University of Southern California Dissertations and Theses

Evolution & ecology of Mesozoic birds: a case study of the derived Hesperornithiformes and the use of morphometric data in quantifying avian paleoecology

(USC Thesis Other)

Evolution & ecology of Mesozoic birds: a case study of the derived Hesperornithiformes and the use of morphometric data in quantifying avian paleoecology

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Evolution & Ecology of Mesozoic Birds:
a case study of the derived Hesperornithiformes and the use of
morphometric data in quantifying avian paleoecology

Alyssa K.A. Bell

A Dissertation Presented for the Doctoral Degree

Department of Earth & Planetary Sciences
University of Southern California
&
The Dinosaur Institute
The Natural History Museum of Los Angeles Co.

Advisement committee:
Dr. Luis M. Chiappe, Dr. David J. Bottjer,
Dr. Frank Corsetti, and Dr. Jill McNitt-Gray
i

Table of Contents 1
2
Institution Abbreviations……………………………....………………………………p. ii 3
4
Abstract…………………………………………………………………………………...p. iii 5
6
Part I: The Hesperornithiformes 7
Chapter 1. Overview of the Hesperornithiformes……………………….………….p. 1 8
Chapter 2. The Hesperornithiformes: a cladistic analysis………………………….p. 99 9
Chapter 3. The Hesperornithiformes: a morphometric analysis……………………p. 204 10
Chapter 4. The Hesperornithiformes: a taxonomic revision ………………………p. 320 11
12
Part II: Quantitative Paleoecology of Mesozoic Birds 13
Chapter 5. Description and ecologic analysis of a Late Cretaceous bird from 14
the Gobi Desert (Mongolia).………………………..…………………p. 348 15
Chapter 6. Statistical approach for inferring ecology of Mesozoic birds ..……… p. 375 16
17
Acknowledgements…………………………………….……………………………..p. 408 18
References ……………………………………………….……………………………..p. 409 19
ii
INSTITUTION ABBREVIATIONS 1
AMNH – American Museum of Natural History, New York, USA 2
BGS – British Geological Survey, Kent, England 3
BMNH - British Museum of Natural History, London, England 4
CAGS – Institute of Geology, Chinese Academy of Geological Sciences, Beijing, China 5
CFDC – Canadian Fossil Discovery Center, Canada 6
DNHM – Dalian Natural History Museum, Dalian, China 7
FHSM – Fort Hays Sternberg Museum, Kansas, USA 8
FMNH – Field Museum of Natural History, Chicago, USA 9
GMV – National Geological Museum of China, Beijing, China 10
HKScM – Hong Kong Science Museum, China 11
IVPP – Institute of Vertebrate Paleontology and Paleoanthropology, Beijing, China 12
JM – Jura Museum, Eichstatt, Germany 13
KUVP –University of Kansas Museum of Paleontology, Kansas, USA 14
LACM – Natural History Museum of Los Angeles County, USA 15
LPM – Liaoning Provincial Museum of Paleontology, Liaoning, China 16
MB - Museum fur Naturkunde Berlin, Berlin, Germany 17
NIGP – Nanjing Institute of Geology and Paleontology, Chinese Academy of Sciences, Nanjing 18
NMC – National Museum of Canada; China 19
RSM – Royal Saskatchewan Museum, Canada 20
SMC – Sedgewick Museum of Geology, Cambridge University, England 21
UCMP – University of California State Museum of Paleontology, California, USA 22
UNSM – University of Nebraska State Museum, Nebraska, USA 23
USNM – United States National Museum, Washington, D.C., USA 24
YORYMG – York Museum of Geology, England 25
YPM – Yale Peabody Museum, Connecticut, USA 26
ZIN – Zoological Institute of Russia, Russia 27
28
iii

ABSTRACT
The discovery of abundant Mesozoic avian fossils, beginning with Archaeopteryx in the
1800’s and increasing dramatically since the 1990’s with the discovery of numerous Chinese
fossils, has provided researchers with sufficient quantities of specimens to study the evolution
and ecology of ancient birds using phylogenetic and morphometric methods. This study
approaches the evolution and ecology of Mesozoic birds from two perspectives – an in-depth
comprehensive analysis of the Hesperornithiformes, a highly specialized group of diving birds,
and a series of morphometric analyses of modern and Mesozoic birds designed to find
correlations between ecologic niche partitioning and morphometric trends.
Despite being one of the most taxonomic, geographic, and stratigraphically diverse
groups of Mesozoic birds, the Hesperornithiformes have received virtually no comprehensive
study since the initial discovery of Hesperornis, Baptornis, and Enaliornis in the late 1800s. This
lack of study has resulted in a confusing array of taxa organized into a taxonomic framework
beset by errors in description, contradictions, and redundancy. Furthermore, little work has
focused on evolutionary relationships among hesperornithiforms, leading to virtually no
understanding of their phylogenetic interrelationships. While hesperornithiforms have an
extensive fossil record that would be appropriate for morphometric analysis, none have been
performed to date, despite the use of numerous morphological features that could be described
quantitatively, but instead are treated qualitatively as diagnostic features in the current taxonomic
framework. Therefore, the objectives of the first portion of this dissertation are to evaluate and
update the current taxonomic framework of hesperornithiform birds (Chapter 1), conduct the first
cladistic analysis of the Hesperornithiformes (Chapter 2), identify morphometric trends that may
have diagnostic utility (Chapter 3), and integrate these studies to develop a new taxonomic
iv

framework informed by the phylogenetic relationships and morphometric patterns identified
(Chapter 4).
Among modern birds, morphometric data has been used in a variety of ways in an
attempt to correlate ecology with morphology. This study seeks to build on previous work first
through the analysis of a new Late Cretaceous ornithuromorphs, Hollanda luceria, from
Mongolia (Chapter 5), and then through a broader analysis of a wide variety of Mesozoic birds
(Chapter 6). The goals of these studies are to first test the correlation of ecologic niches with
fore- and hind- limb measurements using multivariate statistics, and then to analyze Mesozoic
birds in relation to the modern avian morphospace.

1

Chapter 1. Overview of the Hesperornithiformes 1
INTRODUCTION 2
Our understanding of extinct primitive birds and their evolution is limited by their sparse 3
fossil record. While numerous Cretaceous birds are known from a single locality or time period, 4
very few clades are known from a wide geographic and stratigraphic range. One of these groups, 5
the Hesperornithiformes, were flightless aquatic birds first discovered in the 19
th
century from 6
Late Cretaceous deposits in Kansas (USA). Numerous specimens have since been documented 7
around the globe from sites dated 113 to 66 million years old. Despite this abundance of fossil 8
material, little rigorous research has been undertaken to investigate the evolution and 9
interrelationships of hesperornithiforms. Therefore, the objective of the first half of this 10
dissertation is to review and revise the taxonomy, evaluate the evolutionary relationships among 11
species, and investigate the use of morphometric data in describing hesperornithiform diversity. 12
The results of these analyses can be used to better understand the stratigraphic and geographic 13
distribution of hesperornithiforms as well as the development of their unique diving morphology. 14
This project is significant for a number of reasons. Hesperornithiforms were the first 15
birds to develop specializations for a fully aquatic lifestyle, and therefore are an excellent case 16
study for avian adaptation in the Mesozoic. Furthermore, the abundance of fossil material across 17
the globe enables the investigation of broad scale divergence patterns not available for many 18
other Mesozoic bird groups. While several researchers have worked on either North American 19
(Marsh, 1880; Martin and Tate, 1976; Tokaryk et al., 1997; Martin and Lim, 2002) or Eurasian 20
(Seeley, 1876; Nessov and Yarkov, 1993; Rees and Lindegren, 2005; Dyke et al., 2006) 21
hesperornithiform fossils, very few combined analyses of known fossil material from across the 22
globe have been published. Finally, their morphological diversity makes these fossil birds an 23
2

excellent group for evaluating different sources of phylogenetic information as well as different 24
phylogenetic methods. This will improve our understanding of the phylogenetic utility of fossil 25
data as well as enhance our ability to address questions of phylogenetics with the fossil record. 26
This project will thus provide the first geographically integrated approach to studying these 27
cosmopolitan birds. This is crucial to advancing our understanding of hesperornithiform 28
evolution, as the majority of hesperornithiform species, and in some cases genera, are restricted 29
to a single continent. 30
History of Hesperornithiform Paleontology 31
Shortly after the discovery of Archaeopteryx, the first Mesozoic bird discovered and the 32
oldest known bird, a lineage of unusual birds was described by O. C. Marsh from Late 33
Cretaceous marine sediments of Kansas (1870). Marsh first described Hesperornis regalis as a 34
swimming bird over five feet long that he characterized as morphologically similar to modern 35
grebes (1872), but considered to be most closely related to the ratites (1880). Marsh went on to 36
describe four additional hesperornithid birds from Kansas, Hesperornis crassipes and 37
Hesperornis gracilis (1880), Baptornis advenus (1877), and Coniornis altus (1893). Meanwhile, 38
Seeley presented two new Early Cretaceous diving birds from marine deposits in England, 39
Enaliornis sedgewicki and Enaliornis barretti (1876). While Marsh did not consider Enaliornis 40
closely related to the North American diving birds, today they are seen as members of the same 41
order (Galton and Martin, 2002). Thus by the end of the 19
th
century the Hesperornithiformes 42
were the most diverse lineage of Cretaceous birds known, with a wide geographic and 43
stratigraphic distribution and ranging in size from a bird the size of a grebe to birds over five feet 44
in length. 45
3

The first half of the twentieth century saw the discovery of a single taxon of 46
hesperornithiform, Hesperornis montana (Shufeldt, 1915), and for the most part very little 47
research was done on these birds. The next 50 years saw several papers reporting the discovery 48
of new remains (e.g. Russell, 1967; Martin and Tate, 1967; Fox, 1974) or re-describing the 49
morphology of these birds (Martin and Tate, 1976); however very little evolutionary research 50
was undertaken. More recently, research across the globe has identified many more 51
hesperornithiform taxa: Parahesperornis alexi (Martin, 1984); Hesperornis bairdi, H. chowi, H. 52
conlini, H. macdonaldi, and H. mengeli (Martin and Lim, 2002); and Brodavis americanus, B. 53
bailey, and B. varneri (Martin et al., 2012) from the United States; Pasquiaornis hardei and P. 54
tankei (Tokaryk et al., 1997) and Canadaga arctica (Hou, 1999) from Canada; Asiahesperornis 55
bazahanovi from Kazakhstan (Nessov and Prizemlin, 1991; Dyke et al., 2006); Hesperornis 56
rossicus from Russia and Sweden (Nessov and Yarkov, 1993; Rees and Lindgren, 2005); 57
Brodavis mongoliensis (Martin et al., 2012) from Mongolia; and Enaliornis seelyi from England 58
(Galton and Martin, 2002) (Fig. 1). Over-arching evolutionary research exploring the 59
relationships among these groups is non-existent. 60
As we understand them today, the Hesperornithiformes comprise a morphologically and 61
geographically diverse group of birds. Twenty-nine species in thirteen genera and four families 62
have been described (Table 1); however the validity of a number of these species and genera is 63
debated. Hesperornithiform taxonomy is fraught with problems regarding the accurate 64
description of morphology, the precise identification of various specimens (and confusion of one 65
specimen with another), and the rigor with which species are defined and criteria established to 66
identify them as unique. Without a clear understanding of taxonomy, determining evolutionary 67
4

relationships within the hesperornithiforms is impossible. Studies that have attempted to do so 68
thus far are fraught with errors, creating a definite need for more research. 69
70
TAXONOMY OF THE HESPERORNITHIFORMES 71
Taxonomic Problems within the Hesperornithiformes 72
Our understanding of hesperornithiform taxonomy is plagued by a host of problems 73
common to paleontology, such as the renaming of previously described taxa (Lucas, 1903; Rees 74
and Lindgren, 2005), taxa described from highly fragmentary material (Shufeldt, 1915; Martin 75
and Lim, 2002; Martin et al., 2012), and subjective, unspecific, or incorrect characters used for 76
diagnosis (Tokaryk et al., 1997; Martin, 1984), reliance upon which may result in further 77
confusing the assignment of fragmentary taxa (Olson, 1992; Martin et al., 2012). The majority of 78
hesperornithiform species have been described from fragmentary material (Fig. 2). Of the 79
twenty-nine described species, only four include specimens preserving more than five elements. 80
Eighteen species were described and remain known from a single bone. Whether or not all of 81
these species are valid taxa has rarely been questioned. For example, debate over the synonymy 82
of Coniornis altus (Marsh, 1893), Hesperornis altus (Shufeldt, 1915), and Hesperornis montana 83
(Shufeldt, 1915) as well as that of Hesperornis gracilis (Marsh, 1876), Hargeria gracilis (Lucas, 84
1903), and Parahesperornis alexi (Martin, 1984) has appeared repeatedly in the literature 85
without ever being resolved in the form of a concise diagnostic description and justification of 86
the ‘valid’ taxon. 87
From the very first descriptive work on the group, confusion has been common to 88
hesperornithiform research. In his first detailed diagnostic description of Hesperornis regalis, 89
Marsh incorrectly listed the following characters as diagnostic of the group: teeth in sockets, 90
5

biconcave vertebrae, sternum with keel, and wings well developed (1875). As Marsh went on to 91
describe the Icthyornithiformes as having teeth in a groove, a keel-less sternum and rudimentary 92
wings, he apparently switched the descriptions of hesperornithiforms and ichthyornithiforms. He 93
later published a correct diagnosis, adding an undivided proximal articular head of the quadrate 94
as an additional diagnostic feature (1877). A similar type of error occurred much more recently 95
when YPM PU 17208 was designated as the holotype specimen of H. chowi (Martin and Lim, 96
2001; p. 166), yet later in the same publication the type specimen of H. bairdi was also 97
designated as YPM PU 17208 (Martin and Lim, 2001; p. 166). Clearly a single specimen cannot 98
be the holotype for two species. 99
Perhaps due to difficulties arising from the fragmentary nature of the fossil record, a 100
number of taxa have been poorly or inaccurately described. An example of one such recurring 101
error and source of much confusion is the presence or absence of the proximal foramina on the 102
cranial surface of the tarsometatarsi of hesperornithiforms. In his main descriptive work on 103
hesperornithiforms, Marsh (1880) did not mention the presence of these foramina in Hesperornis 104
or Baptornis. More recent descriptive work has specifically pointed out the lack of these 105
foramina in numerous species of hesperornithiforms (Martin and Tate, 1976; Martin, 1984; 106
Galton and Martin, 2002). This has led to the presence of these foramina to be used, in part, as 107
justification for the designation of Baptornis varneri (Martin and Cordes-Person, 2007) and for 108
the exclusion of Neogaeornis wetzeli (Lambrecht, 1922) from the Hesperornithiformes (Olson, 109
1992). Furthermore, the relative degree of development has been used as a diagnostic feature of 110
two species of Brodavis (Martin et al., 2012). However, closer examination of specimens of H. 111
regalis, H. gracilis, H. crassipes, P. alexi, B. advenus, and other unidentified hesperornithiform 112
shows that in all cases proximal foramina are present on the cranial surface of the tarsometatarsi 113
6

