planktic (Acarina)) and benthic (Nuttallides and Cibicidoides) foraminifera for Site 738C (Figure 5.5). These stable isotope records document the pronounced step increase in isotope values at the onset of the LPTM, known as the carbon isotope excursion (CIE). The higher-resolution stable isotope decrease at Site 690B is very abrupt, with lighter 81 8 0 values preceding lighter 81 3 C values. In contrast, 51 s O and 51 3 C values at Site 738C show what appear to be a much more gradual onset of the CIE, roughly equivalent in its rate of increase to its subsequent decrease. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 248 Figure 5.4 Paleoenvironmental framework for Site 690B. Isotopic data from Kennett and Stott (1990). Ages based on paleomagnetic and biostratigraphic data as discussed in text. Sedimentological data from Kennett (pers. comm.). White triangles represent stratigraphic position of 15 cc samples examined for ostracodes. Pre-LPTM, LPTM, and Post-LPTM boundaries are based on the onset of the CIE and the return of values to relatively stable levels. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. % Sand Ma 54.400 Clay Minerals 54.425 ■ ■ Post- LPTM 55.450- 55.475 LPTM 55.500 • Pre- LPTM 55.550 55.575- N. truempyi Subbotina spp. M. praepentcamerata — Smectite Kaolinite — Illite 55.600 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 %0 % % %0 2 4 9 250 Figure 5.5 Paleoenvironmental framework for Site 738C. Isotopic data from Barrera (pers. comm.). White triangles represent stratigraphic position of 5 cc samples examined for ostracodes. Pre-LPTM, LPTM, and Post-LPTM boundaries are based on the onset of the CIE and the return of values to relatively stable levels. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mbsf A A M A A A A A AAA2A A t ■ ■ ‘ 1 1 1 ' o ■ ■ 1 ■ ■ in ■ ■ 1 © in in CO 00 00 00 00 C M C M CM C M in o CO 0 0 00 C M C M Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 8 6 .5 + 252 The apparently more gradual isotopic recovery at Site 738 is considered an artifact caused by no isotopic data between 284.95 and 284.52 mbsf. Future isotope analyses will likely reveal a relatively rapid and early isotopic recovery consistent with the Site 690B pattern and the inter-site faunal patterns presented below. These isotope stratigraphies served to partition the records into pre-LPTM, LPTM, and post-LPTM intervals, with breaks at the onset of the CIE and its subsequent return to relatively stable, if elevated, values. For Site 738C, the LPTM/post-LPTM boundary was arbitrarily defined just below the oldest isotopic data after the gap in data, which is equivalent to pre-LPTM values. Sedimentological data for the sites are scarce; only clay mineral (Robert and Kennett, 1994) and percent sand (Kennett, unpublished data) data exist for Site 690B. 5.2.4. Faunal identification and tallying All ostracode valves and valve fragments were picked from the >63 urn size- fraction of 22 Site 690B samples (15 cc sample volumes; mean sampling interval = 23±13 cm; 6±3 kyr) and 19 Site 738C samples (5 cc sample volumes; mean sampling interval = 12±10 cm, excluding the uppermost sample - see Figure 5.5). The total time-interval examined for Site 690B ranged from 55.427-55.549 Ma (167.35- 172.28 mbsf) and the interval examined for Site 738C ranged from 284.48-286.56 mbsf, with a relatively much higher sample at 283.56 mbsf. Valve material was assigned to genus, and where possible species, through reference to primary literature, unpublished works (e.g., Millson, 1987; Balman, 1997), and systematic collections in the Smithsonian NMNH. Individuals in each taxon were sub-divided into four valve types: left adult, right adult, left juvenile, right juvenile; sexes were not differentiated. Valve fragments exceeding -50% completeness were tallied as complete individuals (e.g., adult left valve). Less complete, but still identifiable, valve fragments were Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 253 tallied by equating two 25-50% complete fragments (F1 in Appendix 2) or three <25% complete fragments (F2 in Appendix 2) to one complete valve. Fragments deemed unidentifiable were retained, but not considered further. Left/right and adult/juvenile ratios show no obvious temporal trends in any taxon. Given the paucity of comparable high-southern latitude Paleocene ostracode systematic studies and the short geologic interval examined, formal systematic description was not pursued. Morphologically cohesive groups of individuals not confidently assignable to known taxa were defined as operational taxonomic units using the letters A-C. For many deep-ocean ostracode genera, only one species is generally present at a given locale; exceptions in this study include, at Site 690B, Pennyella, Propontocypris, Eucytherura, Munseyella, Krithe, Cytheropteron, and Cytherella spp., with the first four represented by two species each and the later three represented by an indeterminate number of species. At Site 738, Eucytherura and Pennyella are represented by the same two species as at Site 690B, while Krithe and Cytherella are represented by an indeterminate number of species. To maximize specimen abundance and information content, all faunal analyses were based on the summation of each taxon’s left adult, right adult, left juvenile, and right juvenile valves, along with fragment- based “individuals” from the F1 and F2 categories rounded up to the next whole number. This summation approach likely increases the relative abundance of taxa with relatively abundant juveniles, but ensures that rarer taxa are not discounted. 5 .3 . Results 5.3.1. Primary data Faunal abundance matrices of ostracode taxa (i.e., left adult, right adult, left juvenile, right juvenile, F1, and F2) are presented in Appendix 2. Absolute (total number of individuals) and relative (% of sample) abundances are plotted in Figure 5.6. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 254 Figure 5.6 Absolute and relative abundances of taxa at Site 690B and 738C. Absolute values in black; relative values in gray. For each site, the upper-axis hash-marks indicates 10 individuals and lower-axis hash-marks indicate 10% of the sample. Note that absolute hash-mark distance vary between taxa, but are constant between sites for each taxon (except for Dutoitella, Cytherella spp., and Krithe spp.). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 255 U l 00 a O s o > <0 o o C O o o G O £ £ to lA Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 287.0 Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. M i 55.40 ...f° 9 'tten/ra * Henryhowella Peiococythere Mayburya Rimacytheropteron Indeterminate Genera Site 690B Eucythewra sp. B Philoneptunus Cytheropteron spp. Profundobythere Palmoconha A B C - 1.0 0.0 1.0 (-■ 55.42 • 55.44 55.46 55.50 5554 55.56 • 5556 nM Site 738C 283.5 - ....................................... 284.0 284.5 285.5 286.0 286.5 287.0 I t • ■ I- )-... -f— H -t i Algulhasina r - 2 5 6 257 Most taxa may be placed within three distributional patterns with respect to the LPTM. These patterns are generally consistent at both sites for any given taxon; emphasis is placed on the more robust record at Site 690 (i.e., higher overall abundance, etc.), with exceptions noted for Site 738. These three distributional patterns are: 1. Largely unaffected — Roughly 25% (7/29) of all taxa from the two sites appear largely unaffected by the LPTM, except for moderate changes in absolute and relative abundance through time. These taxa consist of Propontocypris sp. A, Pennyella sp. B, Krithe spp., Eucytherura sp. A and sp. B, Cytheropteron spp., Mayburya, and indeterminate genus B. Most of these taxa show abundance minima coincident to the total abundance minimum just prior to the onset of warming and the carbon isotope excursion (CIE). Exceptions and notable details to this general pattern include the abundance maximum of Propontocypris sp. A during and after the LPTM interval, the relative abundance maxima of Pennyella sp. B at the CIE, the decreased relative abundance of Krithe spp. during the LPTM, and the progressively decreased abundance of Eucytherura spp. Aversovalva hydronamica is provisionally included in this pattern, being rare prior to and throughout the LPTM, after which it is largely absent. 2. Predominately pre-CIE — Over 60% (10/26) of all taxa common prior to the CIE are absent to extremely rare following its onset. Taxa with local LAD coinciding with the CIE at Site 690B are Bairdoppilata, Philoneptunus, Rimacytheropteron, and the indeterminate genera A and C.. Taxa that are rare (i.e., isolated post-CIE occurrences at very low abundance) following the CIE are Propontocypris sp. B, Bairdoppilata (at Site 738C), Bairdia, Dutoitella, Henryhowella, Pelecocythere, and Algulhasina (present only at Site 738C). In some cases, taxa largely absent from post-CIE intervals show a recovery pattern towards pre-CIE abundances in the later portions of the two records (e.g., Dutoitella at Site 690B; Bairdoppilata, Bairdia at Site 738C). At Site 738C, this recovery coincides with the most abundant and taxonomically rich sample, likely Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 258 reflecting some combination of ecological recovery and sedimentological changes affecting relative accumulation rates of ostracode-diluting grains (e.g., planktic foraminifera). Note that this last sample occurs very much later in the section, probably after an absolute time interval equivalent to the entire LPTM interval. 3. Predominately post-CIE — Some taxa are largely restricted to portions of the post-CIE interval or within LPTM interval. Eucythere relative abundance increases drastically at the CIE, followed by a slight decrease as total abundance recovers. Cytherella spp. reaches its maximum abundance in the later LPTM interval. Munseyella sp. A and sp. B are restricted to the LPTM interval, where this genus and Pennyella sp. B predominate the fauna (-80% ). Palmocohna first appears slightly later than Munseyella spp. in the LPTM interval, is most abundant early in the Post-LPTM interval, and disappears from Site 690B in the last two samples. Total valve abundance and richness data are provided in Figure 5.7. At Site 690B, abundance plummets from an average of -200 valves to less than five at the onset of the CIE, recovers unevenly towards initial values through the LPTM, then decreases again followed by an increase during the post-LPTM interval. Site 738C ostracode abundance shows a similar inflection pattern to that at Site 690B, although significantly lower overall and with later LPTM abundances returning to pre-LPTM levels. Note that the relative rate of the CIE onset appears much less rapid at Site 738C. If the injection and incorporation of methane-derived light 81 3 C into the ocean-atmosphere system was geologically instantaneous and the isotopic determinations are accurate for both sites, then a significant increase in Site 738C accumulation rate, a decrease, hiatus, or unconformity at Site 690B are highly possible. Richness, the simple tally of the number of taxa per sample (Peet, 1974), generally varies with abundance (Figure 5.7). At Site 690B, richness is high (-20 taxa), rapidly decreases at the CIE, and then increases for the remainder of the LPTM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 259 Figure 5.7 Ostracode abundance and richness at Sites 6906 and 738C Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Nuttallides truempyi Abundance > 0 100 Ma 54.400 Nuttallides truempyi Abundance 200 300 mbs* ■"t 283.5 100 54.425 -■ Post- LPTM 284.0- 55.450- 284.5- £ 55.475 ■■» 285.0 LPTM 55.500 285.5 55.525 > 286.0- 55.550 ■> Pre- t> LPTM LPTM I 286.5-> 55.575- 287.0 55.600 - 2.0 - 1.0 0.0 1.0 r; 10 0.0 0 - 2.0 - 1.0 1.0 0 0 1 0 .) \K .hf ti 260 261 interval and post-LPTM, followed by a slight decrease. Site 738C richness shows an uneven decrease preceeding the CIE and then a return to initial richness in the latest LPTM interval. As at Site 690B, there is a slight decrease in the early LPTM interval. 5.3.2. Sample size and richness relationships: Rarefaction assessment Comparison of the above ostracode abundance and richness records demonstrate a persistent phenomenon: sample size can strongly affect observed richness (Figure 5.8; Connor and McCoy, 1979). This “species-area effect” precludes direct richness comparison for samples of differing abundance. The rarefaction method estimates how many taxa would be “captured” if n specimens were randomly sub-sampled from a sample population of N abundance (Sanders, 1968; Hurlbert, 1971). Plotting the “rarefied” richness of different samples for n = 1 to each sample’s N produces a set of hypergeometric curves that allow richness comparisons at any common n. Rarefaction values are deterministic, being a direct function of number of taxa per sample and their relative frequency (i.e., evenness). The method is applicable here given the similar taxonomic composition, mode of collection, and paleoenvironmental context of samples (Raup, 1975; Tipper, 1979). The analytical expressions for rarefaction (E (S n )) is: s E<S„ )= I i = l 1 - N — N i n N n (Hurlbert, 1971), where: S is the number of taxa in the sample population (richness), N is the number of individuals in the sample population, Nj is the number of individuals in the / th taxon, and n is the number of individuals in the rarefied sub-sample population. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 262 Figure 5.8 Ostracode abundance versus richness for Sites 690B and 738C. Site 738C samples are less diverse and abundance relative to Site 690B, which may relate to a 33% larger sample volumes of Site 690B. Note that neither show a strong asymptotic trend to some stable richness with increasing abundance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Richness 263 20 - 18 | ! i 16 ! ! A A A 14 • A A 1 2 A A • A A 10 A A • A • A A A 8 ! • A A • I 6 • i • 4 i I 2 |« • 0 e---------- A690B • 738C 25 50 75 100 125 150 175 200 Valve Abundance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 264 In the E(Sn ) equation, the binomial coefficient N ~ Ni represents the number of n ways of choosing n specimens from N - specimens (i.e., all specimens not in the Ah species), whereas the binomial coefficient N represents the number of ways of choosing n specimens from all specimens N. The entire pre-summation term represents the probability of choosing a given taxon / when sub-sampling n specimens. Summing these probabilities from I = 1 to S provides a deterministic estimate of richness if n specimens were sub-sampled from the original population N. £(S n )values were calculated using the computer program Mathematical program code is provided in Appendix 3. For each site, rarefaction curves were calculated for pooled sample populations using a calculation series of n = 10, 20, 30, . . . N. In addition, individual samples were rarefied using a calculation series of n = 1, 2, 3, . . . N. Rarefaction-based total richness of Site 690B exceeds that of Site 738C at all common sub-sample sizes (Figure 5.9a). At Site 690B, richness decreases from the pre-LPTM interval through the LPTM interval and into the post-LPTM interval (Figure 5.9b). In contrast, Site 738C richness is roughly equal during pre- and post-LPTM intervals, with the earlier LPTM interval samples much lower (Figure 5.9c). These patterns are also reflected in rarefactions of individual samples (Figures 5.10 and 5.11). As discussed previously, the Site 738C pattern is strongly controlled by the isotope-based divisions. Based on the site’s general faunal pattern, it is likely that samples LPTM-9 through LPTM-12 (marked with asterisks in Figure 5.11) would likely be classified as post-LPTM given additional isotope analyses and would show abundance and richness patterns very similar to the pre- and post-LPTM samples. The species-area effect was also examined by a bootstrapping approach (Appendix 4), which produced results consistent with those from rarefaction. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 265 Figure 5.9 Rarefaction curves for entire sites (a) and for LPTM divisions each site (b, c). . Values are based on a calculation series of n = 1, 2, 3, . . . N and stratigraphic divisions are as in Fig. 5.4. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R i c h n e s s R i c h n e s s R i c h n e s s 266 A. 3 0 2 5 20 1 5 10 5 0 6 9 0 B (n=22) 7 3 8 C (n=19) Rarefaction of Sites 1 5 0 0 2000 2 5 0 0 3 0 0 0 5 0 0 0 1000 n 3 0 2 5 20 1 5 10 5 0 __________________ Pre-LPTM LPTM < n=8> "(n=8) " ▼ K r Post-LPTM V (n=6) Rarefaction of Site 690B by LPTM Event 2 5 0 5 0 0 7 5 0 n 1000 1 2 5 0 1 5 0 0 3 0 2 5 20 1 5 10 5 0 P o s t - L P T M ( n = 2 ) T L P T M ( n = 1 2 ) f P r e - L P T M Rarefaction of 1 ( n = 5 ) Site 738C by LPTM Event 2 5 0 5 0 0 7 5 0 1 0 0 0 1 2 5 0 1 5 0 0 n Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 267 Figure 5.10 Rarefaction curves for individual Site 690B samples. Values are based on a calculation series of n = 1, 2, 3, . . . N and stratigraphic divisions are as in Fig. 5.4. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Richness Richness Richness 268 Site 690B Rarefaction of Individual Samples 25 r 20 Post-LPTM o 50 100 200 250 300 150 25 r 20 • 1 5 5 0 0 50 100 200 150 250 300 25 20 Pre-LPTM o 50 100 150 200 250 300 Sub'Sample Interval n Reproduced with permission of the copyright owner. Further reproduction prohibited without permission 269 Figure 5.11 Rarefaction curves for individual Site 738C samples. Values are based on a calculation series of n = 1, 2, 3,... N and stratigraphic divisions are as in Fig. 5.5. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Richness Richness Richness 270 Site 738C Rarefaction of individual Samples 14 12 10 8 6 4 Post-LPTM 2 0 150 25 50 100 125 0 75 14 12 Note: Samples 1, 2, 4 not shown given very low abudance and richness. 10 8 6 4 2 0 25 50 125 150 0 75 100 14 8 6 4 Pre-LPTM 2 0 25 50 125 0 75 100 150 Sub-Sample Interval n Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 271 5 .3 .3 . Relation of faunal abundance and richness to paleoenvironment Faunal abundance and richness of each stratigraphic sample were compared through bivariate plots to their 51 8 0 and 51 3 C values (Figure 5.12). This approach provides an alternate, non-stratigraphic view of faunal-isotopic relationships. Only faunal samples with coeval stable isotope data are presented (i.e., 73% of 690B samples, 58% of 738C faunal samples). At Site 690B, the major drop from high abundance and richness coincides with the onset of the 5,80 excursion and precedes the onset of the 81 3 C excursion. Both abundance and richness are lowest at the negative 5,80 extrema and then return irregularly toward, but never attain, initial values (but neither does 81 8 0 ). A similar pattern is present with respect to 51 3 C. At Site 738C, abundance and richness show a less-pronounced relationship to the isotopic records, with their minima generally coinciding with the initial LPTM decrease in isotopic values, followed by increasing values through the 5,80 and S1 3 C relative minima. 5.3.4. Diversity analysis In all of ecology, the term “diversity” itself probably contains the greatest diversity of potential meanings beyond that of simple taxonomic richness (sensu strictu). Its plethora of denotations and connotations has grown from the general concept’s application in both taxonomy (i.e., number of taxa; e.g., “richness”) and ecology (i.e., resource partitioning; e.g., “evenness”). In both applications, the term may be invoked at three hierarchical levels (alpha, beta, gamma), each addressing different spatial scales and ecological concepts. Although diversity indices are a powerful means to standardize and summarize large amounts of ecological information, care must be taken to select indices appropriate to the question at hand. For a single sample (i.e., community), alpha diversity indices exist for richness (i.e., Margalef’s index, Menhinik’s index, log-log index, Fisher’s alpha), evenness (i.e., Simpson’s Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 272 Figure 5.12. Stable isotope values versus ostracode abundance and richness for Sites 690B and 738C. Plotted points represent only those faunal samples with coeval stable isotopic data (i.e., 16 of 22 [73%] for 690B; 11 of 19 [58%] for 738C). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S ta b le Is o to p e R ela tio n s h ip s t o O s tra c o d e A b u n d a n c e a n d Richness 2 73 CD O 0> C O c o o © c\ i o o O CO o o o o o o o o t o o i n o t o W (M ( \ i r r O O o o c v i aouepunqv C D O O C O 05 C O < 0 N O o o i n o o O 00 o ^ i n o o o m o i n o o o o o o m o i n C O C M C M »** ^ aouepunqy O C O ( O ^ C \ J O C O ( O ^ C \ I O C\l ssauqojy C D CJ 0 C O 01 C O < £ > N Q O O ( O ^ O I O qO ( O ^ C M O W sseuipiy Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0 .5 0 0 .0 0 -0.50 -1.00 -1.50 -2.00 2.0 0 1 .0 0 0 .0 0 -1.00 -2.00 81 8 0 613C 274 index, Berger-Parker index, McIntosh index), and a combination of the two (i.e., Shannon index, Brillouin index). Each index’s appropriate use, as well as strengths and weaknesses are discussed at length by Margalef (1958), Magurran (1988), and Hayek and Buzas (1997). A problem common to all indices is that they are sampled from unknown distributions, which precludes assignment of probability values. Consequently, there is no basis for evaluating the statistical significance of diversity index differences between any two samples. Note also that these indices are only sensitive to net changes in relative proportions — between two samples, a complete faunal turnover that produced a similar proportional distribution would not produce a change in index values. To examine ostracode community structure through the LPTM event, the abbreviated Shannon information function (H), its maximum value (Hm a x ), and evenness (E) were calculated for each sample using s h = - £ Pi in jp i) (Shannon, 1948), i = 1 H max = lniS) (Shannon, 1948), e* e = — (Buzas and Gibson, 1969), s where S is richness and p, is the proportion of the / th species (Figures 5.13). Shannon information function values are plotted with their maximum potential value to compare actual and potential values — convergence between these two measures indicates increased evenness. At both sites, relative maxima in evenness and H— Hm a x convergence coincide with the onset of the LPTM, indicating that total abundance decreased relative to richness. At Site 690B, Shannon information function values show a broad, weakly defined increasing then decreasing trend through the entire record, whereas Site 738C contains a reversed pattern with local maxima in the later portion of the LPTM interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 275 5.13. Site 690B and 738C stratigraphic variations in faunal evenness (E), Shannon information function (H), and its maximum potential value for a given sample (Hm a x ). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 690B > . . . i . . . i 54.400 54.425 55.450 55.475 55.500- 55.525- 55.550 ■ ■ 55.575- 55.600 0 1 2 3 4 Diversity Indices mbsf Site 738C Post- LPTM 284.0 284.5 285.0 LPTM ■ 285.5 286.0 Pre > LPTM 287.0 - 2.0 - 1.0 0.0 % 0 1 2 Diversity Indices 2 7 6 277 5.3.5. Faunal Turnover Turnover, or change in the constituents of a community through time, is another metric of potential use for examining ecological response to environmental change. Turnover measures are generally calculated from data collected by one of two distinct sampling strategies: pairwise, where censuses taken at the beginning and end of an otherwise unsampled interval are compared, and cumulative, where all appearances and disappearances are tallied within an interval and adjacent intervals are compared. Pairwise turnover was calculated for each site using the following indices: Appearances(A = > B) + DisappearancefeA => B) Intersample Turnover=---------— ----------------------------------, £ (TaxapresentinA andB) and Appearances(A => B ) + Disappearance^ => B ) Turnover rate= —— -------------------------------------------, £ (TaxapresentinA andB ) x Censuslnterva where A = older stratigraphic sample, and B = younger stratigraphic sample. At Site 690B, the number of taxonomic appearances and disappearances from each older sample to the stratigraphically adjacent younger sample, as well as the directly related turnover and turnover rates, all show maximum values coinciding with the CIE that decrease slightly to approximately pre-LPTM values by late in the LPTM interval (Figure 5.14). In contrast, Site 738C turnover shows much less change, with only a slight increase in appearances and disappearances at the CIE and a relative maximum in appearances late in the LPTM interval (Figure 5.15). The major increase in appearances from the penultimate to last sample is difficult to interpret given their large stratigraphic distance, which is nearly equal to the entire LPTM interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 278 Figure 5.14. Site 690B intersample ostracode appearances, disappearances, turnover, and turnover rate. Turnover calculated from number of appearances and disappearances from the previous sample divided by the total number of taxa in the previous and current sample. Turnover rate calculated from number of appearances and disappearances from the previous sample divided by the total number of taxa in the previous and current sample times the census interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Intersample Disappearances Turnover (per kyr) 11 0.2 Nuttallides truempyi Ma 5 4 .4 0 0 0.0 0.3 0.4 54.425- 55.450- 5 5 .4 7 5 -> 5 5 .5 2 5 ID IA 5 5 .5 5 0 - > 55.575- 5 5 .6 0 0 0.0 1.0 - 2.0 - 1.0 0 5 10 0 10 1 5 5 1 5 % o Intersample Appearances Turnover (Intersample) 279 280 Figure 5.15. Site 738C intersample ostracode appearances, disappearances, turnover, and turnover rate. Turnover calculated from number of appearances and disappearances from the previous sample divided by the total number of taxa in the previous and current sample. Turnover rate calculated from number of appearances and disappearances from the previous sample divided by the total number of taxa in the previous and current sample times the census interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nuttallides truempyi Intersample Disappearances 281 C M tfi in o in a ^ a. °i i n i n co C M in co co C M o o <b oo C M in co CM C O C M C O C M C O C M C O C M Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Intersample Appearances Turnover (Intersample) 282 5.3.6. Cluster analyses — stratigraphic and taxonomic Similarity coefficients and clustering algorithms were used to examine correlations between (1) each stratigraphic sample pair (i.e., taxa as variables) and (2) each taxon pair (i.e., samples as variables) at each site. To minimize effects due to sample size, rank abundance, etc., variable values were transformed into presence/absence form (i.e., 1/0) prior to similarity coefficient calculation (Cheetham and Hazel, 1969). Taxa occurring in a single sample (i.e., “stratigraphic singletons”) and stratigraphic samples containing no individuals were culled from presence/absence matrices. The Simple Matching and Jaccard similarity indices, two common and complementary ecological indices, were used to calculate similarity matrices for Sites 690B and 738C: Simple matching: sS M = —a + c— t a + b + c Jaccard: S j = —- — , a + b where: a = number of variable pairs with common occurrence (i.e., positive match), b = number of variable pairs with only one occurrence, c = number of variable pairs with no occurrence (i.e., negative match). The simple matching coefficient includes negative matches between two compared samples in both the numerator and denominator, whereas the Jaccard coefficient includes only positively matched pairs. Thus, simple matching coefficients are effective for examining stratigraphic (community) structure through time, whereas Jaccard coefficients are effective for identifying ecologically similar taxa within a range of (perhaps unrecognized or unquantifiable) paleoenvironmental conditions (i.e., not based on their mutual absence) (Kaesler, 1966; Maddocks, 1966; Mello and Buzas, 1968). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 283 Both the Jaccard and, to a lesser extent, the simple matching coefficients are affected by the aforementioned “species-area” effect, where rarer taxa present during a time interval are not recorded during its sampling. For example, given three taxa of congruent distribution, but order-of-magnitude abundance differences, the two more abundant taxa will tend to share higher similarity coefficients if sampling is incomplete. Similarity matrices calculated from presence/absence matrices for Site 690B and 738C stratigraphy and fauna are presented in Appendix 5; all matrix calculations were conducted through Systat 8.0 (PC-version). Clustering algorithms allow graphic examination of similarity matrices through hierarchical grouping of most similar variables into dendrograms. The method is largely exploratory, lacking the probabilistic underpinnings of related approaches such as discriminant function analysis, where the number and members of groups are known a priori. Often some similarity level is chosen above which clusters are considered “significant” (e.g., “> 0.5”), but these levels are arbitrary. Integration of bootstrapping into cluster analysis potentially provides a more robust means to evaluate the robustness of dendrogram branch points, but was not pursued for this study (Nemec and Brinkhurst, 1988). Stratigraphic and faunal similarity matrices for Sites 690B and 738C were clustered using average-linkage algorithms through Systat 8.0 (PC-version), with individual samples iteratively agglomerated into larger, more inclusive groups until all were contained within a single group. Average-linkage, also known as unweighted pair- group method with arithmetic averages (UPGMA), iteratively merges groups in order of the average similarity value of their constituent samples. For Site 690B, average- linkage of simple matching coefficients for stratigraphic samples produced three major sample clusters that typify pre-LPTM, LPTM, and post-LPTM intervals (Figure 5.16). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 284 Figure 5.16. Average-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 690B. Sample numbering consists of location within pre- LPTM, LPTM, or post-LPTM interval followed by relative stratigraphic position from base of given interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. L P T M Post-LPTM Pre-LPTM 285 Site 690B SM Coefficient— Average Linkage Pre-5 Pre-4 Pre-3 Pre-6 Pre-2 Pre-1 Pre-8 Post-8 Posi-5 Post-3 Post 4 Post-2 Pcst-1 LPTM-6 LPTM-8 LPTM-3 LPTM-7 LPTM-5 LPTM-4 LPTM-2 LPTM-1 Pre-7 I ------- 1 -------1 -------1 ------- 1 -------1 0.0 0.2 0.4 0.6 0.8 1.0 Distance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 286 Exceptions to these “correct” classifications are samples either stratigraphically adjacent to the interval within which they clustered (e.g., Post-1) or with very low richness (e.g., Pre-7, Pre-8, LPTM-1). This clustering hierarchy implies a return of a faunal composition similar to that of the pre-LPTM interval after the end of the LPTM. For Site 738C, the average-linkage dendrogram contains a more complex and difficult to interpret pattern, consistent with the less robust distributional patterns likely related to “missed” occurrences given the site’s smaller sample volumes, lower abundances, and poor post-LPTM sampling (Figure 5.17). A cluster of three of the five pre-LPTM samples is nested between two LPTM clusters, with the later including one post-LPTM sample each. The two LPTM clusters consist of older samples 2, 3, 5, 6, 7, 8 and younger samples 9, 10, 11, 12, with their cluster distance indicating significant faunal compositional change between LPTM-8 and LPTM-9. The average-linkage dendrogram of Jaccard coefficients for Site 690B taxonomic samples independently confirms the abundance-based general distributional patterns described earlier (Figure 5.18). Rimacytheropteron, Bairdoppilata, Philoneptunus, Henryhowella, and Indeterminate Genera A and C constitute a distinct cluster (1) that joins other taxa at a distance of ~0.75 — all of these taxa are restricted to the pre-LPTM interval. Taxa persisting through the entire study interval form a large cluster (4), while Profundobythere and Aversovalva form a small distinct cluster (5) consistent with their persistence through, and general absence following, the LPTM interval. Taxa restricted to the LPTM interval (i.e., Pennyella sp. B, Munseyella sp. A and sp. B) and the later LPTM/early post-LPTM intervals (i.e., Palmocohna) form the distinct cluster 6. In contrast, Site 738 shows little structure given the spotty distribution of many taxa (Figure 5.19). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 287 Figure 5.17. Average-linkage dendrogram of simple matching coefficient matrix for stratigraphic samples of Site 738C. Sample numbering consists of location within pre- LPTM, LPTM, or post-LPTM interval followed by relative stratigraphic position from base of given interval. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 288 Site 738C SM Coefficient— Average Linkage LPTM-9 LPTM-10 LPTM-11 LPTM-12 Pre-3 Pre-2 Pre-5 Pre-1 LPTM-8 LPTM-7 LPTM-6 — LPTM-3 LPTM-2 Pre-4 — LPTM-5 — Post-1 LPTM-4 I --------1 ------- 1 ------- 1 ------- 1 --------1 0.0 0.2 0.4 0.6 0.8 1.0 Distance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 289 Figure 5.18. Average-linkage dendrogram of Jaccard coefficient matrix for taxonomic samples of Site 690B. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 290 Rimacytheropteron i Indet. Gen. C ' Bairdoppilata Philoneptunus Henryhowella Indet. Gen. A Pelecocythere Propontocypris sp. B Mayburya Bairdi Dutoitella h Indet. Gen. B Pennyella sp. B Krithe spp. Propontocypris sp. A Cytheropteron spp. — Eucythere Cytherella spp. Eucytherura sp. A Eucytherura sp. B Profundobythere Aversovalva Pennyella sp. A Munseyella sp. B Munseyella sp. A Palmoconha Site 690B Jaccard Coefficient— Average Linkage ® ® h ® ® ® - ® i -------------1 ------------ 1 ------------ r~ 0.0 0.2 0.4 0.6 Distance 0.8 1.0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 291 Figure 5.18. Average-linkage dendrogram of Jaccard coefficient matrix for taxonomic samples of Site 738C. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 292 Site 738C Jaccard Coefficient — Average Linkage Algulhasina-------------------------------------- — Bairdi------------------------------------------ Bairdoppilata--------------------------------------------------— Cytheropteron spp.-------------------------------------------------------- Eucytherura sp. A ----------------------------------------------------------- — Aversovalva------------------------------------------------ Munseyella sp. A ------------------------------------------------ Cytherelloidea------------------------------------ Propontocypris sp. A ------------------------------------ Pennyella sp. A ------------------------------------------ Dutoitella------------------------ Eucythere------------------------ Cytherella spp.------------------------------ Krithe spp.------------------------------ Pennyella sp. B ------------------------------------------------- Eucytherura sp. B ------------------------------------------------------------------------------- I -------1 -------1 -------1 -------1 -------1 0.0 0.2 0.4 0.6 0.8 1.0 Distance Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 293 5.3.7. Principal components analysis — stratigraphic and taxonomic Principle components analysis (PCA) is a multivariate statistical technique for reducing a data matrix of potentially correlated variables into an uncorrelated (i.e., orthogonal in multi-dimensional space) set of indices termed principal components (PCs). For a given data matrix, the first PC is oriented in the dimension of maximum multivariate variance, the second PC is oriented in the dimension of the maximum remaining multivariate variance orthogonal to the first, and so on. The relative amount of original multivariate variance accounted for by each PC is represented by its eigenvalue. The degree of inter-variable correlation directly controls the ranked magnitude of PC eigenvalues: highly correlated data matrices produce eigenvalue spectra skewed towards higher PCs, whereas less correlated data matrices produce more even eigenvalue spectra. The contribution of each original variable to each PC is represented by its eigenvector coefficient. The linear position of each case along each PC axis is represented by its factor score for that PC. PCA provides PC indices and associated eigenvalues, eigenvector coefficients, and factor scores that reflect a mathematically unique solution based on maximized orthogonal variance rankings. However, this solution is but one of an infinite number that would describe the original data matrix with equal precision, but not parsimoniously with respect to maximum orthogonal variance ranking. To facilitate interpretation of a PC as a composite variable for multiple original variables, PC indices are often orthogonally rotated post-hoc to coincide more closely with clusterings or arrays of eigenvector coefficients. PCA parameters are then recalculated, maximizing specific eigenvector coefficients on specific Pcs and improving PCA explanatory power, but no longer from a perspective of maximum multivariate variance. Numerous rotation methods exist, with Kaiser’s (1958) varimax method Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 294 being the most common. Additional theoretical background and computational procedures for PCA and rotation methods are discussed by Manly (1994). PCA was performed on absolute, log-transformed, and standardized (%) matrices for Site 690B and 738C using stratigraphic samples as cases and taxa as variables (i.e., Q-mode); all calculations were performed through Systat 8.0 (PC-version). Log- transformation used the natural log of each variable value (plus one) to minimizes variance due to relatively abundant taxa. Standardization converted each variable's value into its percent of a given sample (i.e., 5 Xus yus in a sample containing 20 specimens = 0.25). Samples containing no specimens, and taxa occurring in only one sample (i.e., “stratigraphic singletons”), were culled prior to PCA. Each of the three matrices was analyzed using non-rotation and varimax rotation techniques. Only the varimax rotation of the log-transformed matrices are discussed below as they most succinctly summarize and simplify the general structure of each site’s faunal history. In all analyses, PCs with eigenvalues less than one are not reported or considered further given their relatively low explanatory power. Eigenvalue structure of the six PCA analyses of Site 690B is presented in Figure 5.20. Absolute and log-transformed eigenvalues have a generally similar structure, with PC-1 accounting for 19-38% of the original matrix variance, and subsequent PCs a lesser amount (except for some later varimax-rotated PCs). The distribution of non rotated values are right-skewed relative to the varimax-rotated values; this results from the later’s orthogonal rotation to maximize a small number of high eigenvector coefficients on each rotated PC, which produces a more equitable distribution of variance explained among rotated PCs. In contrast, PC-1 eigenvalues for the standardized (%) matrix account for over 65% of the total variance. These high PC-1 eigenvalues result from most taxa loading highly on the first axis due to the transformation technique and presence of most taxa in the earliest and latest portions of the record. Subsequent PCs Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 295 Figure 5.19. Q-mode PCA eigenvalues for Site 690B faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 690B — PCA Eigenvalues — Q-Mode Principal Component Absolute Abundances Loo-Tran stormed Abundances Standardized (%> Abundances No Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Axis Eiaenvalue % Variance % r Eiaenvalue % Variance % E Eiaenvalue % Variance % E Eiaenvalue % Variance % E Eiaenvalue % Variance % I m 1 I % Variance % E 1 9.37 36.05 36 05 5.14 19.77 19.77 9.61 37 72 37 72 6 8 3 26 26 26 26 17.71 68.13 68.13 1693 6512 65.12 2 4 39 1689 52 94 3 83 14.72 34 49 4.99 1920 56 92 354 1361 39.87 292 11.21 79 34 3.55 13 64 78 76 3 2 74 10.53 6347 2 75 1056 45 05 286 11 00 67.92 3 17 12.19 52 06 242 9 2 9 88 63 1 96 7 55 86.31 4 1.91 7.35 70.82 4.53 1742 6247 1.62 622 74.14 396 15 22 67.28 1.47 565 94 28 196 752 93 83 5 1.72 6 63 77.46 1 71 6 59 69 06 1.37 5.28 79.41 3.03 11.66 78.94 1 12 431 98 59 1 24 4.76 98 59 6 1.34 5 16 6261 2 31 887 77.93 1 14 438 83 79 1 26 4 86 83.79 — — — — — — 7 1.14 4 39 87.00 2 36 907 87 00 — — - — — — — — — — — — 100 s c 3 Q . X U i ® o c ( 0 *c s o □ Absolute Abundances - No Rotation ■ Absolute Abundances - Varimax Rotation ■ Log-Transformed Abundances - No Rotation ■ Log-Transtormed Abundances - Varimax Rotation □ Standardized (%) Abundances - No Rotation □ Standardized (%) Abundances - Varimax Rotation r U n i ■ ! 11 Principal Component Axis 2 9 6 297 each explain a relatively small proportion of the total variance and are highly loaded by taxa sharing discrete intervals of relatively high abundance (e.g., Indeterminate Genus C and Rimacytheropteron for PC-3, Munseyella sp. A and B for PC-4, and Palmoconha for PC-5). Eigenvector coefficients for each taxon in the Site 690B log-transformed matrix are presented in Figure 5.21 The varimax-rotation significantly alters the original loading distribution, “successfully” producing small suites of taxa loading highly on each PC. PC-1 is highly loaded by taxa that are most abundant within the pre-LPTM interval (i.e., Dutoitella, Bairdia, Mayburya, Krithe spp.), or largely or wholly restricted to the pre-LPTM interval (i.e., Philoneptunus, Pelecocythere, Henryhowella, Propontocypris sp. B). Factor scores for these eigenvalue coefficients are presented in Figure 5.22. PC-1 factor scores are commensurate within the pre-LPTM interval, followed by lower scores throughout the LPTM interval, and slightly higher scores in the latest portion of the post-LPTM interval. PC-2 is highly loaded by taxa that are common in the pre-LPTM interval and early portion of the LPTM interval, after which they become locally extinct (i.e., Profundocythere, Eucytherura sp. A, Aversovalva hydronamica), with commensurably high, but variable factor scores in the early portion of the record and low, stable scores in the later portion. PC-3 is highly loaded by taxa that are most abundant around the end of the LPTM interval (i.e., Cytherella spp., Eucythere, Indeterminate Genus B) and Palmocohna, which appear in the middle of the LPTM interval, reach maximum abundance in the early post-LPTM interval, and then again disappear later in the interval. Factor scores for PC-3 reflect the general stratigraphic distribution of these taxa, with a large increase to stable, high scores in the later LPTM interval. PC-4 is highly loaded by taxa either restricted to the LPTM interval (i.e., Munseyella sp. A and B) or most abundant within the LPTM interval (i.e., Propontocypris sp. A, Pennyella sp. A and B, Cytheropteron spp.), whereas PC-5 is Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 298 Figure 5.20. Q-mode eigenvalue coefficients (“loadings”) of taxa for non- and varimax-rotated PCA of Site 690B log-transformed abundances. Taxa are sorted and grouped in a step-wise fashion in order of decreasing loadings and increasing PC. Eigenvalues in dark gray cells are 0.75-1.00 and in light gray are 0.50-0.74. Note that little loading structure exists in PC1-PC4, which accounts for most of the total variance. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 690B — Eigenvector Coefficients — Q-Mode Loq-Transformed Abundances — No Rotation Log-Transformed Abundances — Varimax Rotation Taxon 1 2 3 4 5 6 1 2 3 4 5 6 Philoneptunus c. 1 . tricostatus Pelecocythere Dutoitella Bairdi Henryhowella Propontocypris sp. B Mayburya _______ Krithe spp.________ Profundocythere Eucytherura sp. A Aversovalva ct. hydronamica Cytherella spp. Eucythere Ind. Gen. B Palmoconha Munseyella sp. A Munseyella sp. B Propontocypris sp. A Pennyella sp. B Cytheropteron spp. Pennyella sp. A Indet. Gen. C Rimacytheropteron Bairdoppilata -0.17 -0.16 0.22i 0.43 0.27 0.01 0.14: 0.25|^ -0 .8 0 -0.11 0.01 0.19! -0.07 -0.27 0.20 0.28 0.07 -0.16 0.01 0.06 -0.27 0.19 -0.04 0.07 -0.20 -0.17 0.24 0.10 0.13 -0.21 0.04 0.06 -0.14 •0 .5 2 0.01 0.06 0.01 0 .6 2 -0.46 0.16 0.13 -0.34 -0.18 0.16 -0.30 0.16 0.01 -0.61 -0.27 0.24 0.09 -0.35 -0.13 -0.18 0.06 -0.43 -0.43 -0.07 0.14 -0 .5 2 ! -0.29 -0.34 -0.29 0.33 0.30 0.06 -0.03 -0 .3 3 1 0.04 0.44 0.24 0.01 -0.10 -0.40, 0.10 0.02 0.01 0.02 -0.16 0.24 0.29 0.17 0.19 0.15 0.02 0.34 0.34 -0.24 0.22 -0.02 0.05 0.19 0.00 0.02 0.38 0.37 0.40 -0.21 -0.11 0.33 0.41 -0.08 -0 .6 3 -052 -0.13 0.21 0.04 0.41 0.00! -0.05 -0.20 -0.07 -0.27 - 0.02 -0.19 - 0.11 -0.05 0.07 - 0.10 -0.04 -0.15 - 0.11 -0.23 0.08 -0.17 -0.02 0.17 0.17 -0.09 -0.05 - 0.12 0 .3 6 1 0.42' 0.39 0 .10; 0.02J 0.37 -0.09 -0.19 0.23 Eucytherura sp. B Ind. Gen. A - 0.11 - 0.22 -0.38 -0.05 -0.28 0.29 0.10 0.38 0.45; 0.45i 0.13 0.04 0.48 -0.33 0 .2 6 : 0.39 0.51 -0.15 = > 0.75 Eigenvector Coefficient = 0.50-0.74 Eigenvector Coefficient 2 9 9 300 Figure 5.22. Varimax-rotated Q-mode PCA factor score plots for Site 690B log- transformed abundances. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S it e 6 9 0 B — P C A — Log-Transformed Abundance — V arim ax Rotation 55.42 55.44 55.46 55.48 55.50 55.52 55.54 55.56 301 CO CM O CM I I 9 J O O S Jopej Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S it e 6 9 0 B — P C A — Log-Transformed A bundance — Varim ax Rotation 55.42 55.44 55.46 55.48 55.50 55.52 55.54 55.56 302 8 a. - j t 8 a. 8 a. (O £ o o o « Q _ Q- Q _ 'Zo t 0 < < < < < < < < < < < < < < m C M T- o »- C M co • I I 0joos Jopej Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 303 highly loaded by taxa restricted to the pre-LPTM interval (i.e., Indeterminate Genus C, Rimacytheropteron, Bairdoppilata). Stratigraphic factor scores for the two PCs reflect the abundance patterns of their constituent taxa with high factor scores in the LPTM and pre-LPTM interval, respectively. Eigenvalue structure of the six PCA analyses of Site 738C is presented in Figure 5.23. Although a small number of PCs accounts for most of the variance in the absolute and log-transformed abundance matrices, eigenvector coefficients (Figure 5.24) and factor scores (Figure 5.25) demonstrate that faunal response to the LPTM is either less pronounced and/or sampling effects (i.e., low abundances, spotty distribution, inclusion of much younger final sample, etc.) preclude as clear an interpretation as for Site 690B. PC-1 is highly loaded by taxa common prior to the CIE (i.e., Algulhasina, Bairdia, Krithe spp.), whereas PC-2 is highly loaded by taxa most abundant within or restricted to the LPTM interval (i.e., Dutoitella, Cylherella spp., Eucythere). Sample factor score plots for PC-1 and PC-2 are essentially opposites of one another with their cross-over point in the early-middle LPTM interval. Subsequent PCs are highly loaded with less abundant taxa of spottier distribution. Factor score plots for subsequent PCs show no clear trends through the event, with the exception of higher scores overall in the last sample as many taxa absent from earlier samples reappear. Each Site 690B and 738C matrix was also subjected to non- and varimax- rotation R-mode PCA analyses, using taxa as cases and stratigraphic samples as variables. In all Site 690B analyses of absolute and log-transformed matrices, four or fewer PC accounted for >80% of the total original variance, while the standardized matrix required six PCs to account for the same percentage with a much more equitable distribution of variance (Figure 5.26). In the varimax-rotation of absolute and log- transformed matrices, there is a strong association of high loadings of specific intervals with specific PCs: in the absolute matrix, pre-LPTM with PC-1, LPTM with PC-2, and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 304 Figure 5.23. Q-mode PCA eigenvalues for Site 738C faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 738C — PCA Eigenvalues — Q-Mode Principal Component Axis Absolute Abundances Loo-Transformed Abundances Standardized (%) Abundances No Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Eiaenvalue % Variance % 1 Eiaenvalue % Variance % £ Eiaenvalue % Variance % I Eiaenvalue % Variance % I Eiaenvalue % Variance % E Eiaenvalue % Variance 1 7.60 47.52 47.52 557 34.81 34.81 6.53 40.80 40.80 2.64 1651 1651 3.45 21 53 21.53 2.31 1444 1444 2 2.57 1609 63 60 294 1836 53.17 252 1578 56.58 3.78 2365 40.16 255 15.94 37.47 2 16 1348 27 93 3 1.75 1092 7453 338 21.09 74.27 1.69 1057 67 14 3 29 20 53 60.69 1.96 1223 49.70 2 20 13 75 41 68 4 1.21 753 82.06 1.25 7.79 82 05 1.41 882 75 96 1.43 8 9 4 69 63 1.76 1099 60 70 1.77 11.05 52 72 5 — — — — — — 1.06 659 82 56 2.07 1293 82 56 1 44 897 6967 1 57 981 62 53 6 — — — — — — — — — — 1 35 8.43 78 09 1.85 11.54 74.07 7 — — — — — — — — — — — 1 06 6 62 84 71 1 54 9 64 83 70 8 — — — - — — — — — — — — 1 02 6 36 91.07 1 18 737 91 07 □ Absolute Abundances - No Rotation ■ Absolute Abundances - Varimax Rotation ■ Log-Transformed Abundances - No Rotation ■ Log-Transformed Abundances - Varimax Rotation □ Standardized (%) Abundances - No Rotation □ Standardized (%) Abundances - Varimax Rotation 4 5 Principal Com ponent Axis 3 0 5 306 Figure 5.24. Q-mode eigenvalue coefficients (“loadings”) of taxa for non- and varimax-rotated PCA of Site 738C log-transformed abundances. Taxa are sorted and grouped in a step-wise fashion in order of decreasing loadings and increasing PC. Eigenvalues in dark gray cells are 0.75-1.00 and in light gray are 0.50-0.74. Note that little loading structure exists in PC1-PC4, which accounts for most of the total variance. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 738C — Eigenvector Coefficients — Q-Mode Log-Transformed Abundances — No Rotation Log-Transformed Abundances — varmax Rotation -0.06 -0.06 0L22 0.54 -0.11 - 0.11 -0.13 -0.26 -0.13 -0.07 -0.38 -0.42 - 0.20 - 0.10 - 0.12 -0.09 -0.07 -0.17 Taxon Alguhasina Bairdi Krithe spp. Eucy there Cytherelloidea Cytherella spp. Dutoitella Munseyella sp. A Eucytherura sp. A Cytheropteron spp. Aversovalva Propontocypris sp. A Pennyella sp. B Eucytherura sp. B Bairdoppilata Pennyella sp. A -0.26 •0.65 -0.39 -0.28 -0.05 -0.13 - 0.22 -0.27 0.13 0.46 0.11 0.12 0.08 0.34 0.18 0.21 -0.06 -0.03 0.14 -0.11 0.38 0.17 0.00 -0.01 0.31 0.05 0.09 0.34 -0.14 0.23 0.60 0.19 -0.30 - 0.01 -0.04 -0.06 -0.13 -0.35 -0.19 •0.59 -0.46 -0.26 -0.40 -0.08 0.01 0.19 0.24 0.43 = > 0.75 Eigenvector Coefficient = 0.50-0.74 Eigenvector Coefficient 3 0 7 308 Figure 5.25. Varimax-rotated Q-mode PCA factor score plots for Site 738C log- transformed abundances. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S it e 7 3 8 C — P C A — Log-Transformed A bundance — Varim ax Rotation 283.5 284.0 284.5 285.0 285.5 286.0 286.5 287.0 309 C L Q. Q- T o -j CM CO o CM ■ 0joos Jopej Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S it e 7 3 8 C — P C A — Log-Transformed A bundance — Varim ax Rotation 283.5 284.0 284.5 285.0 285.5 286.0 286.5 2 87.0 310 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 1 Figure 5.26. R-mode PCA eigenvalues for Site 690B faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 690B — PCA Eigenvalues — R-Mode Principal Absolute Abundances Loo-Transformed Abundances Standardized (%t Abundances Component No Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Axis Eioenvalue % Variance % r Eioenvalue % Variance % £ Eioenvalue % Variance % £ Eioenvalue % Variance % £ Eioenvalue % Variance Eioenvalue % Variance %1 1 13.32 60.53 60 53 9.17 41.68 41.68 11.24 51.11 51.11 5.14 23 38 23.38 6 57 29.86 29 86 3.72 16.90 16.90 2 361 1640 7694 4 52 20 52 62.20 442 20.08 71.18 587 26 69 50.07 4.19 1906 48 92 4 47 20.31 37 21 3 1 50 6 84 83.77 4 52 20.56 82 76 1.65 7.51 78.69 6 25 28.39 78.46 2.10 9 55 58 47 3.69 16.75 53 96 4 1.12 5 10 68.87 1.34 6 11 08.87 1.03 4 69 83 38 1.08 4 92 63 38 1.88 8 53 67 01 1.91 8 66 62 62 5 — — — — — — — — — — — — 1.61 7.32 74 33 2.27 10.34 72 95 6 — — — — — — — — — — — — 1 31 5 93 80 26 1 61 730 80 26 100 j 0 3 ro 5 50 □ Absolute Abundances - No Rotation ■ Absolute Abundances - Varimax Rotation ■ Log-Transformed Abundances • No Rotation ■ Log-Transformed Abundances - Varimax Rotation □ Standardized (%) Abundances - No Rotation □ Standardized (%) Abundances - Varimax Rotation j z f c a O l D 3 4 Principal Component Axis 313 post-LPTM with PC-1 as well as PC-1 and, in the log-transformed matrix, pre-LPTM with PC-2, LPTM with PC-1, and post-LPTM with PC-3 (Figure 5.27). The standardized matrix produces a weaker correlation overall, with the majority of LPTM samples having their highest coefficients in PC-1 and post-LPTM samples in PC-2. These associations reflect the general shift in faunal composition through the event previously illustrated in the faunal distribution chart and cluster analyses. Application of R-mode PCA to each Site 738C matrix produced a slightly less skewed set of eigenvalues relative to Site 690B (Figure 5.28). Only the log- transformed matrix produced a distribution of maximum eigenvector coefficients showing a trend correlated with the CIE (Figure 5.29). In the log-transformed matrix, pre-LPTM samples tend to load highly on PC-3, followed by early LPTM samples loading highly on PC-1, and finally a shift in late LPTM samples to highest laoding on PC-2. 5.4. Discussion 5.4.1. Overview Paleobiologists face an apparent paradox in reconstructing the history of life. On one hand, we embrace uniformitarianism, commonly using the ecological and environmental distribution of living individuals to reconstruct the paleoenvironments of their fossil counter-parts (e.g., Murray, 1995). On the other hand, we appreciate that taxa evolve, with genomic, ecological and environmental shifts not necessarily accompanied by morphologic (i.e., taxonomic) change (e.g., Schopf et al., 1975). Resolution of this paradox is neither straightforward, unique, nor inevitable — it is case-by-case and based on an eclectic mix of experience, data, and assumption. This study is no exception as it attempts to move beyond the above documentation of deep-ocean ostracode faunal changes during the LPTM to imbue these data with paleoenvironmental and paleoecological meaning based on the ecology of living Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 314 Figure 5.27. Site 690B R-mode eigenvalue coefficients (“loadings”) of taxa for varimax-rotated PCA of absolute, log-transformed, and relative (%) abundance matrices. Samples are presented in stratigraphic order, with eigenvalues in dark gray cells greater than 0.75 and in light gray between 0.50-0.75. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Absolute Abundances - 0.11 0.48 0.48 0.45 - 0.20 Site 690B — Varimax Eigenvector Coefficients — R-Mode Sample Log-Transtormed Abundances -0.04 - 0.10 -0.03 - 0.02 -0.48 -0.23 -0.13 •0.50 -0.04 Relative (%) Abundances Age (Ma) : Interval 1 57.087 57.109 57.141 57.149 57.172 57.201 57.232 57.262 57.271 57.277 57.284 57.292 57.300 57.315 57.322 57.329 57.346 57.368 57.384 57.406 57.428 57.459 Post-LPTM Post-LPTM Post-LPTM Post-LPTM Post-LPTM Post-LPTM LPTM LPTM LPTM LPTM LPTM LPTM LPTM LPTM Pre-LPTM Pre-LPTM Pre-LPTM Pre-LPTM Pre-LPTM Pre-LPTM Pre-LPTM Pre-LPTM -0.07 0.27 0.21 0.54 0.04 0.09 0.05 -0.11 0.07 0.07 -0.01 -0.01 0.17 -0.03 0.07 0.24 -0.01 0.11 0.23 -0.05 0.02 0.20 -0.02 -0.02 0.20 0.53 0.28 0.45 0.35 0.26 0.25 0.24 0.22 0.05 0.31 0.13 0.46 0.00 0.15 -0.06 -0.03 0.00 -0.09 -0.05 -0.18 - 0.01 -0.14 0.24 0.09 0.33 0.13 0.55; 0.07 0.46 0.11 0.07 0.11 -0.09 -0.11 0.10 -0.06 0.32 0.24 0.09 -0.17 -0.31 -0.03 0.05 0.38 0.41 -0.48 0.22 -0.32 0.06 0.00 0.45 0.42 -0.45 0.36 -0.27 0.29 0.06 0.41 0.32 -0.37 -0.19 -0.22 0.25 -0.01 0.24 0.17 -0.14 -0.06 0.37 0.00 0.17 0.21 •0.51 0.09 0.14 = > 0.75 eigenvector coefficient = 0.50-0.74 eigenvector coefficient 0.08 -0.01 0.04 0.41 0.03 0.06 0.02 0.09 0.06 -0.07 0.11 0.05 0.04 0.06 0.09 0.04 0.04 -0.24 0.19 0.11 0.66 0.26 0.13 0.68 0.12 0.49 0.13 -0.20 0.07 0.09 0.06 0.13 -0.01 0.34 0.16 0.02 0.20 0.12 0.72, 0.13 I -0.17 0.07 0.05 m m 0.09 ! -0.24 -0.04 0.01 0.11 ! 0.11 0.23 0.17 0.01 0.04 ; -o.io 0.02 -0.28 -0.72 i = 0 3 cn 316 Figure 5.28. R-mode PCA eigenvalues for Site 738C faunal matrices. Three data matrices (absolute abundance, log-transformed abundance, relative abundance) were each analyzed by PCA using no rotation and varimax rotation methods. Eigenvalues for each PC of each of the six analyses are presented, along with the percent of eigenvalue variance explained and the cumulative percent of eigenvalue variance explained. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Site 738C — PCA Eigenvalues — R-Mode Principal Component Axis Absolute Abundances Loo-Transformed Abundances Standardized <%) Abundances No Rotation Varimax Rotation No Rotation Varimax Rotation No Rotation Varimax Rotation Eioenvalue % Variance Eioenvalue % Variance % £ Eioenvalue % Variance Eioenvalue % Variance % r Eioenvalue % Variance i • > I UJ % Variance % £ 1 15.09 8383 83.83 1482 8232 82 32 11.22 62 33 62 33 6.89 38.27 38 27 3 72 2067 20.67 262 1455 14.55 2 1 47 8.14 91.97 1 74 965 91 97 2 10 11.67 74.00 4 18 2324 61 50 2 66 14 76 35 43 2.97 16.51 31 06 3 — — — — — — 1.57 8 74 82 74 382 21.23 82.74 2.36 1309 48 52 2.40 13 34 44 40 4 — — — — — — — — — — — — 2.18 12 11 60 63 2.30 1276 57.16 5 — — — — — — — — — — — — 2.10 11.64 72.27 2 28 1264 69.80 6 — — — — — — — — — — — — 1 60 8 91 81 17 2.05 11 37 81.17 100 □ Absolute Abundances - No Rotation ■ Absolute Abundances - Varimax Rotation ■ Log-Translormed Abundances - No Rotation ■ Log-Translormed Abundances - Varimax Rotation □ Standardized (%) Abundances ■ No Rotation □ Standardized (%) Abundances - Varimax Rotation 3 4 Principal Com ponent Axis u > 4 31 8 Figure 5.29. Site 738C R-mode eigenvalue coefficients (“loadings") of taxa for non- and varimax-rotated PCA of absolute, log-transformed, and relative (%) abundance matrices. Samples are presented in stratigraphic order, with eigenvalues in dark gray cells between 0.75-1.00 and in light gray between 0.50-0.74. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. mbsf Sample Interval 2 83.56 Post-LPTM 284.48 ! Post-LPTM 284.58 LPTM 2 84 .68 LPTM 2 84 .75 LPTM 2 84 .85 LPTM 284.93 LPTM 285.00 LPTM 285.11 LPTM 285.22 j LPTM 285.31 LPTM 2 85 .35 LPTM 285.43 LPTM 285.52 Pre-LPTM 285.65 | Pre-LPTM 285.91 ' Pre-LPTM 286.34 Pre-LPTM 286.56 Pre-LPTM Site 738C Varimax Eigenvector Coefficients — R-Mode Absolute Abundances Log-Transformed Abundances Relative (% Abundance 0.33 0.22 0.44 0.68 > 0,68 0.06 0.36 0.42 - 0.02 0 .48 0 .05 0.01 -0.35 -0.03 - 0.20 -0.37 -0.03 0.3 6 -0.08 0.2 6 - 0.11 - 0.20 -0.06 -0.04 -0.09 -0.08 - 0.02 -0.15 -0.14 -0.05 -0.14 - 0.02 0.03 0.03 - 0.11 -0.06 -0.06 - 0.01 0 .0 3 0 .2 3 -0.17 -0.04 J0.06 [ -0.06 -0.03 0.04 0.12 -0.25 -0.13 0.04 0.28 -0.07 = > 0.75 Eigenvector Coefficient = 0.50-0.74 Eigenvector Coefficient 3 1 9 320 representatives and reconstructed paleoecology based on independently derived paleoenvironmental conditions. Although modern ecological data may not be directly applicable to Paleogene individuals given evolutionary ecological shifts, it is assumed that paleoecological limits of specific taxa did not change significantly during the geologically short interval of study. Thus, the following discussion is based on an amalgam of independent paleoenvironmental data, modem ecological data, and ancient faunal data — the strong signals recorded in the first provide an opportunity to test the applicability of the second to the third. I first address aspects of ostracode abundance, richness, diversity, and faunal structure, then conclude with the pronounced pattern of some key taxa, their relation to shallower Site 689B, and paleoceanographic implications for the nature of the LPTM event. 5 .4.2. Ostracode abundance, richness, and diversity Fossil abundance is the net sum of myriad factors of biology (e.g., reproductive fecundity, success, and rate), ecology (e.g., standing crop, food availability, competition), sedimentology (e.g., “dilution" by other grains), taphonomy (e.g., fragmentation, transport, in situ post-mortem dissolution), and post-depositional processes (e.g., compaction, deep-burial diagenesis). Given these myriad factors, constructing unique and testable explanations for observed changes in fossil abundance is often difficult. This is particularly difficult for the studied LPTM intervals as they have been subjected to little sedimentological study of seawater corrosivity, paleoproductivity, or percent abundance of various grains. Given the lack of site-specific sedimentological data, ostracode valve abundance may only be compared to the general observations and hypotheses on dissolution, productivity, and dissolved oxygen dynamics during the LPTM, which are based largely on bethic foram analyses (see Thomas, 1998 for comprehensive review). The LPTM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 321 coincides with a pronounced and global interval of intensified carbonate dissolution from shallow to abyssal depths (e.g., King, 1989; Kaiho eta/., 1996). This increased corrosivity is consistent with a massive carbon input to the ocean-atmosphere system through clathrate dissociation, given that simple bottom-water warming alone would decrease carbonate solubility. However, this sequence is problematic as dissolution appears to precede the CIE at many sites (Thomas, 1998). At both study sites, the precipitous drop in ostracode abundance occurs just prior to the CIE and benthic foram extinction horizon (Figure 5.7). In these very low abundance samples, valves are generally thinner and fragmentation higher. Valve thinness may be biological (i.e., biomineralization effectiveness was reduced in undersaturated waters) or taphonomic (i.e., valves were partially dissolved after epicuticle decay) — neither can be precluded based on limited SEM analyses (c.f., Carbonnel and Tolderer-Farmer, 1988). Intense seawater corrosivity likely influenced valve abundance both biologically and taphonomically, however establishing relative importance is difficult without additional ground-truthing of modern field studies, such as Swanson (1995). After the initial portion of the LPTM, Site 690B abundances increase to roughly two-thirds that of pre-LPTM abundances and vary strongly through the remaining study interval, whereas Site 738 abundances increase to, and in some cases exceed, pre-LPTM abundances. Direct comparison of the relative rates of change at the two sites is difficult given the absence of an age-model for Site 738C. Richness within a given community is related to numerous factors, including niche partitioning, competitive exclusion, and disturbance. Virtually no data exist on the community structure and dynamics of deep-ocean ostracodes, with recognized trophic strategies consisting of filter versus deposit (i.e., actively foraging) feeding based on soft-part morphology. The class is rarely considered in most “holistic" benthic studies given their low densities and small size, often easily passing through the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 322 finest sampling screen. It is the author’s opinion that most deep-ocean ostracodes are “hyper-k-selected” given their (1) long pre-sexual ontogeny of seven to nine molts over a period of up to two years and possible much longer and (2) low fecundity given a relatively low body-size:sex-cell ratio (Cohen and Morin, 1991). This is a hypothesis in desperate need of testing. If so, individual ostracodes may rarely “ see” one another ecologically, implying little intra-class competition, with abundance and richness controlled by other biological (e.g., predation, meiofaunal competition, etc.) or environmental (e.g., temperature, dissolved oxygen, corrosivity, etc.) factors. Rarefaction analyses demonstrate that the Maud Rise fauna was overall taxonomically richer than that of Kerguelen Plateau and suffered less loss during the LPTM. However, Kerguelen Plateau richness recovered after the LPTM, whereas Maud Rise richness remained low, but less variable overall. This pattern may reflect lingering deleterious conditions at Maud Rise compared to Kerguelen Plateau. Unfortunately, the isotope-proxy record is incomplete for the later LPTM interval at Kerguelen Plateau, so testing this explanation is not currently possible. If the isotope stratigraphy reflects that of Maud Rise, the Maud Rise pattern may instead reflect slower re-colonization due to the relative isolation of Maud Rise. 5 .4 .3 . Whole-fauna patterns through the LPTM Whereas the above taxa show distinct stratigraphic restrictions and inferred paleoenvironmental preferences, whole-fauna analysis via similarity coefficients, clustering algorithms, and R-mode PCA indicate that the Site 690B record consists of three distinct faunas that largely coincide with the a priori isotope-based division of the record (Figure 5.16 and 5.22). Clustering “errors” largely consists of samples of very low abundance and richness at the onset of the CIE (i.e., Pre-7, Pre-8, LPTM-1) and samples stratigraphically adjacent to their “ wrong" cluster (i.e., Post-1). LPTM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 323 and post-LPTM faunas tend to be more similar to one another than either is to pre-LPTM faunas. However, a complete-linkage dendrogram (not shown) gives some indication that the latest post-LPTM faunas begin to shift towards those present in the pre-LPTM (i.e., Post-5 and Post-6 clustering in the pre-LPTM cluster). This “recovery” is also evident in the R-mode PCA eigenvector coefficients, in which the youngest sample in the post-LPTM interval shows its highest loading on the pre-LPTM-dominated axes, particulary when absolute abundances are analyzed. Analogous whole-faunal data for Site 738C are more difficult to interpret (Figure 5.17 and 5.25). This difficulty likely relates to increased “ noise” through less complete census of faunal composition given small sediment volume and commensurate abundance and richness. As in Site 690B, three distinct clusterings may be identified, but they do not coincide with the a priori isotope-based divisions of the record. Instead, a distinct faunal discontinuity occurs in the LPTM interval between samples LPTM-8 and LPTM-9, with each of these LPTM sub-groups internally very similar in composition. This pattern is clear in the dendrogram and R-mode PCA eigenvector coefficients, with the later LPTM sub-group more similar to the pre-LPTM fauna. This disparity between fauna and isotope-based divisions likely relates to the way isotope intervals were defined: because no isotopic data exist between the LPTM-8 and Post-1 horizon (284.95-284.52 mbsf), the LPTM/post-LPTM boundary was arbitrarily drawn just below the Post-1 sample (-285.52 mbsf), which has isotopic values most similar to those preceding the LPTM. Subsequent isotope analyses will likely reveal a relatively rapid and early return during this interval, which would be consistent with the pattern at Site 690B. These whole-fauna patterns indicate that the LPTM produced a major, but transient, reorganization of deep-ocean ostracode communities, with short lived “outages” of many common deep-ocean taxa and epiboles of certain taxa as described in detail above. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 324 5.4.4. Podocopids, platycopids, and benthic foraminifera: constraints on paleooxygenation levels Benthic foram studies have implicated some combination of increased temperature and decreased dissolved oxygen as the cause of the mass extinction at the LPTM (e.g., Thomas, 1989; Kaiho, 1994). Clearly identifying one over the other as a “ smoking gun” has been hampered by the lack of consensus on modern environmental controls (e.g., organic supply versus dissolved oxygen) and the paleoecology of predominately post-extinction taxa such as Bulimina spp. and Tappanina selmensis. Regardless, post-extinction benthic foram faunas show much greater inter-site differences than pre-extinction faunas and also differ more by region than depth (Thomas and Shackleton, 1996; Thomas, 1998). Regardless, warmer deep-waters would have inherited lower dissolved oxygen concentrations from the surface. This initial concentration would only decrease through respiration with higher organic supply, greater source distance, and slower circulation. Examination of total ostracode abundance and the ratio of platycopid filter-feeders (Cytherella, Cytherelloidea) to podocopid deposit-feeders provides some metazoan-based constraints on these potential extinction mechanisms. Little experimental data on oxygen-demand exists for either benthic forams or ostracodes. Of the two benthic groups, ostracodes likely have a much higher demand give their multicellular organization, relatively high surface:volume ratio, and diffusion- based gas exchange system. Therefore, if the LPTM event involved major reductions in dissolved oxygen sufficient to cause the benthic foram extinction, one might expect absolute ostracode abundance to decrease and remain depressed until conditions ameliorated. Extreme dysoxic conditions would likely cause an ostracode faunal “outage” at some geographic or bathymetric scale, with post-dysoxia re-population time limited Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 325 by the migration distance and mechanism (e.g., random crawling, physical transport, biological transport by fish, etc.). Assuming relatively constant sedimentation through the LPTM event, ostracodes at both sites are effectively absent during the rapid CIE (Figure 5.30). Immediately thereafter, however, abundance steadily recovers to pre-CIE values, returning pre-CIE values (and higher at Site 738C) while 81 8 0 remains -0.75 % o more negative than pre- CIE 51 8 0. This pattern supports a scenario of short-lived temperature/oxygenation extremes sufficient to severely affect benthic metazoan biomass, but limits their effects to times when temperatures exceeded by > -4 C° that of pre-LPTM temperatures. In addition, the assumption of constant sedimentation is questionable given that in-progress LPTM studies at Blake Nose examining Milankovitch-scale sedimentation and Fe- variations support locally and perhaps globally increased sedimentation rates, which may partially explain this decreased ostracode abundance through dilution (Norris, pers. comm.). Temporal variations in the percentage of filter-feeding platycopids Cytherella and Cytherelloidea provide additional metazoan-based constraints on the LPTM. Platycopids are obligate filter-feeders that fundamentally differ in respiration, feeding, and reproduction from the generally more abundant deposit-feeding Podocopida (see Whatley, 1991, 1995 for detailed discussion). Nearly all benthic ostracodes lack a circulatory system or specialized respiratory organs, instead molecular gas exchange (as well as osmoregulation) occurs directly across the uncalcified inner lamella. This exchange is assisted by epipodial respiration plates (ERPs) that flux seawater through the domociliar cavity, maintaining partial pressure gradients that drive the exchange of seawater dissolved 0 2 and metabolism-derived COz across the thin epicuticle (McMahon and Wilkens, 1983; Vannier and Abe, 1995). These ERPs are present on the mandibles and maxillula of all marine benthic ostracodes, but platycopids possess additional plates Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 326 Figure 5.30. Stratigraphic abundance of ostracode valves and percent abundance of filter-feeding podocopid taxa for Sites 690B and 738C. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 328 on the maxilla and first thoracic appendages. This greater number of ERPs increases the fluid flux through the domiculum where specialized sets of pilate setae sieve out suspended organic matter. An exaptation of this enhanced circulation for suspension feeding is more efficient gas exchange, allowing platycopids to preferentially survive, and in some cases thrive, over podocopids in reduced oxygen conditions. In addition, all platycopids brood and ventilate their young through multiple early instars, providing a relatively well- oxygenated microenvironment for early development and decided advantage over podocopids. Platycopids often constitute the most common taxa within modern oxygen- minimum zones, with well-documented examples from the southeastern USA (Cronin, 1983) and South Africa (Dingle et al., 1989; Dingle and Giraudeau, 1993). Ancient examples of platycopid-dominant dysaerobic intervals occur in the Eocene (Whatley and Arias, 1993), Cretaceous Oceanic Anoxic Events (OAEs) (Jarvis et al., 1988; Horne et al., 1990; Babinot and Crumiere-Airaud, 1990), Jurassic (Boomer and Whatley, 1992), and Late Devonian (e.g., Lethiers and Whatley, 1994). At both LPTM sites, platycopids comprise less than 5% of pre-CIE faunas, and show no increase in relative abundance during the early LPTM interval, but maintain a relatively constant rate of increase through the event to a stable relative abundance of -20% at the return to stable (although slightly negative relative to pre-CIE) 51 8 0 values (Figure 5.30). This pattern is inconsistent with an LPTM model where dysoxia is greatest in the earliest LPTM, during which % platycopids would be expected to be somewhat elevated over podocopids. Thus, these ostracode faunal data do not support a long interval of suppressed oxygenation as the prime extinction mechanism for benthic foram extinction during the LPTM, although this may have occurred “instantaneously” relative to the study samples. Absolute abundances may reflect some combination of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 329 deleterious warmth and increased sedimentation (as well as dissolution at the CIE horizon). The steady relative increase of platycopids through the event may reflect increased flux of organic matter from the surface following the event. 5.4.5. LPTM taxa: implications for paleoecology and dispersal Faunal distributions through the LPTM may serve as natural experiments registering the relative environmental tolerance of ostracode taxa in two ways. First, for specific taxa, evolving ecological and environmental preferences may be documented through stratigraphic examination within a robust environmental framwork. Second, for taxa largely lacking documented paleoecological or paleoenvironmental preferences, relative values may be extracted from comparison to other taxa through well- documented, short-lived environmental perturbations such as the LPTM. Krithe is examined here as an example of the first approach. In the Late Cretaceous and early Paleogene thermospheric oceans, the genus was geographically widespread and highly correlated with thermospheric taxa such as Cytherelloidea (Boomer and Whatley, 1995; Dingle, 1981). For example, Majoran et al. (1997) analyzed Maastrichtian faunas of the South Atlantic and found Krithe most common at paleotemperatures of about 15 °C, the highest of the study. This paleoecological regime stands in stark contrast to its modern temperature limit of generally less than 10° C (often predominant in <5 °C) and global psychrospheric distribution (Whatley and Quanhong, 1993; Coles et al., 1994). Stratigraphic abundance of Krithe at Site 690B and 738C is consistent with high thermal tolerance into the latest Paleocene. The genus is a predominant component of the LPTM fauna at both sites, but becomes relatively rare at the extreme of the LPTM warming (currently estimated at -20 °C). Thus, the genus appears to have maintained a relatively high thermal tolerance through at least the late Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 330 Paleocene, with paleoenvironmental restriction likely developing during the general cooling and development of the psychrosphere in the later Cenozoic. As an example of the second approach, the relative paleoenvironmental and paleoecological preferences of poorly understood, but widely distributed Munseyella and Palmocohna may be inferred from their relative abundance during the LPTM event. Munseyella spp. is completely restricted to the LPTM interval at both sites, comprising up to 15% of the fauna while most taxa suffer greatly decreased abundances (Figure 5.6). Thus, Munseyella is interpreted as thermophyllic relative to most taxa in the late Paleocene and this datum may serve as a starting point for tracking paleoecological evolution. At Site 690B, Palmocohna also shows a restricted stratigraphic distribution, being limited to the later LPTM through earlier Post-LPTM intervals (Figure 5.6). This taxon’s initial appearance and increased abundance correlates with the proportional rise in platycopids (Figure 5.30). This correlation may indicate unique paleoenvironmental conditions during this post-CIE interval, perhaps related to increased surface productivity and carbon export coupled with elevated, but not extreme temperatures. The above hypothesized environmental preferences require testing at additional LPTM intervals. If similar stratigraphic distributions of specific taxa are common at geographically widespread sites, then a relatively homogeneous deep-ocean expression of the LPTM is implied by a common, but distinct metazoan response compared to the regionally heterogeneous response of benthic foram faunas (Thomas, 1996). In addition, the degree of synchroneity of such epiboles between sites has important implications for the rapidity of dispersal of deep-ocean ostracodes and the thermal stratification of past oceans. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 331 5.4.6. Comparison of LPTM records on Maud Rise: Sampling constraints and environmental comparison Faunal patterns at Site 690B may be compared to those of Steineck and Thomas (1996) at Site 689B, approximately 50 kilometers away and 1,000 meters shallower, for potential bathymetric aspects of ostracode response to the LPTM. Thomas (1996) found little difference in the benthic foram response between these two sites. Before comparing these faunal responses, site sampling differences must be examined. For the common study interval of 55.43-55.55 Ma, Steineck and Thomas (1996) examined the >150 pm size-fraction of five samples (2-6 in Figure 5.3; -17 kyr/sample) compared to the > 63 pm size-fraction of 22 samples 6±3 kyr/sample) examined in this study. Such differences in size-fraction must be appreciated as smaller taxa would not have been tallied for Site 689B, biasing faunal interpretation towards larger constituents. Comparison of percent abundance values accounts for differences in sample volumes between sites (15 cc for Site 690B compared to 20-50 cc at Site 689B). Comparing Sites 689B and 690B (Figures 5.3 and 5.7), there is a good correspondence between percentage faunal pattern in the platycopid Cytherella and the common genus Krithe, which appears to recover in its predominance slightly earlier at the deeper site. In addition, Cytheropteron shows slightly greater faunal predominance at the shallower site {i.e., -20% compared to -10%) through the LPTM. A major discrepancy between the two sites is the pre-LPTM occurrence of Propontocypris. While its “blossoming” within and just after the LPTM interval appears consistent between sites, the genus is present in small amounts at the deeper site throughout the pre-LPTM (averaging -10 individuals and -5 %), but is completely absent in the shallower site. Based on this Site 689B absence and the modern association of Propontocypris with high food availability (Maddocks and Steineck, 1987; Steineck et a!., 1990), Steineck and Thomas (1996) interpret this pattern to indicate increased Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 332 organic flux to the bottom at the onset of the LPTM. However, the robust pre-LPTM presence of the genus at deeper Site 690B is not consistent with this model. As an alternative, prior to the LPTM, Site 689B may have been too cool for Propontocypris, which was established in the warmer water-mass bathing deeper Site 690B (assuming the reality of warm, saline bottom waters). Subsequent warming at the shallower site with the onset of the LPTM allowed the genus to migrate upwards. During the relatively warmer LPTM and post-LPTM, similar thermal conditions (and perhaps organic supply) produced similar faunal patterns between the two sites. However, this model is not supported by the synchronous, environmentally restricted occurrence of Munseyella and Palmocohna between the two sites. 5.5. Conclusions The study documents the response of deep-ocean metazoans to the extreme global warming associated with a major carbon cycle perturbation at the end of the Paleocene. At widely separated Southern Ocean sites on the Maud Rise and Kerguelen Plateau, deep- ocean ostracode abundances drop precipitously in association with the onset of the carbon isotope excursion and benthic foram extinction. After the initial portion of the LPTM, abundances increase to roughly two-thirds to completely to pre-LPTM amounts. The Maud Rise fauna was overall taxonomically richer than that of Kerguelen Plateau and suffered less loss during the LPTM. Kerguelen Plateau richness recovered after the LPTM, whereas Maud Rise richness remained low, but less variable overall, perhaps reflecting lingering deleterious conditions. Cluster and PCA faunal analyses demonstrate that Site 690B consists of three distinct faunas largely bounded by the CIE and recovery to relatively stable, if elevated, isotopic values. A similar, but less robust, faunal pattern is present at Site 738C, which would benefit from additional stable isotope data and larger sample volumes. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 333 Prolonged dysoxia of bottom-water as a mechanism for benthic foram extinction is not supported by the sampie-based continuity of the higher-oxygen-demand ostracodes. In addition, the proportion of dysoxia-tolerant filter-feeding ostracode taxa does not increase coincident with, or immediately following the CIE, as would be expected from physiology and other documented intervals of dysoxia. Instead, the proportion of filter-feeders increases steadily as overall abundance recovers, supporting increasing suspended food availability in the later LPTM. The restriction of specific taxa, such as Munseyella and Palmocohna, to particular intervals of the LPTM event provides provisional paleoenvironmental preferences for testing at other LPTM sites and other intervals. For example, these two taxa show similar distributions at Maud Rise sites separated by 1,000 m depth, while a third, Propontocypris, is present before the LPTM only at the deeper site. This comparative approach provides feed-back on both environment and ecology for more holistic paleoecological reconstructions. 5.6. Future Directions Improved understanding of the deep-ocean ostracodes and the Late Paleocene Thermal Maximum could be achieved through numerous means. First and most straightforward is increased sampling of Site 689B to allow more precise comparison of shallower faunal dynamics and timing to the higher-resolution Site 690B record presented here. Second, examination of any sites at any paleolatitude or paleodepths, particularly those with good stable isotope stratigraphies and benthic foram faunal analyses (e.g., equatorial Pacific Allison Guyot Site 865; Bralower et al„ 1995), would improve our picture of ostracode biogeographic sensitivity and response to the event. However, good sections of the LPTM are relatively rare in existing deep-ocean cores and ostracodes are reportedly rare within these candidates (Thomas, 1998; Thomas, pers. comm.). Third, little sedimentological study has been conducted on LPTM sections Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 334 relative to stable isotope and benthic foram studies. Analyses of dissolution indices and weight % of different sedimentary grains (e.g., benthic forams, biogenic opal), would provide a more complete environmental context for interpretation of both ostracode and foram abundance, susceptibility to dissolution, and response to surface productivity changes. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 335 6. Summary of major conclusions 6.1. No evidence for temperature change in Kerguelen Plateau bottom-waters through the latest Eocene to earliest Oligocene The Eocene-Oligocene transition is marked by the pronounced stable isotope event termed Oi-1, in which bethic foraminiferal 8,80 decreased rapidly by an average of -1.5 %o. This decrease reflects some combination of development of major ice-volume on Antarctica and thermal cooling of bottom-waters, but partitioning the two controls is difficult without independent constraints on ice-volume, seawater 5,80 , or bottom water temperatures. To address this uncertainty, the recently documented temperature-dependence of Mg/Ca in the low-Mg calcitic valves of Krithe ostracodes was applied to material from the Kerguelen Plateau using inductively coupled plasma mass-spectrometry (ICP-MS). Comparisons of valve morphology, area, mass, and translucence to determined elemental ratios show no strong correlations, while diagenetic screening, principally by cathodoluminescence (CL) and limited SEM study, shows no positive evidence for diagenetic alteration except for minor dissolution. Individual Krithe Mg/Ca shows relatively high intersample variability and no net trends through the oxygen isotope excursion. Simplistic interpretation of this pattern supports the hypothesis that the 51 s O decrease largely reflects major Antarctic ice-sheet development in response to thermal isolation of the continent following its isolation by oceanic gateways. This absence of a major temperature factor at Kerguelen Plateau is consistent with on-going benthic foram-based Mg/Ca studies at Maud Rise. IRRegardless, the field of elemental paleothermometry is in its infancy with major questions on the primary (i.e., physiological) and secondary (i.e., diagenetic) controls upon modern and ancient biological carbonate remaining unanswered. Thus, these data Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 336 are considered a small experimental exercise rather than robust evidence for the nature of climatic dynamics during the Eocene-Oligocene transition. 6.2. Common and reversed faunal changes at Kerguelen Plateau and Maud Rise coincide with the onset of Oi-1 Ostracodes are common constituents of Eocene-Oligocene sediments on both Maud Rise (689B) and Kerguelen Plateau (744B), which were separated by roughly 90° paleolongitude, but positioned at similar paleodepths (1,400-1,800 m) and distances from Antarctica. Major changes in the faunal abundance, richness, and evenness coincide with the onset OF the major Oi-1 step increase in 61 8 0 marking the onset of significant Antarctic glaciation and possibly thermohaline cooling. Abundance changes reflect possible dilution effects at Site 744A, and biologically or preservationally deleterious conditions related to seawater corrosivity at both sites, particulary Site 689B, where corrosivity is high throughout the interval. Following the event, Site 744A faunaL composition shifted slightly back towards initial compositions, whereas Site 689B faunas remained relatively distinct from earlier faunas. This major increase in high-latitude “refrigeration" likely led to the initiation or increase of thermohaline-driven circulation, and certainly significant increases in primary productivity, likely highly seasonal in nature. The Oi-1-coincident faunal changes among the sites include both uniform (e.g., loss of Cytherella and Bairdioppilata at both sites) and reversed (e.g., Site 689B loss and Site 744A appearance of Trachyleberis and Actinocythereis). Together, these faunal patterns imply a common site response to the increase in surface productivity (e.g., Cytherella and Bairdoppilata), as well as a disparate site response to either subtle productivity intensity differences or changing bottom water conditions (e.g., Trachyleberis and Actinocythereis). In contrast, previous benthic foram studies have found little faunal Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 337 changes at these sites coinciding directly with Oi-1 (Thomas, 1992). Thus, ostracodes appear more sensitive to Oi-1-coincident environmental changes than benthic forams. Ostracodes show no faunal response to the sedimentology-based intervals of proto-AABW and WSBW in the upper and lower portions of the study interval at Site 744A. 6.3. Algulhasina quadratica shows significant ontogenetic size increase in response to early Eocene surface productivity increase Morphometric examination of seven discrete ontogenetic stages in the ostracode Algulhasina quadrata at Site 744A documented significant size increases in most instar stages after the onset of the Oi-1 event. The size increase was not accompanied by obvious valve architectural changes or increased valve-secreting cells in the epidermis. The significant positive correlation of size with biogenic opal supports a model where increased food availability allowed increased instar size through sequestering of greater metabolic reserves. However, such hypotheses are difficult to test without independent data on ontogenetic timing of development. This body size:surface productivity linkage is also consistent with the biogenic opal as a primary signal of surface productivity and not a preservational artifact. Across the Oi-1 transition from relatively low-organic carbon/carbonate- saturated/slightly warmer to relatively high-organic carbon/carbonate- undersaturated/possibly slightly cooler bottom-waters, increased food availability appears to have “ won the battle” over decreased carbonate availability to produce generally larger individuals throughout the later ontogeny of Algulhasina quadrata. 6.4. Common faunal changes at Kerguelen Plateau and Maud Rise support temperature increase, but not prolonged dysoxia, during LPTM The study documents a major response of deep-ocean metazoans to the extreme global warming associated with a major carbon cycle perturbation at the end of the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 338 Paleocene. At widely separated Southern Ocean sites on the Maud Rise and Kerguelen Plateau, deep-ocean ostracode abundances drop precipitously in association with the onset of the carbon isotope excursion and benthic foram extinction horizon, with valves generally thinner and fragmentation higher. After the initial portion of the LPTM, abundances increase to roughly two-thirds to completely to pre-LPTM amounts. The Maud Rise fauna was overall taxonomically richer than that of Kerguelen Plateau and suffered less loss during the LPTM. Kerguelen Plateau richness recovered after the LPTM, whereas Maud Rise richness remained low, but less variable overall, perhaps reflect fingering deleterious conditions. Cluster and PCA faunal analyses demonstrate that Site 690B consists of three distinct faunas largely bounded by the carbon spike and recover to relatively stable, if elevated, stable isotope values. A similar, but less robust, faunal pattern is present at Site 738C, which would benefit from additional stable isotope data. Prolonged dysoxia of bottom-water as a mechanism for benthic foram extinction is not supported by the sample-based continuity of the higher-oxygen-demanding ostracodes. In addition, the proportion of dysoxia-tolerant filter-feeding ostracode taxa does not increase coincident with, or immediately following the CIE, as would be expected from physiology and other documented intervals of dysoxia. Instead, the proportion of filter-feeders increases steadily as overall abundance recovers, supporting increasing suspended food availability in the later LPTM. The restriction of specific taxa, such as Munseyella and Palmocohna, to particular intervals of the LPTM event provides provisional paleoenvironmental preferences for testing at other LPTM sites and other intervals. For example, these two taxa show similar distributions at Maud Rise sites separated by 1,000 m depth, while a third, Propontocypris, is present before the LPTM only at the deeper site. This Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 339 comparative approach allows feedback on both environment and organisms for more holistic paleoecological reconstructions. As the nature of these climate extrema become better understood, these events will provide a powerful tool for examining evolving paleoecologies and relative physiological limits of ostracode taxa in the deep-ocean. For example, if the potential of combined elemental-isotopic analyses is realized and provides a robust means to reconstruct more accurate paleotemperatures, then such climate extrema may serve as historical experiments through which the evolving thermal tolerance of specific taxa may be compared through time. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 340 Bibliography Aubry, M. P., 1999, Stratigraphic (dis)continuity and temporal resolution of geological events in the upper Paleocene-lower Eocene deep sea record, in Aubry, M. P., Lucas, S. G., and Berggren, W. A., eds., Late Paleocene-early Eocene climatic and biotic events in the marine and terrestrial records: New York, NY, Columbia University Press, p. 37-66. Aubry, M. P., Berggren, W. A., Stott, L., and Sinha, A., 1996, The upper Paleocene- lower Eocene stratigraphic record and the Paleocene-Eocene boundary carbon isotope excursion; implications for geochronology, in Knox, R. W. O. B„ Corfield, R. M., and Dunay, R. E., eds., Correlation of the early Paleogene in Northwest Europe: London, United Kingdom, Geological Society of London, p. 353-380. Ayress, M., Neil, H., Passlow, V., and Swanson, K., 1997, Benthonic ostracods and deep watermasses: A qualitative comparison of Southwest Pacific, Southern and Atlantic Oceans: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 131, p. 287-302. Babinot, J. F., and Airaud, C., 1990, The effect of global events on the evolution of Cenomanian and Turonian marginal Tethyan ostracod faunas in the Mediterranean region, in Whatley, R. C., and Maybury, C„ eds., Ostracoda and global events: London, England, Chapman and Hall, p. 25-40. Baldauf, J. G., and Barron, J. A., 1990, Evolution of biosiliceous sedimentation patterns, Eocene through Quaternary; paleoceanographic response to polar cooling, in Bleil, U., and Thiede, J., eds., Proceedings of the 1988 NATO advanced research workshop on Geological history of the polar oceans; Arctic versus Antarctic, Volume 308: Dordrecht, Germany, D. Reidel Publishing Company, p. 575-607. Balman, K„ 1997, Cainozoic deep sea Ostracoda from the South Atlantic and Indian Oceans [Ph. D. Dissertation thesis], University of Wales - Aberystwyth. Barker, P. F., Kennett, J. P., eta!., 1988a, Site 689, in Barker, P. F., Kennett, J. P., et al., eds., Proceedings of the Ocean Drilling Program, Initial Reports, Volume 113: College Station, TX, Ocean Drilling Program, p. 89-181. Barker, P. F., Kennett, J. P., et al., 1988b, Site 690, in Barker, P. F., Kennett, J. P., et al., eds., Proceedings of the Ocean Drilling Program, Initial Reports, Volume 113: College Station, TX, Ocean Drilling Program, p. 183-292. Barrera, E., and Huber, B. T., 1991, Paleogene and early Neogene oceanography of the southern Indian Ocean; Leg 119 foraminifer stable isotope results, in Barron, J. A., Larsen, B., et al., eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 119: College Station, TX, Ocean Drilling Program, p. 693-717. Barrera, E., and Huber, B. T., 1993, Eocene to Oligocene oceanography and temperatures in the Antarctic Indian Ocean, in Kennett, J. P., and Warnke, D. A., eds., The Antarctic paleoenvironment: A perspective on global change: Washington, DC, American Geophysical Union, p. 49-65. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 341 Barron, J., Larsen, B., et al., 1989a, Site 744, in Barron, J., Larsen, B., et al., eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 119: College Station, TX, Ocean Drilling Program, p. 477-504. Barron, J., Larsen, B., et al., 1989b, Site 738, in Barron, J., Larsen, B., et al., eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 119: College Station, TX, Ocean Drilling Program, p. 477-504. Barron, J. A., Barrera, E., et al., 1991a, Biochronologic and magnetochronologic synthesis of Leg 119 sediments from the Kerguelen Plateau and Prydz Bay, Antarctica, in Barron, J. A., Larsen, B., et al., eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 119: College Station, TX, Ocean Drilling Program, p. 813-847. Barron, J. A., Larsen, B., and Baldauf, J. G., 1991b, Evidence for late Eocene to early Oligocene Antarctic glaciation and observations on late Neogene glacial history of Antarctica; results from Leg 119, in Barron, J. A., Larsen, B., et al., eds., Proceedings of the Ocean Drilling Program; Scientific Results, Volume 119: College Station, TX, Ocean Drilling Program, p. 869-891. Bartek, L. R., Sloan, L. C., Anderson, J. B., and Ross, M. I., 1992, Evidence from Antarctic continental margin of late Paleogene ice sheets: a manifestation of plate reorganization and synchronous changes in atmospheric circulation over the emerging southern ocean, in Prothero, D. A., and Berggren, W. A., eds., Eocene- Oligocene Climatic and Biotic Evolution: Princeton, NJ, Princeton University Press, p. 131-159. Beck, J. W., Edwards, R. L., et al., 1992, Sea surface temperature from skeletal Sr/Ca ratios: Science, v. 257, p. 585-716. Benson, R. H., 1959, Ecology of Recent ostracodes of the Todos Santos Bay region, Baja California, Mexico: Kansas University Paleontological Contribution, v. 23, 80 p. Benson, R. H., 1972, The Bradleya Problem, With Descriptions of Two New Psychrospheric Ostracode Genera, Agrenocythere and Poseidonamicus (Ostracoda; Crustacea): Washington, DC, Smithsonian Institution, 100 p. Benson, R. H., 1975a, The origin of the psychrosphere as recorded in changes of deep- sea ostracode assemblages: Lethaia, v. 8, p. 69-83. Benson, R. H., 1975b, Morphological stability in Ostracoda: Bulletin of American Paleontology, v. 65, p. 13-46. Benson, R. H., 1981, Form, function, and architecture of ostracode shells: Annual Reviews in Earth and Planetary Sciences, v. 9, p. 59-80. Benson, R. H., 1982, Comparative transformation of shape in a rapidly evolving series of structural morphotypes of the ostracod Bradleya, in Bate, R. H., Robinson, E., and Sheppard, L. M., eds., Fossil and Recent Ostracods: Chichester, England, Ellis Horwood Ltd. Publishers, p. 147-163. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 342 Benson, R. H., 1983, Biomechanical stability and suddden change in the evolution of the deep-sea ostracode Poseidonamicus: Paleobiology, v. 9, p. 398-413. Benson, R. H., 1984, Estimating greater paleodepths with ostracodes, especially in past thermospheric oceans: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 48, p. 107-141. Benson, R. H., 1988, Ostracods and palaeoceanography, in De Deckker, P., Colin, J. P., and Peypouquet, J. P., eds., Ostracoda in the earth sciences: Amsterdam, Netherlands, Elsevier, p. 1-26. Benson, R. H., 1990, Ostracoda and the discovery of global Cainozoic palaeoceanographic events, in Whatley, R„ and Maybury, C., eds., Ostracoda and Global Events: London, England, Chapman and Hall, p. 41-58. Benson, R. H., Berdan, J. M., et al., 1961, Part Q, Arthropoda 3, Crustacea, Ostracoda: Lawrence, KS, Geological Society of America and University of Kansas Press, 442 P- Benson, R. H., Chapman, R. E., and Deck, L. T„ 1984, Paleoceanographic events and deep-sea ostracodes: Science, v. 224, p. 1334-1336. Benson, R. H., Chapman, R. E., and Deck, L. T., 1985, Evidence from the Ostracoda of major events in the South Atlantic and world-wide over the past 80 million years, in Hsu, K. J., and Weissert, H. J., eds., South Atlantic Paleoceanography: Cambridge, England, Cambridge University Press, p. 325-350. Benson, R. H., DelGrosso, R. M., and Steineck, L., 1983, Ostracode distribution and biofacies, Newfoundland continental slope and rise: Micropaleontology, v. 29, p. 430-453. Benson, R. H., and Peypouquet, J., 1983a, The upper and mid-bathyal Cenozoic ostracode faunas of the Rio Grande Rise found on Leg 72 Deep Sea Drilling Project, in Barker, P. F„ Johnson, D. A., et al., eds., Initial reports of the Deep Sea Drilling Project, Volume 72, p. 805-818. Benson, R. H„ and Peypouquet, J. P., 1983b, The upper and mid-bathyal Cenozoic ostracode faunas of the Rio Grande Rise found on Leg 72 Deep Sea Driiling Project, in Barker, P. F„ Johnson, D. A., et al., eds., Initial reports of the Deep Sea Drilling Project, Volume 72: College Station, TX, Ocean Drilling Program, p. 805-818. Berger, W. H., 1973, Deep-sea carbonates; Pleistocene dissolution cycles: Journal of Foraminiferal Research, v. 3, p. 187-195. Berger, W. H., 1976, Biogenous deep sea sediments; production, preservation and interpretation, in Riley, J. P., and Chester, R., eds., Chemical oceanography; Volume 5: New York, NY, Academic Press, p. 265-388. Berger, W. H., 1982, Climate steps in ocean history—lessons from the Pleistocene, Climate and Earth History: Washington, DC, National Academy Press, p. 43-54. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 343 Berggren, W. A., Kent, D. V., Swisher, C. C., Ill, and Aubry, M. P., 1995, A revised Cenozoic geochronology and chronostratigraphy, in Berggren, W. A., Kent, D. V., et al., eds., Geochronology, time scales and global stratigraphic correlation, SEPM Special Publication 54: Tulsa, OK, Society for Sedimentary Geology, p. 129-212. Berggren, W. A., Lucas, S., and Aubry, M., 1999, Late Paleocene-Early Eocene climatic and biotic evolution: an overview, in Aubry, M., Lucas, S., and Berggren, W. A., eds., Late Paleocene-Early Eocene climatic and biotic changes: New York, NY, Columbia University Press, p. 1-17. Billups, K., and Spero, H., 1996, Reconstructing the stable isotope geochemistry and paleotemperatures of the equatorial Atlantic during the last 150,000 years: Results from individual foraminifera: Paleoceanography, v. 11, p. 217-238. Bodergat, A. M., Carbonnel, G„ Rio, M., and Keyser, D., 1993, Chemical composition of Leptocythere psammophila (Crustacea: Ostracoda) as influenced by winter metabolism and summer supplies: Marine Biology, v. 117, p. 53-62. Boersma, A., Premoli, S. I., and Hallock, P., 1999, Trophic models for the well-mixed and poorly mixed warm oceans across the Paleocene/Eocene epoch boundary, in Aubry, M. P., Lucas, S. G., and Berggren, W. A., eds., Late Paleocene-early Eocene climatic and biotic events in the marine and terrestrial records: New York, NY, Columbia University Press, p. 204-213. Boomer, I., and Whatley, R., 1992, Ostracoda and dysaerobia in the Lower Jurassic of Wales: the reconstruction of past oxygen levels: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 99, p. 373-379. Boomer, I., and Whatley, R., 1995, Cenozoic Ostracoda from guyots in the Western Pacific; Holes 865B and 866B (Leg 143), in Winterer, E. L„ Sager, W. W., et al., eds., Proceedings of the Ocean Drilling Program; Scientific Results, Volume 143: College Station, TX, Ocean Drilling Program, p. 75-86. Bralower, T. J., Zachos, J. C., et al., 1995, Late Paleocene to Eocene paleoceanography of the equatorial Pacific Ocean: Stable isotopes recorded at Ocean Drilling Program Site 865, Allison Guyot: Paleoceanography, v. 10, p. 841-865. Brass, G. W., Southam, J. R., and Peterson, W. H., 1982, Warm saline bottom water in the ancient ocean: Nature, v. 296, p. 620-623. Bremer, M. L., and Lohmann, G. P., 1982, Evidence for primary control of the distribution of certain Atlantic Ocean benthonic foraminifera by degree of carbonate saturation: Deep-Sea Research, v. 29, p. 987-998. Broecker, W. S., 1982, Glacial to interglacial changes in ocean chemistry: Progress in Oceanography, v. 11, p. 151-197. Broecker, W. S., and Peng, T.-H., 1982, Tracers in the Sea: Palisades, NY, Lamont- Doherty Geological Observatory, 690 p. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 344 Broecker, W. S., Peteet, D. M., and Rind, D., 1985, Does the ocean-atmosphere system have more than one stable mode of operation?: Nature, v. 315, p. 21-26. Brown, J., Colling, A., et al., 1989, Ocean Circulation: Oxford, England, Open University and Pergamon Press, 238 p. Bruland, K. W., 1983, Trace elements in sea water, in Riley, J. P., and Chester, R., eds., Chemical Oceanography: London, England, Academic Press, p. 157-220. Brusca, R. C., and Brusca, G. J., 1990, Invertebrates: Sunderland, MA, Sinauer Associates, Inc., 922 p. Buzas, M. A., and Gibson, T. G., 1969, Species diversity; benthonic foraminifera in western North Atlantic: Science, v. 163, p. 72-73. Cadot, H. M., Jr., and Kaesler, R. L., 1977, Magnesium content of calcite in carapaces of benthic marine Ostracoda: University of Kansas Paleontological Contributions Paper, v. 87, 23 p. Cadot, H. M., Van Schmus, W. R., and Kaesler, R. L., 1972, Magnesium in calcite of marine Ostracoda: Geological Society of America Bulletin, v. 83, p. 3519-3522. Cande, S. C., and Kent, D., 1995, A revised geomagnetic polarity timescale for the Late Cretaceous and Cenozoic: Journal of Geophysical Research, v. 100, p. 6093- 6095. Carbonnel, P., and Tolderer-Farmer, M., 1988, The ostracode carapace as a hydrochemical source of information at water-sediment interface, Developments in Paleontology and Sedimentology, Volume 11, p. 341-353. Chamberlin, T. C., 1906, On a possible reversal of deep-sea circulation and its influence on geological climates: Journal of Geology, v. 14, p. 363-373. Cheetham, A. H., and Hazel, J. E., 1969, Binary (presence-absence) similarity coefficients: Journal of Paleontology, v. 43, p. 1130-1136. Chivas, A. R., De Deckker, P., and Shelley, J. M. G., 1983, Magnesium, strontium, and barium partitioning in nonmarine ostracode shells and their use in paleoenvironmental reconstructions - a preliminary study, in Maddocks, R. F., ed., Applications of Ostracoda: Houston, TX, University of Houston Geoscience, p. 238-249. CLIMAP Members, 1986, Seasonal reconstructions of the Earth's surface at the last glacial maximum: Geological Society of America Map Chart Series, v. MC-36. Clyde, W. C., and Gingerich, P. D., 1998, Mammalian community response to the latest Paleocene thermal maximum; an isotaphonomic study in the northern Bighorn Basin, Wyoming: Geology, v. 26, p. 1011-1014. Cohen, A. C., and Morin, J. G., 1990, Patterns of reproduction in ostracodes: a review: Journal of Crustacean Biology, v. 10, p. 184-211. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 345 Coles, G. P., 1990, A comparison of the evolution, diversity and composition of the Cainozoic Ostracoda in the deep water North Atlantic and shallow water environments of North America and Europe, in Whatley, R. C., and Maybury, C., eds., Ostracoda and Global Events: New York, NY, Chapman and Hall, p. 71-86. Coles, G. P., Ayress, M., and Whatley, R., 1990, A comparison of North Atlantic and Pacific Cainozoic deep-sea ostracoda, in Whatley, R. C., and Maybury, C., eds., Ostracoda and Global Events: New York, NY, Chapman and Hall, p. 287-306. Coles, G. P., Whatley, R. C., and Moguilevsky, A., 1994, The ostracod genus Krithe from the Tertiary and Quaternary of the North Atlantic: Palaeontology, v. 37, p. 71- 120 . Colwell, R. K., and Coddington, J. A., 1994, Estimating terrestrial biodiversity through extrapolation: Philosphical Transactions of the Royal Society of London, v. 345, p. 101-118. Connor, E. F., and McCoy, E. D., 1979, The statistics and biology of the species-area relationship: American Naturalist, v. 113, p. 791-832. Corfield, R. M., and Norris, R. D„ 1996, Deep water circulation in the Paleocene ocean, in Knox, R. W. O. B., Corfield, R. M., and Dunay, R. E., eds., Correlation of the early Paleogene in Northwest Europe: London, England, Geological Society of London, p. 443-456. Corfield, R. M., and Norris, R. D., 1999, The oxygen and carbon isotopic context of the Paleocene/Eocene epoch boundary, in Aubry, M. P., Lucas, S. G., and Berggren, W. A., eds., Late Paleocene-early Eocene climatic and biotic events in the marine and terrestrial records: New York, NY, Columbia University Press, p. 124-137. Correge, T., 1993, Late Quaternary palaeoceanography of the Queensland Trough (western Coral Sea) based on Ostracoda and the chemical composition of their shells [Ph.D. thesis], Australian National University. Correge, T., and DeDeckker, P., 1997, Faunal and geochemical evidence for changes in intermediate water temperature and salinity in the western Coral Sea (northeast Australia) during the Late Quaternary: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 313, p. 183-205. Craig, H., and Gordon, L. I., 1965, Isotopic oceanography: Deuterium and oxygen-18 variations in the ocean and the marine atmosphere, Proceedings of the Symposium on Marine Geochemistry, Volume 3, University of Rhode Island, p. 277-374. Cronin, T. M., 1979, Late Pleistocene marginal ostracodes from the southeastern Atlantic Coastal Plain and their environmental implications: Geographie Physique et Quaternaire, v. 33, p. 121. Cronin, T. M., 1983, Bathyal ostracodes from the Florida-Hatteras Slope, the Straits of Florida, and Blake Plateau: Marine Micropaleontology, v. 8, p. 89-119. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 346 Cronin, T. M., 1985, Speciation and stasis in marine Ostracoda: climatic modulation of evolution: Science, v. 227, p. 60-63. Cronin, T. M., DeMarino, 0. M„ Dwyer, G. S., and Rodriquez-Lazaro, J., 1999, Deep- sea ostracode species diversity: response to late Quaternary climate change: Marine Micropaleontology, v. 37, p. 231-249. Cronin, T. M., Dwyer, G. S., et al., 1996, Deep-sea ostracode shell chemistry (Mg:Ca ratios) and Late Quaternary Artie Ocean history, in Andrews, J. T., Bergsten, W. E. N., and Jennings, A. E., eds., Late Quaternary Paleoceanography of the North Atlantic Margins: Boulder, CO, Geological Society of America, p. 117-134. Cronin, T. M., and Hazel, J. E., 1980, Ostracode biostratigraphy of Pliocene and Pleistocene deposits of the Cape Fear Arch region, North and South Carolina: US Geological Survey Professional Paper, v. 1125-B, p. 1-25. Cronin, T. M., and Raymo, M. E., 1997, Orbital forcing of deep-sea benthic species diversity: Nature, v. 30, p. 624-627. Cronin, T. M„ T. R. Holtz, J., and Whatley, R. C., 1994, Quaternary paleoceanography of the deep Arctic Ocean based on quantitative analysis of Ostracoda: Marine Geology, v. 119, p. 305-332. Crowley, T. J., and North, G. R., 1988, Abrupt climate change and extinction events in Earth history: Science, v. 240, p. 996-1002. Dalingwater, J. E., and Mutvei, H., 1990, Arthropoda (Crustacea and Trilobita), in Carter, J. G., ed., Skeletal biomineralization: patterns, process and evolutionary trends: New York, NY, Van Nostrand Reinhold, p. 101. Delaney, M. L., 1989, Uptake of cadmium into calcite shells by planktonic foraminifera: Chemical Geology, v. 78, p. 159-165. Delaney, M. L., Be, A. W. H., and Boyle, E. A., 1985, Li, Sr, Mg, and Na in foraminiferal calcite shells from laboratory culture, sediment traps, and sediment cores: Geochimica et Cosmochimica Acta, v. 49, p. 1327-1341. Derry, L. A., and France-Lanord, C., 1996, Neogene growth of the sedimentary organic carbona reservoir: Paleoceanography, v. 11, p. 267-275. Diacron, P., and Efron, B., 1983, Computer-intensive methods in statistics: Scientific American, v. 248, p. 116-130. Dickens, G. R., O'Neil, J. R., Rea, D. K., and Owen, R. M., 1995, Dissociation of oceanic methane hydrate as a cause of the carbon isotope excursion at the end of the Paleocene: Paleoceanography, v. 10, p. 965-971. Diester-Haass, L., 1991, Eocene/Oligocene paleoceanography in the Antarctic Ocean, Atlantic sector (Maud Rise, OOP Leg 113, Site 689B and 690B): Marine Geology, v. 100, p. 249-276. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 347 Diester-Haass, L., 1993, Late Eocene-Oligocene sedimentation in the Antarctic Ocean, Atlantic Section (Maud Rise, ODP Leg 113, Site 689): Development of surface and bottom water circulation, in Kennett, J. P., and Wamke, D. A., eds., The Antarctic paleoenvironment: A perspective on global change: Washington, DC, American Geophysical Union, p. 185-202. Diester-Haass, L., 1995, Middle Eocene to early Oligocene paleoceanography of the Antarctic Ocean (Maud Rise, ODP Leg 113, Site 689): change from a low to a high productivity ocean: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 113, p. 311-334. Diester-Haass, L., 1996, Late Eocene-Oligocene paleoceanography in the southern Indian Ocean (ODP Site 744): Marine Geology, v. 130, p. 99-119. Diester-Haass, L., Robert, C., and Chamley, H., 1993, Paleoceanographic and paleoclimatic evolution in the Weddell Sea (Antarctica) during the middle Eocene-late Oligocene, from a coarse sediment fraction and clay mineral data (ODP Site 689): Marine Geology, v. 114, p. 233-250. Diester-Haass, L., Robert, C., and Chamley, H., 1996, The Eocene-Oligocene preglacial- glacial transition in the Atlantic section of the Southern Ocean (ODP Site 690): Marine Geology, v. 131, p. 123-149. Diester-Haass, L., and Zahn, R., 1996, Eocene-Oligocene transition in the Southern Ocean: History of water mass circulation and biological productivity: Geology, v. 24, p. 163-166. Dingle, R. V., 1971, Some Cretaceous ostracodal assemblages from the Agulhas Bank (South African continental margin): Transactions of the Royal Society of South Africa, v. 39, p. 393-418. Dingle, R. V., 1981, The Campanian and Maastrichtian Ostracoda of South-East Africa: Annals of the South African Museum, v. 85, p. 1-181. Dingle, R. V., 1985, Turonian, Coniacian, and Santonian Ostracoda from South-East Africa: Annals of the South African Museum, v. 96, p. 123-237. Dingle, R. V., and Giraudeau, J., 1993, Benthic Ostracoda in the Benguela System (SE Atlantic): A multivariate analysis: Marine Micropaleontology, v. 22, p. 71-92. Dingle, R. V., and Lord, A. R„ 1990, Benthic ostracods and deep water-masses in the Atlantic Ocean: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 80, p. 213-235. Dingle, R. V., Lord, A. R., and Boomer, I. D., 1989, Ostracod faunas and water masses across the continental margin off southwestern Africa: Marine Geology, v. 87, p. 323-328. Dingle, R. V., Lord, A. R., and Boomer, I. D., 1990, Deep-water Quaternary Ostracoda from the continental margin off south-western Africa (SE Atlantic Ocean): Annals of the South African Museum, v. 99, p. 245-366. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 348 Douglas, R., and Woodruff, F., 1981, Deep-sea benthic foraminifera, in Emiliani, C., ed., The oceanic lithosphere: New York, NY, John Wiley and Sons, p. 1233- 1327. Dromgoole, E. L., and Walter, L. M., 1990, Iron and manganese incorporation into calcite: Effects of growth kinetics, temperature and solid solution: Chemical Geology, v. 81, p. 311-336. Dwyer, G. S., Cronin, T. M., et al., 1995, North Atlantic deep-water temperature change and late Quaternary climatic cycles: new insight from ostracode shell chemistry: Science, v. 270, p. 1347-1351. Efron, B., 1979, Bootstrap methods: another look at the jackknife: Annual Statistics, v. 7, p. 1-26. Efron, B., and Gong, G., 1983, A leisurely look at the boostrap, the jackknife, and cross- validation: American Statistician, v. 37, p. 36-48. Ehrmann, W. U., and Mackensen, A., 1992, Sedimentological evidence for the formation of an east Antarcticisce sheet in Eocene/Oligocene time: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 93, p. 85-112. Emiliani, C., 1955, Pleistocene temperatures: Journal of Geology, v. 63, p. 538-578. Epstein, S., Buchsbaum, R., Lowenstam, H. A., and Urey, H. C., 1953, Revised carbonate-water isotopic temperature scale: Geological Society of America Bulletin, v. 64, p. 1315-1326. Fischer, A. G., 1981, Climatic oscillations in the biosphere, in Nitecki, M. H., ed., Biotic Crises in Ecological and Evolutionary Time: New York, NY, Academic Press, p. 103-131. Fischer, A. G., 1984, The two Phanerozoic supercycles, in Berggren, W. A., and Van Couvering, J. A., eds., Catastrophes and Earth History: Princeton, NJ, Princeton University Press, p. 129-150. Fischer, A. G., and Arthur, M. A., 1977, Secular variations in the pelagic realm, in Cook, H. E., and Enos, P., eds., Deep Water Carbonate Environments: Houston, TX, Society of Economic Paleontologists and Mineralogists, p. 18-50. Foldvik, A., and Gammelsrod, T., 1988, Notes on Southern Ocean hydrography, sea-ice and bottom water formation, in Olausson, E., ed., The Southern Ocean-the Antarctic; present and past: Amsterdam, Netherlands, Elsevier, p. 3-17. Foster, D. W., and Kaesler, R. L., 1988, Analysis of heterochrony; Shape analysis ideas from the Ostracoda, in McKinney, M. L., ed., Heterochrony in evolution; a multidisciplinary approach.: New York, NY, Plenum Press, p. 53-69. Freeman, K. H., and Hayes, J. M., 1992, Fractionation of carbon isotopes by phytoplankton and estimates of ancient C 02 levels: Global Biogeochemical Cycles, v. 6, p. 185-198. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 349 Gibbs, M. T., and Kump, L. R., 1994, Geochemical erosion during the last glacial maximum and the present: sensitivity to changes in lithology and hyddrology: Paleoceanography, v. 9, p. 529-543. Gooday, A. J., 1993, Deep-sea benthic foraminiferal species which exploit phytodetritus; characteristic features and controls on distribution: Marine Micropaleontology, v. 22, p. 187-205. Gotteli, N. J., and Graves, G. R., 1996, Null Models in Ecology: Washington, DC, Smithsonian Institution Press, 368 p. Gould, S. J., and Calloway, C. B., 1980, Clams and brachiopods: ships that pass in the night: Paleobiology, v. 6, p. 383-396. Guernet, C., 1982, Contribution a I'etude des faunes abyssales; les Ostracodes paleogenes du bassin des Bahamas, Atlantique Nord (D.S.D.P., Leg. 44) [Abyssal fauna; Paleogene ostracods of the Bahamas Basin, North Atlantic, DSDP, Leg 44]: Revue de Micropaleontologie, v. 25, p. 40-56. Guilderson, T. P., Fairbanks, R. J., and Rubenstone, J. L., 1993, Tropical temperature variations since 20,000 years ago: modulating inter-hemispheric climate change: Science, v. 263, p. 663-665. Hardie, L. A., 1996, Secular variations in seawater chemistry: an explanation for the coupled secular variation in the mineralogies of marine limestones and potash evaporites over the past 600 m.y.: Geology, v. 24, p. 279-283. Hastings, D. W., Russel, A. D„ and Emerson, S. R., 1998, Foraminiferal magnesium in Globigerinoides sacculifer as a paleotemperature proxy: Paleoceanography, v. 13, p. 161-169. Hayek, L. C., and Buzas, M. A., 1997, Surveying Natural Populations: New York, NY, Columbia University Press, 563 p. Hazel, J. E., 1988, Determining Late Neogene and Quaternary palaeoclimates and palaeotemperature regimes using ostracods, in De Deckker, P., Colin, J.-P., and Peypouquet, J.-P., eds., Ostracoda in the Earth Sciences: Amsterdam, Netherlands, Elsevier, p. 89-101. Heip, C., 1976, The life-cycle of Cyprideis torosa (Crustacea, Ostracoda): Oecologia, v. 24, p. 229-245. Hemming, N. G., Meyers, W. J., and Grams, J. C., 1989, Cathodoluminescence in diagenetic calcites: The roles of Fe and Mn as deduced from electron probe and spectrophotometric measurements: Journal of Sedimentary Petrology, v. 59, p. 404-411. Herguera, J. C., and Berger, W. H., 1991, Paleoproductivity from benthic foraminifera abundance; glacial to postglacial change in the west-equatorial Pacific: Geology, v. 19, p. 1173-1176. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 350 Herman, P. M. J., and Heip, C., 1982, Growth and respiration of Cyprideis torosa Jones 1850 (Crustacea Ostracoda): Oecologia, v. 54, p. 300-303. Herman, P. M. J., Heip, C., and Vranken, G., 1983, The production of Cyprideis torosa Jones 1850 (Crustacea, Ostracoda): Oecologia, v. 58, p. 326-331. Horne, D. J., and Ian; Rosenfeld, A., 1990, Recovering from the effects of an oceanic anoxic event; Turanian Ostracoda from S.E. England, in Whatley, R. C., and Maybury, C., eds., Ostracoda and global events: London, England, Chapman and Hall, p. 123-138. Hurlbert, S. H., 1971, The nonconcept of species diversity: A critique and alternative parameters: Ecology, v. 52, p. 577-586. Ikeya, N., and Cronin, T. M., 1993, Quantitative analysis of Ostracoda and water masses around Japan; application to Pliocene and Pleistocene paleoceanography: Micropaleontology, v. 39, p. 263-281. Ishizaki, K., 1975, Morphological variation in Leguminocythereis ? hodgii (Brady), Ostracoda (Crustacea), from Japan: Bulletins of American Paleontology, v. 65, p. 245-262. Jarvis, I., Carson, G. A., et al., 1988, Microfossil assemblages and the Cenomanian- Turonian (Late Cretaceous) Oceanic Anoxic Event: Cretaceous Research, v. 9, p. 3-103. Kaesler, R. L., 1966, Quantitative re-evaluation of ecology and distribution of Recent Foraminifera and ostracoda of Todos Santos Bay, Baja California, Mexico: Kansas University Paleontological Contributions, v. 10, 50 p. Kaesler, R. L., 1975, Morphology of Cypridopsis vidua (O. F. Muller): Variation with environment: Bulletin of American Paleontology, v. 65, p. 225-244. Kaesler, R. L., and Lohmann, K. C., 1976, Phenotypic variations of populations of Krithe producta with environment: Abhandlungen des Naturwissenschaftlichen Vereins in Hamburg, v. 18-19, p. 279-286. Kaiho, K., 1994, Planktonic and benthic foraminiferal extinction events during the last 100 m.y: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 111, p. 45-71. Kaiho, K., Arinobu, T., et al., 1996, Latest Paleocene benthic foraminiferal extinction and environmental changes at Tawanui, New Zealand: Paleoceanography, v. 11, p. 447-465. Kaiser, H. F., 1958, The varimax criterion for analytic rotation in factor analysis: Psychometrika, v. 23, p. 187-200. Kamiya, T., 1988, Contrasting population ecology of two species of Loxoconcha (Ostracoda, Crustacea) in Recent Zostera (eelgrass) beds; adaptive differences between phytal and bottom-dwelling species: Micropaleontology, v. 34, p. 316- 331. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 351 Kamiya, T., and Hazel, J. E., 1992, Shared versus derived characters in the pore- system of Loxoconcha (Ostracoda, Crustacea): Journal of Micropalaeontology, v. 11, p. 159-166. Kasting, J. F., and Richardson, S. M„ 1985, Seafloor hydrothermal activity and spreading rates: the Eocene carbon dioxide greenhouse revisited: Geochimica et Cosmochimica Acta, v. 49, p. 2541-2544. Katz, A., 1973, The interaction of magnesium with calcite crystal growth at 25° - 90° C and one atmosphere: Geochimica et Cosmochimica Acta, v. 37, p. 1536-0000. Keating, B. H., and Sakai, H., 1991, Magnetostratigraphic studies of sediments from Site 744, southern Kerguelen Plateau: Proceedings of the Ocean Drilling Program, Scientific Results, v. 119, p. 771-794. Keen, M. C., 1982, Intraspecific variation in Tertiary ostracods, in Bate, R. H., Robinson, E., and Sheppard, L. M., eds., Fossil and Recent Ostracods: West Sussex, England, Ellis Horwood, Ltd., p. 381-405. Kelly, D. C., Bralower, T. J., and Zachos, J. C., 1998, Evolutionary consequences of the latest Paleocene thermal maximum for tropical planktonic Foraminifera: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 141, p. 139-161. Kelly, D. C., Bralower, T. J., et al., 1996, Rapid diversification of planktonic foraminifera in the tropical Pacific (ODP Site 865) during the late Paleocene thermal maximum: Geology, v. 24, p. 423-426. Kennett, J. P., and Barker, P. F., 1990, Latest Cretaceous to Cenozoic climate and oceanographic developments in the Weddell Sea, Antarctica: An ocean-drilling perspective, in Barker, P. F., and Kennett, J. P., eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 113: College Station, TX, Ocean Drilling Program, p. 937-960. Kennett, J. P., Houtz, R. E., etal., 1975, Cenozoic paleoceanography in the Southwest Pacific Ocean, Antarctic glaciation, and the development of the Circum-Antarctic Current: Initial Reports of the Deep Sea Drilling Project, v. 29, p. 1155-1169. Kennett, J. P., and Shackleton, N. J., 1976, Oxygen isotope evidence for the development of the psychrosphere 38 m.y. ago: Nature, v. 260, p. 513-515. Kennett, J. P., and Stott, L. D., 1990, Proteus and Proto-Oceanus, Paleogene oceans as revealed from Antarctic stable isotopic results; ODP Leg 113, in Barker, P. F., and Kennett, J. P., eds., Proceedings of the Ocean Drilling Program; Scientific Results, Volume 113, p. 865-880. Kennett, J. P., and Stott, L. D., 1991, Abrupt deep-sea warming, palaeoceanographic changes and benthic extinctions at the end of the Paleocene: Nature, v. 353, p. 225-229. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 352 Keyser, D., 1990, Morphological changes and function of the inner lamella of podocopid Ostracoda, in Whatley, R. C., and Maybury, C„ eds., Ostracoda and global events: London, England, Chapman and Hall, p. 401-410. Koch, P. L., Zachos, J. C., and Dettman, D. L., 1995, Stable isotope stratigraphy and paleoclimatology of the Paleogene Bighorn Basin (Wyoming, USA): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 115, p. 61-89. Kornicker, L. S., and Wise, C. D., 1960, Some environmental boundaries of a marine ostracod: Micropaleontology, v. 6, p. 393-398. Kort, V. G., 1962, The Antarctic Ocean: Scientific American, v. 207, p. 113-132. Kvenvolden, K. A., 1993, Gas hydrates, geological perspective and global change: Reviews of Geophysics, v. 31, p. 173-187. Larson, R. L., 1991, Latest pulse of Earth: Evidence for a mid-Cretaceous superplume: Geology, v. 19, p. 547-550. Lasaga, A. C., Bemer, R. A., and Garrels, R. M., 1985, An improved geochemical model of atmospheric C 02 fluctuations over the past 100 million years, in Sundquist, E. T., and Broecker, W. S., eds., Natural variation in carbon dioxide and the carbon cycle: Washington, DC, American Geophysical Union, p. 397-411. Latifa, G. A., 1987, Effect of temperature on the development of embryonic and postembryonic stages of Heterocypris incongruens (Ramdohr, 1808): Crustaceana, v. 52, p. 95-100. Lawver, L. A., Gahagan, L. M., and Coffin, M. F„ 1992, The development of paleoseaways around Antarctica, in Kennett, J. P., and Barron, J. A., eds., The Antarctic Paleoenvironment: A perspective on global change: Washington, DC, American Geophysical Union, p. 7-30. Lethiers, F., and Whatley, R., 1994, The use of Ostracoda to reconstruct the oxygen levels of late Palaeozoic oceans: Marine Micropaleontology, v. 24, p. 57-69. Liebau, A., 1975, The left-right variation of the ostracode ornament: Bulletin of American Palaeontology, v. 65, p. 77-86. Liebau, A., 1977, Homologous sculpture patterns in Trachyleberididae and related ostracods: Belgrade, Yugoslavia, Nolit Publishing House, 93 p. Lohmann, G. P., 1978, Abyssal benthonic foraminifera as hydrographic indicators in the western South Atlantic Ocean: Journal of Foraminiferal Research, v. 8, p. 6-34. Lu, G., and Keller, G., 1993, The Paleocene-Eocene transition in the Antarctic Indian Ocean: Inference from planktic foraminifera: Marine Micropaleontology, v. 21, p. 101-142. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 353 Lu, G., and Keller, G., 1995a, Planktic foraminiferal faunal turnovers in the subtropical Pacific during the late Paleocene to early Eocene: Journal of Foraminiferal Research, v. 25, p. 97-116. Lu, G., and Keller, G., 1995b, Ecological stasis and saltation: species richness change in planktic foraminifera during the late Paleocene to early Eocene, DSDP Site 577: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 117, p. 211-227. Lutze, G. H., and Coulbourn, W. T., 1984, Recent benthic foraminifers from the continental margin off northwest Africa: Community structure and distribution: Marine Micropaleontology, v. 8, p. 361-401. Lyle, M., 1997, Could early Cenozoic thermohaline circulation have warmed the poles?: Paleoceanography, v. 12, p. 161-167. Maddocks, R. F., 1992, Ostracoda, in Harrison, F. R„ and Humes, A. G., eds., Microscopic anatomy of invertebrates, Volume 9: New York, NY, Wiley-Liss, p. 415-444. Maddocks, R. F., and Steineck, P. L., 1987, Ostracoda from experimental wood-island habitats in the deep sea: Micropaleontology, v. 33, p. 318-355. Maddocks, R. M., 1966, Distribution patterns of living and subfossil podocopid ostracodes in the Nosy Be area, Northern Madagascar: University of Kansas Paleontological Contributions, v. 12, 72 p. Magurran, A. E., 1988, Ecological Diversity and Its Measure: Princeton, NJ, Princeton University Press, 487 p. Majoran, S., Kucera, M., and Widmark, J. G. V., 1998, Maastrichtian deep-sea ostracods from DSDP/ODP Sites 327, 356, 525, 528, 529, and 690 in the South Atlantic: Revista Espanola de Micropaleontologia, v. 30, p. 59-73. Majoran, S., Widmark, J. G. V., and Kucera, M., 1997, Palaeoecological preferences and geographical distributions of Late Maastrichtian deep-sea ostracods in the South Atlantic: Lethaia, v. 30, p. 53-64. Manabe, S., and Stouffer, R. J., 1988, Two stable equilibria for a coupled ocean- atmosphere model: Journal of Climate, v. 1, p. 841-866. Manly, B. F. J., 1994, Multivariate Statistical Methods (2nd Edition): New York, NY, Chapman and Hall, 215 p. Margalef, D. R., 1957, Information theory in ecology: Memorias de la Real Academia de Ciencias y Artes de Barcelona, v. 23, p. 378-449. Martens, K., 1985, Effects of temperature and salinity on postembryonic growth in Mytilocypris henricae (Chapman) (Crustacea, Ostracoda): Journal of Crustacean Research, v. 5, p. 258-272. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 354 McCorkle, 0. C., Martin, P. A., Lea, D. W., and Klinkhammer, G. P., 1995, Evidence of a dissolution effect on benthic formainiferal shell chemistry: 51 3 C, Cd/Ca, Ba/Ca, and Sr/Ca results from the Ontong Java Plateau: Paleoceanography, v. 10, p. 699-714. McIntyre, K., Ravelo, A. C., et al., 1997, Ground truthing the Cd/ Ca-carbon isotope relationship in foraminifera of the Greenland-Iceland-Norwegian seas: Marine Geology, v. 140, p. 61-73. McKinney, M. L., 1990, Trends in body-size evolution, in McNamara, K. J., ed., Evolutionary trends: Tucson, AZ, University of Arizona Press, p. 75-118. McKinney, M. L., McNamara, K. J., and Zachos, L. G., 1990, Heterochronic hierarchies; application and theory in evolution: Historical Biology, v. 3, p. 269-287. McMahon, B. R„ and Wilkens, J. L., 1983, Ventilation, perfusion and oxygen uptake, in Mantel, L. H„ ed., The biology of the Crustacea, vol. 5: internal anatomy and physiological regulation: New York, NY, Academic Press, p. 289-372. Mead, G. A., Hodell, D. A., and Ciesielski, P. F., 1993, Late Eocene to Oligocene vertical oxygen isotopic gradients in the South Atlantic: Implications for warm saline deep water, in Kennett, J. P., and Wamke, D. A., eds., The Antarctic Paleoenvironment: A perspective on global change: Washington, DC, American Geophysical Union, p. 27-49. Medlin, W. L., 1968, The nature of traps and emission centers in thermoluminescent rock materials, in McDougal, D. J., ed., Thermoluminescence of Geological Materials: New York, NY, Academic Press, p. 193-223. Mello, J. F., and Buzas, M. A., 1968, An application of cluster analysis as a method of determining biofacies: Journal of Paleontology, v. 42, p. 747-758. Miao, Q., and Thunell, R. C., 1993, Recent deep-sea benthic foraminiferal distributions in the South China and Sulu seas: Marine Micropaleontology, v. 22, p. 1-32. Miller, K. G., 1992, Middle Eocene to Oligocene stable isotopes, climate and deep-water history: The terminal Eocene event?, in Prothero, D. R„ and Berggren, W. A., eds., Eocene-Oligocene Climatic and Biotic Evolution: Princeton, NJ, Princeton University Press, p. 160-177. Miller, K. G., and Fairbanks, R. G., 1985, Oligocene to Miocene carbon isotope cycles and abyssal circulation changes, in Sundquist, E. T., and Broecker, W. S., eds., The carbon cycle and atmospheric C 02 ; natural variations Archean to present: Washington, DC, American Geophysical Union, p. 469-486. Miller, K. G., Fairbanks, R. G., and Mountain, G. S., 1987, Tertiary oxygen isotope synthesis, sea-level history, and continental margin erosion: Paleoceanography, v. 2, p. 1-19. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 355 Miller, K. G., Wright, J. D., and Fairbanks, R. D., 1991, Unlocking the ice house: Oligocene-Miocene oxygen isotopes, eustacy, and margin erosion: Journal of Geophysical Research, v. 96, p. 6829-6848. Millson, K. J., 1987, The palaeobiology of Palaeogene Ostracoda from Deep Sea Drilling Project cores in the South West pacific [Ph. D. Thesis thesis], University of Wales. Morse, J. W., and Bender, M. C., 1990, Partition coefficients in calcite - Examinations of factors influencing the validity of experimental results and their application to natural systems: Chemical Geology, v. 87, p. 265-277. Mucci, A., 1987, Influence of temperature on the composition of magnesian calcite overgrowths precipitated from seawater: Geochimica et Cosmochimica Acta, v. 51, p. 1977-1984. Mucci, A., 1988, Manganese uptake during calcite precipitation from seawater: Conditions leading to the formation of a pseudokutnahorite: Geochimica et Cosmochimica Acta, v. 52, p. 1859-1868. Mucci, A., and Morse, J. W., 1990, Chemistry of low-temperature abiotic calcites: Experimental studies on coprecipitation, stability, and fractionation: Reviews in Aquatic Sciences, v. 3, p. 217-254. Murray, J. W., 1995, Microfossil indicators of ocean water masses, circulation and climate, in Bosence, D. W. J., and Allison, P. A., eds., Marine Palaeoenvironmental Analysis from Fossils: London, England, Geological Society of London, p. 245-264. Nemec, A. F. L., and Brinkhurst, R. O., 1988, Using the boostrap to assess statistical significance in the cluster analysis of species abundance data: Canadian Journal of Fisheries and Aquatic Sciences, v. 45, p. 965-970. Nurmberg, D., Bijma, J., and Hemleben, C., 1996, Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperature: Geochimica et Cosmochimica Acta, v. 60, p. 1654-1668. Okada, Y., 1981, Development of cell arrangement in ostracod carapaces: Paleobiology, v. 7, p. 276-280. Okada, Y., 1982, Structure and cuticle formation of the retifulated carapace of the ostracode Bicornucythere bisanensis: Lethaia, v. 15, p. 85-101. Oomori, T., Kaneshima, H., Maezato, Y., and Kitano, Y., 1987, Distribution coefficient of Mg2 *- ions between calcite and solution at 10-50 degrees C: Marine Chemistry, v. 20, p. 327-336. Orgell, L. E., 1955, Phosphorescence of solids containing the manganous or ferric irons: Journal of Physical Chemistry, v. 23, p. 1958. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 356 Owens, R. M., and Rea, 0. K., 1985, Sea floor hydrothermal activity links climate to tectonics: the Eocene greenhouse: Science, v. 227, p. 166-169. Peet, R. K., 1974, The measurement of species diversity: Annual Review of Ecology and Systematics, v. 5, p. 285-307. Peterson, L. C., and Prell, W. L., 1985, Carbonate dissolution in Recent sediments of the eastern equatorial Indian Ocean; preservation patterns and carbonate loss above the lysocline: Marine Geology, v. 64, p. 259-290. Pingitore, N„ Eastman, M„ eta!., 1988, The coprecipitation of manganese (II) with calcite: An experimental study: Marine Chemistry, v. 25, p. 107-125. Popp, B. N., Anderson, T. F., and Sandberg, P. A., 1986, Brachiopods as indicators of original isotopic compositions in some Paleozoic limestones: Geological Society of America Bulletin, v. 97, p. 1262-1269. Prentice, J. L., and Matthews, R. K., 1988, Cenozoic ice-volume history: development of a composite oxygen isotope record: Geology, v. 17, p. 963-966. Prothero, D. R., and Berggren, W. A., 1992, Eocene-Oligocene Climatic and Biotic Evolution: Princeton, NJ, Princeton University Press, 568 p. Rathburn, A. E., and Corliss, B. H., 1994, The ecology of living (stained) deep-sea benthic foraminifera from the Sulu Sea: Paleoceanography, v. 9, p. 87-150. Raup, D. M„ 1975, Taxonomic diversity estimation using rarefaction: Paleobiology, v. 1, p. 333-342. Rea, D. K., 1990, Aspects of atmospheric circulation: the Late Pleistocene (0-950,000 yr) record of eolian deposition in the Pacific Ocean: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 78, p. 217-227. Robert, C., and Kennett, J. P., 1994, Antarctic subtropical humid episode at the Paleocene-Eocene boundary: Clay-mineral evidence: Geology, v. 22, p. 211 - 214. Rosenfeld, A., and Bein, A., 1978, A preliminary note on Recent ostracodes from shelf to rise sediments off Northwest Africa: Meteor-Forschungsergebnisse, v. 29, p. 234-242. Rosenthal, Y., Boyle, E. A., and Slowey, N., 1997, Temperature control on the incorporation of magnesium, strontium, flourine, and cadmium into the benthic foraminiferal shells of Little Bahama Bank: Prospects for thermocline paleoceanography: Geochimica et Cosmochimica Acta, v. 61, p. 3633-3643. Salamy, K. A., and Zachos, J. C., 1999, Latest Eocene-early Oligocene climate change and Southern Ocean fertility; inferences from sediment accumulation and stable isotope data: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 145, p. 61- 77. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 357 Sanders, H. L., 1968, Marine benthic diversity: A comparative study: American Naturalist, v. 102, p. 243-282. Schindel, D., 1980, Microstratigraphic sampling and the limits of paleontologic resolution: Paleobiology, v. 6, p. 408-426. Schmitz, W. J., Jr., 1995, On the interbasin-scale thermohaline circulation: Review of Geophysics, v. 33, p. 151-174. Schnickter, D., 1980, Global paleoceanography and its deep water linkage to the Antarctic glaciation: Earth Science Reviews, v. 16, p. 1-20. Schnitker, D., 1994, Deep-sea benthic foraminifers: Food and bottom water masses, in Zahn, R„ ed., Carbon Cycling in the glacial ocean: Constraints on the ocean's role in global change: New York, NY, Springer-Verlag, p. 539-554. Schopf, T. J. M., Raup, D. M., Gould, S. J., and Simberloff, D. S., 1975, Genomic versus morphologic rates of evolution: influence of morphologic complexity: Paleobiology, v. 1, p. 63-70. Schweitzer, P. N., and Lohmann, G. P., 1990, Life-history and the evolution of ontogeny in the ostracode genus Cyprideis: Paleobiology, v. 16, p. 107-125. Semenova, L. H„ 1979, Key features in the biology of Candona Candida (Ostracoda, Crustacea): Hydrobiological Journal, v. 15, p. 21-24. Shackleton, N. J., 1967, Oxygen isotope analyses and Pleistocene temperatures re assessed: Nature, v. 215, p. 15-17. Shackleton, N. J., Corfield, R. M., and Hall, M. A., 1985, Stable isotope data and the ontogeny of Paleocene planktonic foraminifera: Journal of Foraminiferal Research, v. 15, p. 321-336. Shannon, C. E., 1948, A mathematical theory of communication: Bell System Technical Journal, v. 27, p. 379-423; 623-656. Sheppard, L. M., 1978, The exploration application of the range tables, in Bate, R. H., and Robinson, E., eds., A Stratigraphic Index of British Ostracoda: Special Issue - Geological Journal, v. 27, p. 473-523. Signor, P. W., and Vermeij, G. J., 1994, The plankton and the benthos; origins and early history of an evolving relationship: Paleobiology, v. 20, p. 297-319. Simpson, G. G., 1963, Historical science, in Albritton, C. C., Jr., ed., The Fabric of Geology: New York, NY, Addison-Wesley Publishing, p. 24-48. Skinner, D. M„ 1962, The structure and metabolism of a crustacean integumentary tissue during a molt cycle: Biological Bulletin, v. 123, p. 635-647. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 358 Skinner, D. M., 1985, Molting and Regeneration, in Bliss, D. E., and Mantel, L. H., eds., Integument, Pigments, and Hormonol Processes, Volume 9: The Biology of Crustacea: New York, NY, Academic Press, p. 43-146. Sloan, L. C., and Thomas, E., 1998, Global climate of the Late Paleocene Epoch: Modeling the circumstances associated with a climatic "event", in Aubry, M., Lucas, S., and Berggren, W. A., eds., Late Paleocene-Early Eocene Climatic and Biotic Events in Marine and Terrestrial Records: New York, NY, Columbia University Press, p. 138-157. Smith, W., and Grassle, F., 1977, Sampling properties of a family of diversity measures: Biometrics, v. 33, p. 283-292. Sokal, R. R., and Rohlf, J. F„ 1995, Biometry: New York, NY, W. H. Freeman and Co., 885 p. Spiess, V., 1990, Cenozoic magnetostratigraphy of Leg 113 drill sites, Maud Rise, Weddell Sea, Antarctica, in Barker, P. F., Kennett, J. P., et al„ eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 113: College Station, TX, Ocean Drilling Program, p. 261-315. Steineck, P. L., Dehler, D., Hoose, E. M., and McCalla, D., 1988, Oligocene to Quaternary Ostracods of the Central Equatorial Pacific (Leg 85, DSDP-IPOD), in Hanai, T., Ikeya, N., and Ishizaki, K., eds., Evolutionary Biology of Ostracoda: New York, NY, Elsevier, p. 597-617. Steineck, P. L., Madocks, R. F., etal., 1990, Xylophile Ostracoda in the deep sea, in Whatley, R. C., and Maybury, C., eds., Ostracoda and Global Events: London, England, Chapman and Hall, p. 307-320. Steineck, P. L., and Thomas, E., 1996, The latest Paleocene crisis in the deep sea: Ostracode succession at Maud Rise, Southern Ocean: Geology, v. 24, p. 583-586. Stoll, H. M., Schrag, D. P., and Clemens, S. C., 1999, Are seawater Sr/Ca variations preserved in Quaternary foraminifera?: Geochimica et Cosmochimica Acta, v. 63, p. 3535-3547. Stott, L. D., 1992, Higher temperatures and lower oceanic pC02: A climate enigma at end of the Paleogene Epoch: Paleoceanography, v. 7, p. 395-404. Swanson, K. M., 1995, Strategies for, and evolutionary implications of, carapace conservation in some living, benthic, marine Ostracoda: Senckenbergiana lethaea, v. 75, p. 193-213. Thomas, E., 1989, Development of Cenozoic deep-sea benthic foraminiferal faunas in Antarctic water, Origins and Evolution of the Antarctic Biota: London, England, Geological Society of London, p. 283-296. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 359 Thomas, E., 1990, Late Cretaceous through Neogene deep-sea benthic foraminifers (Maud Rise, Weddell Sea, Antarctica), in Barker, P. F., Kennett, J. P., et al., eds., Proceedings of the Ocean Drilling Program, Scientific Results, Volume 113: College Station, TX, Ocean Drilling Program, p. 571-594. Thomas, E., 1992, Cenozoic deep-sea circulation: Evidence from deep-sea benthic foraminifera, in Kennett, J. P., and Wamke, D. A., eds., The Antarctic Paleoenvironment: A Perspective on Global Change: Washington, D. C., American Geophysical Union, p. 141-166. Thomas, E., 1999, Biogeography of the late Paleocene benthic foraminiferal extinction, in Aubry, M. P., Lucas, S. G., and Berggren, W. A., eds., Late Paleocene-early Eocene; climatic and biotic events in the marine and terrestrial records: New York, NY, Columbia University Press, p. 214-243. Thomas, E., and Gooday, A. J., 1996, Cenozoic deep-sea benthic foraminifers: Tracers for changes in oceanic productivity: Geology, v. 24, p. 355-358. Thomas, E., and Shackleton, N. J., 1996, The Paleocene-Eocene benthic foraminiferal extinction and stable isotope anomalies, in Knox, R. W. O. B., Corfield, R. M., and Dunay, R. E., eds., Correlation of the Early Paleogene in Northwest Europe: London, England, Geological Society of London, p. 401-441. Thunell, R. C., 1976, Optimum indices of calcium carbonate dissolution in deep-sea sediments: Geology, v. 4, p. 525-528. Tipper, J. C., 1979, Rarefaction and rarefiction — the use and abuse of a method in paleoecology: Paleobiology, v. 4, p. 423-434. Travis, D. F., 1955, The molting cycle of the spiny lobster Panulirus argus Latreille. II. Pre-ecdysial histological and histochemical changes in the hepatopancreas and integumental tissues: Biological Bulletin, v. 108, p. 88-112. Turpen, J. B., and Angell, R. W., 1971, Aspects of molting and calcification in the ostracod Heterocypris: Biological Bulletin, v. 140, p. 331-338. Van Andel, T. H., 1975, Mesozoic/ Cenozoic calcite compensation depth and the global distribution of calcareous sediments: Earth and Planetary Science Letters, v. 26, p. 187-194. Van Harten, D., 1975, Size and environmental salinity in the modern euryhaline ostracod Cyprideis torosa (Jones, 1850), a biometrical study: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 17, p. 35-48. Van Harten, D., and Van Hinte, J. E., 1984, Ostracod range charts as a chronoecological tool: Marine Micropaleontology, v. 8, p. 425. Vannier, J., and Abe, K., 1995, Size, body plan and respiration in the Ostracoda: Palaeontology, v. 38, p. 843-873. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 360 Vincent, E., and Berger, W „ 1985, Carbon dioxide and polar cooling in the Miocene: the Monterey Hypothesis, in Sundquist, E. T„ and Broecker, W. S., eds., Natural variations in carbon dioxide and the carbon cycle: Washington, DC, American Geophysical Union, p. 99-110. Whatley, R., and Arias, C. F., 1993, Palaeogene Ostracoda from Tripoli Basin, Libya: Revista Espanola de Micropaleontologi'a, v. 25, p. 125-154. Whatley, R., and Quanhong, Z., 1993, The Krithe problem; a case history of the distribution of Krithe and Parakrithe (Crustacea, Ostracoda) in the South China Sea: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 103, p. 281-297. Whatley, R. C., 1991, The platycopid signal; a means of detecting kenoxic events using Ostracoda: Journal of Micropalaeontology, v. 10, p. 181-183. Whatley, R. C., 1995, Ostracoda and oceanic palaeoxygen levels: Mitt hamb zool Mus Inst, v. 92, p. 337-353. Whatley, R. C., 1996, The bonds unloosed: The contribution of Ostracoda to our understanding of deep sea events and processes, in Muguilevsky, A., and Whatley, R. C., eds., Microfossils and Oceanic Environments: Aberystwyth, Wales, University of Aberystwyth Press, p. 3-25. Whatley, R. C., and Coles, G. P., 1987, The Late Miocene to Quaternary. Ostracoda of Leg 94, Deep Sea Drilling Project: Revista Espanola de Micropaleontologi'a, v. 19, p. 33-97. Whatley, R. C., and Coles, G. P., 1990, Global Change and the biostratigraphy of North Atlantic Cainozoic deep water Ostracoda: Journal of Micropalaeontology, v. 2, p. 8 3-104. Whatley, R. C., Downing, S. E., Kesler, K., and Harlow, C. J., 1986, The ostracode genus Poseidonamicus from the Caniozoic of D. S. D. P. Sites in the S. W. Pacific: Revista Espanola de Micropaleontologi'a, v. 18, p. 387-400. Whatley, R. C., and Stephens, J. M., 1977, Precocious sexual dimorphism in fossil and Recent Ostracoda, in Loffler, H., and Danielopol, D., eds., Aspects of the ecology and zoogeography of Recent and fossil Ostracoda: The Hague, Netherlands, Junk Publishing Co., p. 69-91. Wing, S. L., Alroy, J., and Hickey, L. J., 1995, Plant and mammal diversity in the Paleocene to early Eocene of the Bighorn Basin: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 115, p. 117-155. Wingstrand, K. G., 1988, Comparative spermatology of the Crustacea Entomostraca. 2. Subclass Ostracoda: Biologiske Skrifter, v. 32, p. 1-149. Wise, S. W., Breza, J. R., Harwood, D. M., and Wei, W„ 1991, Paleogene glacial history of Antarctica, in McKenzie, J. A., Muller, O. W„ and Weissert, H., eds., Controversies in Modern Geology: New York, NY, Academic Press, p. 131-171. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 361 Wright, J. D., and Miller, K. G., 1993, Southern Ocean influences on late Eocene to Miocene deepwater circulation, in Kennett, J. P., and Wamke, D. A., eds., The Antarctic paleoenvironment; a perspective on global change: Washington, DC, United States, American Geophysical Union, p. 1-25. Zachos, J. C., Lohmann, K. C., Walker, J. C. G., and Wise, S. W., 1993, Abrupt climate change and transient climates during the Paleogene: A marine perspective: The Journal of Geology, v. 101, p. 191-213. Zachos, J. C., Quinn, T. M., and Salamy, K. A., 1996, High-resolution (104 years) deep- sea foraminiferal isotope records of the Eocene-Oligocene climate transition: Paleoceanography, v. 11, p. 251-266. Zachos, J. C., Stott, L. D., and Lohmann, K. C., 1994, Evolution of early Cenozoic marine temperatures: Paleoceanography, v. 9, p. 353-387. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 362 Appendices Appendix 1. Age Model Data Appendix 2. Ostracode Faunal Data Appendix 3. Rarefaction Code Appendix 4. Bootstrapping Description, Code, and Results Appendix 5. Algulhasina Morphometric Data Appendix 6. Algulhasina Morphometries Macro Appendix 7. Taxonomic Plates Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 363 Appendix 1. Age model data Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Eocene-Oligocene Age model data for Sites 744A and 689B Datum Top (Ma) 689B MBSF* 744A MBSF* Bottom (Ma) 689B MBSF* 744 MBSF* C12n 30.479 104.38 119.36 30.939 106.88 124.89 C13n 33.058 116.71 139.25 33.545 119.70 146.64 Cl5n 34.665 124.09 155.65 34.940 125.07 158.10 C16n.1n 35.343 128.33 161.60 Absolute ages from Berggren et al.t 1995; Cande and Kent, 1995 Site 744A paleomagnetic data from Keating and Sakai, 1991 Site 689B paleomagnetic data from Speiss, 1990 Late Paleocene Thermal Maximum Age model data for Site 690B Datum Age (Ma) 690B MBSF Reference NP10/NP9 BFE C25n.top 55.0 148.9* 55.5 170.48 55.904 185.47 Aubry et al., 1996 Aubry et al., 1996 Speiss, 1990; Berggren et al., 1995; Cande and Kent, 1995 * MBSF values are mean depth between measured intervals where reversal occurs. 3 6 4 365 Appendix 2. Ostracode faunal data Site 744A Site 689B Site 690B Site 738C Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 366 <3 * co - - - - - - C M - 1 3 C M - - - - - - CM a . - - C M - C M - lZ - C M « - .w 2 2 •c £ 3 o c t - - - - - - m C M C M o 3 - - - - - - - - - - C M - C M - C M u . C O C M C M co co - - C M - - - - - - - - - - - - u_ - C M C M C M C O - C M C M - - - - C M - - - - - 1 £ C M n — C M C M (T) C M _ _ n »_ _ C O »_ »_ 6 s 3 n C M C M C O - C M C O C M C M - - - - - - CO C M - ■§ 0) £ - - - - - - - C M C M - C O - - - 1 < _ t C M C M C M * 40 C M co * C M C M C O - - - - - - C M u . C M C M m o> C O 60 co o * - cn - C M C O - fs Oi co co m co C M co co m m V n CO r s u l C M C M m co C O C O co co C O - C M - - C M - C O - C M C M - CM - C M £ r s C M o o C O ® as fs m m in co C M m C M C O - CO C M - - - C M - C M - C O CM s § ■c a 3 £ co C M co C M C O lO * 2 1 5 eo rs. «n in co * “ - C M C M C M C O <- C M C O C M CM * - C M C M S 1 < — i C M - - C M - C M - - - - - - < a 2 h» (O rs in 03 rs cO m C O C O co co m ® C O co 0 ) o> rs. 03 in o o C O o r s m o C O C O o C M r s © co ® o o o C O co C M r s C O m CO CO Oi r s C O C O ’ T O C M rs C M C M C M C M C O C M m C M C O m C M ® in C M C O < B C M » r 03 C M CO CO CO f t CO r s m CO CO r s CO a» r s C O S . < C M C O C M C O c\i C O C M co csl C O C M C O C M C O C M C O co C O co C O C O C O C O C O CO C O C O C O C O C O C O co co’ C O C O C O C O C O C O C O CO co C O C O CO CO C O C O C O C O co’ C O C O co co C O C O C O C O co C O co C O C O CO CO CO co CO CO co’ CO CD C M r s a CO co 03 ® o ® C M ® ® cO 03 ® rs o C M co co to o fs 03 ao 03 o C O in * co co m ® a * o C M C M C O co co C M rs Oi C O 0 ) o C M C O C M r s C O <o Oi co r s i n <o r s co o Z i 1 »s C O rs C O r s C O r s co ® C O cd C O ® C O C D CO ® C O 03 C O d C O d C O d C O d C O d co o o o o o * * C M n C M V C M C M n C M C O co’ CO M n n c C O r s C M 68 co r*. 68 1 1 0 9 1 2 6 1 4 8 a> rs C O 5 4 6 3 ® r s lo o t 03 1 3 9 2 5 C O * 6 5 8 9 1 0 2 CO C M CO n C M 6 1 a> C O 5 1 5 8 1 0 1 1 1 3 Oi r s CO 5 8 8 3 03 t / i - * - ’ « - - * - « - * - - C M C M CM CM CM C M CM C M c o * CO . CO CO c o * co CO co * ^ " 1A m i n m m i o x <o X <o X CO I co I * C O X co X < £ > X C O X co X C O X CO X to X C O X C O X CO x' co X * co X to X © X C O X * co X C O X C O X <0 X C O X * X to X co X to X co X * to x' t o x’* CO X CO X CO x' to i < ' * f v < '*• rs < • r s < r s < * fs. < ' * rs < rs < «T < r * < ’ n r s < n r * < n rs < * * r s < ' * r s < * * rs < ♦ fs < «e r s « r r s « r * r s * r» -< + n r s < < * + r s r . < «* r s < * rs <* * * r * ' < * r s ~ < * r s < * rs r s r s r s < r s r s < * r s < r s < ' «*• r s Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 367 C M *• » - n ID C O C M ^ m a> uo ^ «c rs co O ) o ^ cm co ^ o > o m c m m rs «* 0 0 9 )Q Q >0 0)0 >0 C M ^ C O (A rs. • - • - < d ~* C M If) N O C M O © © © — C O C O C O C O C O C O C O a »• (O IA N fl) f t x i m i n m i a t o C M C O (D < b (A ^ fs. ^ 81 O CM t lAlAiAlACO<0<O<0COCO<O<Orsrs rs rs. h. rs. h. « N O ) N S S O ® » - N * - S S S O O N tt O O f N t i A N IS O N 000000 ) 0) 010 ) 0) t t t n t t t t m i o m m » - r s . » r s . o ) ^ ^ ^ o ) r « » r s r * - 2 ® ^ — •“ fs* r" r'w r s * - ^ t - » - ~ ~ > * - « - C M C M C M C M C M C M C M C M C O C O C O C O x‘ J J J J J r ' x ’ x* x'x*x*~x*x ‘ x * x * i * i *i <' <• <• <• < — < - < • < - <* <■ <• rs rs rs rs rs rs rs rs rs rs s rs rs s rs rs < ’ < < * < * < rs k s. rs rs Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 368 e» Oi u . - - 0) - C M lZ 04 - - 04 04 - * - - 2 04 04 - - - - - 3 2 C M < J - - 04 U. E £ - - - - £ I - - — • " *“ C M - 04 U_ U. 2 3 5 < -J 04 U. 04 - - - C M C M C M C M © © lZ - - - C M 2 3 t - - - - - C M - C M - C M - 5 - - - - - - 05 - - - OI U. li_ - - 2 - 3 i - - - - 5 - - - OI u . - 05 04 C M C M - C M - C M C M lZ - C M C M - - 05 C M C M - - - 05 C M - C M £ - 04 04 - - - - - 05 - 03 - 05 C M C M C M - - 3 OI - - - - - C M - C M C M - C M 03 C M - - - $ - 04 - - - - - 05 C M - C M - - C M < 04 - 04 04 - C M 05 - 05 - - C M - - - OI u. A 04 05 05 04 z C O 05 1 4 5 05 sr C M co 2 3 C O 04 rs C M (O 2 6 C M C O 03 05 3 5 O * • sf C O 03 *“ 85 C M (4 5 C M O C M 2 8 © (4 5 C M £ V rs «r 05 to (4 5 05 C O o o C O <0 05 rs co fs C M C M (4 5 n C M * • o (4 5 s f fs Oi £ V 04 ts 2 0 fs. 05 co z 05 1 4 1 9 (4 5 04 C M (4 5 04 C O (4 5 C M 05 sf C M C M C M C M (4 5 rs co to fs Oi 3 rs 05 C O 05 05 1 3 C D 35 to co 2 2 85 C O C O 04 05 1 5 C M 05 C D C M C M C M C O C O fs C M z O z fs - A s - (4 5 - - - - - - - - - sT - - C M - - - ST - < -1 - 04 OI - C M - - - C M 05 - - - C M C M n C M - C M < 0 2 C O * n rs 05 05 O 04 04 |s 04 35 rs 04 (45 05 04 C M 05 rs 05 o rs 05 rs 83 05 <o 05 05 C O o C M n * o rs C O 05 C M (4 5 C O (4 5 fs (4 5 03 05 (45 o C M C O (45 C O fs co C D fs fs C O 05 fs fs (4 5 rs to fs fs 05 05 rs A A & < 05 05 05 S T 05 05 05 ST 05 sf 05 05 05 05 * • 05 05 n 05 05 05 05 V 03 S T 03 03 V 05 05 03 V 05 05 05 © © © © % 05 * o 0. 05 00 rs o o> 04 C M 1 4 5 rs C O C O C O 05 O C O 04 05 * fs. (4 5 N. rs. so 85 o 0) 04 05 * 05 C O 05 G O fs © Oi C M 05 * fs C O « C M fs to A O 1 (4 3 U5 U 5 04 1 4 5 04 (4 5 04 (4 5 C M (45 04 (4 5 05 1 4 5 05 (4 5 05 (4 5 05 (4 5 05 (4 5 05 (4 5 03 (4 5 * (4 5 (4 5 * (45 (45 (4 5 (4 5 (4 5 (45 (45 (45 (4 5 (45 (45 co (45 C O (45 <o (4 5 to 1 4 5 to (45 fs (4 5 In l. Oi C O o 1 2 9 , 1 4 7 1 9 42 5 7 70 6 6 1 1 8 1 3 1 1 3 9 fs * z 2 7 4 8 I 6 9 05 85 1 0 9 1 2 9 fs V 05 05 05 5 7 O A fs © V C M c/> 05 05 05 05 n V * ▼ n * sr (4 5 (45* (4 5 (4 5 (4 5 (4 5 (4 5 (45 * C O C O co co * to to fs fs rs £ o o X c o X C O X 3) X 35 X C O X A X 35 X 85 X c o X 85 X 85 X 00 X 83 X 83 X 00 X 85 X 83 X 83 X 83 X C D X 85 X . 85 X 85 X 85 X 85 X A X A X 83 X A X A I < * rs < * rs < sf * h. < * rs < y t r» < * fs < ' s r fs < * rs. < • * * rs. < ’ rs fs. < ' r * < ’ * rs < * s* + fs « r* fs * fs s* fs '< * fs '< ■ «T >* fs. < ' fs < ’ fs < * fs fs ' < * fs < " fs < * fs < * * fs < ' ▼ fs “ <* « r fs ~ < fs Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 369 2 3 S < 2 3 I 5 2 3 s 2 3 $ < 04 04 04 co m to n v m »- ® co v n o n ® ® d > 0 ) 0 > o o o co 04 ® o co ® ^ C *5 ® C O V N II) ID N (O O ^ ^ C M C M r ^ ® ® ® ^ ) ® ® ® ^ N V C O ® O N « r t D 8 ) O M N N N B t t C O I O S ® ® w c o f j w o r t o p j n r t o 1 1 1 ® lO O OJ co <0 o a> co * s c o o j c j o ^ o i o o o c o * - * - n f l n n t t t t t t t C O C O < D (O + «o m ® oi 04 04 04 04 04 o o w w w «- co C O ^ C O C O C O C O C O o n w i O ® o o i t i o s ® M ® ® ^ a ) ^ ( o o > N o n C M N S S * - ^ * - 0 4 0 4 0 4 0 4 0 4 C O to CO C O s 2 co to co r - O, C M < £ > C D « © N J — « a) O *“ *- f*» ® co ** O K ) ® ® 04 04 04 O 4 O 4 O 4 O 4 O 4 C O C 0 C 0 C 0 C 0 C O C 0 C 0 x x x x x x x x x r x x x (D IO <0 IO IQ 10 to (O (O (O ID (O ID m m to tn id ' x ' r x ® ® to < * < ■ < ’ <• < ~ « r <■ < ’ <•<■ < ' <* <* < ~ < * <■ N N N N r» N N N N N N N N N N N N N N N N N N N N f*» K N ^ fs N fs fs ^ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 370 cn » - cn cn cn «-* C M cn cm CM — e o f l ) O T * o j n « r ( C o o O) O in «■ > m cm cn m co 0 a t r* W if) tf) m ia m to (0 ®®®a>®mcMin ^ r** ® *- in to « N- s n s in o> cm ^ r» a i co a) cb ooo b ) 0i o ) o o o o cn c n cn cn cn cn cn O) ** 0) s os N 8 O M V N S ) O ( M « < 0 « r- o ® — c m (y ^ in S # O N N N O CO ® Q O ) O ) O ) 0 S O ) O O O £ »- ® ® 3 - cm -f mmmin<O( 0 < 0 <o<o X ' X ' X ‘ O C M < * * «r ® ® p * r» c n m ^ ^ ^ Z ^ . ’- C M ' f ® N ’- » ‘ ^ * - » - ^ ^ » - C M C M ( M ( M CM CM CM n cn n (*) ? ? ?