(Clarke, 2004; Everhart and Bell, 2009). Incomplete preparation of the bones may be to blame 114
for the foramina being overlooked by previous authors. Additionally, the appearance of these 115
foramina appears to be closely tied to preservation quality. 116
Errors such as these are common in the descriptive literature of hesperornithiforms, 117
mandating a revision of the taxonomy of these birds before more extensive evolutionary research 118
can be undertaken. In order to develop a synthesis of our understanding of hesperornithiform 119
birds as well as update and revise the taxonomy of the order, this study presents a two-fold 120
approach. First, the current taxonomic framework is compiled and analyzed to identify those 121
features that are valid well as those in need of further work or revision. Second, a comprehensive 122
morphometric database is developed to explore the use of quantitative data in hesperornithiform 123
taxonomy. This is then combined with a more conventional character-based diagnostic approach 124
in order to develop an updated and comprehensive taxonomy of the Hesperornithiformes. 125
Current Taxonomic Framework of the Hesperornithiformes 126
The first portion of this study - a complete taxonomic review - is presented here, within 127
the framework of the current taxonomy. For each taxonomic unit, all proposed diagnostic 128
features have been evaluated. The results of this analysis are found in 2, organized by taxonomic 129
group. Characters that are supported are described in Appendix I. A number of characters are not 130
supported as diagnostic of the group for which they were originally proposed. Characters were 131
rejected for a number of reasons: inaccurate morphological description (the proximal foramina 132
example discussed above is this type of feature), inappropriate application of an observed 133
morphology (for example, a feature present in one genus but not another cannot be justified as 134
diagnostic of a family that includes both genera), plesiomorphy (the presence of an ancestral 135
state found in basal birds such as Archaeopteryx cannot be considered apomoprhic to 136
7

hesperornithiform taxa), lack of independence (diagnostic features should describe an 137
apomorphy in the taxon being considered, not in relation to the presence of the feature in another 138
taxon), redundancy (for example, if a character has been proposed that a specific element is 139
slender in overall form, then another character describing some particular feature of that element 140
as slender is redundant), or some combination of these reasons. Appendix II presents all 141
characters unsupported by the analysis with a justification of their exclusion from 142
hesperornithiform taxonomy. Additionally, a number of characters are identified that were based 143
on implied morphometric definitions (for example, the tarsometatarsus of numerous species has 144
been described as being ‘slender’ – this is essentially a qualitative way of describing the length to 145
width ratio of the element). These characters are identified in Table 2, but will be fully discussed 146
in Chapter 3, which presents a number of morphometric analyses of hesperornithiform birds. The 147
following discussion is intended to provide an introduction to hesperornithiform taxonomy and 148
reduce the diagnostic features of each taxonomic unit to only those for which there is support. 149
Order Hesperornithiformes. To date, the order Hesperornithiformes has been defined 150
by morphologic features from the cranial and postcranial skeleton. However, the veracity of this 151
collection of characters has never been evaluated. Whether or not early characters conflict with 152
later discoveries, if all species are adequately diagnosed to modern standards, and how the 153
discovery of new species changes the diagnosis of more inclusive groups are all questions that 154
have not been investigated. To date, twenty-six characters have been proposed as diagnostic of 155
the Hesperornithiformes. Of these, the present analysis finds support for nine, while sixteen 156
cannot be supported, and a character is reserved for morphometric analysis (Table 2). 157
Family Baptornithidae. The Baptornithidae was erected to place Baptornis advenus 158
within the Hesperornithiformes but separate from the Hesperornithidae (Martin and Tate, 1976). 159
8

Since the establishment of the family, several new genera and species have been added to the 160
group. A small baptornithid, Judinornis nogontsavensis, was described from a single thoracic 161
vertebrae discovered in alluvial deposits of Mongolia (Kurochkin, 2000). The genus 162
Pasquiaornis, consisting of two species from the Cenomanian of Canada, has also been added to 163
the Baptornithidae (Tokaryk et al., 1997). These studies, along with research into baptornithid 164
specimens from Canada (Tokaryk and Harrington, 1992) and Kansas (Everhart & Bell, 2009) 165
have added to the diagnostic features of the family. Eighteen characters have been proposed as 166
diagnostic of the Baptornithidae, of which four are supported, twelve are unsupported, and two 167
are reserved for morphometric analysis (Table 2). 168
Genus Baptornis. The mono-specific genus Baptornis was first erected in 1877 by O.C. 169
Marsh for an isolated tarsometatarsus from the Niobrara Formation of Kansas. While a number 170
of isolated elements have been identified that appear to differ slightly from those of Baptornis 171
advenus, due to the fragmentary nature of the remains erecting new species has been deemed 172
inappropriate (Tokaryk and Harrington, 1992; Kurochkin, 2000; Rees and Lindgren, 2005; 173
Everhart and Bell, 2009) (Table 3). While a second species was named (Baptornis varneri, 174
Martin and Cordes-Person, 2007), the holotype and sole specimen (SDSM 68430) was later re- 175
assigned to the Brodavidae as Brodavis varneri (Martin et al., 2012). Three features have been 176
proposed as diagnostic of the genus, of which all are rejected by the present study (Table 2). 177
Baptornis advenus. Baptornis advenus was first described by Marsh (1877) at the same 178
time he erected the genus for an isolated tarsometatarsus (YPM 1465). When Baptornis advenus 179
was first erected, Marsh described the holotype as “a nearly perfect tarso-metatarsal (sic) bone” 180
(1877, p. 86), however neither figures nor a specimen number was provided. Marsh later echoed 181
this description and provided the specimen number of YPM 1465 as that of the holotype in his 182
9

extensive monograph on the hesperornithiforms, where he also presumably published line 183
drawings of the element, however they were not labeled (1880). YPM 1465 has not survived the 184
intervening years in its original condition – today it preserves only the distal end. A separate 185
proximal end was originally included in YPM 1465 in the collection of the Yale Peabody 186
Museum (2008, pers. obs.), however it has since been removed from that specimen and assigned 187
to Baptornis indeterminate for reasons that are unclear. 188
Today, a number of specimens have been assigned to B. advenus and the skeleton is 189
almost completely known, with the exception of most of the cranium (Martin and Tate, 1976). 190
These specimens are known entirely from the Niobrara Formation (Coniacian) of Kansas (Table 191
3). The diagnosis of B. advenus is currently the same as that of the genus. 192
Genus Judinornis. Judinornis was described from an isolated thoracic vertebra from the 193
Nemegt Formation (Maastrichtian) of Mongolia (PO 3389; Nessov and Borkin, 1983 as 194
translated by Kurochkin, 2000), making it one of the youngest hesperornithiforms. The vertebra 195
appears to be a caudal thoracic, and judging from the smooth ventral surface may correspond to 196
the 23
rd
and caudal-most thoracic vertebra (in all hesperornithiforms for which the caudal-most 197
thoracic vertebra can be identified, the ventral surface is smooth, while in more cranial thoracic 198
vertebrae a keel is developed to varying degrees). However, these observations are dependent 199
upon the quality of the published drawings (Fig. 4 - Nessov, 1992). If the reported dimensions 200
are correct (centrum length of 14.1 mm; Kurochkin, 2000), then PO 3389 falls within the size 201
range exhibited by specimens assigned to Baptornis. Three characters were proposed as 202
apomorphic (Nessov and Borkin, 1983). Of these, none can be supported (Table 2). The genus is 203
monospecific, with the diagnosis for Judinornis nogontsavensis the same as that for the genus. 204
10

Genus Pasquiaornis. Pasquiaornis was erected to unite two species of small 205
hesperornithiforms from the Belle Fourche Formation (Cenomanian) of Canada (Tokaryk et al., 206
1997). These taxa are only known from unassociated and disarticulated elements found in a bone 207
bed deposit (Cumbaa et al., 2006). Tokaryk et al. (1997) determined that a combination of size 208
and select morphological differences could be used to separate the disarticulated elements 209
assigned to Pasquiaornis into two species, P. hardei and P. tankei. While P. tankei was 210
described as the larger of the two taxa, specific size differentials used to separate these species 211
have never been quantified. Furthermore, as there are no associated elements known within the 212
genus, it is unclear how different elements are assigned together in one of the two species (Table 213
4). Of the four morphological features that have been proposed as diagnostic of the genus, three 214
are unsupported by the present analysis and one is reserved for further morphometric study 215
(Table 2). 216
Pasquiaornis hardei. A left tarsometatarsus (RSM P2077.117) was assigned as the 217
holotype of Pasquiaornis hardei, along with a right femur designated as a paratype, and 218
additional elements referred to the species (Tokaryk et al., 1997) (Table 4). As mentioned above, 219
despite relying on size for the assignment of specimens to one of the two species, no size 220
differential was included in the diagnosis (Tokaryk et al., 1997). It is also unclear how sizes were 221
compared across the variety of elements recovered, and on what basis different elements were 222
assigned to this species. Furthermore, figures or published measurements of the described 223
material are scarce (Tokaryk et al., 1997; Sanchez, 2010). Of the features proposed as diagnostic 224
of this species, three are unsupported and a fourth is reserved for further morphometric study 225
(Table 2). 226
11

Pasquiaornis tankei. Like P. hardiei, P. tankei has a partial tarsometatarsus (SMNH 227
P2077.113) designated as the holotype and a femur as a paratype (SMNH 2077.108), with a 228
number of additional elements referred to the species (Tokaryk et al., 1997) (Table 4). As 229
discussed above, it is unclear how these elements were assigned to P. tankei. Of the three 230
morphological features proposed as diagnostic, none are supported by this study (Table 2). 231
Genus Parascaniornis. Parascaniornis was a monospecific genus erected for an isolated 232
thoracic vertebra from Sweden, Parascaniornis stensoi (MGUH 1908.214; Lambrecht, 1933). 233
Originally allied with the modern Ciconiiformes (Lambrecht, 1933), the specimen was later 234
recognized as a hesperornithiform (Nessov and Prizemlin, 1991). Further work identified the 235
vertebra as belonging to the genus Baptornis, thus rendering Parascaniornis a junior synonym of 236
Baptornis and Parascaniornis stensoi a nomen dubium (Rees and Lindgren, 2005). 237
Family Brodavidae. The Brodavidae is the newest addition to the Hesperornithiformes, 238
erected to unite four fragmentary specimens into a single family on the basis of two features of 239
the tarsometatarsus (Martin et al., 2012). Both of these features refer to the dimensions of the 240
tarsometatarsus, and so are reserved for further morphometric analysis (Table 2). As the family is 241
monogeneric (Brodavis), the diagnosis for the genus is the same as that of the family. The four 242
species assigned to the genus have a broad geographic and stratigraphic distribution (Fig. 1, 243
Table 5), similar to that seen in the Baptornithidae or Hesperornithidae. 244
Brodavis americanus. B. americanus was described from a partial tarsometatarsus 245
lacking the extreme proximal and distal ends collected from the Frenchman Formation 246
(Maastrichtian) of Saskatchewan, Canada (RSM P2315.1; Martin et al., 2012), making it one of 247
the youngest hesperornithiforms. The holotype and sole specimen appears fairly well-preserved, 248
lacking only the proximal-most end. However, this is dependent upon the accuracy of the 249
12

published drawing (Fig. 1A-C in Martin et al., 2012). Four features were proposed as diagnostic 250
of the species, three of which are reserved for morphometric analysis and one of which is not 251
supported (Table 2). 252
Brodavis baileyi. B. baileyi was described from a partial tarsometatarsus lacking the 253
proximal and extreme distal end, found in what was reported as the Hell Creek Formation of 254
South Dakota (UNSM 50665; Martin et al., 2012). However, comparison with other finds in the 255
South Dakota School of Mines collections from the same location (described as 13 miles south 256
and 6 miles west of Buffalo, SD, USA), shows that this area consists of Pierre Shale exposures, 257
not Hell Creek. This would change the age of B. baileyi from the latest Maastrichtian to early 258
mid-Campanian and from an estuarine environment (contra Martin et al.’s freshwater 259
interpretation) to one of marine deposition (Weimer, 1960). Therefore, Martin et al.’s (2012) 260
claim that B. baileyi represents the first known freshwater hesperornithiform (quoting Martin, 261
1983) is incorrect. Even if B. baileyi were from the Hell Creek Formation, a number of other 262
hesperornithiforms are known from the deltaic and estuarine Hell Creek and Judith River 263
Formations in North America (Marsh, 1872; Tokaryk and Harrington, 1992) as well as Brodavis 264
mongoliensis from the alluvial Nemegt Formation of Mongolia (Nessov and Borkin, 1983; 265
Kurochkin, 2000). Therefore, the presence of hesperornithiforms in environments that have been 266
interpreted as estuarine or freshwater is not unique, despite the greater abundance of material 267
from marine sediments. Of the five characters proposed as diagnostic of B. baileyi, two are 268
unsupported and three are reserved for morphometric analysis (Table 2). 269
Brodavis mongoliensis. The holotype of B. mongoliensis, a partial tarsometatarsus 270
lacking the distal end (PIN 4491-8), was first described as a small, possibly volant 271
hesperornithiform (Nessov, 1992 as translated in Kurochkin, 2000) and later designated as the 272
13

holotype of B. mongoliensis within the Brodavidae (Martin et al., 2012). PIN 4491-8 is from the 273
Nemegt Formation (Maastrichtian) of Mongolia. The initial description noted a number of 274
features diagnostic of the Hesperornithiformes as well as two features (short and stout with a 275
transversely expanded shaft) identified as separating this specimen from other 276
hesperornithiforms, however no new taxonomic designations was erected (Nessov, 1992). Martin 277
et al. later used these features as well as others to unite PIN 4491-8 with other brodavids and 278
proposed five features as unique to B. mongoliensis (2012). Of these, two are unsupported and 279
two are reserved for morphometric analysis, while one feature could not be analyzed at this time 280
(Table 2). 281
Brodavis varneri. B. varneri was initially described as Baptornis varneri (Martin and 282
Cordes-Person, 2007), as discussed above. Martin et al. (2012) tentatively re-assigned the 283
holotype specimen (SDSM 68430) to Brodavis, however character evidence was not provided to 284
justify this re-assignment. Instead, Martin et al. pointed to a similar resemblance “in the overall 285
shape” of the tarsometatarsus (Martin et al., 2012, p. 62) with the other species of Brodavis as 286
justification for inclusion in the genus while also noting that it “should probably have its own 287
generic designation” (Martin et al., 2012, p. 62). While Martin et al. did not identify any 288
diagnostic features of the taxon (2012), the original description identified twelve diagnostic 289
characters (Martin and Cordes-Person, 2007). Of these, one is supported, nine are unsupported, 290
and two are reserved for future morphometric analysis (Table 2). 291
Family Enaliornithidae. The Enaliornithidae from the Cambridge Greensand (Albian) of 292
England are the oldest hesperornithiform family. They are known entirely from the Early 293
Cretaceous, while all other hesperornithiforms are from Late Cretaceous deposits. First reported 294
in 1859 by Lyell, they were more fully described by Seeley (1864, 1876). Two species were 295
14

identified on the basis of size, but without quantification of those differences (Seeley, 1876). 296
While Marsh did not consider Enaliornis closely related to his ‘swimming ostriches’, the 297
Hesperornithiformes, Seeley saw them as closely related (1876). As the remains are highly 298
fragmentary and consist entirely of disarticulated and unassociated remains, determining the 299
exact taxonomic diversity of this group has proven difficult (for example: Seeley, 1876; Storer, 300
1965; Martin and Tate, 1976; Elzanowski and Galton, 1991; Galton and Martin, 2002). Brodkorb 301
identified a lectotype and numerous paralectotypes for each species (Brodkorb, 1963) (Table 6). 302
Later, extensive evaluation of all known material by Galton and Martin (2002, 2002b) proposed 303
a number of diagnostic features of the family as well as each species. They also identified a third 304
species, E. seeleyi, on the basis of size (Galton and Martin, 2002). It remains unclear how 305
unassociated elements were combined into a single species (Table 7). As the family is presently 306
monogeneric, the diagnosis for the genus Enaliornis is the same as that of the family. Of the 307
seven features proposed as diagnostic of the family and genus, none are supported (Table 2). 308
Enaliornis barretti. E. barretti was one of two species of Enaliornis initially reported by 309
Lyell (1859). Seeley (1864, 1876) later elaborated on this description and differentiated E. 310
barretti as the larger of the two species without providing any specific size criteria. Additional 311
material was reviewed by Galton and Martin (2002), who proposed diagnostic morphological 312
features and further investigated size variation within Enaliornis. While a holotype was not 313
originally designated by either Lyell (1859) or Seeley (1864, 1876), Brodkorb designated the 314
distal tarsometatarsus BMNH A477 as a lectotype and a number of specimens as paralectotypes 315
(1963) (Table 6) while Galton and Martin assigned additional specimens to the species (2002) 316
(Table 7). Of the five proposed diagnostic features of E. barretti, three are unsupported and two 317
are reserved for morphometric analysis (Table 2). 318
15

Enaliornis sedgewicki. Like E. barretti, E. sedgewicki was initially reported by Lyell 319
(1859) and then further studied by Seeley (1864, 1876), who described E. sedgewicki as the 320
smaller species. Later work by Brodkorb identified the proximal tibiotarsus SMC B55314 as a 321
lectotype and designated a number of paralectotypes (1963) (Table 6). Galton and Martin (2002) 322
identified four characters allegedly apomorphic to E. sedgewicki, of which one is unsupported 323
and three are reserved for morphometric analysis (Table 2). 324
Enaliornis seeleyi. E. seeleyi was described by Galton and Martin (2002) as the medium- 325
sized member of Enaliornis; however specific size criteria were not established. The distal 326
tibiotarsus BGS 87935 was designated as the holotype and a number of specimens previously 327
assigned to either E. barretti or E. sedgewicki were moved to E. seeleyi (Table 7). Five 328
diagnostic features were proposed, of which two are unsupported and three are reserved for 329
morphometric analysis (Table 2). 330
Family Hesperornithidae. The Hesperornithidae was erected by Marsh (1872), and at the 331
time was monogeneric and monospecific. Since that time the genera Parahesperornis (Martin, 332
1984), Asiahesperornis (Nessov and Prizemlin, 1991), and Canadaga (1999) have been added to 333
the family, as well as several additional species of Hesperornis (Marsh, 1877; Nessov and 334
Yarkov, 1993; Martin and Lim, 2002) (Table 1). Unfortunately Marsh did not diagnose the 335
family separately from the genus Hesperornis, and subsequent work has focused on individual 336
species or genera and not diagnostic features of the entire family. Therefore, only two diagnostic 337
features exist at this taxonomic level. Of these, both are supported by the present analysis (Table 338
2). 339
Genus Asiahesperornis. Asiahesperornis was first described from several unassociated 340
elements, with a tarsometatarsus and tibiotarsus retained as the holotype (Nessov and Prizemlin, 341
16

1991; translated by Kurochkin, 2000). However, the elements identified as the holotype were not 342
associated and most likely came from different individuals (Dyke et al., 2006). Therefore, only 343
the tarsometatarsus was retained as the holotype, and only features from that element are valid as 344
diagnostic of the group (Dyke et al., 2006). A number of other have been assigned to 345
Asiahesperornis, but it is not certain they are con-specific (Dyke et al., 2006) (Table 8). All 346
known specimens of Asiahesperornis come from the Zhuravlovskaya Svita (Maastrichtian) in the 347
Priozernyi Quarry of northern Kazakhstan (Dyke et al., 2006), putting them among the youngest 348
hesperornithiforms. As the genus is monospecific, the diagnosis for the species, A. bazhanovi, is 349
the same as for the genus. Of the eleven diagnostic features originally proposed, two have been 350
invalidated because they pertained to the tibiotarsus (Dyke et al., 2002). Of the remaining nine 351
features, the present analysis finds one to be supported, five to be unsupported, and three in need 352
of further morphometric analysis (Table 2). 353
Genus Canadaga. Canadaga was erected to describe a number of mostly disarticulated 354
elements discovered on Bylot Island in Canada (Hou, 1999). While the stratigraphic setting was 355
not described, a Maastrichtian age was reported (Hou, 1999). These specimens are among the 356
most northern occurrence of hesperornithiform birds (Hou, 1999). A series of three articulated 357
posterior cervical vertebrae (vertebrae 15 ~ 17, only 16 is complete) were identified as the 358
holotype (NMC 41050) and an additional caudal vertebrae and two femora were assigned to the 359
species (Hou, 1999). As there was no association of these elements and the holotype they should 360
be considered Hesperornis indeterminate. Since the initial description, another series of three 361
articulated vertebrae (vertebrae 16-18) have been assigned to the species from the Kanguk 362
Formation (Coniacian) on Devon Island (Wilson et al., 2011) (Table 9). There is a long gap in 363
time between the Devon Island (Coniacian) and holotype (Maastrichtian) specimens despite 364
17

they’re geographic proximity. It may be that the reported Maastrichtian age of the holotype 365
(Hou, 1999) was in error. As the genus is monospecific, the diagnosis for C. arctica is the same 366
as that for the genus. Of the eight proposed diagnostic features, one is supported, six are 367
unsupported, and one is reserved for morphometric analysis (Table 2). 368
Genus Hargeria. Hargeria is an invalid taxon that was erroneously erected by Lucas 369
(1903) as a re-classification of a species originally described by Marsh (1880), Hesperornis 370
gracilis. Lucas based his decision not on study of the holotype (YPM 1473), which he apparently 371
never saw, but instead on observations of a fairly complete specimen from the University of 372
Kansas, KUVP 2287 (Lucas, 1903). Unfortunately, this specimen is in no way a member of the 373
same species as the holotype of Hesperornis gracilis, as Lucas mistakenly believed. All of the 374
differences between the tarsometatarsi of KUVP 2287 and H. regalis that Lucas described are 375
also present between KUVP 2287 and the holotype of Hesperornis gracilis. This confusion was 376
later sorted out and resolved, with KUVP 2287 becoming the holotype of a new genus and 377
species, Parahesperornis alexi (Martin, 1984). 378
Genus Hesperornis. Hesperornis was the first genus of the Hesperornithidae described, 379
with H. regalis designated as the type species (Marsh, 1872). Two additional species were 380
referred to the genus, H. gracilis (Marsh, 1876) and H. crassipes, originally described as 381
Lestornis crassipes (Marsh, 1876) and quickly re-assigned to Hesperornis (Marsh, 1877b). H. 382
montana was later added after the discovery of the first hesperornithid outside of the Niobrara 383
Formation of Kansas and Nebraska, an isolated vertebra from the Claggett Shale (Campanian) of 384
Montana (Shufeldt, 1913). For over fifty years new taxa were not added to the genus, while 385
discovery of additional Hesperornis material was constantly being added to collections. 386
18

More recently, an analysis of material from the Pierre Shale resulted in the designation of 387
four new species of Hesperornis: H. bairdi, H. chowi, H. macdonaldi, and H. mengeli (Martin 388
and Lim, 2002). While H. bairdi is based on two elements (YPM PU 17208A, but see below), 389
the others are all represented by a single element. Discovery of large hesperornithids from Asia 390
resulted in identification of the new species H. rossicus (Nessov and Yarkov, 1993), members of 391
which were later identified from Sweden as well (Rees and Lindgren, 2005) (Table 10). 392
While abundant material has been referred to H. regalis over the years, all other species 393
of Hesperornis remain more poorly known (Table 10). Despite the amount of research that has 394
been done erecting new species, virtually no work has expanded on Marsh’s original description, 395
which identified a single diagnostic character. This character, a deeply excavated intercondylar 396
groove of the tibiotarsus, will be further investigated in the morphometric analysis. 397
Hesperornis altus. The holotype of H. altus (YPM 515) was initially described as 398
Coniornis altus (Marsh, 1893), as discussed above. The specimen was found in the Claggett 399
Shale (Campanian) of Montana. Marsh identified C. altus as distinct from H. regalis because of 400
different distal extents of the condyles as well as a smaller size (1893). Shufeldt later re- 401
evaluated the holotype, which is still the sole representative of the species, and determined the 402
differences between it and the tibiotarsus of H. regalis were insufficient to warrant generic 403
distinction (1915). Shufeldt also questioned whether there were sufficient differences to warrant 404
species distinction, however he did not invalidate the species, and re-designated the holotype as 405
Hesperornis altus (1915). Despite retaining the species distinction, Shufeldt invalidated both 406
features that had been proposed as apomorphic (1915); the distal extent of the condyles and the 407
overall smaller size (Table 2). Therefore, there appears to be no justification for retaining 408
Hesperornis altus as a valid taxon. 409
19

Hesperornis bairdi. H. bairdi was recognized by Martin and Lim (2002) along with three 410
other new species of Hesperornis. Unfortunately, the elements used as the holotype of this 411
species as well as that of H. chowi were unclear, as mentioned above. Examination of the 412
original material at the Yale Peabody Museum revealed that the collection number YPM PU 413
17208 applies to several individuals, as indicated by the presence of at least three tarsometatarsi. 414
Of these, one tarsometatarsus was identified by Martin and Lim (2002) as the holotype of H. 415
chowi while another tarsometatarsus and a partial pelvis were identified as the holotype of H. 416
bairdi, and given the revised catalogue number of YPM PU 17208A. It is unclear how the partial 417
pelvis was associated with the tarsometatarsus, as there is no information with YPM PU 17208 to 418
explain why the elements are grouped together with a single specimen number. Martin and Lim 419
(2002) did not provide any information about this. Therefore, the partial pelvis assigned to YPM 420
PU 17208A should be removed as part of the holotype, and considered Hesperornis 421
indeterminate pending further investigation. All of the YPM PU17208 material is from the lower 422
Sharon Springs Member of the Pierre Shale (Campanian) of South Dakota. 423
Aside from the confusion regarding specimen numbering and associations, Martin and 424
Lim (2002) proposed two characters of the tarsometatarsus as apomorphic of H. bairdi, one of 425
which cannot be supported and the other will be further analyzed in the morphometric study 426
(Table 2). 427
Hesperornis chowi. H. chowi was identified by Martin and Lim (2002) as having PU 428
17208 as the holotype, the same as for H. bairdi, as discussed above. A single tarsometatarsus 429
from the multi-individual specimen was selected as the holotype. However, unlike the case in H. 430
bairdi, no alteration was made to the specimen number. Therefore, the holotype tarsometatarsus 431
of H. chowi is still lumped with at least five other elements from an unknown number of 432
20