*"x i i * x i x z i * x x i x * x x * x * x * x z ® ® ® ® ® o o ® ® ® ® ® < < < < < < ' < • < < < <• < < < < << ■<■ < ■ < < < < < < < < < <■< < < < < < < N N N N Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 371 04 U. - - - 04 - - - 05 - - Z 04 - - - C M 04 - - - - - 5 C M 05 Q £ 3 -■ 05 • - C M •- 05 C M O 2 ^ . C M 05 1 < - -- - -4 ~ ~ *“ C M *” 04 U. - - - 05 - - - - C M - - - - - F I 05 04 - - - - C M C M * - - C M C M - - 05 m fs C M 2 04 05 - 04 05 OI 05 — 04 C M 05 05 05 m - 05 C M •> 05 C M •S C M 4 3 05 04 04 m •» 05 C M fs C O rs C M C M 5L ■ Q $ 04 - C M - - C M - C M - - 3 Q Q < - 04 - - C M * - - - C M - - - C M u. *“ u. 2 - - 3 □ I 04 04 — 05 * - 04 05 - - • - C M * - — C M s- — »- — - * - cj 3 Ui 3 04 - - - - - - 05 C M - 04 04 05 C M C M - - 04 U. Z £ o c 3 $ 2 1 < • ” 04 U. - rs 05 - m co 04 C M - co n - - 05 C M - - - - U. - - 05 04 05 05 05 04 m 05 - 04 C O 05 C M c r - - C M 05 - £ - 04 04 C M 05 C M C M - - 3 - - - - C M - C M - C M - - 3) 3 2 _ _ W 5 « j — 04 U. 1 U. c £ & Q. 3 O Q > 2 - - C M - C M 1 < - - - - - - - 04 U. Z - £ - § 3 _ 3 3 ,g 2 < Q. _ i a ao 00 o Ps C O in fs o fs C O co o C O _ C O 05 o m fs fs C O fs C O 05 oo S rs 05 04 ts 05 04 V fs C O 05 o OI fs 05 C M * fs 05 C M co 05 m fs 05 04 04 04 04 05 05 05 05 05 n in m in m 9 C O C O fs fs fs fs fs a> & V W ■ n V V n n n M - V V * T n n W V *T V V V V V w W V V < 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 (3) o ' O i * fs O i * C M fs C O 05 05 _ * 05* fs ^ * f s * 00 a ‘ 05 05* O J ’ fs 05 * 05 fs ^ ' 03 rs n % N C O O 04 in C O C O O 04 * * m fs C D o 04 C O 00 o 04 <0 »- C M C O a o m »-* 04 04 o i 04 04 05 05 05 05 05 05 05 w V * ’ * 1 4 5 in in 1 4 5 co C O cb co cb IS 3 in m m m m m m in m m m m m m in m m in m m m m m n in 1 4 5 m in in m c 6 8 1 1 0 1 2 9 1 4 7 05 04 n fs m C O fs 6 6 1 1 8 1 3 1 05 05 fs rs 04 C D 6 9 8 9 1 0 9 1 2 9 f * 1 9 05 05 5 7 1 0 1 (O fs 05 * C M to 05 05 05 05 «r n * in m * m m in m * m m C O C O co C O C O C O fs fs fs s r x' x' x' X * X * X x ’ X * x* x* x'* x ‘ x' x" X * X * x” X * x ‘ x' X * I ~ ”x ‘ X * x" X * X * X * X O o 00 ao 00 C O ao 05 00 00 05 05 co ® ao ao C O ao <s a> 00 a t 00 05 o ao C O ao <s 35 a) 00 o < * < ' < ’ < ■ < ' < ‘ < ’ < ’ < * < ’ < * < * < * < * < ' < * < ~ ■<■ < * < " < * < 3 + * * «*• + * * «r ■ * * * f n n * n «r n * * * * «e «r * * T ps o» rs f» rs. rs fs fs fs fs fs rs fs fs fs fs fs fs fs fs fs fs fs fs fs f - fs fs fs fs Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 372 Q. 2 3 $ < 2 3 § 1 <£ s 3 £ < 2 3 s < 2 3 £ < 2 3 2 3 n n is o n id N N C O C O 0)000 < 0 r- ® o M n U) <o fv *- O •- CM C M C M C M m m o o o > < * > to r- N C M W M t M N n n n n n I D S C O I O C O O I D C O O I N N C C O O I O N C I O I O O N r ^ r - . r - o c o c D c o i B O ) 0 ) w w r t o o r t o w o o o no ooo — *n«omo)CM<o«£)cMO) 0) ^ f s . < D O C * ) ^ ( O Q O C M ^ C O K C O O CMCMCMCMOCOC*)^^ CM (O S 3 ) o» r-* ^ r> <o O O i O) _ o *- c * ) ^ ifl (0 Ifl O) O CM ^ CM ( M N N C M N n n n n n n n n x x x x x x x x x x x i x i x x x x x x x x x x x x x 1 0 ( 0 1 0 ( 0 ( 0 ( 0 ( 0 ( 0 1 0 0 <0 <0 (0( O( 0l0 <0 (0( 0< 0< 0ID <O( 0( DI0 I <* <• <• <*<*«<* < ’ < * “< * < * < • < h . r - » r>p s. K r » ^^r» . h » N.r«.f>.p«»r— r-.p,- r ,«»r> »r-.p-.r-»r>.f>.p».p< - to to to co to '<* < ’ < * < * '< * < ’ < * < Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 373 £ 3 £ 3 § £ 3 $ < £ 3 s < £ 3 $ £ 3 2 < £ 3 cn cn * ^ w n in io ® o i •■ o — co ^ m m m «n <o s > o > q o m n m ^ N « t - ^ N O N in <A (O N a * c * to w- < 0 ^ c * > cn co co ® ® o o w n w rt < P Q } Q ( S ( 0 0 ) d > 0 ) 0 ) 0 0 0 0 6 ^ V V V n w p) o n n n s oi p n n u) k) m vi io to to n ^ ^ ^ «r <o ® ° 3 5 ~ (O 1 0 IO (O ( 0 N N S B O N t S B O « t < C ( D N N Q D 0 O ) O O a } O ) O ) O ) O ) O ) O * cm * (O r*. •- •- O J co to t - © cn co • • • • • ■ ^ i - N N N N N N N P i c n n n n x ' x ‘ x ' x ' x ' x * x ' x ‘ x ' x ' x ‘ x x ' x ’ x ' x ' x ‘ x ‘ x ‘ x ‘ x C O Q C D ® 00 00 ® 00 00 ® ® 00 C O 00 ® © C O C O ® Q C O < < ' < ’ < ' < • < * < * < • < • < * < * < * < • < * < ' < ' < ' < ’ < • < * < ~ < * < * < -< < Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 374 ® ^ c o o N i f l ' n * - s o N i o « < ^ o i o * * i C ' - n o i n f ^ O J C M ^ r - o c ' j n r N . ® a > o c M ^ f >“ OJC'j nrs» o o j ^ n n n n w p j H o n o n n r t w n r t n n o o r t o ^ r - co o 0 ) N * * N ( 0 0 ) 0 ) 0 ) 0 ) N O ) 0 ) •* rtn r - oo o c g ^c o ®o c 'j^ <£> « - c o r -. V ^ co co n ® s ® •- C O C O C O C O C O (O * ■ N t ( O o o o in id to < o to c o n i/i ® in in ® 1 / 1 in o> O 03 » “ eg 0) cm n ® a « ^ a n »■ n co a> a f- o e g ♦ a> o> 0 0 ^ ” i ^ t l f t N O > ^ ^ ^ ^ * * W ia in m m c o < o x ’ z ’ x ’ x ' x ’ i ' i ' x ' r* r * x ' r ' x ' x ’ x ’ x ’ x ' x" x* x 'x'x'x‘ Q 8 8) 00 1 8 ® ® q ^ CD 8 CO 8 00 9 ® ® ® S N S r ’ x^x* r* x* x* x 00 8 00 CO 00 ® C O < *■<* < < * < ' < * < • < • < ~ < ~ < * < • < * < * < * <* < <• <* < ^ < * ^ ♦ t v ♦ t «r ^ ^ fS . fS . h » f. f-. h » K Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 375 to o > m < o r *» o cn m ® o> o> a> o © o CO Ifl (S Ol C O Q Ol <0 « O OJ cn cm ® o <n ® IO S CO O * - N C M ® cm cn ^ C M C M C M cn ® in in C M C M n • - cn oj cn ^ cm cn cn oo>a>~ '< n<oi n a>c M<0t ocMO> o> r s . ® o c n ^ © ® o c M n ® r - . ® o o o o o o ^ ^ ^ C M t M C M C M C M c n c n c n ^ C M <0 Ps o cm <r ® ® * ^ t o ® in Z . _ — cm * co 0 ) rs. ® cn cm cm cm C M C M C M C M C M e n c n c n c n e n c n c n c n Z X « to z z x z x x z x co<oc0<0<0co(0<0 J: ^ •- cn in ao o> m m in m m ' " x * I * X ’ X ‘ f s P * » p s f l» p s f ^ r > s P s f * s p s N N S N S S N S fs h. r>» N K h» N S N N s s N Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 376 s 3 $ < s 3 $ < 2 3 £ < £ 3 2 < in cn o ' C M 2' r. in to a> cm cm cn in cn 2 3 $ < 3 i o r - . * - ^ r ® o o o > ao A a) in cv ia IT. Ifi (O S o eg f- O) co a) o d co n n i/i to o oi mi r t i Ammcococot t ) oo ao <o co o> 0> eg * a i o i 0 ) 0 ) 0 0 0 0 0 * - «o ^ eg ^ h* f» r» cm ^ m r**. ao o eg r*. © r- _ *- CO *T « t <o co J; iA iA iA iA( 0< 0( 0 I 0 <0 ( 0( 0( 0 SN N s ^ ^ f. v. * . , . f f c < i N N N N N N N n n n n x* x‘ ±’ x’ x* x^ x* x' x' r*~x~"x* x'x* x ‘ x* x ‘ x"x* x' x' < *<*<* <* < ~ < * ’< ' < * < * < ’ < * < * < < * < * <* <' Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 377 C M Ik lZ 2 - - 1 M uns eyell a 3 2 < C M - - - - C M U. u . a c s 3 - - 3 1 2 < _J - C M Ik C M - C M - C M sr - - C M - C M C O - C M - C M l Z C M cn C M s- sr - - - C M C M s- - - C O - - - £ - C M s- c 3 * - § . cj ■6 2 - C M s- - - C M - C M G Q < -J C M - - - - C M C M C O C M U. C M C M - - C M - ® C M C O to - o C O - C M C M z - co C M C O s- - co - lZ C O <0 Z s r s r C M ® C O C O t o to to 05 (O * s r t o C M t o c n C M - 2 C O C M «T co m fs. s r ® C M fs - n C O sr C M - O i t o t o C O in fs C M C M C M - 3 C M ® + sr sr to ® C M M- fs sr s r to ® C M to z O i tO o o to C O C M - - « 2 in tO co sr C M C O C M <0 t o to < o to t o C M to C O O tn ® t o t o « to fs C M C O * < -J rs r-» C O rs C O rs C M sr lO sr t o t o C O O i C O * C M t o m fs - m fs C O C M C O C O C O - C M u . sr C M C M C M C O C M C O lZ C M C O - - C O ® - - - C M - C M C M C M - - C M s r - - C M £ C M C M - - - C M - - C M - - - - C M C M lO sr C M C O - □ - - - - - C M co sr - C M - 2 - - - - C M - S- C M C M S- - i5 © < - - C M - - - - - C M LL * “ § Ik £ C M _ _ C M C M C O C M V ) 5 3 _ C O C M _ C M C O C M co _ C M » _ _ S 0) 2 to _ 1 < - - - - - - C M C M C M - C M C M U. lZ £ 5 3 - a o 2 Oi * 5 - - - - - * - < 8 2 ® rs « O i o C M C M fs. sr C M ® rs C M to O i C M C M C O fs V C O o fs c n fs ® C O t o O i C O t o © m ; * C M sr * m; O fs * to O i n C M to t o sr tO fs to C O 05 m o C M t o to t o fs t o t o fs rs t o c n rs fs to fs t o fs fs O i O i fs ® ® & < V C O V" C O sr C O C O V C O s f c n V C O co sr C O V c n r c n C O m - C O n c n W c n M > C O n C O co V C O V co v r c n r C O V c n w C O n C O n C O r C O V co V c n w & O i o ’ rs ® ’ « fs. o O) C M C M lO fs. t o ® ’ ® O i' © ® ' C M s r O i' sr fs to fs fs ao s ’ o O) C M a t O i t o O i * ® fs o O i' C M O i + fs t o z ® C M fs s r ’ co n ' a o S in lO C M to C M ® C M to C M tO C M IO C O to C O in C O tn C O to C O to C O lO C O to w to n t n V to V to in to in in in in lO in tn t O i n t o n cd to co tO t o i n fs m O i ® o a C M fs. O i C M * fs IO ® rs O i O i ® C O 139 fs * - fs C M ® sr O i to O i ® O i o O i C M fs a t a c n fs m o ® fs C O * * C M <0 C O C O co C O n * sr + ♦ * r M * ■ M to lO to to to * to to tO to t o t o t o t o t o fs fs fs I 0 1 x ‘ C O X * C O X * ® x ’ ® X ' ® X * ® X ® x ’ ® X * ® x ’ ® x ' ® x ’ ® X * ® X * ® X * ® ® l " ® X* ® x ” ® X * a X * ® X * ® X « X™ ® x ’ ® x ’ ® X * « x ’ ® X * a X ® < ■ rs < * * fs < * sr rs < ' rs < ' fs. < * fs. < * s r fs < • M- n fs < * M' * fs < * fs < • • r M' fs < * M' * fs '< ■ * fs < * sr fs < * * fs < fs < * sr fs < * fs < ■ * fs < ■ sr fs < * ♦ fs fs < s r fs < * fs < * sr fs < ' * fs s r fs < ' fs < ' * fs < fs Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Bairdoppilata Dutoitella tndet. Gen. H Indet Gen. 1 Hole Core S Int. M BSF Age (Ma LA RA LJ R J F I F2 LA RA U R J F I F2 LA RA LJ RJ F1 F2 L A l m l u RJ F1 F2 744A 16H e 137.28 32 767 1 744A 16H 27 137.47 32.795 744A 16H 66 137.88 32.856 1 3 744A 16H 76 137.96 32.868 744A 16H 88 138.08 32.885 2 744A 16H 109 138.29 32.916 744A 16H 126 138 46 32.944 1 744A 16H 148 138 68 32.974 1 744A 16H 2 1 9 138.89 33.005 744A 16H 2 37 139.07 33.031 744 A 16H 2 54 139.24 33.057 744A 16H 2 63 139.33 33.063 744A 16H 2 76 139.46 33.072 744A 16H 2 100 139.70 33.088 744A 16H 2 119 139.89 33.100 744A 16H 2 139 140 09 33.113 744A 16H 3 1 1 140.31 33.128 1 744A 16H 3 25 140.45 33.137 _ 2 744A 16H 3 46 140.66 33.151 744A 16H 3 65 140.85 33.163 744A 16H 3 89 141.09 33.179 744A 16H 3 102 141.22 33.188 744A 16H 3 126 141.46 33.204 744 A 16H 3 146 141.66 33.217 744A 16H 4 2 141.72 33.221 744A 16H 4 19 141.89 33.232 744A 16H 4 39 142.09 33.245 744A 16H 4 51 142.21 33.253 744A 16H 4 58 142.28 33.258 744A 16H 4 101 142.71 33.286 744A 16H 4 113 142.83 33.294 744A 16H 5 19 143.39 33.331 744A 16H 5 37 143.57 33.343 1 1 744A 16H 5 58 143.78 33.357 744A 16H 5 83 144.03 33.373 744A 16H 5 91 144.11 33.378 3 7 8 379 cnj n c v cv tn 0 5 00 05 o n n ^ o r*- co r-. o j <*5 i n <n a> ^ co o o 05 r-* o S C O O (V t ^ ^ m m ® in m ao tn h» n . 05 »- n m ^ tn tn in os ^ c m cn i n to to cn c n aoc D05eotnc Mm s a p » s o w in ® cn r- C O CO IS to 05 05 o o o © n n n w 0 5r ^ 05 f ^r s « r « . o a 0 p - c ^ * “ f>*h-r>* C O O N t i f l C O p W ^ i O N C O O N ® o oj ^ ^ ( 0 ( 0 ( 0 ( 0 0 0 0 ) 0 ) 0 ) 0 0 ) P Q U5 S S QJ W ^ X T l A A i n m t o i o i o i o i o i o c o i o r-. v - p - o j o j o j o j o j o j o j c s i c o c n cn cn x x x x x x x x x x x x x x x c n c o c o c o c o c o c o t n c n c o c d c n c o c o c o X X X X X X X Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 380 C M u . iZ 2 s C D 3 # ] ■§ 5 - C M u . u . - - X 2 C M c 8 J 2 - cn — C M - - 1 < - - - - - - C M U. a . 2 - s s o 3 2 - Q -J - - C M u. LL « £ 3 C M C M I 5 - - - 3 < _ j cn - - cn C M - - - - C M - - C M - - - - < a 5 C O *T *r rs a a> o C M C M rs. C M C O fs C M m O) C M C M cn rs cn o rs cn fs ao cn C O o> cn co o s r V C M V o rs co Oi *r C M m C O i/> rs m cn Oi t n o C M C O in co rs co co ps ps C O cn rs rs m rs C O rs Ps 03 Oi rs co a O M < sr cn cn sr cn sr cn sr cn cn n cn cn cn cn cn < cn cn cn cn * • cn sr cn sr cn s r cn cn sr cn cn sr cn sr cn n cn cn sr cn cn cn s r cn 03 o rs 05 co rs © a> C M C M m rs co co C O 03 o S3 C M Oi sr rs m rs Ps o> a o o C M Oi sr o> C O o> 00 fs o 03 C M 03 * rs C O 00 C M ps s r C O C O o i m m m C M tn C M in C M m C M in C M m cn m cn m cn in cn in cn m cn m cn m m n m m 1 *3 n in m m m m m m m m co m co m C O m C O m co m fs lA c 0) co o 129 sr 03 C M rs. m C O rs 0) a> 03 cn Oi cn rs sr Z fs C M ao ♦ O) co Oi C O (33 O Oi C M fs 03 03 cn ps in o 03 rs cn sr C M •V <0 cn cn cn cn sr sr sr T sr * * sf *n m in m m 03 m m C O co C O C O co co Ps rs rs e o O X 00 X C O X C O X O X 03 X co X C O X <0 X C O X 03 X C O X ao X C O X C O X co X 03 X 00 X C O X 00 X C O X 03 X C O X 00 X 03 X oo X G O X G O X oo X 00 X a < sr rs < «* rs < ps < sr sr rs. < sr ps < s r fs < sr sr rs < * rs < * * fs < rs < s f rs < * r - < rs < sr rs < * fs * fs < n n rs < n . < ♦ s r fs < sr n rs < rs < rs < * ps < rs < n rs < sr sr ps < ■V Ps < + rs < * rs < * Ps Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 381 s 3 2 £ 3 2 C *5 * - »- C M C M ^ n « n ^ *- n * - C M U "5 (D O C M ^ i n u i i n u ) t o ( 0 <o *- «- co o os co n in a ^ ifl ^ N N m n n t n co v m o s a S S N C O ^ O C O C M C D l o c o r ^ r - c o o o o o a s o co co co 0 5 — co <o r*» ** c *j cn o c m ^ c o c o c o g o a a a a C O C O C O C O C O C O C O C O * • 0 0 ( 0 ( 0 oo co »* co *- CO *» © w V J 0 5 O C O o a a t cm co m a ^ T3 a> co m i A m i n u ,) i o m < o < 0 < o < O ( o < 0 p < - CM CM CM CM CM CM C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C D f f i C S f f i f f i f f l f f i C D o o o o o o o o <0 co co co co <o <o ( D C D ( D ( S C D C D ( S C D ( S f f l C D ( D ( f i f f i O) o) 05 a a a a a a a a a a a ffi o oo co co co co co o < o co o ® ® _ _ _ _ _ _ _ _ ( o t o t o c o c o c o i o c o c o c o c o i o i o i o ( O c o Q c o c b o c o c o c o i o c o c o c o c o ( OO C D S O C D O f f i f f i f f l O O C D f f i 0505050050505050050)00)05 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 382 co v CM C *> ^ w ~ CM CM CM CM »- f W CO CM CM ® CO CO *- CO CM CM ^ CM CM CO CO CM CO *- CM CM CM ^ oo co •* 10 CM CO ^ N n m co oi in »- o ^ oo ^ co ^ v «n io CO CO CO CO CO N ^ o C M C O CO CO CO CO CO tf) 00 N * o *- ® ® r- a> CM CM CM CM CM n ^ ^ C M « ^ ( O M ( M l O N O ( O N N S V C O I O O I « « 0 ) I O ^ lOifllOVNNQOOlO*''- lOlAiAlOiAOiAiAifllOiAiAiAiAiAlOcdcO C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O r - ^ CO CM CO «- »• CM CM CO m is 8) n t CM CO 00 o ii n oi v n n « n O ^ C M C O C O O C M T N N S N S Q f l 0 O C 0 0 ) 0 > 0 ) C M C M C M C M C M C M C M C M C M C M C M C M co co “ co co X. ® * - ® co CM CO CO CO CO co c o c o i n i n i A i A i c i m i A i o < O ( O ( O ( O ( 0 ( O ffl CD CD S CD 010)0)0)171 C O ^ m cO CO ® ® Q) CO ' ® 00 (O (O (O. < 0 < o IO *C if) < fi (D (O CD ffl CD 0) CD CD O Q Q o £Q flj Q qq 0 ) 0 1 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) (O(OIO(0(O<D<O(OlO(OlO(O(O(D ffi CD CD O CD CD 0)0)0)010) O) ® OQ Q QQ (O co t o co co c o co Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 383 ^ N tfl (O i/> m to t o <© to o c< i *r t o ■e * - to 0) cn to o •- *- o> co * - O) to * - •— m o> «- u> c n oj m cm n n ^ « 1 0 s s o cm ^ co O 00 <3> O) Q O) O N Ol N Ifl Ol N o <£> cm <o o m co oi » © o — n w o n n w 03 CM ^ t o N- 0)0 0 0 0 co cn ^ n ^ ifl (O 0 0 ) 0 ) ^ <m n ifl a m l o m m t n t n t n m t o t o t o c o c o t o r s . CM CM CM CM CM CM w c p o w o w r t c o r t w o w r t w r t r t w o w o c o o oo co co ao co ao S O O O O O O Q C D I D f f i O i S I & f f l f f l 0) 0) 0> 0) 0) 0) 0) 0) 0) 0) 0) 0> 0) 0) 0) 0) o c o c o a o o c o c o a o O C D ( D O l S I D B ) O D Q 8 1 f f i O f f i g ) q > o a o o o o ) 0 i 0 ) 0 ) 0 > o > o t O t O t O t O t O t O t O t O <0 10 (O (D (D IO ( 0 ( 0 ( 0 ( 0 1 0 1 0 ( 0 1 0 l O ( 0 ( D < D < b < O l O ( O l D l O I O ( D I O ( 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 384 2 3 S 2 3 £ m ® o> m *- o v ao v co n n m m ^ n w n w n w ^ O M (O o w n ® co n o n 0) n n n n w eg eg eg eg eg tO h . *- c o o r » .< n *- * - C ' i o c o c M t o ^ c o e g w ^ n ^ w e g e g c o n n i n i n ® eg in eg n o ® o >o o n d o> o n eg M l O O O N P i m N a> eg o» n co f*» n . . c o i n r > . ® * - e g ® © o e g n N N 0 0 8 ) ( 0 0 > 0 ) 0 ) c g c M c g c g e g o g e g c s i c g co o o n p) p j * N P) P5 P5 P) P) P) P) i A i n m i A l A i D i n ( 0 ( O l O l O ( 0 ( C ( O N N Ol 91 O) < 0 < O< O< O< 0 (0 (D IO V ID 0 } C O ( D C O C O O O D Q O ( D C O (O (O <O (D ( O ( O VC D lO ( O l O ( O < 0( D ( O I0 lO C O C O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 386 i n co o i 1 /1 ^ o v oo * oo n ^ *9 «/> to n n w M * <*5 n n <n r*- gr © cm co © C M C O h* eo v o n a <o S CO O Q *• V W p) pj n n w /) o s v o ^ n i/i n oi v «r m* CM CM CM CM CM » *co r*- »- « or s. r> *-^ 0 ) 0 < O N < 0 ^ < O C M C M S a * - » - c M w n n t i / ) i / i eg ® f io v o n ® co co r* r- 0 0 ) 0 ) 0 ^ O) O) O) ^ CM CM co O O CM CO CO O « - C M C 0 O Q C M ^ CM CM CM V U-) O 0 0 ^ 2 O O) ^ $2 c Mc o c n n o n o n t A m m t n t n m m c o c o c o c o c o c o c o ffi S) ID O (0 0 ) 0 ) 0 ) 0 0) 00 GO CO 00 03 00 00 03 CO oo < O < O < O 1O 1O < 0 < 0 < 0 < 0 < 0 ffi ffi ffi ffi ffi O ® 0) O) O) CO CO CO CO CO ffi ffi ffi O ffi ffi O ffi ffi ffi ffi ffi ffi ffi ffi 0 ) 0 0 ) 0 ) 0 ) 0 0 ) 0 ) 0 ) 0 0 ) 0 ) 0 3 0 ) 0 ) COCOCOCOCOCQCOCPCOCOcDCO'O'PfP Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R e p r o d u c e d w it h p e r m is s io n o f t h e c o p y r ig h t o w n e r . F u r t h e r r e p r o d u c t io n p r o h ib it e d w it h o u t p e r m is s io n . 388 C M C O r«* cm 40 (O C M co • - * - * - i/) co' O ) m »• o co v ® >r v ^ m in (O C O C O C O C O N ^ O N IO O cm v < 0 r-. n a co co c j C M C M C M C M C M c o f - » . - c o © r - c o — — O C O C M C O * - C D C M C M S ' • ’■ N C M n n t m m ^ cm 1 0 cm o co ^ ^ O) co — co oo o > o > o ^ ^ cm co oo o cm co m N S N C O O D O D O D O ) C M C M C M C M C M C M C M C M n w n n n w o o m m c o m i n i n t n . c o c o c o c o c o c o c o r ' . r ' - O ffi O O CD 83 CD ffi O ffi ffi O O CD CD CD O CD ffi ffi O > a ) 9 > O ) O ) O ) O ) O ) O ) O ) O ) O ) 0 > O ) O ) O ) O ) O ) O > O } c o c o c o c o c o co co co co co co c Q c Q c o c o c o c o c o c o c o c o c o c o c p r o c o c o c p c p c o c D Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 389 e g u . - C M C M - - C M - © C M © u . © - C M © © 2 © r > . © r * » © 3 O © © r * . 5 C M C M C M - < © - - C M - - C M u . u _ 2 - - 3 - C M C M 5 - - < - C M U . u . 2 3 s 2 © C M C M © © C M © © © O J u . l Z 2 3 I c o C M - - c o C M C O C O C M r -. * © - © * © © © © 2 c o C O C O C O C M ri C M c o © C O C M C M o C O r * » © C M © © r » C M L L - t Z - - 2 - - - - 3 - - - C M $ c o C M C M - - - - - < _ j - C M - C M U _ C M C O - © d © C M in - - © © - £ C M - - © © C M © © © 3 © C O C M C M © © © © © © $ - - *r C M - © < - i - - © - - © s T - * C M n C O n n C O *r C O © C M C O C M r - 0 0 a in © © 0 0 * o © * © © * r » » © o o © C M o © © o © © © C M © o © © © © © © © © a t < in © in m in © in m in in i n m in in © in in m • n in i n i n in if) in © in © © © © © © © © © © © © © © © © © i n C O C O c o c o o c o C O m © c O C M m C O r * « . r * - . i n © © © © o r - © © © © © t - » . © © C M © © © © © C M I r * » < o r v i C O d < o a t c o a t c o C O C O a t C O a t c o a t c o a t c o a t c o 6 N o r » o f s » o N . d r * . o h . p « . r - r * . N . C M f*. a t r-. r - o C M © C O at a t n a t 0 0 a t C M © © © o © r ^ . * © © © © o © © o © © © © in * C O r - c o C O C M c o in C O < o in C M C O in in C O C M © © © © C O © © c o © © © © © S e c t - - - - C M C M C M C M C M C M © © © © © C O © © © * C o r e © a t a t a t at a t at a t a > © © © © © © © © © © © ® © | H o le m * o a t < 0 m* o ® C O m* o a t in CD* o at C O m * o at c o m o a t co ~ ffl* o at C O a* © a t CO «* o © CO CD* © © CO a © © co s o © co ffi o © (O ffi o a © ffi O © to ffi o © © ffi' o © © f f i * o © © a ! O 1 © © f f i * o © © f f i * O © © . ffi o © © Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 390 m co co in cm co O C O G O O i O ) O ) o c o m co m « C O M < 0 S S C O to C O co co co co co 0)0)0)0 0 0 0 0 4 co co o o — m m in o w co m in m s co s co to co m p*. o cm m oo cm o O N O O 0 ) ^ - 0 ) ffl s n n o o cm * <r co o s o a; » r r ' » « - » - o ^ c o * - » - » - i n » - c * ) * i n co mocococoincoinr^m^incoinin n ^ r>. *- cm t « N n t ^ n t io co co m m m * - * - * - r - w c \ J N C M M N n n « * ) c * i n co co n co «e ^ v 0)0)0)0)0)0)0)0)0)0)0)00)0)0) 0 ) 0 0 0 0 0 0 f f i S l C D l S S C D l S f f i Q l t S S I f i O Q f f l O ffi O ffl ffl O CD o o o o o o o _ _ _ _ _ © © o o _ _ _ o o o o o o o o O o o o C O C O C O <o C O C O C O C O C O C O C O C O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 391 C M U . 1 1 . s s 3 0 3 I $ 0 3 o 0 3 c o 0 0 - 0 1 C M 0 0 - C M - - - C M U J < - J C M( 0 < 0 c o V C M C M0 C M0 0 0 C M C M - o C O 0 0 0 C M U . C M< 0 C O p » 0 C M 0 * 0 C M 0 • “ C M u . < 0 C O ( 0 C M 0 0 o 0 C MC M 0 0 0 0 * o 0 C M < 0 0 3 0 0 0 0 0 0 X C M* - C M — P * * C M 0 0 C M O ) C MC M0 0 C M C M 0 0 0 0 P . 0 0 0 0 □ C M C M V C M 0 C M Q > £ $ C M 0 * 0 C M 0 0 o C MC M - £ < C M - - 0 C M 0 0 0 C MP ^ . C M U . c g u . « & s S X z 3 u s C M e < r 5 - C M u . l Z s a i i 3 < 3 4 - - - - 0 0 C M* - ! 2 C M* C M C M u . l Z s < n 3 b -- a • o 4 - - - - - 0 I 5 - 0 - C M 0 C M - - C M u . l Z £ * c 3 - - C M i < -j - 0 - ■ f O 2 r » - C MC O 0 V 0 0 C M 0 C M r * » 0 0 0 0 0 0 o 0 0 0 P - 0 O o C M O 0 o 0 C M O 0 0 0 0 n n V 'f V * V n V 0 0 0 0 0 0 0 0 s u O i n i n i n i n i n 0 0 0 0 i n i n 0 0 0 0 i n 0 0 0 0 0 < m • n o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C Oc o 0 0 0 0 0 r * » 0 0 0 r * 0 0 0 p » 0 P » 0 0 CD & C O< 0 o v 0 C M0 p » 0 0 3 o 0 0 p s . O C M0 0 C M f f i p - r s ! C Oc d c d 0 0 0 ) 0 0 3 0 © o o o o b C M 2 < 0 ( 0 c o 0 0 0 0 0 0 0 0 p - r * p » p » p * . P * P - P » P - P > » * * * ” • “ * “ * “ * “ * " * " * “ * " • " * “ * ” * “ • “ * “ * " * " 0 o o 0 0 0 O r * . C MC O 0 0 C M* 0 o p - 0 0 o 0 0 0 • r M T P - • “ 0 0 •- 0 ~ - 0 * • 0 0 0 ** • • 0 — i n C OC Oc o 0 0 0 0 p i . 0 0 < d 0 0 « e s 0 i d 0 0 0 ♦ P » C M * 0 0 0 C M* 0 0 0 V 0 S C O - - - - C M0 0 C MC MC M0 0 0 0 0 0 0 0 0 « T S 0 ) 0 3 0 3 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C J * * * * * * * “ * " * * “ * " * “ * “ * " * " * " $ ffi* C D * m ’ ffi’ ffi* f f i * ffi‘ f f i * C D * ffi* f f i ’ m *ffi* ffi* ffi f f i * f f i * f f i * f f i *f f i *f f i * C D 1 0 3 0 3 0 3 0 ) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ( 0 < 0 C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 392 cn ti IS ll. 2 s i 3 1 * - 04 - - O l - to 0 3 - C M 0 3 C M - 3 U J 5 - 0 3 - 04 0 3 C M 0 3 to C M - C \J u. lZ $ a •C S — C M — s. -• -Q o 3 — *- C M *“ C M — c £ to - C M - o a. < - - 0 3 e g u. u. 2 - 0 4 «- «- 9 to 3 0 4 - o 5 2 5 — C M — *- ffi *- C M « 3 > to < C M - C M 0 4 U » to l Z Q . to 2 to | 3 | 5 - - to to 0 3 - - C M 0 3 0 3 C M ffi to lu 3 - 0 4 0 4 - - 0 3 O J ffi C M C M C O - C M 0 3 in O l u. l Z 2 C M O £ 3 C M < 3 £ < O l u. - - 0 4 - u. O l m - - 2 - - 0 3 to - to •c e 3 to _ to » _ 0 4 o u o e 2 0 4 - 0 4 O J 3 < - O l 0 4 p- V to to 0 3 0 4 0 4 to in ffi O 0 3 r- O C M ffi to O ffi 0 3 2 O J 0 3 to to in to P * . to to a ffi 0 3 0 3 0 3 © O o C M 0 3 0 3 to to to to to to to to to to to to to to to in in tn n in in in in & in in in m in « n in in in in in in in in in in in in 1 4 3 in in in < m m m m m m in in m m n in in in in m in in in in in in in 0 3 to < o (0 m to m rv m n to p » to to in r» to p * . a o C O c o 0 3 to o »- to g o 0 4 to « 0 3 o 0 3 to in o C M in c o C M m P - K ® c o cd ® 0 3 0 3 0 3 0 3 0 3 b o o b b o C M to to (O < o to co to c o to to to r- r« . h » h- p * » f-. r-. P * . p » » p- P * o o 0 3 0 3 0 3 o 0 3 P * » 0 4 0 3 0 3 0 3 0 4 to to ffi o h* a > ffi o to 0 3 0 3 0 3 — to r** *- •- 0 3 to ffi in 0 3 to m to 0 3 to 9 — in 0 3 to to to in to in rs . in to in ffi in in to to in td in in in to p- 04 to to 0 4 0 3 to C M to in to ffi 0 3 *“ to a o 0 3 £ o CJ 0 3 0 3 19 0 3 0 3 0 3 19 0 3 0 3 0 3 0 3 0 3 19 19 03 19 0 3 0 3 0 3 0 3 0 3 19 m CD* ffi f f i ' ff i* C D * f f i ' f f i ' f f i ' f f i ' f f i ' ff i* ffi ffi ffi ffi ffi ffi' "ffi* ffi' ffi' ffi o O o o o o o o O o o o o o o o o o O O o o 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 0 3 to to t© C O C O ffi to to to to to to to to to to to to to C O to to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 393 ffi & 5 I i O J li. u. £ □ 1 2 C O (0 I 6 6 * m m O J O J u. u. & £ aj 3 4 5 ft — — O J — * - — 5; 2 ID - O J C O O J - r-. C O C O O J O J tn a o o 05 rs. O C M 35 C O O a o 05 O J 05 TT in c o r- C D 05 e o < s 05 05 05 O O o C M 05 TT O ' * * m ID <n m m m Si m iD ID l D m m iD in m in m m m m in in m m m m m m < m in m in m m m m m m in in m m m in m m m m ID C O co co m co m o. m in co p*. co co m p- co p- a o G O a o u. 05 cO o »- ® C M co o. a o 05 o *- 05 m o C M m C M M r-. on e o a o C O 05 05 05 05 05 o o o o o o w- w C M C O C O C O co C O co C O C O C O co C O r*. r - r*» p- r - p « » r*» *“ *“ *“ * " *“ *“ *~ *“ ** *“ *“ *“ *“ o o 05 05 05 o OJ eo 05 05 C M c o o r*» a o a o o 05 tT f - »- 05 «T G O »- in ■- 05 n m < o 05 » “ — ID C O C O C O co m C O m r- m in co m in C O m co m in m V 0* w OJ * © O J C O * w* C M m C O 05 05 n •“ *“ ** G _ O J OJ O J OJ O J OJ cn 05 05 05 05 05 05 05 05 *0 w C D o > o > 05 05 0) o > 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 o •" -* *” •“ •* *“ *" *“ *” ** *“ *“ *“ *“ f f i f f i f f i m m f f i f f i ffi f f i ffi f f i f f i f f i ffi f f i ff i ffi ffi f f i f f i f f i f f i o o o o o o 05 O o © O 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 05 (O C O C O C O co co co co c o C O C O C O C O C O C O C O co C O C O co Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 394 1 s « > 1 C M li. U. s 3 i < «- C M ^ C M •- »- co c m co 1 0 *- «- ® ~ ^ ^ ^ C M C M C M C M 2 1 1 C M L L u. 2 3 i 3 C M < £ > *- Bairdi C M U. u. 2 3 $ < ^ C O c m eo k rt C M »- Propontocypris sp. / C M U . l Z s 3 s < < 0 W - C M *- eo c m ~ •- Ponnyella sp. A C M U. u. £ 3 < C O C M *“ •“ * “ *- C M *- C O Pennyella sp. B C M U. iZ a 3 s < V C O C O «- «- «- r*. *- C M •- Algulhasina C M L L l Z £ 3 5 < «j < 0 *- ® 0 0 ® — C M lO « J3 2 283.56 284.48 284.58 284.68 284.75 284.85 284.93 285.00 285.11 285.22 285.31 285.35 285.43 285.50 285.52 285.65 285.91 286.34 286.56 I H o le C o re S: Int. 738C , 1 1 1 , 15 738C 1 1 1 108 738C 1 1 1118 738Ci 1 1 J 1 j 128 738C 1 1 1 ■ 1 135 738C | 1 1 1 145 738C 1 1 2 2 738C| 1 1 2 9 738C 1 1 : 2 | 20 738Cj 1 1 j 2 31 738C 1 1 2 ‘ 40 738C 1 1 2 44 738C; 1 1 \ 2\ 52 738C 1 1 2 59 738C 1 1 2 61 738^ 1 1 2 74 738Ci 1 1 ! 2 ;1 0 0 738Cj 1 1 2;143 738C 1 1 3 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 395 If" <0 C M < 0 — in co ” m «- r t w m cj »- 1 0 *- n < o to ® <•» to ^ v v v t f i n m i A m m i n i A i n m i n Q C O G O ® O O Q Q C O ® Q 03 tO Q G O G O O G D C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M < 0 (0 G D Q C M C M C M C M CM C M O f N O) t t m m <o ° * *- CMCV’ CMCMCMCMCMCMCn O U U O U O O O U O O O O U U O O O U GO O O G O O o r t p jo w n o w o w w w w w w o o r t o Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 396 c o c o c o a j m m c n © tfcntntnrtmmminmcoco 2] ® 03 qq ^ CO CO C O C O <s ^ C O o «- o oj n n lA U) IO N N M O J N N O J O I M O J O i n U O O O U O O O C J O O O O O O O O O O G O Q CD G O CO CO CO G O CO G O G O GO CO CO C O G O CO C O n n o r t o w r t r t M M c o r t o o M O w o r t Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 397 Appendix 3. Rarefaction Code Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 398 Presented below is the Mathematics program used for all rarefaction calculations in this study. Values for blanks A through D of this generalized form are listed in subsequent appendices for each sample population. Note that A, B, and C are inputs by the user and D is output from the program. (* Rarefaction (E(Sn) *) (* This Mathematics program rarefies samples by abundance data using Raup 1975, Paleobiology 1:333, Eq. 1. The only user inputs are the data matrix and the n increment value . Rows are samples, columns are taxa listed by abundance. *) (* O R D E R :____________ A_____________ *) incr = B ; DataMatrix = { _______ C_______ }; (‘PARAMETER CALCULATION*) Size = Dimensions[DataMatrix]; Samples = Size[[1]]; Species = Size[[2]]; (•HOLDING ARRAY SETUP*) Rowsum = Table[0, (Samples)]; For [j = 1, j <= Samples, j++, Rowsum [□]]= Sum [ DataMatrix [[j,i]], {i, 1, Species}]]; (•COMPUTE NUMBER OF SPECIMENS IN ROW j, WHICH IS N IN FORMULA*) Rarefactions = Table[0, {Samples}, {Max[Rowsum]}]; (•OUTER LOOP - INCREMENTS THROUGH EACH ROW TO CALCULATE ROWSUM*) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 399 For [j = 1, j <= Samples, j++, (*INNER LOOP*) For [n=1,n<= Rowsum[[j]],n+=incr, Rarefactions [[j, n]]= N [ Sum [1- Binomial [ Rowsum[[j]]- DataMatrix ([j.Speciesnumber]] , n] / Binomial [Rowsum[[j]],n], {Speciesnumber, 1, Species}], 4]; ]: (•COMPUTES INNER SUM*) (*END INNER LOOP*) ]; (•END OUTER LOOP*) Print [Rarefactions]; D Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 400 Appendix 4. Bootstrap Description, Code, and Results Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Description 401 As an alternative to the rarefaction approach, the “ species-area effect” was also examined at all sites and intervals using a bootstrap approach. Bootstrapping consists of calculating a “ bootstrap sample” through random sub-sampling of a data matrix drawn from an unknown probability distribution (Effron, 1979; Diacron and Efron, 1983; Efron and Gong, 1983). The mean of such bootstrapped samples approximates the mean of the parent population and the standard deviation approximates the standard error of a unknown parent population repeatedly sampled without replacement (Sokal and Rohlf, 1995). Consequently, specification of the underlying probability model is unnecesary and the statistical moments of bootstrap samples from different sample populations may be compared. The bootstrap technique was applied to site presence/absence (1/0) faunal matrices sub-divided into the same temporal groups as for rarefaction. Boostrap samples were calculated through the computer program Mathematica using the code provided below. Each matrix sub-division was randomly sub-sampled with replacement 5,000 times for each integer n from one to the total number (N) of stratigraphic samples (i.e., the first bootstrap consisted of 5,000 random sub-samplings of one stratigraphic sample each, second bootstrap consisted of 5,000 sub-samplings of two stratigraphic samples each A/th-1 bootstrap consisted of 5,000 sub-samplings of A /-1 stratigraphic samples each, A /th boostrap consisted of 5,000 sub-samplings of N stratigraphic samples each). All bootstrap-based richness curves show relationships consistent with those of rarefaction-based richness curves and are present on the following page. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 402 Code Presented below is the Mathematics program used for all bootstrap calculations in this study. Listed in subsequent appendices are specific sample population values for A (input taxon-sample data matrix in binary form) and B (output of mean and standard deviation bootstrap diversity estimates). Note that 5,000 bootstraps were used run of each n random sub-sampling. (* BOOTSTRAP DIVERSITY ANALYSIS *) (* This Mathematics program bootstraps binary taxon-sample data for each sub sample size up to the total number of samples. Output is a list of means and a list of mean standard deviations for each sample size in increasing order. User enters inputs the data matrix and the number of bootstrap runs desired. Data matrix consists of samples (rows) and taxa (columns). *) t = { ________ A________ }; (* Enter number of bootstrap runs *) Bootstraps=5000; (* Calculates matrix dimensions *) Size = Dimensions[t); Localities = Size[[1 ]]; Species = Size[[2J]; (* Define permutation function *) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 403 f[n_,k_]:=RandomKSubset[Range[1 ,nj,k]; (* Set up holding arrays *) VectorMeans = Table[0,{Localities}]; VectorStDevs = Table[0,{Localities}]; Spp = Table[0, {Bootstraps}]; (* Outer loop: Produces vectors of means and standard deviations *) For [Subsample=1 ,Subsample<=Localities,Subsample++, (* Inner loop: Calculates mean and standard deviation of number of taxa for a single sub set size of localities. Sub-samples for a set number of runs specified by user (i.e., Bootstraps) *) For[i=1 ,i<=Bootstraps,i++, R=f[Localities,Subsample]; Total = Sum [ t[[ R[[j]] ]],{j,1.Subsample}]; Spp[[i]] = Species - Count{Total,0]; ]; (* end of inner loop *) VectorMeans[[Subsample]] = N[Mean[Spp],3]; VectorStDevs[[Subsample]] = N[StandardDeviation[Spp],3]; ]; (* end of outer loop *) Print[VectorMeans, VectorStDevs]; { ________ B________ } Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. Results — Eocene-Oligocene Transition 404 hhiiiiiii|||||||IIH *744a (n=10l) Sites 0 10 20 30 40 50 60 70 80 90 100 110 120 n B . 40 35 30 C O o c $ 25 j§ 20 S 15 10 5 0 10 744A Pre-Oi (n=52) 744A Post-Oi1 (n=49) 689B Pre-Oi 1 689B Post-Oi1 (n=43) (n=23) sites by Oi-1 Boundary 20 30 40 n 50 60 70 A(n=52) E (n=6) D (n=23) * B (n=11) C(n=10) Site 744A by Isotope Intervals A-E Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 405 Results — Late Paleocene Thermal Maximum A. 690B (n=22) w 25 < u £ 20 £ 15 738C (n=19) Sites 0 10 25 5 15 20 n B . Pre-LPTM (n=8) - LPTM (n=8) « 20 £ 15 o a: 10 Post-LPTM (n=6) Site 690B by LPTM Event 0 2 8 4 6 14 n c . Post-LPTM (n=2) Pre-LPTM (n=5) S 20 d ) £ 15 o i f 10 LPTM (n=12) Site 738C by LPTM Event 0 2 4 6 8 14 n Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 406 Appendix 5. Algulhasina Morphometric Data Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 407 o « T in o C O ( O o P * 0 0 o 0 5 C O C O h* C O fs. C M P - . 0 0 C O o o C O C M 0 0 in * — in o co m ( £ > 0 0 0 0 (0 in in 0 0 n p - C O p » o o p * 0 0 ® O © 0 5 0 5 © o p * C O co 0 5 C O m C O 0 0 o m in d d d d in in d d in in in d d in d d d d d d d d d d d T “ T “ *- in 0 5 m C O 0 5 t o C M p- 0 0 0 5 0 0 o p* m C O i n s P » c o o *- C M 0 5 in p - C O o m C O 0 5 in 0 5 m 0 5 N © Nf 0 5 C O 0 5 K C O m o C M < 0 p* in o o o C O C M C O C M in 0 5 C O * T p- o o; in 0 0 C O 0 5 d d d o rv C M 0 5 * d rs ! 