individuals as YPM PU 17208. Two additional specimens (YPM 18589, YPM PU 17193) have 433
been referred to H. chowi (Marton, pers. comm.; Wilson et al., 2011). As neither of these 434
specimens contains a tarsometatarsus, they do not overlap with the holotype and cannot be 435
retained as valid. Of the four diagnostic features proposed for the species, three are unsupported 436
and one will be subject to further morphometric analysis (Table 2). 437
Hesperornis crassipes. H. crassipes was initially described as Lestornis crassipes by 438
Marsh from a partial skeleton (YPM 1474) found in the Niobrara Formation of Kansas (1876). 439
The skeleton is fairly complete and currently inaccessible due to its location on display in the 440
Yale Peabody Museum. Marsh used the large size of the bird as well as features of the 441
tarsometatarsus and sternum to separate H. crassipes from H. regalis (1876, 1880). While an 442
additional isolated tarsometatarsus (AMNH 5102) has been assigned to the species, it is unclear 443
how this assignment was made. Since the specimen does not exhibit morphology described by 444
Marsh as diagnostic of H. crassipes, this specimen should be regarded as Hesperornis 445
indeterminate at this time. Of the four characters proposed as diagnostic, one is supported, one is 446
in need of morphometric analysis, and two cannot be evaluated at this time (Table 2). 447
Hesperornis gracilis. H. gracilis was described by Marsh from an isolated 448
tarsometatarsus (YPM 1473, the holotype) and two more complete specimens (YPM 1478, YPM 449
1679) (Marsh, 1880). Since then no other specimens have been discovered and no additional 450
diagnostic features have been added (Table 10). H. gracilis was briefly renamed as Hargeria 451
gracilis (Lucas, 1903) as previously discussed. A single feature has been proposed as diagnostic 452
of the group, which will be evaluated in the morphometric analysis (Table 2). 453
Hesperornis macdonaldi. H. macdonaldi is an isolated femur from the Pierre Shale 454
(Campanian) of South Dakota (LACM 9728; Martin and Lim, 2002). No other published 455
21

material has since been assigned to this species. The only diagnostic character proposed was the 456
extremely small size of this individual, making it one of the smallest species of Hesperornis. 457
While the element is extremely small, it is also extremely poorly preserved, with the entire 458
surface corroded by calcite formation. There is no way to tell if the articular surfaces are ‘well- 459
formed’, as claimed by Martin and Lim (2002, p. 171). As this was their only evidence for an 460
adult age of the specimen, further histologic work is needed to establish the age of the individual. 461
The size difference proposed as diagnostic will be investigated in the morphometric analysis 462
(Table 2). 463
Hesperornis mengeli. H. mengeli is an isolated tarsometatarsus from the lower Sharon 464
Springs Member of the Pierre Shale (Campanian) of Manitoba, Canada (Martin and Lim, 2002), 465
and is one of the smallest tarsometatarsi assigned to Hesperornis. The original description of the 466
specimen listed the holotype as BO 780106 (Martin and Lim, 2002); however the correct 467
specimen number is CFDC 78.01.08. The ontogenetic status of the holotype has not been 468
discussed in the literature (Martin and Lim, 2000) and cannot be readily evaluated due to the lack 469
of published photographs. While further specimens have not been presented in the literature, 470
Martin also assigned a portion of the chimeric YPM PU 17208 (femur, thoracic vertebrae, and 471
partial pelvis) to H. mengeli (pers. comm., Table 10). As the holotype of H. mengeli is a 472
tarsometatarsus and none of the elements assigned from YPM PU 17208 overlap, there is no 473
basis for this assignment. While possibly made because of the small size of the selected elements 474
of YPM PU 17208, H. mengeli is not the only small Hesperornis (see H. macdonaldi above) and 475
the use of this sort of character is not justified. Four diagnostic features were proposed for the 476
holotype and sole specimen of H. mengeli, of which one is unsupported and three will be further 477
considered in the morphometric analysis (Table 2). 478
22

Hesperornis montana. H. montana was described by Shufeldt (1915) from an isolated 479
thoracic vertebra after comparison with vertebrae belonging to H. regalis and H. crassipes. The 480
paper is unusual in its writing style, as the entire description of the new vertebrae and 481
comparison to other hesperornithids is made by a Dr. Richard S. Lull of the Peabody Museum, 482
who recommended the bone be assigned to Hesperornis regalis or Hesperornis indeterminate 483
(Shufeldt, 1915, p. 292). Shufeldt, however, after reporting Lull’s description and 484
recommendation, went on to assign the vertebrae to a new species, Hesperornis montana, 485
without identifying any features unique to the element (Shufeldt, 1915; p. 294). Given the lack of 486
variation between vertebrae of Hesperornis species, it is recommended that this taxon be 487
invalidated and the specimen assigned as Hesperornis indeterminate, as initially recommended 488
by Lull (Shufeldt, 1915). 489
Hesperornis regalis. H. regalis was the first species of Hesperornis to be described, 490
originally from numerous remains found in the Niobrara Formation (Coniacian) of Kansas 491
(Marsh, 1872). At the time, Marsh did not identify any specific diagnostic features of the species 492
(1872, 1880). Despite extensive research over the subsequent years and the addition of numerous 493
species of Hesperornis from across the globe (Table 10), there is only a single diagnostic feature 494
that has been proposed for the type species (Martin, 1984). The use of this feature as diagnostic 495
is unsupported by this study (Table 2). 496
Hesperornis rossicus. H. rossicus was the first and remains the only Hesperornis 497
discovered in Asia, from the Rybushka Formation (Campanian) of Russia (Nessov and Yarkov, 498
1993). Known entirely from fragmentary, disarticulated remains, the holotype of H. rossicus is a 499
proximal tarsometatarsus (VPM N 26306/2). Several other elements (vertebrae, a tibiotarsus, 500
tarsometatarsi, and a phalanx) from Russia and Sweden have been assigned to the species 501
23

(Nessov and Yarkov, 1993; Panteleyev, 2004; Rees and Lindgren, 2005) (Table 10). In the 502
absence of associations between elements, only the tarsometatarsi should be considered as H. 503
rossicus at this time. Despite the presence of another hesperornithid, Asiahesperornis, from 504
nearby Kazahkstan, no direct comparison of the taxa has ever been presented. A number of 505
diagnostic features were identified when the species was erected (Nessov and Yarkov, 1993; 506
translated by Kurochkin, 2000). Additional work has identified more characters diagnostic of the 507
group (Panteleyev et al., 2004; Rees and Lindgren, 2005). Of these, two are supported, seven are 508
unsupported, and three are retained for further morphometric analysis (Table 2). 509
Genus Lestornis. Lestornis was initially described by Marsh from a fairly complete 510
skeleton from the Niobrara Formation of Kansas (YPM 1474; 1876). Soon after, Marsh re- 511
classified Lestornis as a species of Hesperornis, H. crassipes (1880). Therefore, the genus 512
Lestornis is a junior synonym of Hesperornis. Marsh did not state any reasons for this change 513
(1880); however it may be that after the discovery of more hesperornithid material, the 514
differences between YPM 1474 and Hesperornis specimens was insufficient to justify generic 515
distinction. 516
Genus Parahesperornis. Parahesperornis was initially known from a fairly complete 517
skeleton from the Niobrara Formation of Kansas mistakenly assigned to H. gracilis (Lucas, 518
1903; see discussion above). Since that time a complete description has not been published, but 519
additional specimens have been discovered and assigned to the genus (Table 11). As there is only 520
one species, P. alexi, the diagnosis of the species is the same as that of the genus. In the original 521
preliminary publication, eight diagnostic features were identified, of which one is supported, four 522
are unsupported, and three are reserved for morphometric analysis (Table 2). 523
524
24

CONCLUSIONS 525
Hesperornithiform birds are known from Early and Late Cretaceous deposits across the 526
globe, making them prime candidates for studies of primitive bird diversification. They are one 527
of the earliest known groups of birds to have secondarily lost or drastically reduced their ability 528
for flight, making them an ideal group to inform us of evolutionary processes. Despite this, 529
relatively little research has been done, and that which has is beset with flaws. This chapter 530
presents the first comprehensive analysis of hesperornithiform taxonomy, which reveals that of 531
the thirty-seven valid taxonomic units (twenty-three species in nine genera and 4 families); only 532
eleven are supported by any definitive diagnostic features (Table 12). While some of these taxa 533
may prove invalid, others (such as the genus Hesperornis) are clearly valid taxonomic groups in 534
need of an accurate diagnosis. This demonstrates the need for sound scientific research of the 535
Hesperornithiformes in two main areas: 1) a taxonomic revision, and 2) development of 536
hypotheses regarding the evolutionary relationships within the group. 537
This dissertation meets these needs through a comprehensive study that investigates all 538
species of hesperornithiforms by evaluating both the evolution of their unique skeletal system as 539
well as the use of quantitative data in analyses of fossil taxa, something that has only rarely been 540
done to date (but see Chiappe et al., 2008; Marugan et al., 2001). Chapter 2 presents the first 541
comprehensive and scientifically rigorous cladistic analysis of the majority of described 542
hesperornithiform species. Chapter 3 presents a number of analyses that investigate both the 543
variance of hesperornithiform specimens in multi-dimensional morphospace as well as the use of 544
quantitative data in the diagnosis of hesperornithiform taxa. Chapter 4 then summarizes and 545
combines the results from these studies to present a revised taxonomic framework for 546
hesperornithiform birds based on evolutionary relationships as determined in the cladistic 547
25

analysis. This research presents a better understanding of the evolution of early birds and also 548
facilitates the advancement of avian paleontology through integration of collection material from 549
across the globe with advanced phylogenetic methodologies. Beyond the bounds of avian 550
paleontology, this research provides examples of the use of morphometric data to illuminate 551
questions regarding ecology, taxonomy, and evolutionary relationships that are valuable to many 552
areas of modern biology as well as paleontology. 553
554
26

APPENDIX I. 555
This study has confirmed a number of features as diagnostic of various hesperornithiform 556
taxonomic units. Each supported feature (summarized in Table 2) is discussed here within the 557
current taxonomic framework. 558
559
ORDER HESPERORNITHIFORMES 560
561
Teeth in grooves (Marsh, 1877). Marsh initially and mistakenly switched the 562
descriptions of the teeth in hesperornithids and ichthyornithids (1875), however this mistake was 563
quickly rectified (Marsh, 1877, 1880) and the placement of the teeth in grooves has since been 564
upheld as unique to the Hesperornithiformes (e.g., Kurochkin, 2000). This differs from the 565
condition in Ichthyornis, which has teeth set in individual sockets within the jaw (Clarke, 2004). 566
It should be noted that within the jaw of Hesperornis, septae separate the teeth, however these 567
are not true sockets and cannot be seen externally. 568
Pterygoid process of the quadrate elongate and appressed onto the medial mandibular 569
condyle (Elzanowski, 2000). While most birds (primitive as well as modern) possess a small, 570
peg-like pterygoid condyle, the hesperornithiforms are unique in having an elongate pterygoid 571
process superimposed on the mandibular condyle of the quadrate (Elzanowski, 2000). 572
Sternum without keel (Marsh, 1877). This character was also involved in Marsh’s early 573
mix-up (1875, 1877), however it has been repeatedly upheld as a uniting feature of the 574
Hesperornithiformes (for example: Martin and Tate, 1976; Galton and Martin, 2002). While 575
Pasquiaornis does possess more morphological similarities with flighted birds than other 576
hesperornithiforms (Tokaryk et al., 1997), at this time a sternum has not been reported for 577
27

Pasquiaornis, making analysis of this feature impossible. Therefore, this feature can be used to 578
unite those species of hesperornithiforms for which a sternum is preserved. 579
Long bones nonpneumatic (Martin, 1983). The loss of pneumaticity and resulting 580
increase in density in the hind limb bones of hesperornithiforms such as Baptornis and 581
Hesperornis has long been observed (Marsh, 1880). Due to the fragmentary nature of a number 582
of taxa, determining the extent of pneumaticity is not always possible, however it appears that 583
Enaliornis (Galton and Martin, 2002), Pasquiaornis (Tokaryk et al., 2002; Sanchez, 2010), and 584
Brodavis (Martin et al., 2012) also exhibit reduction of pneumaticity in the hind limbs, thus 585
supporting use of this character as diagnostic of the Hesperornithiformes. 586
Large, trihedral patella, perforated for the ambiens tendon (Martin and Tate, 1976). The 587
unusually large patella present in Hesperornis regalis was noted by Marsh (1880). Later analysis 588
of Baptornis (Martin and Tate, 1976) and Parahesperornis (Martin, 1984) also identified large 589
triangular patellae. Unfortunately the patella is not preserved in many hesperornithiforms, for 590
example in Enaliornis (Galton and Martin, 2002) or Pasquiaornis (Tokaryk et al., 1997). In 591
modern birds the patella is often highly reduced, even among some diving birds (Storer, 1958). 592
Loons (Podiceps) have a very small patella that does not contribute to the area of muscle 593
attachment, and is therefore mechanically insignificant, while grebes (Aechmophorus) possess a 594
more robust patella which functions along with the proximal expansion of the tibiotarsus to 595
increase the area of muscle attachment (Storer, 1958), more similar to the case in 596
hesperornithiforms, although in the latter it is not fused. While the shape of the patella varies 597
widely across hesperornithiforms, from an elongate, laterally-compressed isosceles triangle in 598
Hesperornis regalis to a nearly equilateral pyramid in Baptornis advenus, the presence of a 599
robust, trihedral patella is unique among birds. 600
28

Triangular cnemial expansion on the tibiotarsus (Martin, 1983). A feature common to all 601
hesperornithiform birds - from the earliest named taxa such as H. regalis and B. advenus to more 602
recently described taxa such as Pasquiaornis - is an elongate triangular projection extending 603
proximally from the cranial edge of the proximal articular surface of the tibiotarsus. While the 604
dimensions and the degree of development vary across taxa, the presence of a distinctive, 605
triangular projection is definitive of the order. However, referring to this feature as a ‘cnemial 606
crest’ (as, for example, in Martin, 1983 pp. 313) is slightly problematic, in that accepted avian 607
nomenclature (Baumel and Witmer, 1993) identifies two cnemial crests present on the avian 608
tibiotarsus, the lateral and the cranial. Both of these crests are also present in hesperornithiforms, 609
but they originate on the lateral and medial sides of the cranially-located cnemial expansion and 610
extend distally down the cranio-medial and cranio-lateral shaft to varying degrees. Therefore, the 611
projection is not, in fact, either of the cnemial crests, but an extension of the distal end of the 612
tibia, on which these crests are present. As this configuration is not common among modern 613
birds, Baumel and Witmer (1993) do not address this in their terminology. Therefore, it is 614
recommended that this character be amended to refer to the feature in question as the cnemial 615
expansion rather than cnemial crest. Among other early birds Gansus also has a triangular 616
cnemial expansion, however it differs from that of hesperornithiforms in being more pyramidal, 617
with a prominent cranial, as well as proximal, component. 618
Tarsometatarsus possesses a sharp cranio-lateral ridge leading to the lateral trochlea 619
(Martin, 1984). This character is one of the most common traits used to identify a fossil 620
tarsometatarsus as hesperornithiform (for example: Rees and Lindgren, 2005; Dyke et al., 2996; 621
Bell and Everhart, 2009). This feature has been reported in Baptornis, Parahesperornis and 622
Hesperornis (Martin, 1984) as well as Enaliornis (Galton and Martin, 2002), Asiahesperornis 623
29

(Dyke et al., 2006), Brodavis (Martin et al., 2012), and Pasquiaornis (Sanchez, 2010). This ridge 624
is formed from the way in which the individually laterally-compressed metatarsals align along 625
their length, such that the lateral metatarsal has very high cranio-caudal relief as compared to the 626
other metatarsals. This configuration has also been referred to as ‘shingled metatarsals’ (Galton 627
and Martin, 2002). This feature is highly diagnostic of the group and can be readily used to 628
identify hesperornithiform bones from any of the included families. It should be noted however, 629
that the extent of the cranially projecting portion of metatarsal IV varies widely across taxa. It is 630
restricted to only the proximal portion of the shaft in Pasquiaornis and Brodavis, while the entire 631
length of metatarsal IV is projected cranially in Hesperornis. Therefore, this feature should be 632
amended to “tarsometatarsus possess a sharp craniolateral ridge along a portion of the shaft.” 633
Compressed tarsometatarsus (Martin, 1984). The tarsometatarsus of hesperornithiforms 634
is one of a number of clear indications of the diving lifestyle of these birds (for example: Marsh, 635
1875, 1880; Martin and Tate, 1976; Galton and Martin, 2002). As some features of this bone also 636
differ from those of modern diving birds (Galton and Martin, 2002) and it is one of the most 637
commonly preserved elements of hesperornithiforms, it is important that any diagnostic 638
information from the tarsometatarsus be fully utilized. However, using a term like ‘compressed’ 639
without assigning any more specific meaning is less than desirable. It appears that the 640
compression is the result of the shaft of each individual metatarsal having a midshaft cross- 641
section that is highly ovoid, nearly elliptical, with the dorso-plantar axis longest. In this way, 642
when arranged next to each other the entire tarsometatarsus appears compressed. Therefore, this 643
feature should be amended to “metatarsals with ovoid cross-sections that are dorso-plantarly 644
elongated.” 645
646
30

Family Baptornithidae 647
Small pit lies directly anterior to the diapophysis in the thoracic vertebrae (Nessov and 648
Borkin, 1983). This character was identified to unite Judinornis nogontsavensis, an isolated 649
vertebrae from Mongolia, with the baptornithids (translated in Kurochkin, 2000). While the 650
holotype of Judinornis has not been figured in any publication, a small, round pit anterior to the 651
transverse processes of the thoracic vertebrae is visible in specimens of Baptornis. However, 652
Hesperornis and Parahesperornis possess an analogous pit that is triangular in shape. 653
Unfortunately the known, published vertebrae for Pasquiaornis and Enaliornis are not well 654
enough preserved to observe this feature. At this time, it appears that the presence of a small, 655
round pit is unique to Baptornis while a larger, triangular pit is present in hesperornids. This 656
feature should be amended to “Small, round pit anterior to diapophysis in thoracic vertebrae.” 657
Pelvis with the preacetabular portion of the ilium relatively longer than in 658
hesperornithids (Martin and Tate, 1976). A similar feature, elongation of the pre-acetabular 659
ilium, was proposed as diagnostic for the Hesperornithiformes (Martin and Tate, 1976). That 660
feature is discussed in the morphometric analyses presented in a subsequent chapter. As this 661
feature is not an independent apomorphy of the specimens within this family, it is not supported 662
as a robust diagnostic character. 663
Medial and lateral trochlear ridges are more pronounced on trochlea III than trochlea 664
IV (Everhart and Bell, 2009). This character was proposed as an alternative to less specific 665
characters dealing with toe-rotation (Everhart and Bell, 2009; Martin and Tate, 1976). As toe- 666
rotation is simply inferred from the morphology of the feet, this character is preferable to 667
previous statements regarding the inferred degree of toe-rotation possible in various species. All 668
hesperornithiforms possess well-defined trochlear ridges on both trochlea III and IV. In 669
31

Baptornis, the ridges of trochlea III are more defined than those of trochlea IV. This also appears 670
to be the case in Pasquiaornis, however the studied material is very poorly preserved, making 671
observation of such detail difficult. This morphology differs greatly from hesperornids, where 672
the ridges of trochlea IV are greatly enlarged relative to trochlea III. Enaliornis is differs from 673
other hesperornithiforms in that trochlea III and IV appear to be similarly defined, however the 674
poor preservational quality makes determining this difficult. 675
Intercotylar prominence of the tarsometatarsus is present as a low, rounded bump 676
approximately centered over metatarsal III (Everhart and Bell, 2009). This character was used to 677
unite an isolated, previously undescribed tarsometatarsus with Baptornis (Everhart and Bell, 678
2009). Pasquiaornis, which has been included in the Baptornithidae, also has a reduced 679
intercotylae eminence (Sanchez, 2010). While Galton and Martin (2002) described the 680
intercotylar eminence of Enaliornis as ‘low and gently rounded’ (p 505), this feature in 681
Baptornis and Enaliornis is not similar. The intercotylar eminence in Enaliornis is so reduced as 682
to appear almost absent. This character, therefore, appears to have some utility in separating 683
baptornithids from other hesperornithiforms, however it must be noted that the state in 684
baptornithids is intermediate to that of Enaliornis and hesperornithids. 685
Genus Baptornis. 686
None of the proposed diagnostic characters for the genus Baptornis are supported 687
Baptornis advenus. 688
Diagnosis is currently the same as that of the genus. 689
Baptornis varneri. 690
See Brodavis varneri. 691
Genus Judinornis. 692
32

None of the proposed diagnostic characters for the genus Judinornis are supported 693
Judinornis nogontsavensis. 694
Diagnosis is currently the same as that of the genus. 695
Genus Pasquiaornis 696
None of the proposed diagnostic characters for the genus Pasquiaornis are supported. 697
Pasquiaornis hardei 698
None of the proposed diagnostic characters for Pasquiaornis hardei are supported. 699
Pasquiaornis tankei 700
None of the proposed diagnostic characters for Pasquiaornis tankei are supported. 701
702
Family Brodavidae 703
None of the proposed diagnostic characters for the family Bordavidae are supported. 704
Genus Brodavis 705
Diagnosis is currently the same as that of the family. 706
Brodavis americanus 707
None of the proposed diagnostic characters for Brodavis americanus are supported. 708
Brodavis baileyi 709
None of the proposed diagnostic characters for Brodavis baileyi are supported. 710
Brodavis mongoliensis 711
None of the proposed diagnostic characters for Brodavis mongoliensis are supported. 712
Brodavis varneri 713
Midpoint of tarsometatarsal shaft narrows to waist (Martin and Cordes-Person, 2007). 714
The tarsometatarsus of B. varneri appears waisted in dorsal and plantar view, especially as 715
33

compared to other hesperornithiforms. This may in part be exacerbated by the taphonomic 716
deformation of this element (Martin and Cordes-Person, 2007), however it can still be considered 717
diagnostic at this time. 718
719
Family Enaliornithidae 720
None of the proposed diagnostic characters for the Enaliornithidae are supported. 721
Genus Enaliornis 722
Diagnosis is currently the same as that of the family. 723
Enaliornis barretti 724
None of the proposed diagnostic characters for Enaliornis barretti are supported. 725
Enaliornis sedgewicki 726
None of the proposed diagnostic characters for Enaliornis sedgewicki are supported. 727
Enaliornis seeleyi 728
None of the proposed diagnostic characters for Enaliornis seeleyi are supported. 729
730
Family Hesperornithidae 731
Medial trochlear ridge of metatarsal IV enlarged relative to lateral (Bell and Everhart, 732
2009). This feature was erected to better qualify earlier observations by Martin (1984) regarding 733
the extreme form of the articulation of the tarsometatarsus with the fourth toe, which added a 734
medio-lateral component of motion to the toe joint in addition to the traditional dorso-plantar 735
motion. In Hesperornis, as well as Parahesperornis and Asiahesperornis, the medial trochlear 736
ridge of trochlea IV is lengthened dorso-plantarly. This extreme discrepancy between the lateral 737
and medial ridges of the fourth trochlea is unique to hesperornithids. 738
34

Fourth metatarsal trochlea wider than the third (Tokaryk et al., 1997). In all species of 739
hesperornithids trochlea IV is nearly twice as wide as trochlea III, supporting the use of this 740
feature as diagnostic of the family. 741
Genus Asiahesperornis 742
Prominent medial and lateral grooves are present on the medial portion of the dorsal 743
surface of the tarsometatarsus (Nessov and Prizemlin, 1991 as translated by Kurochkin, 2000). 744
In most hesperornithiforms grooves exist along the tarsometatarsus shaft separating the 745
metatarsals. In all hesperornithid taxa these grooves are particularly well-developed and extend 746
the length of the shaft. In Asiahesperornis, the lateral groove appears to be better developed than 747
in other species of Hesperornithdae, such as Hesperornis regalis or Parahesperornis alexi. 748
Therefore, the possession of a more prominent lateral groove is supported as diagnostic of 749
Asiahesperornis, while the medial groove does not differ from the state in Hesperornis. 750
Asiahesperornis bahazanovi 751
Diagnosis is currently the same as that of the genus. 752
Genus Canadaga 753
Concavitas lateralis large and deep, occupies entire lateral face of centrum (Hou, 1999). 754
The posterior cervical vertebrae and thoracic vertebrae of many hesperornithiforms are marked 755
on the sides by deep concavities. Canadaga is unique in the sharp definition of the borders of 756
these cavities, having well-delineated edges and being nearly perfectly ovoid in shape. In 757
Hesperornis the lateral concavity is much less defined, merging out into the centrum on the 758
ventral and posterior sides with no defined boundary. Furthermore, the lateral concavity of 759
Hesperornis takes up around 75% of the lateral side of the centrum, while that of Canadaga 760
takes up nearly the entire lateral surface. 761
35

Canadaga arctica 762
Diagnosis is currently the same as that of the genus. 763
Genus Hesperornis 764
None of the proposed diagnostic characters for Hesperornis are supported. 765
Hesperronis bairdi 766
None of the proposed diagnostic characters for Hesperornis bairdi are supported. 767
Hesperornis chowi 768
None of the proposed diagnostic characters for Hesperornis chowi are supported. 769
Hesperornis crassipes 770
Five articular projections for rib attachment on the sternum (Marsh, 1876). The holotype 771
of H. crassipes (YPM 1474), includes a well-preserved sternum, which Marsh figured (1880, 772
plate VII). Sternums are rare in other species of Hesperornis, however Marsh provided 773
measurements on YPM 1206 and figures of an unidentified specimen that show four articular 774
projections for rib attachments. As YPM 1206 was not present in the collection during any of the 775
research visits, this was impossible to confirm. If accurate, then this feature can be retained as 776
diagnostic of H. crassipes. 777
Hesperornis gracilis 778
None of the proposed diagnostic characters for Hesperornis gracilis are supported. 779
Hesperornis macdonaldi 780
None of the proposed diagnostic characters for Hesperornis macdonaldi are supported. 781
Hesperornis mengeli 782
None of the proposed diagnostic characters for Hesperornis mengeli are supported. 783
Hesperornis regalis 784
36

None of the proposed diagnostic characters for Hesperornis regalis are supported. 785
Hesperornis rossicus 786
Trochlear condyle of metatarsal II completely behind that of digit III (Panteleyev et al., 787
2004). This feature is similar to that later reported for H. mengeli (Martin and Lim, 2002), 788
however the state in H. rossicus is clearly more developed than that in H. mengeli. In dorsal 789
view, trochlea II is completely hidden behind trochlea of III and IV, while in H. mengeli trochlea 790
II is illustrated as being partially visible in dorsal view (Martin and Lim, 2002; Fig. 5). This 791
character is therefore apomorphic of H. rossicus, as it is not observed in any other species of 792
hesperornithiform for which sufficient material is preserved. 793
Genus Parahesperornis 794
Lacrimal more elongated dorso-ventrally than in Hesperornis (Martin, 1984). The 795
lacrimal of Hesperornis possesses a robust body with a relatively short, stout, dorsally-curved 796
descending process, while in Parahesperornis the descending process is much longer, straighter, 797
and proportionally thinner. 798
799
37

APPENDIX II 800
801
This study has identified a number of features previously proposed as diagnostic of various 802
hesperornithiform taxonomic units that are not valid. Each unsupported feature (summarized in 803
Table 2) is discussed here within the current taxonomic framework. 804
805
ORDER HESPERORNITHIFORMES 806
807
Thoracic vertebrae heterocoelus (Martin and Tate, 1976). The state of this character has 808
been somewhat confused. Initially, Martin and Tate reported that both Hesperornis and 809
Baptornis have heterocoelus vertebrae, while those of Enaliornis are amphicoelus (1976). Later 810
work described the vertebrae of Enaliornis as heterocoelus cervically with the posterior thoracic 811
vertebrae amphicoelus (Martin, 1983), while still later work described the vertebrae as fully 812
heterocoelus (Galton and Martin, 2002). Analysis in the course of this study confirmed that all 813
vertebrae of Enaliornis are heterocoelus. However, in Pasquiaornis the known thoracic vertebrae 814
are amphicoelus (Sanchez, 2010). The retention of the primitive state in Pasquiaornis invalidates 815
the use of the advanced condition, heterocoelus vertebrae, as diagnostic of a group including 816
Pasquaiornis and other hespernrothiforms. 817
Reduced quadrate pneumaticity (Martin and Tate, 1976). Evaluating this character is 818
difficult, as Elzanowski has shown that the quadrate described and figured by Martin and Tate 819
(1976, Fig. 1 c-e) not only does not belong with the specimen to which it was attributed (AMNH 820
5101), it is also most likely not Baptornis advenus and at this time it is impossible to definitively 821
state where the quadrate used by Martin and Tate (1976) came from (2000). Furthermore, 822
38