0 5 0 5 C M in C O d o 0 5 0 5 0 5 * d C O o d d o p^ m p - 0 5 m p* C O C M 0 5 in o C O O o m P * p* C O in C O K o C O 0 4 < 0 0 5 P « C O C O C M p- V p- 0 0 C O ^r T * N m < 0 o o in 0 5 in ( O P* p* C O in ( 0 C O C M C M w fs. n . C N J C O C M C M in in C M C O C M C M 0 0 in m in in *“ C M C M C M * T C O C M C O 0 4 C D o C M 0 0 C M C O C O 0 5 in o o C O 0 0 in P * C M C O C O p * C M C O O 0 5 C M C O 0 0 O O C O C O 0 0 o o O P * h* C O © C M (O ( O p- 0 5 C O C M 0 0 m C O 0 5 < 0 0 5 C M 0 0 ( O 0 0 *- O O p* 0 0 0 5 C M CD o 0 5 C O c v i T " * C O d d C O C O C M C O C O d 0 5 o C M C O d d d 0 5 d d o o o C N J C O in C O m C M *" ( O C O m C M C M C M C M C M C O C O C O C O C O C O C O C O C M in Nf-ni-wrtW(o<ONssN»-rtncoc5*"CM«^tifiioifl<ow P* C O cvi co* C O C O CO 05 CM 10(0 0 C O C O ^ ^ C O C O C O C O oo p» r - o p* o co in K o> C O TT ^ P* P- G O 00 C O V ^ ^ 05 C O 0 0 in p - in C M 04 d p** in in 05 ^ r o C O in f » • co »- C O 00 in N. C O » - in n ^ d 05 d d cvi d «-* cm d co CO C O C O CO C O C O C O N K m co d oo e > 04 p ^ o o TT C O o o C M in p ^ o a in d o 05 o o Length (nm ] U 9 9 | 5 B .3 9 1 0 4 .2 7 0 0 d o 8 6 .3 5 6 7 .3 9 6 9 .8 4 3 2 .7 1 3 3 .3 7 1 0 7 .5 2 6 9 .5 4 1 0 9 .5 9 8 7 .1 6 6 7 .7 0 6 8 .3 6 3 7 .7 8 4 0 .0 7 3 3 .5 3 3 3 .6 9 3 4 .0 2 3 4 .3 5 1 0 0 .9 4 6 2 .1 5 6 9 .5 1 7 1 .3 1 7 2 .4 6 1 0 2 .6 7 8 1 .2 9 7 0 .1 7 5 6 .2 7 5 7 .2 5 4 8 .2 5 d p*. d ^ r 3 9 .7 5 1 8 5 .2 2 05 C M C O C O C O C O C O C O C O C O C O T “ t 05 05 05 05 05 N f C O C M in in in in in in in in C O 0 0 *- * “ »- * “ » “ *- T - * ” r— q o o o o o o o o 05 C / 3 0 ? d p-i d d d d d C O d d d d d d d d d d d d d d d d d d d d d d d d d d d d 1 C O p- * 0 * o* o* n- • o * n ^r C O C O T T 0 0 0 0 0 0 0 0 0 0 0 0 C O 0 0 0 0 o u o N C O C O C O C O C O C O C O C O 05 05 C O C O C O C O C O C O C O C O C O C O C M C M C M C M C M 5 ? C O d d d d d d d d d d w V d T T n d d d d d * T n* ^r N T ^ ■ * o C O C M C O C O C O C O C O C O C O o o in in in in in in m in in in C O C O C O C O C O * T T T n 0 0 0 0 05 *- 05 05 05 05 05 05 05 *- *“ 05 05 05 05 05 05 05 05 05 05 o o o o o o o o o o o o o o C M v O C M T » ■ d d d C M C M * ■ * T “ ^ ' C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M o 05 0 0 C O C O C O C O C O C O C O C O C O o o d o o o * o o o o o o o o ’ o 05* 05 05 05 05 05 05 05 05 ^r C O o 05 05 05 05 05 05 05 »- o o o o o o o o o o o o o o C M £ *- d o o o o o o o d »- *- »- *" »- *- *“ in C O C O C O C O C O C O C O 0 0 0 0 in in m in m in in in ' in in C O c o C O C O co n- m C O 05 in in in in in in in C O co G O C O o o o o 0 0 00 o o o 0 0 0 0 y — y — Tf fv O l N h* 00 o o C O C O C D C O 0 0 0 0 0 0 0 0 o o C O 9 oo G O C D C D C O C O 05 05 05 05 05 q 0 5 05 05 05 05 q 05 05 05 < c v i C M C M C M C N J c v i C M C M C N J C M C M C M C M C M C M C M C M C M C M C M C M C M C M C N J C M C N J C M * C M C M C M C M C M c v i C M C M * C M C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O d C O 0 0 d 0 0 0 0 0 0 c o C O 00* d ’ d o o 0 0 ooco d * d d d * 05 05 05* d 0 0 0 0 a > 0 0 0 0 0 0 d * C O C D * 0 0 u > C M 0 0 C O o o C O 0 0 C O C O 05 05 o o o o o o O o o o C M C M C M C M C M q m ; q m N p ^ K fv fsl d d d d d d d d d d d d d d d d d d d d d d d d d s C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O + ~ r« T - . T " » *• * “ T “ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 408 o p . 00 n C O o C O C O C O 00 C O C O O C O C O in C O 00 O o * r 00 C M 05 00 C O 05 ps C O u C M V C M C M C O C O 0 ) 05 C O 05 05 00 C O p * C O C M C O C O 05 05 ▼ - px C O oo in C O C O in 05 (S O in r t in in in in in in in in in < in *r in in in in in in in **• y- y~ y~ *— T “ y~ y~ T — T — T “ T “ n 05 00 C M m C O 05 o C D C M o C M C M C M 05 co o m C O C O in C O o C M C O in < Q C O O C M V px . C O in o m co m * © m C O N O C O in m y~ m C O * r C M ps. C O 00 C O p^ 8 in N C D cd cd 0 5 * 05 in o 00 d * - 05 o C O P-! 0 5 * C O d in cd is! o in C M rs! V in 05 C O o C O O C M C O C O 05 C O C O © C D C D 05 05 px 2 r— C M oo r*. 05 C M r - 05 x*» C O O C M in N *- C M O 05 m oo in C O px p . C O o C O 00 C O ~? a> C M C O 05 C O h * ® N. T — n C M * - N C D C O C M y- y- C O C O C O C O C M T — * ■ » C M n C M y~ C M T ” C O 05 C O px. C O C M C M in 05 h* 05 C O C M C M * p x . 05 o C M N N n 05 C O o 05 o h - p» . 05 C O C O C O co C O C O m co C O C M 05 00 C M 05 o C M C O * C M C M C O O co V — 00 05 G O C M oo C O © in C O cd px ! C D C D cd in C M cd p-! o in C M cd 00 05 00 P^ 05 in in cd o in cd in ps! o cd in cd X C O in in C O C M C M C M C M in C O C O C M C M in in C M C M C O C O C M C M C M C O in C O C O C O C M in C O ^ c \ j o j n ^ i n i o ( o c o c N i c f) ^ , t f ) ( O t o c \ j w c » ) i / ) i n ^ ^ t f 5 ( O c r) ( 0 ’- c\j c,3c,) P 5 ^ i n c o o j ^ o *- CO 05 C O 0) CO CO O t- CM o CO © c\i CO 05 p. o C O 00 © o> 0 0 *- CM co 05 CO N r*. 0 0 1 *T 0 0 CO CM ® in rv CO 05 in co o in © d co in in o 05 in CM T“ CO in rx. CO CM o CO o fs 05 CO < 05 CO V cd CO r-* CO px. fx- CO fx* _ _ _ 2 "X-' T — < CM Px. CO CO 00 05 Px* T~ CM CO 05 p - CO 00 < O 05 in r .' o 05 05 00 00 CO 05 O) co cd 0) » - W N W C O « 0 ) O » - C O « - N O ^ , * - C O C O ( O t W C O O C M N O C O O J O I r t O O J C M S n N O N O *- O) CO CO in o) o) co m n co co o> o *- ^ CO CO CO N 05 ® in o t - <r CM CO ^ N co O) o in o W in oo co n o 0> 0) cd V coooco«r^ioin*r cm in o px. ^ C O O i N C O ^ i n ^ ^ N O O^(O0)(O^ONNOO0) C M O O ^ ^ ^ C O O ) f l O » " O N v o ) O ) S ( O N i n n ^ 0) i n CO in in CO CO CO 00 CO 05 CM CM CM CM CO CO CO f* *- r - CO CO CM 05 in CO O O O CO CO CO in in in CO CO CO in CO CO in CO CO CO CO CO s P x . CO p x . CO p x . in o a o o o CM o CM p x . ^ r CO c o CO CM o CM O o 05 05 05 in in CO o a o o 0 5 05 05 05 05 o CO CO 00 05 05 in 05 in a o CM CM CM 00 00 CM CM o *- CM CM o O O O o o CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM 00 P x . f x . 05 CO CO 00 p x . CO P ^ O O O 00 00 <0 CO CO CO CM CM o * CM CO CM CM CM CM CM *“ *“ r - r» r**- r - o o o o i n c o o c o o Q c o c o o O O 5 05 05 05 05 05 O 5 0 5 v i n i n i n i n i A i n i n i n ^r c oc oc oco co co coc o i n i n m i n m i n c o c o c o c o c o i n t n i n m i n c n i n i n O 5 0 5 0 5 0 5 O 5 05 05 05 CD 00 CO OO O O O O oooooooo^ r^ r^r^ r T - T - r ’ r f ^ ^ ^ C ’JCOCOW CD O S N N N N o o o o o o o m in o o o o cd cd co co 05 * 05 * 00 00 0 0 0 0 CO CO in in o o o o cd cd co co 05 05 * 0 0 0 0 0 0 0 0 CO CO in : in m in *- o o o o o o o o c o co co co c o c o c o c o c o c o o o o o o o o o o o c o c o n c o w c o ® « N CO *- CM CM CO o ^ ^ ^ ^ ^ ^ 05 05 N N C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO N i - C M i n ( O M ‘ ^ ^ r - f - ^ " C M C O ^ C O 0 5 0 5 0 5 C O C O C M C M C M C M C M C M C M C M C O C O c d c d c d c d c d c d c d c d c d c d c o c o c o c o c o c o c o c o c o c o 05 05 co co 05 05 00 oo h- h- o o h- r*» h- n o o o o C O C O C O C O C O 0 ) » - T - i n 0 5 O 5 CO CO CO CO CO O CO CO t o o (0 CM 05 ® N CO ^ CO CO CO 05 05 h- 00 CD CO CO CO 00,0 0 00 0 0 0 5 0 5 0 ) 05 05 05 05 0 5 0 5 0 5 : 0 5 c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o © *- *- ^ ^ ^ CM CM CM CM CM CO CO ^ ^ ^ V ^ ^ ^ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 409 C M 03 in 03 C M C O o C M C M C O C O rv o fv C M fv in o C O O O C O o 03 C M fv C O in C O C O m C O ^ r o C M u <0 co rv O m q C O IV IV 03 C O C O q C O C M q C M q C O C M o fv ao 03 in in q fv 03 q C O q C O O * r V * V d * V in d in in in in in in in in *T in d in in *“ •“ y - *“ y~ y - w * T “ in f» C O C O C O C O © C O C O 03 C M C O C M co C O fv in * C M C O o C M C M C O C D 03 C O in o C M 03 03 <0 03 in C O o fv C M |v |v C O C O O * - y - C O q C M C O 0 ) 03 o a> q q fv C O fv C M q q q q q in 03 q in 00 C O in o C M 03 cd 03 03 * 03 IV cd cvi cd cd fv in cd 00 d oo o C M in C O * cd cd 00 co Q C O C O o 03 rv C O C O o in C M O fv m C M 03 C O C O C O C O C M C M o o m C O C O 03 03 C O C O C M C M C O in o C O «* C O C O |v C O C O |v C O o fv C O C M C O rv m in C O co T“ in 03 03 fv f . co fv 03 03 o m C M C M C O C M y - C M C O in y - in C M r - C O C O C M C M in — ao in 03 fv 03 C O o <o C O © C O IV in C O C O fv co C O C O 03 o O IV C O ao fv 03 C O 03 ao C M C O 03 03 o OJ in O O C O C M o 03 G O C M C M 03 co q C M in <0 C M o q C M in in C M C M q C M 03 q 03 q ® in V *T in d cd C O C O cd cd cd in o in cd cd m’ cd in fv 03 cd d cd in d cd cd o X <o in in ^ r in C O C O m C O C M C O C O C O C O co m <r C O C O C O C O C M C M C M in in C M in C M C M C O ar M a le | - -?■ (0 c C M C M C O C O C M C O n C O C M m T _ in C M C O *T in C O fv C M C O C O C O m in O C O O < in C O in q 03 < d C O 03 rv *- o o rv o q o < cd cd 03* cd ^ r C O fv T f C O IV rv 03 O C O o C O in C M o C M in fv < C O 03 03 03 IV cd cd 0 3 * in in in in m in in in in rv in C M in C M C O C O fv C O q q fv < 03 N d r - ’ fv* C O fv C O |v IV fv C O | C M 00 v - < 03 C O ‘ 03 0 3 ~ * C O C O * - 03 o C M < 03 cd C M rv d ao C O C O 00 C O o C O 5 03 cd < o y~ y ~ r - C O C O m in q T - C M < d cd cd d - © 03 03 03 a . r v C O C M f» in C O IV 03 03 O C O V f C O 03 C M |v fv 03 o fv o C O C O in C O o in ’ w ’ C O C O C O C O V IV o C O C O 03 in o q o rv o 00 C M m fv o in o C M q * - 03 fv q o q .c o 03 03 cd N d 03 C M 03 cd 03 03 fv cd cd fv* o V cd cd 03 03 in d ad d IV cd cd d o> c © - J o 00 03 C O rv oo C O in IV in 00 m o in in • O' *“ C O fv in in in in C O C O 03 fv C O C O 03 03 C M oo 00 rv |v |V C M C M C O C O C O C O C O TT •a* *T C O C O C O C O co C O C O co < 5 5 r v q n o o q T f in in m C O C O q C O C M C M q C M C M C M C M C M C M C M q 00 C O 00 C M q V fv UJ s? in 03 03 T f d * cd cd cvi cvi cd cd cd cd cd cd cd cd cd cd rv C M * 1 C M C M 03 03 ® C O C O C O C O o o o 03 03 fv rv |v fv fv fv IV fv fv fv o fv C O C O C O C O a o o |v rv C O C O C O C M C M 00 C O q in m C O C O 00 q q C O C D q 00 00 q 00 o 00 00 00 C O C O q C O fv fv* in in cd cd cd cd cd cd cd in in in in in ^ r o d o o d o o o o o T~ C M *“ o o o C O oo C O C O C O C M C M C O C O co in in ao C O < *• o o o C O o o o fv C O q y - y - C O C O C O C M C M » - y - C M C M C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O q o Zo cvi cvi cvi C M cvi cvi cvi cvi cvi cvi cvi cvi cvi C M cvi cvi cvi T “ o 03 C O C O C O C O |V |v f v ‘ * - fv rv * fv * o ’ o in in C O C O C O * CO* C O C O co C O CO* C O »- CM * C M C M C O C M CM* C M 03 C O q C M C M in in m in in in C O q in in y - o q q q fv q q q q 2o r “ I- ’ i - 1 r» o o o o o d o o ao 03 03 in in C O <o C O fv IV C O C O 00 ao 00 o 00 oo 00 00 00 00 00 00 C O C O in C M C M C M in (D fv O o C O C O C O C O C O <0 rv fv a y C O fv fv fv |v rv fv fv fv rv fv 03 fv o O o C M 03 C O q q q q q q q * T f m ; q q q q q q q q q q q fv fv fv fv q q q q < C O C O C O cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd C O * cd cd cd cd cd cd cd cd cd cd cd cd cd cd C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O 00 a>” ’rv * IV* ao C D CO* in in * fv fv * fv > v ‘ fv IV* IV * v* fv * fv fv fv 03 03 U > in in C O co 03 03 03 C O C O C O in in rv fv fv fv fv fv rv q fv rv fv fv q q C M q q fv fv fv q m n *T r t in in in in in in in in in IV |v fv |v rv fv fv IV IV |v r v 00 ad cd ad cd cd cd od 5 * r * * ^ r * * * * * * r ^ : T “ r * T “ * ■ ” r “ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 410 o C M C M O ) ^ *- © ^ i n ^ ^ i h i n i n c b i n C M C M V ( O C M ( o o o ( j > o i n o ^ ( 0 N C M « C M f l 0 ^ r r V N » - « m- Tf in in * i ^ ^ B ( O o o t n c M n ( 0 ( O N i n N c o c o v o e o o o O N V C M I f i t t r O n O J f f l W l f l N W f P J l f l O r - ^ • ' i n i n i n i n i n ^ r ^ T r i n ^ T r i n i n i n d ^ r ^ ^ r i n t n C M (O C M O c d C M C O C O f-» C O c m c o ^ o m C M O ) rt o ^ ^ ^ C D C O o ^ C M C D O J ? C O z C O * £ 1 “ ^ V 1 - w _ lrtinO ® ^»©CM O O _ ^ o ® * - ^ © © © © CM O rv O in © C O C M ^ ^ Ifl ^ C M f C*5 C M O> 0 _ — lr t,n C O ^ C O W f r » N . n f . W <O ^U lC *3 o o o o m j ; ® J " § o e > i ^ < o N . ^ ^ : g { ; j ; o ) » : u > ' W i n N d n ^ j - j j ^ w o D o j n d c o c y i r t J m W W f - n w ^ f" N ® 2 r t S S ® <e* ® (o to c o 5 j 5 » t » - w « n c c c o N r J S S c M r t c o i n c o o o S K S 0* ^ ® ^ 10 ^ ^ . . ^ N N M f l M C y i C M r r r f ^ ^ ^ W C T C O C M o C O ® c o ^ © c o * fv © rv rv 0 0 © C M © © m * in c d c d rv c o © c o c o C M © c o © © c m © rv ^ o in o © c o © © © go « cm cm in n t ^ c o c o c o i f l c o c o i o a N n i n ^ n O J l O N N i A ^ O N C D C O n O V C O i n C M A ^ N n i A *- *- © © C M C M © © ^ c o © © © C O C O C O *-CMCO©©©®®COCOO®^CMCOCO n c n n c o c M C M C M C M ^ ^ n w c o w w w c M w n m m ^ r t » - ^ ^ C M ^ M , 1 0 © S t * N C * 5 M , ^ ^ ^ ^ © © © N C O C ,J C O ^ ^ t O l O < f i ( O N C M C M C \ j C ,5 ^ © rv © c d © C M C M c v i © © © ® in h - C M C M rv d © © © T f © ▼ - T- © r - * ! © © © C M © ^ *“ © © © N. *- © c v i r v © © © © © r- c m © © © *- © h » © © C M © © © © © © © © © © © © © © © C M c d r v © © © © C M © © © © © © < d *- © © r v © © fv © C M fv © © © c o © © © © © © © © © ^ © h» © © C O C M O ^ - ” © C M © © C M © © r v © © C O © © ® i - r v © © ® i nu j ^ N C M i f l « n © w ^ » - n » - C M © w ® i / ) © C M C M C O © i n © i n C M * - C O * - ® C M ® f f l O l O © w . _ .............................. o © i n © « - © f v i n i n f v © c M © t - © r ^ ' f r © © © © © © © © © ^ © © © f v i n i n ^ r ^ © © © ^ ® ® © © ^ © C M © © r v © o © © © © © r v © © O J © © © © O O O O O O O O O O O O W C M C M C ' J ( M C M C M C M W ( M r ,* S N N N 0 0 0 0 0 0 * - * - * - ' » - * - ' » - ^ ' * - ^ - , ^- *“ *“ © © © © © © © © C O © 0 0 0 © 0 ’ © d r v r v r ^ r v r v f v r v r v r v f v f v r v ® © ® © ® © ® © ® ® © ® ® ® ® rv rv rv C M c v i C M © © © © © © rv rv © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © © o o o o o o o o o o © © © © © » - f - » - * - * - » - » - » - * - 0 0 0 0 © © 0 0 0 0 © © rv rv o o © © S W © © © © l f l © ( 0 n © © n © O © © © © N N N N S N N N N N » ’ »" O ^ - ’- ^ - ’- ^ - ’- O O O O O O O O O O O O O O O O O O O O O O O O © © © © © © ^ © © o d d o © © © o o © © © © © © © © © © © © © © © © © © © © © © © © © i n © in in in ©CD © © © © o o o o o o o o o o o o o o o o o o o o o © © © © © © © © C O © CO«©CO©NNNNN 0 0 0 0 0 0 0 0 0 0 © © CM CM © © cd ©’ © « ©’ ©' © © © © ' n n © rv rv CM n ^ © © © © © © © © © d fv fv co o o © 0) 0) n n n fv fv ^ If © © © cd © © fvrv' o o d © n s ^ ^ n ^ N N S © © © © © © © © © © © © © © © fv fv o> © © O O CM CM CM 0 0 0)0)0 ^ ^ n n fv rv © © cd cd © © d o CM CM d d Tf <r ^ Tf ^ ^ rv rv tv rv rv rv ©©©©©© cd cd cd cd cd cd © C O © © © © o o o o o o CM CM CM CM CM CM O O O O O O ^ ^ M ^ ^ ^ fv fv fv fv fv fv O O O O O Q © GO © © © ©©©©©© © © © © © © O fv fv fv fv fv CM * ^ ^ ^ ^ o o o o o o ^ ^ n rv rv rv rv o o o o © © © © © © © © © © © © r ^ f v ' rv f v * * * o o o o ^ ^ fv CM CM CM CM CM O CM CM CM CM CM © o o o o o © © © © © © © © © © fv fv fv fv fv fv «<r © © © © © o o o o o o ^ ^ ^ ^ ^ ^ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41 1 © 0 ) 05 o 05 rv rv o to o o in 05 o 05 05 in 00 C M to 05 © o C M © rv © © © C M © © T - © in C O q in C M in fv q 05 co m C M m q rv q q C M C O ■ » - C O C M o fv © q © © © © d in d (O in in d d d d d *T d d d d d d ■ M * d d d d © T _ T — T ~ T_ T_ T “ *“ T “ *“ C O to C M in C M M* o Y- A 05 o to C M o C M C O 00 rv o C O C M © o C M © © © © 9 © © q m C O rv C M C M C O U l C M |v o fv r v o q C M 00 in q q n C M © q © q q Q © © ?v C M * • rv d d cd o ■ N 05 d o d d cd d d C M d d © o © ’ d rv • ■ d |v V” fv * <0 « — **“ u5 C O r» in to in C M to in m rv rv m co © C M © o o C M © © O ~v 00 o in fv o C O *T in to rv 05 to rv o r v o o C M © fv C M © rv © © © © T* in C M C M *“ 05 C D C M *- in C O C O CM T“ C O in in © C O © C M C M C M © oo in to rv C O 05 IV © 00 00 05 05 o C O 00 ^ r to to © © © fv o C M © to |V C M o C O to C O q CD q C O to »- o q 05 q C M o © fv r*- fv © © © C M © o © © in o *T cd d d IV cd d cd cd d cd cd cd d 05 o d C M C M C M © cd © fv © C M C M C M C M C M co CO C O C M C M C M * 9 C O C O to in m C O C M C O C M to © © © © <r * r © © C M C M © to rv rv C O C O Tf m to to C O t f * - C M C M C O in in to C M C M © © © *T © © © 0 ) in to C O C O * C O C O n C O in 05 C M o C M d o C M o C O - M - C M C O T- m C D C O co C M © to cd d d fv 00 in C O rv in © q ® o q in in © d 05 ^* IV d d © in in in in in in © © C O T f d o © © * — CD C M »— q © cd 0 5 * d © C M rv to to to rv © fv IV © 05 o © O T J- C O C M rv d © 00 to © © to C O C M rv V 05 d d d o 05 05 © o C M 0 ) in C O iv 00 in C O C O C O rv to 05 00 in in C O C M o © o © © © n © ▼ - in to C O q to o 00 O) C M C M q m tv o ■ M * u> C O C M o © q C M ▼ “ © q © C M © (0 C O d cd 05 d 05 cd d o d IV d rv d r - ' d d d C M * d d o d © d T - C M © © © rv 'O' C O C O to to in in in C O ^ r to in in 05 C O 00 rv m 05 © © © © rv fv © © © rv r v rv rv to to to to co to to to oo C O C O to to to to to to to to © o o o o rv fv N rv rv |v rv fv q q q fv fv fv IV rv rv rv fv v - 00 00 d cd iv IV rv rv fv rv rv IV d d d d d d d d d d d © to to to to C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M C M fv 00 C O 00 00 o o o o o o O o o o o o o o o o o o o © o © o d * - T“ * - * • d C O C O C O co C O C O C O C O to to to 05 05 05 05 05 05 05 05 05 © © © © © © © © © © © © o o o o o o o o o o o o T~ o o O O o o O O 05 © © © © © © © © © © © © * - *— V “ O o o o o o d o d o o © o c o ’ C O C O C O CO* cd* C O CO* CO * C O C O C O C M C M C M C D C O 00 d * d CO* C D C O © © ‘ o * o ' o o © © * © © * o ~ © * © iv r v rv rv K N . rv IV fv |V rv rv o o o 05 05 q 05 05 05 q 05 © © © © © © © © © © © © q o o o o o d o o o o o o O d o o o o o o o d o d o o o o d o * o o o C M C M C M C M fv tv |v r v r v fv r v IV in m in IV r v r v |V tv rv rv rv * * n * r ’• t fv C M C M C M C M ^ r n ^ r n ^ r n rv |v fv 05 05 05 05 05 05 05 05 r - C M © © 05 05 05 05 05 q 05 05 05 05 05 q 05 05 q 05 05 05 05 05 05 o q o q o o o © o o © q o C O C O C O cd cd cd cd cd cd cd cd cd cd cd cd cd cd C O * cd cd cd cd cd Tf d C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O C O w © © © © © © © © © © IV* fs j' rv r v ' N * N* r ^ ‘ r v * r v ’ rv tv * o * o * o * CO* C O o * d * 00* 00* d * d ~ ' t- * T“ ' T- ’ V- * ^ ’ CM to to to to CO © co q 00 q q q CM CM CM CM. CM CM CM CM q q q q q q q q © 05 d 05 05 05. 05 05* 05 05 05 05 05 d o o o o o o o o o o o o o d o o o o o o o o o *r ^r in in in in in in m in m m in in © © © © © © © © © © © © T_ T_ T- *“ T_ T- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 412 o a > 0 5 o C M C O T * * C M C O rv C M 0 0 o C O rv C O C M 0 0 C O T T C O fv 0 5 0 5 C O a o C O 0 5 C O rv a o fv C O in m 0 5 a o fv C O C M 0 0 in a o C O rv C O 0 5 c o c o in C O in C M a o in 0 5 0 5 C O C M in C O a o O in in in T T in in C O C O in m in in in m C O in ^r «- r» * * • *— * “ T _ *~ O ) rv C M 1 0 0 0 rv C M in C O C O rv 0 5 in C O C O o a o 0 5 C O a) C O C O C O < 7 5 0 5 r> C O in C O in o C O C O o rv c o C O o C M C M in C M O G O C O C O 0 5 (0 < 0 C M C M C O C O o o fv < D < o C O C O C M * t C O o 0 0 C O C M r*. C O C O o o C M in C O C O C O C O C O C O C O rv in C O C O C O C M C M C M 0 5 in C O in in O ' C O 0 5 n C O o fv C O C O C O C O C O C O C M C M rv C O C M fv C O o T — rv C O a o o m M - c d fv in C O o in C O 0 5 C O C O » » in C M G O o rv C M C O m C O T “ rv (0 0 5 O fv 0 5 C O fv o o C M rv in 0 5 rv C O in C O in C O C M in *• C O C O a in C M C O r r rv 0 5 C O C O o rv t t co in 0 5 C O 0 0 C O C M C M rv c o 0 5 o c o rv 0 5 C O o C M 0 0 C O in + C O O C O C O C M C O C M fv C O O a > C O C M C O C O O 0 5 < 0 C M rv rv C O C O C O 0 5 C O in 9 X C O C M C M C M C O C M C O C O in C O o in in C M in C M C O C M C O C M in in in C M C O 0 5 C M C M m C M 0 5 C M C O in in C O m C O C O m C O C O C O in C O 0 0 C M C O C M V C M C O in C M C O o c o 1 -*• w ( 0 — • - ( O c C O in C O C O C M C M C M C O C M C M C M C O C O in C O C O rv rv C M C O C O in C O C O C O C O C M C O in < 0 5 C M C O C O in o C O C O in 0 5 C M C O o < C O C O rv C O fv C O O fv C O o C M 0 0 0 5 r“ < C O m < c o o 0 5 in C M 0 5 • < r rv 0 5 C O < C M in C O m C O in rv in C O m C M C M O 0 5 in C O o C O o in 0 5 0 5 < 0) C O in C O C O C O rv C O 0 5 C O o rv | < C O 0 5 C O a o rv C O rv C M C O < O C O © C O C O ® C O C O C O rv a o | < s s o u C O 0 5 o 0 5 m rv in C M C D C M 0 5 C O < o o o o o > 0 5 rv 0 5 o E C M O C M C O o C O C O 0 5 in in C O 0 5 ® 0 5 in o in rv C O N c o rv C M C D o C O o 0 5 in 0 5 C M C O C M C O in in C M C O in 0 5 0 5 0 5 in C M 0 5 rv 0 5 C O C O o C O C M 0 5 C O C O 0 5 £ O J c © _ 1 0 5 C O * r C O C O rv C O d o d o C O o C O < o in C O C M in b o C O C O C O C O C O C O rv C O C O rv C O O 0 5 C M C O C O 0 5 0 5 rv C O 0 5 C O O rv C O in C O in rv in C O in in rv C O o *r fv 0 5 o o r - 1 in in in in ^ T 0 5 0 5 ^ T 0 5 ^r 0 5 0 5 0 5 • O ' 0 5 rv o rv o rv o rv o fv o rv o rv o rv o fv o rv O rv o C M in C M in C M in C M in C M in C M m C M in C M m C M in C M in C M in C O rv C O rv (0 fv s fl 0 s * in in in in o * in in in in in in in in m in in C O C O C O c o C O C O c o C O c o C O c o C O C O c o 1 o rv 0 5 rv 0 5 h * 0 5 iv 0 5 o 0 5 o 0 5 O 0 5 o 0 5 O 0 5 o 0 5 o 0 5 C O C O C O C O c o c o C O C O C O C O C O C O a o C O C O C O C O C O C O C O C O C O o o o o o o o o o o o 0 5 0 5 0 5 o '* o O o o o o O O o o O O o o o o o o o o o O *- * “ ’■ "" *" o o o o C O C O 0 5 C O 0 5 C O 0 5 C O 0 5 rv 0 5 rv 0 5 rv 0 5 rv 0 5 rv 0 5 rv 0 5 fv 0 5 o o o n o o o o o o o o in m in in in in in in in w* in in C O o o o O o o o o O o o *- *- *- *- *- *- *- * - T “ *“ O C O C O 0 0 C O 0 0 C O C O C O C O C M 0 5 C M 0 5 C M 0 5 C M 0 5 C M 0 5 C M 0 5 C M 0 5 C O 0 5 C O 0 5 C O 0 5 C O 0 5 C O 0 5 C O 0 5 c o 0 5 c o c o 0 5 0 5 C O 0 5 C O 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 n- 0 5 0 5 0 5 n 0 5 0 5 0 5 0 5 0 5 0 5 0 5 C O © o o o O o O O O O O o O o o o o o o o o o O o o o o o O o o o O O o o a > O l rv C M o rv C M O fv C M o rv C M O in o in o in o in o in o in o in o fv o n o rv o rv o fv o fv o rv o rv o fv o rv o rv o C O 0 5 o C O 0 5 o C O 0 5 o C O 0 5 o C O 0 5 o C O 0 5 o C O 0 5 o C O 0 5 O C O 0 5 O C O 0 5 o C O 0 5 o C M C M C M < C O C O C O C O ■ < r C O C O C O C O C O C O C O C O rr C O C O C O r r C O C O C O C O C O C O C O C O n C O C O C O C O C O C O *r C O C O C O C O C O C O C O L L C M in C M in C M in C M in rv T “ rv rv rv rv T “ rv T * rv rv c o rv ® rv C O rv C O rv C O rv o o rv a > fv C D fv c o fv a rv C O rv o rv o rv o fv o rv o rv o rv o rv o fv o rv o rv o rv C M fv C M fv C M o in o m o in o in o in o in o in o m o in o in o in o in o in o in © in o in o in o in O in o m o in o in in in in in in T“ in in in in in in in T ” in in * - T _ ■ r - Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41 3 r ^ r ^ i n o o » - c M c o c M ® c M © ® c o ® ® c o c M ® r ^ o > r * T - c o t r * - r ^ c M O * - i n © © c o o > ® ^noo(ooo)n^iA(onifln^o>flooriA(MO)ojNoj(00)0(pf~co^c\isiAt ^ ^ ^ u ) ^ ^ ^ u i u ) i A i A u ) i n u ) ^ i A i A ( D N ( d ( b ^ ^ 7 U ) U ) ( 0 < d u ) ( d u ) ( b i A ^ ^ f O o j ^ w f f l O J O J i n o N t o u i e o o ) - C D f l l O l ^ V S r O l O J N I f l N O N S ^ ^ i r i u i i f i r - N i f l N T - p j c o s f l D f f i ^ B O c o B B B O J c o c M r ^ c O f - O B C M ^ - C M C M C M ® ® ® ® ® ® ® ® K CO K ^ CO CO CM CM CM CM CM CM o o *T O r*. *- co in r-* co p-. ® ^ r*. i n o i n i n ^ ® i n p - ^ ^ n o ) P- <0 CM ^ .n * * * W »*/ «*/ H A W C M r - C O j j N W i n N ^ g ^ d c O C M C M d —rCM^' Oi CM ^ P ^ 0 0 ) * - ^ g ^ ® 0 > ® p r t . f. r F P f . ® N ^ |n |‘> |f l W n C M r ® n * - » - ^ ^ ® C M N C M W o © N n c M N a i o ^ u ) c o o u ) i - m i n o o c o n o o ) N U ) ^ ^ O f - c o o A ( D ( O O 7 ^ f l o e o ^ o o N 9 ( O N 7 i A ^ u ) n o o r r t a ) 0 ) n c v N { ' j r > W T > i - U ) ^ U ) 7 ( 0 l / } ( D ^ N n ^ 0 ) O t > C ’)CNjC)(,) 0 0 M ^ < 0 0 ) 0 > C M 0 ) v - C M ( 0 w^^^^nnrtconnnocMCMCMi-CMWcoin^nwinnncMeMCMco^^^ f « N C M » S N O ^ ^ i n c M c o o c o © d ® ^ r *-'CMCMCOCOCDCO^, ^ ^ r * J ’ *J’ ^ ^ , ^, i n © © P ,*P,' * P ' '» - C M C O ^ T © C M * r ^ r © © © C O C O C O C O CM in K CM © C O C M cvi d CO CO C O o > © C M O p -’ d C O C O © 0 3 is ! CO C O ® C O © C M o > co fs! ^ •o * CO ® C M * - ® © V- © fs . O ) ® ® p - © p - o ^ ^ © © © © © p - i n © © © © © © © co m c o o © © d r- © in tj* cd * - co tt o p * d © © © © © © © p- co co © in ® co cm * - cm © * • CM d C O * d ® ® © p» 03 o i ® C O O O ) i- co co © cm ^J“ P» © 03 *- a)©®C00)03CM©P* l O N o c o c M Q C M i n n S i - i n ^ C O l O C O n ^ l f l i n i O l O l O N O N O C M C M C O ^ C M O C O ^ * - ° ® ® © © © © © © © © © © © © ^ c o c o c o c o c o ^ ® © © ^ © 0 ) P » ®® ^ t ®© © CM©© ©C 0 CM ^ N C O N N C y i t C O ^ © © ^ • n c o © © © N ® < o c c c o < o ® ® i o < o c o < o i o ® ® < o ® < o c e ® < e < o o > o > o > o > a > ^ f T r ’< r ' < r n ^ r * - ’- ’- ’- N N N S N N N S S S N N N S S N N N N S N O l O O O O O O n n n n i f l i n i n i n cod(dd(od(O(dd(d<O(O(O(O0(O(O(O(O(O<O(O(0(O(O(O(O<o<O(O(O<O(O(O(O(O 0 ) 0 ) 0 ) 0 ) 0 ) a ) 0 ) 0 > 0 ) 0 ) 0 ) 0 ) 0 3 0 3 0 ) 0 > 0 ) < 3 ) 0 ) 0 ) 0 ) Q O C O Q O C D d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d o C M © © © © © © P - P - P » P » 000)0)0)0)0)0)0)0)0)0) o o o o 03C30>0>0>C30>0)03030)C30>0>03030)0>0>0303COCOCOCOCO©©©©©©©©©© OiO)©CJO>©O>O3O>O>OidJ©©OiC>OJ03O>O>OiO©O©©0>©0>0>O)O>CD©©® d o d d d d o d d d d d o d d d d d d o d p f ’ p r f d d o d d d o o o o CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM^T ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ▼ — ▼ — ▼ — ▼ — r - T “ 1 — T — » — I - 1 — T — r- r- ^ t t ^ ^ t t ^ t ^ ^ t t t t V ^ t ^ t ^ t V t t t ^ t t ^ n n n o w n n n n n n n n n c o n o n n n n o n n o n w c i o o n c o c o w n n p^p/p* ’ p.‘p.p**r^'ps. fs.'r^'r^ r-.'r^'h-‘h.'h-'r^ r ^ K -'r^’h - d c > o j o io jo o© © o © o > o > 6 > b > CMWCM(MCMCM.CMCMCMCMCMCMCMCMCMCMCMCMCMC\JWV<f^^^SSNNNNOO(0®flO i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n t n i n i n ® ® © © © © © © © © © © © © © © © © Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 414 a o a> O 03 C O C O o C O C O a o in C N J in in O O 03 fx. in C O in C O in ffl a o C O C N J o r - C O C O h - a o 0 0 0 3 0 3 o (0 in C O C N J C O h - C N J o Is * in in C N J in ao © <o C N J C N J 03 C O k 0 3 C O a ? 0 3 C N J o C O C O • in * r C O C N J • * C O fx. C N J f x* in in O C N J m N* C O co C O m ao r * S in 5 d C O C O J ” o C O 00 03 fx- C O o 00 in a o in C O co C O C O C N J C O o in C N J in C N J in C O 0 0 C O C O C O T — 0 3 o o C O h « . f x* m 0 3 C O C O C O C N J T “ C N J C O C N J C M C M — C N J in C O r - N 0 3 C O in < 7 3 r **. h» in C N J C O C N J C O N in in «• C O o C O fs. 03 co 03 o C O C O C M CT 0 0 03 C O C O C N J C O C O a o h- h» 0 3 C N J ao C O C O C O p- r — C D C O in C O C O O h* 03 y~~ rs. a o in C M © m C O in C N J C O C N J in C D 0) 0 3 * - C O C O in in C O C N J C O 00 ao C N J n 03 o o T- o C M X C O C N J C N J C N J C O in m in C O C N J C M C O C O C O C N J in C O C M C N J C N J C N J C N J C N J C N J T “ C O in in o in i- c o »- oi O ) c n m (O O r i» o i- r*- r- ( 0 ^ o o c o i n t < - « > o i o o ) O o o ) O ( D O O 5 N o n N ( O i n w o N ’- i o o ^ ’- p j o i c o ^ ^ i n v i n o i c M C D i n c o N f l o s o ) t i n i n i n c M t t i n i n i n i n i n i o i n i n i n i n i n i D v ^ ^ i n i n i n i n i n i n i o v v i n i r ^ v i f v i n ( O h * T “ C M ( \ j M c o ^ i n i n N P ) n ^ v ^ < o o i ^ i o w i n < D < o ® ® N ^ C M w n r t w n c o CO co o cvi c o C N J C N J cvi c o CO C N J C O o S f l O O O ) C O C O C O C O in ■ « r C O N 03 in co ^ V i n < d ^ ^ ^ C N J o > c d in o c o c o C N J T f T T C O in in in h- m C N J C N J h « ! C O C O C N J C O in 0 3 * c o co o> ^ 0 0 co in c n j co n in cd rsi co co co co C O o »- CO CO CO c d in c o o o c o co C O C O 00 r * N C N J C N J o > c o 03 m ^- o^ rc Nj t-c oc ocN jc vjc Nj CM*-^C0CMC0®*-CMO3*“ s ifi o d s c n in < d s < d < d int^coococoocoinn C O C O ^ O N C O N C O ^ O r - in C N J ^ ^ t- C N J C O C O O C N J C O C O ^ C N I C O C O N ^ W ^ ' C O ^ ^ O C N J ^ ' I - C N i i n o i d d f f l r - ' d s ^ i n c d s o o c o o s c N i N i - T - d i n c d N coocoininin^coin^v^cocortcocoo# 0 0 ▼ ” O ) c o s c o m c n i o o 0 0 c o C O c o c o r- in in in in r CO CO CO CO SNSNSNNNNNSNNNcocowcococonncocoininininininin t - t - * - * - ^ - t - » - t - c o c o c o c o c o c o o > o o > o o ) 0 0 > o o > o > o o o o o o o siCNSNNNNcodidcdcdcoiflioininininininininNNSNNNS h» 03 03 in in c o o o © o r ^ o o o o o o o a o c o c D c o c o c o c o - * - * - » - * - * - * - » - i - * - » - r ^ r ^ r ^ r ^ K r ^ r - . 0 )O 0 0 (0 « Q C 0 Q C 0 C 0 N S S N N N S N N N N N N N N N (0 (0 (0 (0 (0 (0 (0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 m c o M e o c o e o c o c o c o o w n n w o n N S N S S N N N N N v ^ ^ t ^ v ^ ,- © © 0 © 0 © 0 © © .- i- * - * - * - * - © © © © © © © 0 © 0 0 3 0 3 0 3 0 3 0 3 0 3 0 3 r * r * T ~ ^ t - t— ^ ^ r - ^ ^ r - r * » - ^ ^ ^ r " r» ^ ^ ^ r * ^ O O O O O O O in C N J C N J C N J C N J C N J C N J OJCNJCNJNSNNShCOCDCDCOCDCOCOCOCOCOCOCOCOCOCOCOCO ®OOOOOOOOOOa3C3300)0)CO©COCOOCOCO®COCOCOCDflOCDCO®CD 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C O co 03 03 CO CO 03 03 co ao CO o o 03 CNJ CNJ CNJ CNJ d d d. CO CO CO o o o o o o o r ^ h - r ^ C N J C N J C N J C N J C N J C N jC N jn ^ r^ r CNJCNJCNJCNJCNJCNJCNJCNJCNJCNJ cocococococococococo o o r s .r > s .fs .h - h « * 0 3 0 ) 0 3 OOOOOCNJCNJCNJ f- CM CM m m in CNJCNJCNJCNJCNJCNJCNJCNJCNJCNJ i n i n t n i n i n i n i n i n i n i n S S N ID (O (O (O ^ n n r^. r*» C N J C N J C N J C N J C N J C N J C N J d d t d t d d C O C O C O C O C O C O co 03^ 03 * 03* C M * C N J * C N I C M c n j c n j c n j in in in in C N J C N J cvi cvi C N J C N J C N J in in in m in in in t o $ in to < o to m f s N N N N O) C N J C N J C N J C N J C N J C N J C N J ^ d d ^ n' t t CO CO CO C O CO CO C O C N J C N J * C N j ’ C N J * C N J C N I h* in i n i n in i n in to C N J C N J C N J C N J cvi cd C N J in i n in in i n in in in in in in in m 03 03 03 03 03 0 3 C N I C N J C N J C N J C N J C N J d d d d d d C O C O C O C O C O C O r*’ h * h *. k* K* h . 10(0(0(0(0(0 c n j cnj cvi cvi C N J C N J in n in in in in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 415 »■ n f li 3 ) n o i ® ® ® r - ^ in * t in ir ir ^ < o < D < 6 * r * T * r * t c M r - C M B O O B i o r - c O ' - O T - B o r - ^ c o B ^ - o ^ r ' - c o ^ ^ y ® ® ® W S ( 0 ' T N ( 0 ’- N 0 ) 0 J i n n W 0 5 ® O C \ i C ) f - O ^ ( 0 ( M 0 ) C * 1 ( 0 « 0 t f ) ® CM ^y r- ® ® C O ® o CM ® ® o CM C O C O o> O) C O r- C O ® r- © r- co o o ® 0 ) CM C O ® ® 0) C D ® <0 o C O o C O CM ® CM o ® CM ® o C O ® ® C O C O ® C O ® CM o m CM CM o • ® (O CM 0> r- • CM ® O r- <0 h* • CM C O O CM CO ® ® 0) N. ® C D o ® ® CO CM o co CM o CM ® o> 0 ) o ® o O ® CM CM ® CO CM o CM r- CO co ^r ® ® 0) o ® cm ® co o ^ co in r- m S K ® ® S g j 5 j ; n n o O W N c v i o) o a *5 i n O ) D B l f l O N ^ B i n S ^ © O J f f l i n N O l f l © ^ ^ N W O ) N ^ ^ P 3 n O N O ) S O ® ( \ i ( 0 ^ O 0) « - N 0 ) ( 0 (0 (0 O 0) 0) 0) 0) N N 0) N O O l A l O < O ' ” O ® N l f l N * >,lflCM(\l 0 0 c d o N c ,j o > » - d o d c ,) n o i i n < c i © f l d o ) t f O ) ^ C N i < M ^ o c o u ,) N r s o ^ : ^: b ^ » - { N i » - *ycocMCM*-CMCMCM< 0 ® T y o ' ^ ’ CMCMCMCMCM*-®<ro, co<ocococMCMCMCMCMCMCM®*rco c o ^ ® ® r - r ' > » r ,' » r - ^ c M c o c o c o ® ® ® ® < o r - C M C O a o ^ ^ ^ ^ ® ® ® ® ® f s» f > - C M c o ^ © r- h- <0 ^ co co ® CO CO CO CO CO ® CO © ID o CO <0 ® o < o cm in co c o o ® co co ® ® o co ® *- *- in n o > CO CO ^y h- < 0 CO OJ N ^ t- O ® 0) *- CO CO ® N ^ ^ ^ ^ ® CO o > ® *y ® CO <0 *T m a> *- co r- C O *r co r < y i *t ^ ® ® ® ® o r - ® ® ® r- ui (O n <o © r- <o co o in r-’ oo co co C N J C N J r*^ ® C N J o> C O CNJ TT ® ® CO *T ® in q ® o o i n s o o s N i o i f l n o i s ® C M r - * r * “ ( 0 O ) * - < 0 c o c 7 ) * y r - c o f f l c o c o ® c o < 0 ( 0 < 0 ^ y c o c o c o o ) C N j © © r - o c 0 0 ) * “ © © ‘ — ® ^ r- in ffl CO CO CO © co co co co co © m id *t C O < o h* CO ® « C N J ® S O U ) f » n ^ ( O d ) ( O ^ U ) N O < 0 ( O N O c d c d d r - d d c N j r ^ c d ^ - C N j ^ ^ ^ i n r - d ^ - C N j i n ^ r s ! * - ^ ^ * r ^ ^ , c o o > ® ® m ® m ® ^ , ^ c o c o * r co co co < o m ® © © © ® ® ® © © © o o o o o o o o c o c o 0 ) 0 > 0 ) < 3 ) < 5 ) < J > < J > 0 > ^ ^ * “ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ C M C M C M c o ® ® c o c o c o ® ® r - r - r - r - r - r - r - r - r - r - r - . r - r - r ,-CNjcNjCM r^r-r-r-r-r-r-r-r-r-r-r--fs -r-*r-r-r-r-r-<0<0c0<0<0<0<0<0<0<o<0<0<o®i0©i0 r v r - r - r - r - r - r - r - ® ® ® ® ® ® f f l ® ® ® ® ^ f ^ , ^ “ ^ , ^, ^ ’ ^ ^ ’ ^, ^ ' ^ ^ ' ^ ,», o o o ( O ( O 0 ( O ( O ( D ( O < O ® ® U ) ® ® ® ® U ) U ) U ) U ) ® ® Q ® O O ® ® ® ® Q f l O ® C O B N S S 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O O) o- o- ^y 0 ) 0 ) 0 ) ^ ^ y o> o> 0 0 0 0 0 0 0 0 0 0 ) 0 > 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 > 0 ) 0 ) 0 0 0 « - * - * - ^ » - » - » - » - » - © © © © © © © © © © © © © © * “ * - *- CO CO ® ® <0 <0 <0 c o ® a o C O co CO CO co * r * r C O o o O o o o ^ © © 0 0)0 o o o o o o O) o o o o o o o o o o o o O ) 0 0 ) 0 ) 0 ) 0 0 0 f - T « ^ o o o o o o © o * - * - * - * - * - * - * - i - * - t - * - o o o © o o o o o o o © o o i - - * - ® ® o o CM CNJ *y n CO CO r^ *~ K ' CO co csi csi ® ® ® ® ® 0) 0) 0) CM CM. CM O* ^ ^ CO C O co n * r -r^ co! co to 1 CM CM CM ® ® ® ® ® O O) CM CM C O C O cO \ C O CM' CM ® ® ® *- *- O CM CM CNJ CO CO w o w ® C O ® ® CM CM CM ® ® ® CM CM C O C O C O co C D C O ® ® cm' cm ® ® CM CM CM CM CO CO CO CO W ^ ^ w w w w ® CO CO ® ® © ® ® CM CM CM CM ® ® ® ® CM CM CO C O rt rt CO C O C O *co ' ® ® CM CM ® ® i“ N N CM ^ ^ CO CO CO n * CO' CO w ® O) O) <0 0 o CM cd cd ® ® ® n r** C O CO co co o o cd cd ® ® h . r - r - r - r - r - r - r - r - r - o o o ^ y ^ ^ y ^ y n ^ y ^ y ^ y ^ y ^ y r - r - r - w w o w w w w w w o w w o I W W W O W W O O W W O 0) 0) 0) 0) 0) 0) 0) 00 ) 0) ® ® ® O O O O O O O O O O C M C M C M c d c d c d c d c d c d c d c d c d c d c d c d c d ® ® ® ® ® ® ® ® ® ® ® ® ® Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 416 ^ t O n O ) 0 ) N ( 0 ( D i - 0 0 0 0 0 ) u i i A i n ^ i n t b c d ^ u i ^ ^ ^ t n i n t A i A ^ ^ in rv o a > ^ n i n c o N O O ) N O O ) ( O « r f l D l A ( O ( O ( 0 ( O C O n O n ( O S C D n N C O f l O ( O O C D I A ^ l A ( O C O ( O i n t ^ ^ ^ ^ ^ ^ u i w u i « i ^ ^ ^ ^ ^ O CD C*) C M in <o cm * - S : ” S S «“ C O in in in o o o in in C M O) C M 05 ® C O C O o> C O fv ^ ^ co c*) m cm in ^ c o ^ c o o o ^ cmco o ^ ^ ^ ^ c d c M C M i n ^ i n K r s i ; S 2 : ® 3 ^ *“ C M C O in o> o co o m co 05 1- CO co . ° o n J o o 0 5 ^ O * ” in ui C M mnc*)Tf^or!I®ir<o(0(Oc*)VP)inu)co*-S2 n c M C M C M ^ ^ n n w n c M o i M o j ^ ^ T * q CO ^ co ^ co O) in cm m co ^ in rv f v CO iv _ .. ^ co co ^ co m o d rv C M ^ n C M C M fv 05 CO C O v « » - i n N i n c o c v i » - o C D C M C M O > 0 > C M C O O in C M tO C M 05 © 05 O t ifi » - * - * * m co co co 05 »- r v rv v - in co co in ^ cm o f v in ^ co ^ ^ C J J fv r* . m co . w w w w< w< v n v< w « / V<| > • » . . 