Elzanowski (2000) claimed that the preserved portion of the Baptornis quadrate presented by 823
Martin and Tate (1976) was not well enough preserved to determine pneumaticity. Other 824
hesperornithids that preserve a quadrate include Hesperornis, which has a pneumatic quadrate 825
(Witmer, 1990), and Pasquiaornis, which has an apneumatic quadrate (Tokaryk et al., 1997). 826
Therefore, it appears that pneumaticity varied across the Hesperornithiformes and so cannot be 827
used as a diagnostic feature of the group. 828
Quadrate with undivided head (Marsh, 1877). While Marsh based the use of this feature 829
on its presence only in species of Hesperornis (1877), later work identified this feature in 830
Potamornis (Elzanowski, 2000), and this study further identified an undivided quadrate head in a 831
specimen assigned to Pasquiaornis (Tokaryk, pers. comm.). In Enaliornis the head of the 832
quadrate has been reported as being divided into two separate articular surfaces for the otic and 833
squamosal, a state similar to that in modern birds (Elzanowski and Galton, 1991). However, this 834
is an interpretation based on the appearance of the articular surface for the quadrate head in the 835
braincase of Enaliornis (SMC B54404), as the quadrate itself is unknown. Therefore, while an 836
undivided quadrate head may not be used to unite all members of the Hesperornithiformes, this 837
should be regarded as conditional, depending on indirect observation. 838
Combination of short, broad pterygoids with long, narrow palatines (Martin, 1984). This 839
character was initially described by Gingerich (1976) from observations of Hesperornis regalis 840
and later used as diagnostic of the larger group Hesperornithiformes (Martin, 1984). However, 841
extrapolating this morphology to the entire order is inappropriate, as the relevant elements are 842
only preserved for Hesperornis and Parahesperornis. Therefore, it is recommended that this 843
character be removed from the diagnosis of the Hesperornithiformes. It may, however, prove 844
useful as a diagnostic feature of a more exclusive grouping. 845
39

Unfused mandibular symphysis (Martin and Tate, 1976). While first proposed to unite 846
Baptornis and Hesperornis (Martin and Tate, 1976), the necessary elements are only preserved 847
in Hesperornis and Parahesperornis, not Baptornis or any other hesperornithiform. Therefore 848
this feature cannot be extrapolated to the entire order, but instead may prove diagnostic of a more 849
exclusive group. 850
Maxillokinesis (Martin and Tate, 1976). Gingerich initially described the articulation of 851
the jaw of Hesperornis regalis as displaying akinesis (1973) and later revised that to possibly 852
displaying a unique form of maxillokinesis (1976). This feature was later used as diagnostic of 853
the Hesperornithiformes by numerous subsequent papers (for example, Martin and Tate, 1976; 854
Martin, 1984; Galton and Martin, 2002). This is inappropriate, as Gingerich (1976) was only 855
describing the state in Hesperornis regalis, as no other hesperornithiform material preserves the 856
appropriate elements. Therefore, it is recommended that this feature be removed from the 857
diagnosis of the order Hesperornithiformes, while it may prove diagnostic of a more exclusive 858
taxonomic grouping. 859
Acetabulum partly closed (Marsh, 1880). Marsh described the walls of the acetabulum as 860
being narrowed by ossification, but still perforated to a much smaller diameter than known for 861
other birds (1880). This feature was also reported from Baptornis (Martin and Tate, 1976) and 862
Enaliornis (Galton and Martin, 2002). However, after observation of multiple pelvises from a 863
wide variety of hesperornithiforms, including the specific specimens referred to in earlier 864
publications, it is nearly impossible to say to what degree the acetabulum was perforated, as all 865
specimens are broken to some degree. Almost always, the natural edge of the interior perforation 866
of the acetabulum is broken; thus the acetabula could have been completely closed, partly closed, 867
or partly open. Additionally, in Pasquiaornis the acetabulum is clearly open and more closely 868
40

resembles the morphology seen in modern birds. Therefore this feature is unsupported due to 869
both the inapplicability of the feature where it is currently used in the taxonomic framework as 870
well as the lack of definitive fossil evidence. 871
Wings reduced (Marsh, 1877). This is one of the earliest proposed diagnostic features of 872
hesperornithiforms (Marsh, 1877). While it is true that most hesperornithiforms have reduced 873
wings, the degree of reduction varies widely across taxa. Hesperornis exhibits the highest 874
reduction of the wing, such that it has been questioned whether an ulna and radius were even 875
present (Martin and Tate, 1976). Baptornis shows a slightly more developed wing, with a more 876
complex humerus as well as the presence of an ulna and radius (Martin and Tate, 1976). 877
Pasquiaornis exhibits only slight reduction of the wing, having a humerus very similar to that of 878
flighted birds (Tokaryk et al., 1997). Thus a generic statement of ‘reduced wings’ is not useful as 879
a diagnostic feature. Specific characters of individual wing elements that contribute to the overall 880
reduction of the wing would better serve for diagnosis, should any be identified. 881
Coracoid with glenoid facet located on the tip of the scapular end (Martin and Tate, 882
1976). In Baptornis advenus and Hesperornis regalis the articular surface for the humerus is 883
displaced proximally to the very end of the coracoid (Martin and Tate, 1976), as is also the case 884
in Parahesperornis. However, the coracoid of Pasquiaornis appears much more similar to that of 885
modern birds. The omal articulation is located more distally on the dorsal surface of the proximal 886
coracoid. Therefore, this feature is not diagnostic to a group inclusive of both Pasquiaornis and 887
other hesperornithiforms. 888
Humerus lacking distinct distal condyles (Martin, 1984). The distal aspect of the humeri 889
of hesperornithiform taxa vary widely. For Hesperornis and Baptornis, the distal end of the 890
humerus fits this description, as even Baptornis, with its better-developed humerus, lacks 891
41

distinctive distal morphology (Martin and Tate, 1976). However, the humerus of Pasquiaornis is 892
much better developed and retains features similar to those found in flighted birds, such as distal 893
condyles (Tokaryk et al., 1997). Therefore, this character cannot be used to unite all proposed 894
hesperornithiform taxa. Pending the identification of other diagnostic features, this character may 895
prove diagnostic of a more exclusive group. 896
Clavicles unfused (Marsh, 1880). While initially identified by Marsh (1880), this feature 897
was first used to unite Hesperornis and Baptornis (Martin and Tate, 1976). However, to date 898
clavicles have not been discovered in any taxa outside of the genus Hesperornis (including 899
Baptornis). Therefore, the use of this character as diagnostic is not justified. 900
Posterior extremities of the ilium, ischium, and pubis separate (Marsh, 1880). The free 901
distal ends of the main elements of the pelvis were first reported by Marsh, who noted the overall 902
similarity of the pelvis of Hesperornis to that of a modern loon (Podiceps) (1880). Marsh also 903
noted that separation of the ilium, ischium, and pubis is present in modern foot-propelled divers 904
as well as some ratites (Marsh, 1880). Martin and Tate were the first to use this feature as 905
diagnostic of the Hesperornithiformes (1976). However, as this feature is also found in other 906
primitive birds, such as Apsaravis (Clarke and Norell, 2002), it is not recommended for use as a 907
diagnostic feature of the Hesperornithiformes. 908
Tibiotarsus lacking supratendinal bridge (Marsh, 1880). The lack of a supratendinal 909
bridge was first recorded in Hesperornis regalis (Marsh, 1880) and later identified in Baptornis 910
advenus (Martin and Tate, 1976), Enaliornis (Galton and Martin, 2002), Pasquiaornis (Sanchez, 911
2010), and other species of Hesperornis. However, many early birds such as Ichthyornis dispar 912
(Clarke, 2004), Patagopteryx (Alvarenga and Bonaparte, 1992), Gansus (You et al., 2006), and 913
42

others (Chiappe, 1996) also lack a supratendinal bridge, making this feature plesiomorphic and 914
unsuitable for use as a diagnostic feature of the Hesperornithiformes. 915
Tarsometatarsus lacking hypotarsal grooves and proximal foramina (Martin and Tate, 916
1976). The proximal end of the tarsometatarsus of hesperornithid birds differs from that of 917
modern birds in not having prominent hypotarsal crests (Marsh, 1880). Rather, the plantar 918
proximal surface possesses a roughened triangular region centered on the third metatarsal, most 919
prominently seen in Hesperornis (Marsh, 1880), but also present in Parahesperornis, Baptornis, 920
Enaliornis, Pasquiaornis and others. Furthermore, a similar hypotarsal-like region has been 921
documented in other Mesozoic birds such as Apsaravis (Norrell and Clarke, 2001) and 922
Patagopteryx (Chiappe , 1996). Therefore, while hesperornithiforms do lack a well-developed 923
hypotarsus, the absence of grooves is plesiomorphic and not diagnostic of the group. 924
Furthermore, Martin and Tate (1976) claimed that Baptornis and Hesperornis do not possess 925
proximal foramina, forming part of the basis of this proposed character. However, further 926
analysis of a number of specimens of both taxa, as well as re-examination and additional 927
preparation of the specimens cited by Martin and Tate (1976), revealed that proximal foramina 928
are, in fact, present in these birds, including Baptornis and Hesperornis (Clarke, 2004; Everhart 929
and Bell, 2009) as well as other Mesozoic birds such as Apsaravis (Norrell and Clarke, 2001). 930
Therefore this feature is also plesiomorphic and this character cannot be retained. As this 931
character was the sole justification for exclusion of Neogaeornis from the Hesperornithiformes 932
(Olson, 1992), the status of this South American taxon should be re-evaluated. 933
Fourth toe longest (Martin, 1984). This character must have been inferred from some 934
undefined feature of the tarsometatarsus by Martin (1984), as hardly any specimens preserve all 935
pedal phalanges. Identifying the actual morphology on which the inference is based as a 936
43

diagnostic feature is preferable to using an inference. In all observed hesperornithiforms for 937
which pedal phalanges are preserved, the first phalanx of the fourth digit is of a similar length as 938
those of the other toes. While it may be tempting to infer a greater length for the fourth toe given 939
the more robust fourth trochlea of the tarsometatarsus in some, but not all species of 940
hesperornithiform, this inference is unfounded. 941
Feet lobed (Martin, 1984). Modern grebes, unlike other diving birds, possess partially 942
webbed feet with broad lobes on the separate portions of the toes. However, there is no direct 943
evidence hesperornithiforms had lobed feet. Rather, the possession of lobed (as opposed to 944
webbed or separated) toes is inferred from the articular structure of the tarsometatarsus with the 945
first pedal phalanx (Martin and Tate, 1976). This morphology consists of highly disproportionate 946
sizes between the medial and lateral articular surfaces of the trochlea of the fourth metatarsal. 947
This discrepancy occurs to varying degrees throughout the hesperornithiforms – the most 948
dramatic being the state in Hesperornis and Parahesperornis, and to lesser degrees in Baptornis 949
(Martin and Tate, 1976), Enaliornis (Galton and Martin, 2002), and Pasquiaornis (pers. obs.). 950
Whether or not the feet were lobed, the observable morphology is certainly a useful diagnostic 951
character, while the inferred tissue morphology is not. Therefore, it is recommended that this 952
character be amended to reflect the actual morphology of the tarsometatarsus. 953
954
Family Baptornithidae 955
Coracoid more slender than in hesperornithids (Martin and Tate, 1976). Coracoids are 956
known for Hesperornis, Parahesperornis, Baptornis, and Pasquiaornis. Martin and Tate (1976) 957
described the coracoid of Baptornis as much more slender than that of Hesperornis. While the 958
coracoids of Baptornis and Pasquiaornis are certainly more slender than those of Hesperornis 959
44

and Parahesperornis, this is a plesiomorphic condition unsuitable for use as a diagnostic feature. 960
The presence of a robust coracoid may, however, prove diagnostic for Hesperornis + 961
Parahesperornis. 962
Uncinate processes of the ribs turned dorsally (Martin and Tate, 1976). In most modern 963
and basal birds, including hesperornithiforms, the uncinate processes are not fused to the rib, but 964
rather articulate to the rib’s outer surface. In Baptornis, the uncinate processes extend away from 965
the shaft of the rib at an angle, such that they are directed dorsally (Martin and Tate, 1976). In 966
Hesperornis the uncinate processes extend outward from the rib more laterally (Martin and Tate, 967
1976). Unfortunately uncinate processes are not preserved with any other baptornithid taxa, and 968
so the use of this feature as diagnostic of the family is inappropriate. This feature may prove 969
diagnostic at the genus or species level. 970
Patella pyramidal in shape (Martin and Tate, 1976). While all hesperornithiforms have a 971
much more robust patella than that found in other birds (see above), that of Baptornis differs in 972
being pyramidal, with the height and width of the base of roughly similar dimensions (Martin 973
and Tate, 1976). This differs from the state in Hesperornis and Parahesperornis, where the 974
patella forms an isosceles triangle, being extremely elongate and laterally compressed. 975
Unfortunately the patella is unknown for Pasquiaornis, making the use of this feature at the level 976
of the family Baptornithidae inappropriate. 977
Pygostyle long and laterally compressed (Martin and Tate, 1976). Marsh originally 978
described the pygostyle of Hesperornis regalis as possessing three to four fused vertebrae 979
(1880). Martin and Tate later contradicted this, claiming only two centra were fused into the 980
pygostyle (1976). They went on to report five fused vertebrae in the pygostyle of Baptornis, thus 981
claiming a longer pygostyle was diagnostic of the Baptornithidae (Martin and Tate, 1976). It is 982
45

unclear why Marsh’s (1880) original description differs from later observations of Martin and 983
Tate (1976). Neither paper lists which specimens were used for study, so it is difficult to check 984
reported observations. It is possible that, for H. regalis, Martin and Tate (1976) were looking at 985
the holotype YPM 1200, which includes a very poorly preserved pygostyle consisting of two 986
centra. However, this element is clearly broken on the posterior end and does not preserve the 987
entire element, so it cannot be used for this character. The University of Nebraska State Museum 988
has a cast of an H. regalis pygostyle that contains 3 fused centra, further supporting Marsh’s 989
original observations (the original specimen from which this cast was made is unkown). For 990
Baptornis, the pygostyle of UNSM 20030 is broken at both ends and only preserves 3 centra, 991
while the pygostyle of FMNH 395 could not be located on a visit to the collection. Therefore it is 992
not possible to confirm the count of fused centra. Parahesperornis alexi possesses 3-4 fused 993
vertebrae in the pygostyle (KUVP 24090). It is most likely that poor preservation makes 994
accurately counting the number of vertebrae difficult. At this time, it is impossible to confirm or 995
reject this proposed character. Due to the minimal number of specimens for which it is even 996
possible to evaluate the complete pygostyle, this character is of low utility. As for the proposed 997
lateral compression of the pygostyle, observation of the few available specimens (UNSM 20030, 998
YPM 1200, KUVP 24090) does not reveal any significant compression. 999
Margin of the distal foramen of the tarsometatarsus is delineated by indentations in the 1000
necks of the trochlea (Everhart and Bell, 2009). This feature was erected to assign a partial 1001
isolated tarsometatarsus to Baptornis (Everhart and Bell, 2009). In the holotype of Baptornis 1002
advenus (YPM 1465), the intertrochlear incision between trochlea III and IV on the dorsal side 1003
of the bone is a wide ovoid space bordered by deep indentations in the necks of trochlea III and 1004
IV. It appears that just within specimens assigned to Baptornis there is a high degree of variation 1005
46

in the shape of the intertrochlear incision, thus invalidating this feature as diagnostic of the 1006
group. 1007
Circular pits in the articular surfaces of the centra of thoracic vertebrae (Martin and 1008
Bonner, 1977). This feature was identified in an immature specimen of Baptornis advenus 1009
(Martin and Bonner, 1977) and later used by Nessov and Borkin (1983) to unite Judinornis with 1010
other baptornithids (translated by Kurochkin, 2000). However, it appears that the presence of 1011
this pit might be heavily reliant on taphonomy or perhaps age of the individual, as not all 1012
specimens of Baptornis advenus have these pits, and some, but not all, specimens of Hesperornis 1013
and Parahesperornis do preserve these pits. Therefore, this feature cannot be used as diagnostic 1014
of a group more inclusive than the order. 1015
Hypapophysis centered on thoracic vertebrae (Tokaryk and Harrington, 1992). This 1016
feature was identified by Tokaryk and Harrington (1992) as uniting an isolated vertebra from the 1017
Judith River Formation (Campanian) of Saskatchewan, Canada with the baptornithids. 1018
Unfortunately the figures provided do not illustrate the ventral surface, so it is not possible to 1019
directly evaluate this character from the literature (Tokaryk and Harrington, 1992). However, 1020
within both Baptornis and, to a greater degree, Hesperornis, the antero-posterior extent of the 1021
ventral process as well as the location on the ventral surface of the caudo-cervical and thoracic 1022
vertebrae varies with position. More anterior vertebrae tend to have more anteriorly placed 1023
processes, which become progressively centrally located on more posterior vertebrae. As it is 1024
difficult to exactly identify the placement of individual vertebrae and generalizations on vertebral 1025
morphology are not possible, this character is unsupported. 1026
Intracentral bones not fused to caudal vertebrae (Martin and Tate, 1976). The 1027
intracentral bones were described as fused to the caudal vertebrae in Hesperornis but not in 1028
47

Baptornis (Martin and Tate, 1976). Marsh (1880) made no mention of this feature and did not 1029
figure the caudal vertebrae of the Hesperornis as differing greatly from those of Baptornis (as 1030
figured by Martin and Tate, 1976 Fig. 6) in any respect other than the transverse processes. The 1031
intracentral bones are synonympous with the haemal processes of the posterior-most caudal 1032
vertebrae of modern birds (Baumel and Witmer, 1993). Evaluation of this feature is not possible 1033
in any hesperornithiform taxa encountered during the course of this study, as none of the caudal 1034
vertebrae are well enough preserved. 1035
Medial and lateral cotyla of the tarsometatarsus tilt dorsally (Everhart and Bell, 2009). 1036
In Baptornis advenus, the articular surfaces of both the medial and lateral cotyla slope distally at 1037
the dorsal-most edge of the articular surfaces. However, this is not the case in Pasquiaornis, 1038
where the medial and lateral cotyla are fairly evenly horizontal. Therefore, this feature should be 1039
considered diagnostic of the genus Baptornis and not a more inclusive family including 1040
Baptornis and Pasquiaornis. Outside of the Baptornithidae, most hesperornithiforms have a 1041
medial cotyla that slopes distally at the dorsal surface and a lateral cotyla that is fairly horizontal. 1042
Outer trochlea of tarsometatarsus not enlarged (Martin and Tate, 1976). In Baptornis, 1043
the trochlea of metatarsals II and III are similarly sized, as opposed to the greatly enlarged 1044
trochlea IV observed in other hesperornithiforms, such as Hesperornis (Marsh, 1877). However, 1045
this is also the case in other primitive birds and many modern birds. Therefore, this feature is 1046
plesiomorphic and cannot be used as diagnostic of the family. 1047
Distal foramen on tarsometatarsus an open groove (Martin and Tate, 1976). The distal 1048
tarsometatarsus of some specimens of Baptornis (for example, FHSM 6318) possess an open 1049
distal foramen that is not separated from the intertrochlear incision between trochlea III and IV. 1050
This foramen is closed and separated from the trochlear groove in Hesperornis and other 1051
48

specimens assigned to Baptornis (AMNH 5101). Martin and Tate identified an open distal 1052
foramen as diagnostic of Baptornis (1976), however as not all specimens display this 1053
morphology the feature as well as the specimen assignments need to be reevaluated. 1054
Furthermore, while Enaliornis has been reported to possess an open distal foramen as well 1055
(Galton and Martin, 2002), observations during the course of this study fail to confirm this, as all 1056
specimens are too poorly preserved to definitively describe the morphology. While Pasquiaornis 1057
has been reported as having an open distal foramen (Sanchez, 2010), a prominent closed distal 1058
foramen is clearly visible. As this character neither unites members of the Baptornithidae nor 1059
excludes potentially closely-related taxa, it should be removed from the list of diagnostic features 1060
of the Baptornithidae, as presently defined. 1061
Toe-rotation not well developed (Martin and Tate, 1976). As discussed above, 1062
determination of this capability is entirely subjective and unsuitable for use as a diagnostic 1063
feature. Rather, the morphology on which this interpretation is based should be used. This was 1064
the impetus for Everhart and Bell (2009) defining a more specific diagnostic feature, that the 1065
medial trochlear ridge of the fourth metatarsal is reduced relative to the lateral trochlea. 1066
Genus Baptornis 1067
Short sacrum (Lucas, 1903). Lucas tentatively described the sacrum of Baptornis as 1068
containing 10 vertebrae, while that of Hesperornis had 14 (1903). Unfortunately no other 1069
baptornithid taxa preserve a complete, or even nearly complete, sacrum for comparison. Within 1070
basal birds, a shorter sacrum is plesiomorphic (Clarke, 2004). In addition to Baptornis, Lithornis, 1071
Ichthyornis, Apsaravis, and Gobipteryx can all be interpreted as having a sacrum with 10 1072
vertebrae. Use of this character is therefore unsupported. 1073
49

Coracoid lacks procoracoid processs (Lucas, 1903). The absence of a procoracoid 1074
process in Baptornis advenus was described by Lucas (1903), however later work described the 1075
procoracoid as present but differing in morphology from that of Hesperornis (Martin and Tate, 1076
1976). Few coracoids are known for hesperornithiforms. It appears that the only difference 1077
between the humeral ends of the coracoids of Baptornis and Hesperornis is the more robust form 1078
present in Hesperornis. Because both Baptornis and Hesperornis lack a procoracoid process, this 1079
feature cannot be used as diagnostic of Baptornis. 1080
Peg-and-crescent articulations on the distal articular surface of metatarsal IV absent 1081
(Martin, 1984). While the general character of less-advanced toe-rotation was proposed as 1082
diagnostic of the larger group Baptornithidae, this character was proposed separately as 1083
diagnostic of the genus Baptornis (Martin and Tate, 1976). The former character is highly 1084
problematic, as discussed above, and the latter has been revised to describe the enlarged lateral 1085
trochlear ridge as compared to the medial (Everhart and Bell, 2009). As that character has been 1086
shown to prove diagnostic for the larger group Baptornithidae (App. I), this character should be 1087
removed as diagnostic for the genus Baptornis. 1088
Baptornis advenus. 1089
Diagnosis is currently the same as that of the genus. 1090
Baptornis varneri. 1091
See Brodavis varneri (Martin et al., 2012). 1092
Genus Judinornis. 1093
The ventral side of the centrum is distinctly narrowed in the middle but very broad 1094
caudally (Nessov and Borkin, 1983). On all hesperornithiform birds, the ventral surface of the 1095
centra of thoracic vertebrae narrows to a waist and then expands cranially and, usually to a 1096
50

greater degree, caudally. The degree of this ‘waisting’ varies with the position of the vertebrae 1097
and across taxa. Therefore, this feature is not a useful diagnosis for Judinornis. 1098
Articular surfaces of the centra of thoracic vertebrae are trapezoidal in shape and extend 1099
transversely (Nessov and Borkin, 1983). This feature is not unique to Judinornis, as all 1100
hesperornithiforms have trapezoidal cranial and caudal articular surfaces on the thoracic 1101
vertebrae, with the dorsal and ventral margins parallel and the sides angled toward each other 1102
dorsally. The ventral portion of the articular surfaces also extends laterally in all 1103
hesperornithiforms. The only taxa for which this might differ is Pasquiaornis, which have 1104
amphicoelus vertebrae, however the edge of the articular surfaces are too poorly preserved to 1105
describe the shape. 1106
Cranial zygapophyses are located low together on the midline (Nessov and Borkin, 1107
1983). In the thoracic vertebrae of Baptornis, the cranial zygapophyses are also located fairly 1108
low on the vertebra, as compared to the caudal zygapophyses, however they are broadly spaced. 1109
The lateral margins of the zygapophyses extend past that of the cranial articular surface. This is 1110
also the case in Hesperornis, however in both taxa the degree of separation does vary with the 1111
position of the vertebra, so care must be taken with comparisons. It appears that the cranial 1112
zygapophyses are not preserved in Judinornis, making this character invalid (Fig. 4, Nessov, 1113
1992) 1114
Judinornis nogontsavensis. 1115
Diagnosis is currently the same as that of the genus. 1116
Genus Pasquiaornis 1117
Proximal end of the femur relatively less expanded latero-medially than in B. advenus 1118
(Tokaryk et al., 2007). Hesperornithiform birds are characterized by having an enlarged 1119
51

trochanteric crest on the proximal femur. It has been proposed that this crest is not as laterally 1120
projected in Pasquiaornis (Tokaryk et al., 2007), which contributes to an overall less expanded 1121
proximal end. There are a number of problems with this use of this feature as diagnostic. First, 1122
the expansion of the trochanteric crest is derived within hesperornithiforms, and so the reduced 1123
state of this character is plesiomorphic and inappropriate for use as diagnostic. Additionally, 1124
diagnostic features should be independent of the state in other taxa, which this state is not (it 1125
relies on comparison to the state in B. advenus. For these reasons, this feature is not accepted as 1126
diagnostic at this time. 1127
Trochanteric crest of femur closer to the shaft than in Baptornis (Tokaryk et al., 1997). 1128
This character overlaps with another proposed for Pasquiaornis regarding the total medio-lateral 1129
expansion of the proximal end, to which the trochanteric crest is the main contributor. Therefore, 1130
this feature is redundant and will not be further considered 1131
Intercotylar eminence of the tarsometatarsus anteriorly placed, such that it overhangs the 1132
shaft (Tokaryk et al., 1997). Tokaryk and Harrington (1997) claim the intercotylar eminence of 1133
the proximal tarsometatarsus is centered over the shaft in B. advenus but shifted dorsally such 1134
that it overhangs the face of the shaft in Pasquiaornis. The differences between taxa in this 1135
regard are minimal, as all have equivalent degrees of overhang of the tarsometatarsus, thus 1136
invalidating this feature. 1137
Pasquiaornis hardei 1138
Distal rim of femoral head perpendicular to shaft (Tokaryk et al., 1997). In 1139
hesperornithiforms, the head and neck of the femur project medially, with the orientation of the 1140
head varying among some taxa. Examination of P. hardei and B. advenus show virtually the 1141
52