1 « . W W ' » V# . - > C O O 05 C O in ^ C 0 ^ 0 5 M ' 0 > c b c M C M C 0 C 0 C 0 C 0 i n i n * - T - * C M C 0 C 0 C M C 0 C M C 0 i n C D O c d 0 > C M 0 5 C M C 0 ^ C 0 C 0 coconww»“ »-invvirncMCMWCMinininiov^^^co<0(OCMi-CMinininin^^ ^^^in<£iNNCMtococovinco<oNCMCMC\ic\inconn^^in(OfvN»“ CMCMC\i{oco C O o O V 05 co C O C O 05 C M co co ^ in CO C O in C M 05* C O m co O C O a ) co fv o* in 0 5 «r <r co O 05 C M (0 N p ^ ' r in in in in co in co C M co rv cd in in m 0 5 m *- fv fv 05 co co co co *- co <o co cm in co N C D ® O C O co C O fv CM CM O O 0 5 0 5 C O C O o 0 5 CO fv co o n <r in * - cm ^ in C O C O C D 00 s to in O CO 7 in o ’ CO CQ 9) o C O cd 0 5 oo5CMrvinaooocomo5r-o5inco05Qocoo^-_ cos»-vcMO^O)Oin»-cotoco(ONt^inco ^'tinind^t^^NNoid^T-i-cfjT-w^ifjNdtod^cbid C 0 ^-C0 C 0 C 0 CM»-C0 0 5 CMOrv< 0 inCMCM (M lf lC O C O S lO C O C O C D flO O flO tO O i n i n i n ^ r c o c o c o c o c o c o c o i n ^ n ^ c o c o c o c o a o c o c o c o r v i n i n _ T- T- IT ID ^ ^ ^ CO CO O) CD in o 05 05 co 05 co co CMCMCMCMCMCMCM<O<O<O<O<O<0<OCO<OCM CMCMCMCMCMCMCMCOCOCOCOCOC0COCOCOO i f l i f i i f l i n i n i f i i f i i - ^ i ’ i ' t ^ i ’ i ’ ^ n i ' ^ CMC MCM CMC MC MCM CMCMCMCMCMCMCMCMCMCMCMCM C O O O O C O C O O O C O C O Q O C O C O Q O ■O ’ O* n 0“ n n ^ CM CM CM CM CM CM in in in in in in o o rv rv o o O O O O O O 5 O 5 O 5 O 5 O 5 O 5 O 5 O 5 O 5 C 0 C 0 C O C O C O C O C O C 0 C 0 C 0 C 0 C 0 C O C 0 C O C 0 C O C 0 C O C O i v r v r v r v r v < o c o c o c o c o c o c o c o c o N K r v r v r v . r ^ r v r v r v r v r v r ^ r v r v r v r v N r v r v r v ddddddddddd dddd dd dd do dd dd dd od ddod d OOOOOCMNCMCMW CMCMCMCM»*i-f»-^frpffr»-rr^^t^^t * - » - * - * - , - * - ^ - * - * - » - t- t- * “ t- 0 0 0 0 0 0 0 0 0 © 0 0 © 0 0 0 0 0 0 0 ^ ^ r* ^ r- O ) 0 5 0 5 O ) 0 5 0 5 t ^ ^ t t ^ ^ ^ ' f ^ t ^ n v ^ ^ ^ i n i n m i n i n i n 0 5 0 0 5 0 ) 0 5 0 5 0 ) 0 5 0 0 5 0 5 0 5 0 5 0 5 0 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 * - * - * - * - * - * - * - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o rv rv co co 00 00 CM CM cd cd m in o o o o rv t v r v t v CO CO CO CO W «r co co co co O O O O (D 00 CM CM CM CM CO CO CO CO in in in in o rv rv rv rv co co co CO CO CO CO ^ * co co co co © ^ r* r" W V cd cd cd cd m in in in rv rv fv rv rv rv co C D C O 00 C O 00 O O 0 5 C O C O C O C O C O C O C O W ^ ^ *r C O C O C O C O C O C O C O ^ ^ ^ ^ ^ 0 5 ^ ^ ^ C O C O C O C O C O C O C O m in m in in in in < 0 < D < 0 < O < 0 < O < O < O < O < 0 < O < D < O < O < 0 < O C O < D < O 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 0 0 0 0 0 c o c o c o n n c o c o c o c o n w c o c o n ^ ^ ^ ^ ^ c d c d c d c dc d c d cdcdcdcdcdcdcocococdcococo rv r^ rv rv rv rv in m m in id in 0505 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 05 05 wwwwowwrtrtcowonwrtoroww i n i n m i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 417 o i n i - o > i f i ^ i n o ^ o o w i f i o ^ o s ® n ® N i n o « o f f l N r U ) < - a ) ( O l O N i n ( D t ( O l O n f l O n O ) ( O M S ^ ( D ( D O ) ^ N O ) 0 0 0 0 ) M M ( O r t ( M « c o i n n N ( o o o o o o i c \ ) l A N l A C O ’- f ’ N O ( O ( O 0 ) ( O O I f l N O a \ j **/ i~» w »'i w * / wi w i« w r* vv v 'S^QD^^C'lOaitCNIfilOOIhOQ^ ai9AiA^tino.;(Of>’ Ood(bdndiAifliA(bf*ti' O ffllflO n ^ O ^ S n O N O O lO in e jrfiN O lO ^ N N O ^ n t A O O t O ) N N C V C D O O J t - O O ^ in c o c o < 0 o C N J <6 C O (O N S O O O W O ^ ^ (O Ifl S O >u «' r « fli ‘ ' ' » " » » U ) ^ C M C N | ( M C N J W » * ^ ^ W ( O W W W W W * * ^ C O C N J C N J co co 1 C N J N v-CNJO>C0CNJCNJCNJt n<O C\IC)rCDaO»-Oa5 C M i n < o m * - c N j c o i n i n rv ao ao in _ ... _ _. . (OS(*)CO^^O)OCDO)GOOlflOJlflO)0)CO ^CNJVffl(O^(X3CNJ»-O)NCNJlfli-<OD0ffl00 f in u) . in t - r * C O cd(ddTfuiojdojonin(£)(doidw ^ w w w n c M C N j w c N j i n ^ ^ n c N j i o w ( O N r n o i n o o c N j o i o ^ N r A cdcdco^r^rcdadi-dcMCM^i-T^CNjcdinrv.- t ^ ^ ^ o n c M C N j i n m i n m ^ ^ ^ n o c N i P j cn^TfTf^inininNWcnrt^ini-cMwncDnntr^tnNWOJCNjCMnrjn^^inm co C N J o co o ^ C N J O A A ui d n ^ ^ V a CNJ C O TT C O O i r- * t n * n in to to n in in in in in C N J rv in in co o in cd in in s + - *T Tf *■* rv m in in ao co cm ^ cd rv r-> in c n j rv c n j rv C O C O in n rv oo *- co co co co rv c n j in in r* s in d in n co co <o co O) co rv 00 C O ^ C O o rv ao co co O ) P) C O N f rv co ao co C M o o . ^ 0 ! 0 ' r o n w * - ^ i n « i f l A C O A r - N N C D A A i n n i O W W C M d w d c d s i n c o N d N » - w N c D n .................... — . __ _ lominininv^^ncNNin^jfAcococoioNinint i n c N j a o N r - m c o c N j c M C o ^ r v c o c M i n m r v » - c o o ) s c N j A c o t o ^ O i - o o o c o o ^ s i n v ^ ^ n co in s s a) r- tn cd co oi co ci d d in Is* d ^ v c o s c o c o o o c o c o c o i n i n ^ r o * CVJCMCNJCNJCNJCNJCMCMCNJCOCOCOCOCOaOCOOOaOaOCOCOOOaOOaOCNJCMCNjCNJCNJCNJCNJCNJCNJCNJCNJ CMCMCMCMCMCMCMCMCMCOCOCOCOCO^'T^rn'^T^J-^r^r^r^r^rCMCMCMCMCMCMCMCMCMCMCM u) in in in in in' in in in d d ^ d d d d d d d d d d d d id id in in in ui in ui in id ui COlOCOCOCOlOlOCOlO’- ’- r ^ r A A A A A A A A A A A S S S N N N S S S N N N S S N S S S S N f f l f f l a j B a S S S N N N N N N S N N N N N S N N N N S S 0 0 0 0 0 0 0 0 0 0 o o o o o o o o o o o o o o o o o o o o o o o o o o o o n ^ ^ tT r^ o fflB A O s s s s N S S N N N S in in m in in in in in in in in in in 05 O i inintninin inc oc oc oc oco ao co co aoo oa oao co ao aoo oC NJC MC MCv jc MC Nje NjC Njc Njc NjC M 0 0 5 0 3 0 5 0 5 0 5 0 0 0 0 0 0 0 0 0 O © O O O O O O O O O O O O O O O O ( 0 ( 0 ( 0 ( 0 < D c 0 ( 0 ( 0 t t 7 N r t t ^ ^ v ^ 9 n 7 9 ^ O O O O O O O O C M W C N J W C N J ^ ^ t V t V ^ ^ t ^ * * * * * * * ■ * * * * + * * * *r * n - * * COCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCO N N N N N S S N P f - r f r S S N N N N N N N N i n i n i n i n i n i n i n m N S N N N o c q a o a o c p e o o o c q c p a o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n m i n i n n o o o o ^ rv rv rv rv ^ ^ ^ ^ ^ n co co co co co rv oo ao ao oo ao o o o o n d ♦ d d m in in in m o o rv r^ co co coco o o o o rv rv «r ^ co co *'ob*od' © o W ^ in in o o o rv rv rv T ^ ^ ^ ^ ^ C O C O co C O <9 00 o o o in in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 418 . . . .-(0CM»-0)®0)in«ecMio^o*-^0)(Dfl0w,- , , , _ . . . . . . . . . fl)0)<ostoofliicino)^co^o)iflninnBWNiflcDOO)f-c'juin^oicD^nscD A ft M M M l M l M f«k Ik. <M M M M — _ M M <A M l M l M l M IM Ml <0 P* ^ a W N W r ^ I O r - m C O N i n c o o o i f - w i f l r t v a c o ^ (B CO CO 00 in <0 px o CM CO 00 CM © CM CM CO CM px o CM m O i CO CO o CO CM 0) 00 px CO N o o ao CO T— CM o o oo CM CO px CO CO o CM 00 CM in m o Px GO m CM CO CO 00 in in O i o ® < CO V- 00 K px Px CO px in in O i CD O CM P-. O i CM 00 00 00 CM CO T “ m p. © px O i CM p. CO 00 O i co ® CM CO CO CM CO px CO CM 00 CO CM 00 CM NT * r CO CM n CO CO CM Px P» CO ® O i * r CM CO px O i in Px O i o CM o in m O i CO in o CD CM in px CO O i O i CO O i CM CO CM CM m CO O i in px. CO 00 hx CO O i Px CO CM flO 00 O O’ n CO CO O i 00 CO O i in in 00 CO CM O i 00 o in CO CO o CO CO CO O in CO in 00 CO o CM a CM 00 px O CM m o CO O i O i CO in CO o CM CM O CO CM CO CO r* CO CO in CM px m in CO o © O CO •O' CM o CM CO in O i CO CO in CM o CO CO CO CO in CO Tf in CD CD 00 px CD O i CO CO CM CO CO Z CM CM CM CM CO CO CO CO CM CM CM CM CM in m V *T CO CO CO CM CM CM CM CM CM CM CM CM m m in o* ® p. p*. O) K n s oo i0(0(Qtf)v*n^^^io(0(0(0N(\ioj(0nnn^^^inu)u)inu)ui(0c0NMCMCMn CO CO C M CD in c o o > c o ® c v s o N CO 9) OT CO CO t- (0 *- px C S C D CO CO CS p. p-i c o c o O' in o o o c o o C O O 7 ^ (0 (O cvi ^ t in N ^ ^ tt px px px T ** in m in in c n j *- C N J in co on cd cd in in in o c o c o (0 Oi «n m co o C N J C N J o p .' © C O C O C O P » CD CO CO px O i in Px o CO O i CO r- in Px 00 co 00 cs p* px C D O i r* (Jl CO O) N »» C N J N O S S CO fl) O C O C O co px CO p* tj- i > r * Q U O J f l 0 C 0 ^ , ^ i n N w w ^ w w » ' y i ' ‘ i ' * r * W T ' i ' j i « i , v ^ c o n c o c s c i J i n i n i n v n c o ^ w N c o c o c o c o s i n i n i n ^ i - c o c N j i n o ^ o i n m n c N j f - c N j i n f N C M t - o i n o c N j c N j o i n c o o ) 0 o ® i - d d d h * N N p ^ c r J c o c N i O OJ O J O Oi nc O M CO CO Nt O ^ T t l O I O r N f f l N O C O O n t ^ in n i o n ui r ^ i r i co ^ t v t ^ c o c o c o o o o c o c o CNjcsjcvjcNjCNjcNjcNjCNjcNjCNjcNjcNjcNjcM^^r^^^^^^^^^^-n^^^^^r^r^r^f^ CNjcNjwcNjNNSSNNSNSNcocococJcococowconwnconconnncNicMCNicNj i f l i o i n m t ^ v ^ ^ ^ ^ ’t^i'oiowiflwiniflioioiotfiio ujduiiflioin^t px px p* px p. px t ^ ’r ’r ’rn^nn^inininininininininininininininininincocococo f f l ( J O ® C O C O C O ® C O C O O C O C O C O C O C O C O C O C O C O C O C O C O ® ( ® O O C O C D C O I s* N N N o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o i n i n i n i n i n i n i n m i n i n i n i n i n i n o o o o o o o o o o o o o o o o o o t - ^ - t - T - * - * » ^ - ^ © 0 © © © © 0 © © © * ~ ^ ^ ^ ^ ^ ^ * ix t - ^ ' ^ v ^ ^ ^ v ^ v- © © © 0 CNJ CNJ CNJ CNJ o o o o CNJ CNJ CNJ CNJ O) 0) 0> 0) CNJ CNJ CNJ 0) O ) O) CNJ CNJ CNJ O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O i O) N N 0)0 0 ) px px O i O i « - * - * - f - 0 © 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o o o N N N S n o; ^ o- ^ ^ ^ 1 “ O ' CO CO CO CO co* do' co' co o o o o ^ O ' ^ ^ i n m i n i n C O <0 C O C O O i O i 0) Oi ^ ^ ^ ^ * T C O C O C O C O O i O i O i O i CNJ CNJ CNJ CNJ i n i n m i n (0 C O C O 0 ) 0 ) 0 ) ^ W ^ c o c o c o 0) 0) 0) CNJ CNJ CNJ O' i n i n i n c o c o c o 0) 0) 0) ^ ^ O ' CO CO CO O i O i O i CNJ CNJ CNJ CNJ CNJ CNI i n i n m r t r t CO CO CO O i O i O i ' T ' T ' T * * i n i n i n CNJ CNJ CNJ CNJ CNJ CNJ i n i n i n i n m i n n n C O C O C O C O C O C O 0) 0)0 0) 0 ) 0 ) ^ ^ ^ ^ ^ ^ ^ * O' "O' ^ i n i n i n m i n in CNJ CNJ CM CNJ i n i n i n i n *r ^ O ' W c o c o c o c o 0 0)0 ^ ^ ^ v ^ ^ v i n m i n i n CM CM i n i n O' CO CO o o’ o* * T * m i n CM CM i n i n i- < o < o cm ^ ^ i n i n i n C O C O C O C O C O o o m i n o o o ^ ( 0 ( 0 ^ m i n i n <0 (0 ^ NT in in ^ o * c o c o o' o <0 (0 in m Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 419 u © rv 00 o o fv o C O o © ^ r O i o C M rv 00 o rv rv rv O) C O in C O m ^ r o © rv 9 rv in © © © © © © © © © © © © IV © ^ r © © © © © © C M © s © O n in in in in m in m <r in ^ r © © © © © r— * - * - *“ * - * - *“ *“ *“ « O C O iv C O V © in C O m o C M G O o C O C O C M C M C O C O C O O i O C O C O m rv in 9 9 in in C O © o © © © o © © o o r«* rv © © C M © C M © © C M © © © © © © < © co C M C M C O O in n O i in oo o C O *• in m m rv O i C O rv O i in co O i C O o 0 01 O i in r t C M m C O o C O C O O i C O C O co C M in C O C O (0 C O C M 05 0 ) in o i C O o © © C M © o C O C M C M © C M © © © C M © © © © © © © © © © o y — © © 05 © © rv © rv ^ r © rv © © © ^ r © C M © © © C M O © C M o © £ a C O C O O i C O 00 00 O i C O o> O i r r o C O rv O i co C O C O © C O o C O O i C O C M rv 00 C O rv © O C O © C O o o © o o C M O o © © © C M © * • © © © © © C M O © © o © © I TT C O C O C M C O C O C O in C O in C O oo C M rv C M C O C M rv C M C O C M O i C M co 05 n o in C M in C M in Tf in in rv in C M C M © © C M rv C M © C M © C M © C M © C M © © © © C M C M © a r M a le | (A C C O in in in © in m C M C M C M C M C M C M C M C O C O © © © © © © © © C M © © © © C O < in o < C M *T C O C O in C O 00 O i C O co C O rv * r © © © © © C M r - ^ r < n n m in n fv © © •*• © •< T © •*r < *5 C O O i O i C M C O •O ' O i rv C M < C M in C M in C O in in m C O in C M © 00 C O © C O fv © rv © © rv © © © © © © < rv C O © © © © © © © © © rv © rv © rv © 5 © © < in © 00 O i C O C O rv m o in in © < O i rv o oo C M C O m 9 rv 9 rv 9 © © 5 < < 9 8 . 9 3 9 5 . 8 2 E 3 o 00 oo O i O i C M C O •o* O i rv C M C O C O in C O 00 O i C O C O C O rv C O O i 00 O i Tf C O C O rv in o in in © © C O © C O rv © rv © © © © 9 © C M * r o C M © © © fv © © © © © © O) c r - C M in C M in C O in m in C O in Tf O* in <r in ^r rv 9 O i O i rv o C O C M 00 m 9 rv 9 rv 9 © O i C O © © © © © © © © ■ « r © © 9 C M © © © © © © rv © rv © rv © © — J i C M rv C M rv C M rv C M rv C M rv C M rv C M rv C M rv C M rv C M rv C M fv C M O i T “ Oi O i *■ * 05 O i O i O i O i © © © © © © © O i O i © © © © © © © © © © © © © O* O ’ n •o* ^r rt C O © © © © © I o 00 rv 00 rv 00 rv 00 rv 00 rv oo n 00 rv 00 rv 00 rv 00 fv C O fv 9 rv o> rv O i rv 05 rv 05 rv O i rv O i rv O i rv © rv © rv © rv © rv © rv © rv © rv © rv © rv O i rv © rv © rv © rv © rv © rv © rv © rv 5^ o o o o o o o o o o O o o o o o o o o o o o o o o o o o o o o o o o o o "o 00 r “ O o o o o o o o o T ~ o o o in in in in in in in © © © © © © © © © © © © o © o © o © o © o © o oO » “ * - *“ *- »- *“ *** * - *“ T— * - * - i - * - *- * - *- *- * - * - * - O C O |v © K O i rv O i rv O i rv O i rv O i rv O i rv O i rv O i rv 0 ) rv O i rv o> C M O C M O C M o C M O C M O C M O C M o C M O C M o C M O C M O C M O C M O C M O C M O C M O C M o C M O C M © C M © © © C M © © © © © CO o © o © o o © o o o © o V “ V* * - • - * - * - * - * - * - o O o o o o © o <o in C O in C O in C O n in C O n in C O in C O in C O in co in C O * t © C O in C O ^r in rv m rv in fv in rv m rv in rv m rv in rv © rv © T• rv © rv © » *• rv © rv © © rv © fv © rv © |v © C O © © © © © © © © © © © © © © © © © < C O C O Tf C O C O C O C O C O C O C O C O C O C O ^ r C O C O C O C O •^r C O C O C O C O C O * C O © © © C O ^r © © C O C O C O © © © © ^ r © ft © (O O i C O O i C O O i C O O i C O O i O i co co O i C O © C O © C O 05 C O O i C O O i O i C O 00 05 C O 05 oo O i 9 O i 9 O i 9 © 9 O i © O i 9 © © © © © © O i 9 © © © © O i 9 O i 9 rv o rv o rv O ' rv o rv o rv o 7 in in in m in in in Tf in in in in *T in in * • m m m in in in © < * ■ © Tf © © © © n © © © * r © * © © © © © © © © © © © © © r« T “ T _ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 420 o in rv o C O © o C M C M v - o 00 C O in O i y C O in O) f". C M C M C O 0) C O C M C O rv O i o 00 O) q q q q q o q V C O q C O in O i q C M q o q 00 q h* r*. q q O) q C O y q q U d d N T in in in in in in in in d O' d d O' d d in in in in d * r in in cd T “ y * - y y y T - * * y co rv O i rv C O C O o o rv C M ao C O C M m o o h* C O C M C O o . in < 0 q 00 in rv o; rv C M C O C O C O o o © C O oo q rv o rv m O i in o m r^. C O O i O i o y - q O) C M *— o q q s cd cd od o C M o i rv od cd cvi cvi cd cvi 0) o cd c m ' cd d O i cd in cd d o o in cvi rv in 00 C O C M C O y |A C O C M m o O i y 0) o C O 00 o C O 00 C O C M 00 >A O) C O J{J 00 C O co m Tf o o o o ID *2 o C M in in C O C O in m C O in 00 m 00 00 C M C O C O o r - 10 o C O C O C M C M y O i * - y - y - m C O C M C M C M C M C M O i »■ C O C O C O C O C M C M C M C O 0) *“ m C O 00 co o in in * r rv in 0) C M V C O in O) C O O i C O O i in N. in y “ O i C M o C O N. O i co o 00 O) O i * - C M y - O i o q in o V oo y C O C O q O i C M C O q m ; q N. q q O i O) V h* C M C O C O © d d cd N. o> cd o i o i d o cvi yS d d in d d in in cd o cd d cd r — cvi 0) cd O) cd y cd o X C O C M C M C M C M C M C M C M C M n C O C O C O C M C O C M in m in n C O C M C M C M C M co in co co i n i n i o i o i f l ( O N N n n n ( 0 « ^ t ^ i f l i f l « w w N n o n n ^ i o i n i f l S f ( M Tt »- o in co ^ r co co in co C O C O r - o 0 0 < d co in ^ ^ o ) co C O *“ * - C M c m m in cd I s* ^ ^ ^ ^ r** co rv c o * cd T f 0 0 *- ^ c o C O co oo ^ ^ ^ C M C O cd in N t t O *T C i * t n in cd in in m o o in C M y ~ 00 C O © rv rv in cm * • co < 0 rv o in co ^ in o i d o i cm co <o co C O l*v o O i m rv O) C O i - oo cd Is* * - cd (O C O N N 0> C D ^ o in co cd rv CO O CO o C M C M in 00 C O C O 0 0 ^ cd c v i in in cd h - ^ O ) C O C O ^ r C M S O o rv rv co C O C O t f ) W i - c D ( O N n c o s N0l0(0^n0)«(0 id d O) o i cvi d id cb cd ( OC Ot Ot ON i n i A i A t cm co ^ O) in h »■ (O o co ^©oconoomrv N C O O i n C O O C O f Q cd d n d n cd n f cd cd n td cd ^ C O C O C O C O t O t f l N N l f i ^ ^ ^ (0C0C0C0(0C0C0<0<0C0(00)0>0)0)O0)a>(7)0)0)0)CMCMCMCMCMCMCMCMCMCMCMCMC0C0 m in in in in in in in in in i n in in i n in i n in m in i n in in in in i n in in in in i n in in in in i n in cdcdcdcdcdcdcdcdcdcdcdcMCMCMCMCMCMCMCMCMCMcvicdcdcdcdcdcdcdcdcdcdcdcdcdcd inininininmininininincocococo<ocococ0<ococoo)0)0>o>o>o>o>o>o>o>o>o>0)0) N N S S S N S f ' S N N C D C O f l O C D C O O O C O t a C D C D C O N N N N N N N N N N N S t C © 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 i n i n i n i n i o i n i n i n i n mi n i o c o c o c o i o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o i o c o c o r v r v o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o CM CM OO CD CM CM CO CO CM CM « 0 0 C M C M CO 00 CM CM CO GO CM CO O) rv rv r*. 0 0)0 rv rv. rv r» rv rv rv o o o o o o o C O C O CO CO CO CO co e co co CO CO C O co CO CO CO CO CO « CO CM CM co o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o CO CO o o in in CO co K* rv © o in in in in CO CO O i O i in in d * CO CO rv * fv ‘ o o in in in m c o c o o> O i m m : d d co co © o in in in in co co O i a m in ■ d d CO CO fv ' rv o o in in in m co co O i O i in in co o O i CM in q d d co co rv h - o o in in in in in in in in o o o CM . CM CM to to <o d d d CO CO CO 0)0)0) CM CM CM in in in in in m © o o o o o o m i n i n i n i n i n i n i n m i n i n i n r v r v C M C M C M C M W W C M ^ ^ ^ ^ ^ ^ ^ ^ V ^ ^ ^ t O O ( O ( O ( O ( 0 ( O < 0 ( D ( O ( O ( O ( O t O t O ( O ( O < D ( O ( O ( O ( l ) ( D d d d d d d d d d d d d d d d d d d d d d c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o O i O i O i O i O i O i O i CM CM CM CM CM CM CM i n i n i n i n i n i n i n i n i n i n i n m i n m i n i n O i O i O) O) O) in in in in m in in in O i O i * * in in in in O i O i in in m in in in in m O i N N ^ <0 co in in in in in in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 421 o * - > n o c M ^ s a s N S c o i n f - T - r v i n r v o o * - o ^r o oo ® o rv *- a > * O i O 00 O i O < 0 » - rv * - in CO CD CM *- CO CO O i 1- cst h* co n* ^ O W Ifl ffl o id s cn co o t- o »- oj o r** f"** C O C O ^ 00 co oo tfl N f 0 0 0 0 ID S S S S S ^ S S S S S S s ; ® S3-RgSS2.SSS!SSg!822as2SSSii-55«!5S.SSSg ^ (O O i ^ CO 0 ) 0 ( 0 ^ 00 C D O C D O C O ^ C O s o S C O C M O W I O O C O ^ ^ f J j S J COCO^TCMCMCMCM^* " ^* ^1 . - . - ^ i di o^ w o . C VJ C MW^ ^ ^ rv rv in O CO (O S CO ^ ID ID in CO CO CM CM CM rv ID 00 ^ fti o 0) © in ” oj O) oi * S ID ^ N ® C O C O C O O i O i O O ^ O O i O i CM K) »• t O © B CMO' COoj cM^CDCMCMCOO^r COO' I D I O I D C O ^ ^ T ^ C O C O C O C O C M C M C M w i w ^ O B l f l A B N n ^ CO*“ C M C O lD C O lD C M C D C O » - c o o ^ - c D c D ^ r i n c D o - o - ^ O J c o c o a jc o « « v « w w . c o o * - c o c M C M i n c q c o e o c o ^ r c o » - c o * - i-f-nninri©«^*-wco»“^coco^^w»-o in CM CM CM C M C M C M C O C O C O C O ^ f ^ ' ^ l D C O C O C D C O N . f - C O C O C O ^ O ' l D C O f ^ v - ^ ' C M C M C O C O C O C O C M C M C M CM CD O r- CO CO Tf CO B r O) in rv t- o o s’ aj a ^ CO CO CO in rv O) co CM o O' CD O' O i •n * CM CO o ID C O CO CO ® co < CM CO ID ID ID in in ID ID O i ® in CO T_ ® fv CM ID ® rv n co CO O CO ® < CM ID fv < r rv o ® ® O i ® (O < o fv ® rv rv CO ® ® © 0 ID 01 CM r- CM O id rv O D 00 ID *- © rv id in 00 oo CO O C M ^ CO C M CM CM ID 00 00 00 5S S i CM t- CM O ID O i 00 i- © © in rv in cm in rv oo 00 O) <o <o co t CM CO CO CO CO CM CO Tf id in in » - ® * - o i n c M c o * - c O T “ ©i n o >i n a o *- * !2 in * -c o s c n n n t ^ N t - © o o o ^ f f l s c o c i r - ^ ^ s o ’- n i n ^ c i v c o t C O C O C O ^ C O W B N N l D i n ^ D C O O ) o rv o n ^ t n o CM 00 ® CM n © CM i n m c o c o o o c M C M i n c o c o ® © © n ® ® ® c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o * j , o, ^ ' ^ f ^ * J , * T ^ 0 ) O ) 0 ) O ) 0 ) 0 ) 0 ) 0 )rvrs.fN. i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n i n © © © © © © © © i n i n i n i n t n i n i n i n c M C M C M O p j c ' i p j c i c O l O W n W W W C O O W O r t ^ r r r r r r r r p r r r r r ^ r r r O i O i ® ® rv rv o o CM CM O i O i a ] 0 ) O ) O l B 0 ) B 0 ) O 0 ) 0 ) 0 ) a ) B 0 ) B B B B B ( 0 © © n ® ® ® ® ® ® ® ® ® ® ® ® ® ® ® ® c o ® c o < o c o ® c o r v d d d d d d d d d d d d d d o d d d d d d d d d N S S S h N N N N N N N N N N O I O J O > ® O l O I O ) 0 > i n o o o o o o o o o o o o o o o o o o o o o o o * - CM CM O i O i CM CM O i O i CM CM O i O i CM CM CM 0 ) 0 )0 CM CM O i O i CM CM O O CM CM O i O i O O rv rv o o o o o o n n n rv rv o o o o o o o o o o o o o o o o o o o o o o o o o COCOCOCOCOCOCOCMCMCM f v N N N N S N © © © 0 0 0 0 0 0 0 0 0 0 i n i n i n i n i n i n i n r v f v N «— ^ • ^^ • ^ ^^ O O C O C O ^ *“ T- *• T* (O (O (O 0 ) 0 ) 0 ) 0 ) 0 0 0 0 0 0 0 0 0 0 0 0 0 ^ ^ ^ N IV <0 C O <0 C O O' ^ co co n rv CO ® ID ID in m n rv CO (O co q C O C O rv' rv* co ® ID ID in in IV N co < o C O C O rv rs . co q ID ID in in |V IV (0 ® co ® CO CO ■ r v ' r v * C O q ID ID in in rv rv rv co ® ID co ® to W ^ oo o y co fv rv * rv co ® ® in in id in in in rv rv co < o q q c o c o rv * rv* co ® ID ID in in rv rv C O < o C O ® V n C O co rv * r v ’ co ® ID ID in in N N CO ® q q co c o rv' rv q q ID in in w N |v n rv <0 CO C O C O ID ID fv rv fv rv rv rv ^ ^ ^ ^ ^ r^ rs. rv fs. fs» fs. W n n ^ C O C O C O C O C O C O <0 CO <0 ® ® ® in m in in m in ® c o rv co ‘ CD C M <d ID <0 C O C O © C O C O C O C O rv rv rv rv ^ ^ CO CO CO C O CO « ® ® C M C M C M C M co cd cd cd in in in in ® <o ® rv rv rv co co co in in in rv rv rv rv rv rv ^ v ^ C O O O C O C O C O C O ® ® ® fv fv fv C V J C M C M ^ ^ ^ co ® ® ® ® cd in in in in in in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 422 2 m in T * a » o O O i in a> fv co T “ n C M O i o C O G O O i fv C O C O co o * in * m m * t - ^ r ^ r T “ fv C M in rv 00 in C M rv O i C O fv rv C M C O <0 C O rv rv G O o 0 01 o C M C O C O C M C O in 2 < C O C N J 00 co 00 h- O C O C O C M C M in C M fv rv C O r “ C O rv o cd C O fv o fv rv C O C M m C M C M m in C M C O o in C M C M o in C M C O in C O 0) in T“ in O i o C M £ O J 00 C M O) C O m C M O O D O i o> rv C D C O C M C O C M C M in o o in C O 00 co co 00 o C O 9 I co in m C M C O V m C O C O C M C O C M C O in + C O O’ C O in C O C M C O rv in Mate I In s ta r i C N J C M C O C O in C O C M C O C O C O C O ^ r - C M <0 < in o < o C O O i < in co C M O i rv •*r in < C O in m in in C N J C O O i C O in C M * C M K C O C M o < Tf C O 00 C O in C O C O C O C O C O O i C O | fv o < O i T- 00 C O in o in C M < in 00 in C D 0 01 | < < 9 0 1 0 1 | 00 CO fv O CO O i CO in CM O i CO O i o ^r in o CM CM n CO CM o fv in ^r CO o in CM O) in CO o> CO 00 CO CO in in o in CO m co C O CO ® CO O i CO «r in in in o 0 01 <D <8 0s h- CM fv CM rv CM fv CM h- CM fv CM rv CM CM C M C M CM CM T* CM r» CM CO o CO O T- T “ *~ I n CM CO CM CO CM CO CM CO CM CO CM CO CM CO CM CO CM CO C M CO CM CO CM CO CM CO CM CO in CO m CO O O o o o o O o o o o O o O o o o CO fv CO IV CO rv CO h » CO fv CO rv CO fv CO CO CO CO C O CO C O r** CO - - (C *- *- T~ *- *- *- *- V“ r- *- O n CO o CO o CO o CO o co o CO o CO o o > O i O i O i t“ O i 0> T“ O i CO o CO o (O *- *- T“* *- o O O o o O o »- ® o fv in rv in rv rv m fv fv in fv fv m fv fv in N fv in fv CO rv rv CO rv rv CO n rv CO rv N co rv rv CO rv fv CO rv rv CO GO CO GO < CO CO CO CO CO CO CO CO CO C O CO CO *• CO CO Tf CO CO LL rv rv rv fv rv rv rv co co * CO n CO CO CO CO o o 1 CO in CO in CO in CO in T“ CO in CO in T“ CO in CO in co m CO in CO in CO in CO in CO in rv in rv in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 423 Appendix 6. Algulhasina Morphometries Macro Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 424 This is a ALI program for transfering data collected within Optimas 5.5. to specific spreadsheet cell in Excel 5.0. The program consists of a series of nested loops that prompt the user to collect certain measurement for a given specimen. After all measurements have been collected, the user is prompted to start the measurement cycle again with a new specimen at which time the program steps down one Excel row and reinitiates the measurement cycle. If (Prompt ("Initiate new connection to Microsoft Excel?",0x1002)) xobject xobjXL = RegisterXObjectfExcel.Application"); If (Prompt ("Open existing Excel file?",0x1002)) xobjXL.Workbooks.Open("C:\\WINDOWS\\DESKTOP\\SAS\\AlgulhasinaData"); Char SEMLoc; CalName; FileName; SetExport (PtPoints, 1, True); SetExport (ArSampledPoints, 1, True); SetExport (ArArea, 1, True); SetExport (ArCentroid, 1, True); SetExport (ArCircularity, 1, True); ExcelRow = Prompt ("Enter row number to begin session:", "Real"); Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 425 Integer SaveFlag = 1; Do { If (SaveFlag = 2) { ClearScreen(False); Scan(True,,True); } Else { ClearScreen(True); Scan(True,,False); } If (Prompt ("Extract 4 primary dimension X,Y coordinates followed by outline data collected clockwise from anteriodorsal landmark?\rNote: Capture order: anteriodorsal, ventral, anterior, posterior", 2)) { CreatePoints (); Extract (); Lan_PtPoints = PtPoints [0,] : PtPoints [1,] : PtPoints [2,] : PtPoints [3,]; xobjXL.Range(xobjXL.Cells(ExcelRow,8),xobjXL.Cells(ExcelRow, 15)).value = Lan_PtPoints; Create Area (); Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 426 Extract (); xobjXL.Range(xobjXL.Cells(ExcelRow,16),xobjXL.Cells(ExcelRow, 16)). value = ArArea; xobjXL.Range(xobjXL.Cells(ExcelRow,17),xobjXL.Cells(ExcelRow, 17)). value = ArCircularity; xobjXL.Range(xobjXL.Cells(ExcelRow,18),xobjXL.Cells(ExcelRow,19)).value = ArCentroid; xobjXL.Range(xobjXL.Cells(ExcelRow,20),xobjXL.Cells(ExcelRow+1,83)). value = ArSampledPoints; } ExcelRow++; ExcelRow++; } While (Prompt ("Analyze another specimen?",2)); Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 7. Taxonomic Plates Eocene-Oligocene - Sites 744A and 689B LPTM - Sites 690B and 738C Scale length for all specimens is 10 pm Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 428 Plate 1 Eocene-Oligocene 1. Actinocythereis 2. Algulhasina quadrata 3. Aversovalva 4. Bairdia 5. Bairdoppilata (inner valve) 6. Bosquetina Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 430 Plate 2 Eocene-Oligocene 1. Cytherella spp. 2. Cytherelloidea spp. 3. Cytheropteron 4. Dutoitella 5. Henryhowella asperima 6. Henryhowella philofelicula Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 432 Plate 3 Eocene-Oligocene 1. Indeter. Gen. B 2. Indet. Gen. A 3. Indet. Gen. F 4. Indet Gen. C 5. Indet. Gen. D 6. Indet. Gen. E Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 434 Plate 4 Eocene-Oligocene 1. Pelecocythere foramen 2. Penny el la 3. Bradley a dictyon 4. Bradleya sp. 5. Argilloecia 6. Eucythere Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 436 Plate 5 LPTM 1. Aversovalva 2. Bairdoppilata 3. Cytheropteron spp. 4. Dutoitella (corroded post-LPTM) 5. Dutoitella (robust pre-LPTM) 6. Eucythere Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 438 Plate 6 LPTM 1. Eucytherura sp. A 2. Eucytherura sp. B 3. Henryhowella 4. Indeterminate Genus A 5. Indeterminate Genus B 6. Mayburya Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 440 Plate 7 LPTM 1. Munseyella sp. A 2. Pelecocythere 3. Pennyel la sp. A 4. Pennyella sp. B 5. Philoneptunus 6. Profundocythere Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 442 Plate 8 LPTM 1. Rimacytheropteron 2. Propontocypris sp. A 3. Algulhasina 4. Propontocypris sp. B 5. Bairdia 6. Cytherella spp. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 444 Plate 9 LPTM 1. Cytherelloidea 2. Pennyella sp. B 3. Indeterminate Genus C 4. Palmoconcha Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Early Jurassic reef eclipse: Paleoecology and sclerochronology of the "Lithiotis" facies bivalves
PDF
The Triassic organic-rich flat clam biotope: A synthesis of paleoecological and climate modeling analyses
PDF
Evolutionary paleoecology and taphonomy of the earliest animals: Evidence from the Neoproterozoic and Cambrian of southwest China
PDF
Structure and origin of echinoid beds, unique biogenic deposits in the stratigraphic record
PDF
Oxygen isotope constraints on the petrogenesis of the Sybille intrusion of the Proterozoic Laramie Anorthosite Complex
PDF
Large epifaunal bivalves from Mesozoic buildups of western North America
PDF
Systematic analysis of crustal anisotropy and attenuation using seismic data associated with the 1999 Chi-Chi, Taiwan, earthquake
PDF
The structure and development of Middle and Late Triassic benthic assemblages
PDF
Paleoecology and depositional paleoenvironments of Pleistocene nearshore deposits, Las Animas, Baja California Sur, Mexico
PDF
Investigating the influence of atmospheric changes on the variability of the North Pacific using paleoproxy data and a fully coupled GCM
PDF
Integrated geochemical and hydrodynamic modeling of San Diego Bay, California
PDF
Investigation of arc processes: Relationships among deformation, magmatism, mountain building, and the role of crustal anisotropy in the evolution of the Peninsular Ranges Batholith, Baja California
PDF
Restriction of a late neoproterozoic biotope: Ediacaran faunas, microbial structures, and trace fossils from the proterozoic-phanerozoic transition, Great Basin, United States of America
PDF
Late Pleistocene ventilation history of the North Pacific as seen through delta carbon-13 records of the Southern California borderland
PDF
Tectonic evolution of a portion of the southwestern Panamint Mountains, Death Valley region, California
PDF
Paleoseismology of the North Anatolian fault, Turkey: Towards a long-term record of seismic moment release in time and space
PDF
The origin of enigmatic sedimentary structures in the Neoproterozoic Noonday dolomite, Death Valley, California: A paleoenvironmental, petrographic, and geochemical investigation
PDF
The unusual sedimentary rock record of the Early Triassic: Anachronistic facies in the western United States and southern Turkey
PDF
Groundwater modeling and geochemical tracer (CFC-12 and tritium) distribution in the Abalone Cove landslide, Palos Verdes, California
PDF
Contact aureole rheology: New constraints from fieldwork in selected Cordilleran aureoles and from numerical modeling
Asset Metadata
Creator
Schellenberg, Stephen Allen
(author)
Core Title
Response of deep-ocean ostracodes to climate extrema of the Paleogene: Ecological, morphological, and geochemical data from the Eocene -Oligocene transition and late Paleocene thermal maximum
Degree
Doctor of Philosophy
Degree Program
Earth Sciences
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
geochemistry,OAI-PMH Harvest,paleoecology,paleontology
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-179919
Unique identifier
UC11339329
Identifier
3054898.pdf (filename),usctheses-c16-179919 (legacy record id)
Legacy Identifier
3054898.pdf
Dmrecord
179919
Document Type
Dissertation
Rights
Schellenberg, Stephen Allen
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
geochemistry
paleoecology
paleontology