same feature, making it unclear how these taxa differ. Furthermore, B. advenus appears to have 1142
the distal rim of the head more perpendicular to the shaft than either species of Pasquiaornis. 1143
Medial cotyla of tarsometatarsus deflected toward the shaft (Tokaryk et al., 1997). It is 1144
unclear how the medial cotyla on the proximal surface of the tarsometatarsus can be deflected 1145
toward the shaft of the tarsometatarsus. As a character of the other species of Pasquiaornis, P. 1146
tankei, is the dorso-plantar alignment of the cotyla, it seems reasonable that the state of the 1147
medial cotyla in P. hardiei must deviate from this alignment in some way, however this is not 1148
shown in any of the published literature (Tokaryk et al., 1997; Sancchez, 2010). Furthermore, 1149
none of the tarsometatarsi observed during the course of this study showed significant 1150
differences in the alignment of the cotylae and all were eroded to some degree, which affected 1151
their appearance such that drawing conclusive comparisons is not possible. 1152
Neck of the trochlea of metatarsal III higher anteriorly than that of IV (Tokaryk et al., 1153
1997). Evaluating this feature is difficult, as the appropriate avian terminology for the 1154
tarsometatarsus does not use the relative term ‘anteriorly’ (Baumel and Witmer, 1993). If by 1155
anterior Tokaryk et al. (1997) were referring to the side of the bone facing toward the skull (the 1156
anterior-most point of the body), then this would be analogous to the dorsal surface of the bone 1157
(Baumel and Witmer, 1993). If that is the case, then this feature is not supported. While 1158
specimens preserving both metatarsal trochlea are rare, in P. hardei trochlea IV appears to 1159
extend further dorsally than trochlea III, as is also the case in Enaliornis. Therefore, this feature 1160
is not supported. 1161
Pasquiaornis tankei 1162
Distal rim of femoral head slanted toward shaft (Tokaryk et al., 1997). As discussed 1163
above for the complementary state of this feature in P. hardiei, the identification of differences 1164
53

in state between P. hardiei, P. tankei, and B. advenus is virtually impossible. 1165
Medial cotyla of tarsometatarsus nearly antero-posteriorly aligned (Tokaryk et al., 1166
1997). The use of the term ‘antero-posterior’ (Tokaryk et al., 1997, pp. 173) is difficult to 1167
interpret, however, as discussed above, there is no discernible difference between the taxa that 1168
cannot be separated from erosion and poor preservation. 1169
Neck of the trochlea of metatarsal IV higher anteriorly than that of III (Tokaryk et al., 1170
1997). This feature is difficult to interpret for the reasons discussed above with the alternative 1171
state proposed for P. hardei. As the state described here is also that observed in P. hardei and 1172
Enaliornis, this feature is not diagnostic of a particular species. 1173
1174
Family Brodavidae 1175
No diagnostic features of the family Brodavidae were found to be unsupported, however a 1176
number of characters will be further explored in the morphometrics chapter. 1177
Genus Brodavis 1178
Diagnosis is currently the same as that of the family. 1179
Brodavis americanus 1180
Anterior-distal surface of the trochlea of metatarsal IV is broad and flat (Martin et al., 1181
2012). As the distal end of trochlea IV does not appear to be preserved (Martin et al., 2012; Fig. 1182
1D), it is difficult to understand how this character was diagnosed. The area in question is 1183
preserved in B. varneri as well as numerous other hesperornithiforms, where it does appear to be 1184
relatively broad and flat in B. varneri but marked by an incision between the trochlear ridges in 1185
B. advenus as well as in H. regalis. Furthermore, it cannot be ruled out that the smoothness of B. 1186
varneri is simply a result of the highly eroded surface of the bone. As this feature is not 1187
54

preserved in the only specimen of B. americanus (RSM P2315.1), it cannot be used as diagnostic 1188
of the species. 1189
Brodavis baileyi 1190
The outer anterior ridge of the tarsometatarsal shaft extends further distally (Martin et 1191
al., 2012). Interpreting this character is difficult, as the ‘anterior’ ridge (here assumed to be 1192
synonymous with the dorsal ridge) is formed from the entire shaft of metatarsal IV. Thus it is 1193
unclear how this ridge could not extend to the trochlea distally. If the authors were referring to a 1194
minimum relief of the ridge, this should be more carefully specified. In both B. baileyi and B. 1195
mongoliensis, as well as hesperornithids, the lateral margin of metatarsal IV in dorsal view 1196
appears to form a sharper ridge near the distal end than in B. americanus. Baptornithids have a 1197
morphology more similar to that of B. americanus, where the dorsal face of the distal metatarsal 1198
shaft flattens out just proximal to the trochlea. Therefore, this feature may be reconsidered as 1199
potentially diagnostic of B. americanus, not B. baileyi. 1200
The proximal nutrient foramina are reduced practically to absence (Martin et al., 2012). 1201
The presence and location of the proximal nutrient foramina have a long and convoluted history 1202
as diagnostic tools for the hesperornithiforms. First reported as absent (Martin and Tate, 1976), 1203
their presence in Neogaeornis was used as justification for excluding this bird from the 1204
hesperornithiforms (Lucas, 1903) and in B. varneri as justification of its unique taxonomic 1205
standing (Martin and Cordes-Person, 2007). Later work clearly identified these foramina as 1206
present in all hesperornithiform species (Clarke, 2004; Everhart and Bell, 2009). However, the 1207
size, shape, and general appearance appear to be easily affected by preservation, perhaps 1208
contributing to previous confusion over their presence or absence. It is therefore recommended 1209
55

that except in extreme or quantitatively supported instances, the proximal foramina should not be 1210
relied upon as diagnostic. 1211
Brodavis mongoliensis 1212
Nearly quadrangular transverse section near the middle of the metatarsal shaft (Martin 1213
et al., 2012). It is not possible to evaluate the state of this feature from the presented figure 1214
(Martin et al., 2012, Fig. 1), as the element appears no different from other hesperornithiform 1215
tarsometatarsi, all of which have ellipsoidal or ovoid cross-section. However, if the morphology 1216
is correctly described then this feature would be supported as diagnostic of B. mongoliensis. As 1217
further investigation is not possible, this feature should be regarded with concern. 1218
Lateral cotyla of the proximal tarsometatarsus expanded cranio-caudally, with cranial 1219
and caudal parts inclined distally (Martin et al., 2012). While the proximal tarsometatarsus is 1220
not preserved in most brodavids, comparison with B. varneri is possible. In B. varneri, the 1221
proximal articular surface appears to be nearly aligned with the margins of the shaft, with little or 1222
no expansion. However, B. varneri is so poorly preserved it is difficult to determine where 1223
breakage or erosion has occurred. Outside of the brodavids, the dorso-plantar expansion of the 1224
cotylae is well-documented, being found in Hesperornis regalis, Asiahesperornis, 1225
Parahesperornis, Pasquiaornis, and Baptornis advenus. Therefore the dorso-plantar flare of the 1226
cotylae is not apomorphic of B. mongoliensis. As for the inclination of the cotylae, the published 1227
drawing (Martin et al., 2012, Fig. 1) appears to show the cotylae tilting strongly distally in dorsal 1228
view, perhaps to a greater degree than in other hesperornithiform birds. The degree of inclination 1229
varies across taxa, being stronger in Hesperornis and much shallower in Baptornis and 1230
Pasquiaornis. In plantar view the cotylae of B. mongoliensis seem almost flattened, with only the 1231
lateral tilting slightly distally. This is similar to the state in Hesperornis. As it is not possible to 1232
56

discern any difference in this feature from that seen in numerous other hesperornithiforms, it is 1233
not possible to retain this feature as diagnostic. Furthermore, the only published images of the 1234
holotype and sole representative of the species (PIN 4491-8) are artistic drawings that cannot be 1235
taken as precise and accurate representations of the actual specimen. 1236
Proximal nutrient foramina of the tarsometatarsus are well-developed (Martin et al., 1237
2012). As discussed above, use of the proximal foramina is not advisable as a reliable diagnostic 1238
feature within the hesperornithiforms. 1239
Brodavis varneri 1240
Indistinct antitrochanter of pelvis (Martin and Cordes-Person, 2007). The antitrochanter 1241
of B. varneri was described as much less developed than that of B. advenus (Martin and Cordes- 1242
Person, 2007). With the exception of Enaliornis, for which little material is preserved (Galton 1243
and Martin, 2002), all other hesperornithiforms can be described as having a robust 1244
antitrochanter with marked lateral extension. Furthermore, Martin and Cordes-Person (2007) 1245
describe the antitrochanter of B. varneri as forming a smooth continuum where it merges with 1246
the ischium, with no distinct margin or lateral extension. Unfortunately, this may prove 1247
taphonomic, as the surface preservation of the holotype and sole specimen (SDSM 68430) is 1248
incredibly poor, making it difficult to determine which features are actually reduced as opposed 1249
to eroded away or even broken off. Therefore, this feature is not supported as diagnostic of the 1250
species. 1251
Tibiotarsus with shallow, wide facet for articulation of fibula (Martin and Cordes- 1252
Person, 2007). Cranial to the lateral cotyla of the proximal tibiotarsus, a deep groove is present 1253
in most birds for the articulation of the fibula. This groove has been described as broad and 1254
sloping in B. varneri as opposed to the deep excavation found in B. advenus (Martin and Cordes- 1255
57

Person, 2007) or Hesperornis. However, to what degree this is due to erosion or weathering of 1256
the element, which is poorly preserved, is impossible to say. Furthermore, despite claims to the 1257
contrary, this groove does not appear more deeply excavated in specimens of Baptornis than in 1258
B. varneri, particularly when compared to the deep excavation in Hesperornis. 1259
Head and tuberculum of rib well separated. This feature was identified as unique to B. 1260
varneri (Martin and Cordes-Person, 2007), however the head and tuberculum of Hesperornis is 1261
also well separated. Even Baptornis has space between the head and tuberculum, however due to 1262
the overall size variation between these taxa it is difficult to differentiate changes in body size 1263
from changes in porportions. This feature is therefore not useful as a diagnostic character of B. 1264
varneri. 1265
Femur with broad, smooth popliteal fossa (Martin and Cordes-Person, 2007). The use of 1266
the term ‘smooth’ makes interpretation of this character difficult, as all hesperornithiforms seem 1267
to have a smooth popliteal fossa. Martin and Cordes-Person (2007) might have been referring to 1268
the lack of a distinct ridge present in Hesperornis which gives the popliteal fossa a pocketed 1269
appearance. Given the poor preservation discussed previously, it is difficult to verify this ridge is 1270
truly absent as opposed to simply not preserved, however it does appear that B. varneri has an 1271
un-pocketed popliteal fossa. However, this is also the case in Enaliornis, Pasquiaornis, and 1272
Baptornis and so it is not unique to B. varneri. 1273
Femur with wide and deep intercondylar fossa (Martin and Cordes-Person, 2007). This 1274
term is analogous to the intercondylar sulcus of Baumel and Witmer (1993), but to refer to the 1275
intercondylar sulcus of B. varneri as deep is misleading. Due to the weathering of the bone, 1276
particularly of the condyles, it is impossible to determine the original morphology. Furthermore, 1277
the intercondylar sulcus, as preserved, is not deep, and appears to be even shallower and less 1278
58

well defined than that of B. advenus or Pasquiaornis. The state in B. varneri appears more 1279
similar to that of Enaliornis, which is also a very poorly preserved hesperornithiform. Therefore, 1280
this diagnostic feature is incorrect and should be removed from consideration. 1281
Tarsometatarsus with proximal foramina (Martin and Cordes-Person, 2007). Citing 1282
Martin and Tate (1976), Martin and Cordes-Person (2007) identified the prominent proximal 1283
foramina present in B. varneri as apomorphic. However, later studies have shown that all 1284
hesperornithiforms possess proximal foramina (Clarke, 2004; Everhart and Bell, 2009), and 1285
therefore this feature is not unique to B. varneri. The appearance of the foramina in this taxon is 1286
unusual, as they are present as elongate slits on both the dorsal and plantar surfaces, but the role 1287
of taphonomy as well as a complete comparison to other hesperornithiforms must be carried out 1288
before this can be considered apomorphic. Regardless, the presence of these foramina is not 1289
diagnostic. 1290
Ischium with indistinct anterior fossa. The poor preservation of the holotype and sole 1291
specimen (SDSM 86430) makes evaluation of this feature difficult, as all the features of the 1292
pelvis are indistinct due to erosion. This character cannot be supported as diagnostic. 1293
Fibular crest of the tibiotarsus is long and straight, extending at least half length of shaft 1294
(Martin and Cordes-Person, 2007). The fibular crest of some specimens of B. advenus (FMNH 1295
395) extends about one-third the length of the shaft, but in others it extends half the length of the 1296
shaft (UNSM 20030). Therefore, this feature is not diagnostic of B. varneri. Furthermore, the 1297
extremely poor preservation of the B. varneri holotype makes determination of the actual distal 1298
extent of the fibular crest difficult. 1299
Femur with prominent internal condyle (Martin and Cordes-Person, 2007). The medial 1300
condyle of B. varneri is greatly reduced compared to the lateral, as in all hesperornithiforms. As 1301
59

the overall sizes of the femora of B. advenus and B. varneri are so dissimilar, it is difficult to 1302
claim that the medial condyle is more prominent than it is in B. advenus. Furthermore, the high 1303
degree of erosion of the surface of the element makes determining the original morphology 1304
difficult. Therefore, this feature is not supported as diagnostic. 1305
. 1306
Family Enaliornithidae 1307
Tarsometatarsus with distinct caudomedial ridge leading to trochlea of metatarsal III 1308
(Galton and Martin, 2002). None of the specimens available for direct study or presented in the 1309
literature (Galton and Martin, 2002) display this feature. In most specimens either the trochlea or 1310
the more proximal shaft leading from the trochlea are not preserved, making evaluation of this 1311
feature impossible. The morphology is not different from that seen in other small 1312
hesperornithiforms in any of the available specimens. 1313
Reduced antitrochanter on pelvis, does not project strongly laterally (Galton and Martin, 1314
2002). A single pelvic fragment preserves the antitrochanter of Enaliornis (BGS 87936), where it 1315
does appear to be reduced as compared to that seen in most other hesperornithiforms. However, 1316
this is also the case in B. varneri, where the antitrochanter has most likely been eroded down, as 1317
the holotype and only specimen (SDSM 68430) is very poorly preserved. The specimens of 1318
Enaliornis that have been studied (BMNH 5310, cast of BGS 87936) or illustrated in the 1319
literature (BGS 87963) do not fully preserve the area around the acetabulum, making evaluation 1320
of the antitrochanter difficult. Finally, if the preserved state accurately reflects the original 1321
morphology, then the feature is plesiomorphic, as Archaeopoteryx also has a reduced 1322
antitrochanter. It may be that the opposite state, expansion of the trochanter, can be used to unite 1323
a more inclusive group within the Hesperornithiformes. 1324
60

Absence of a distinct femoral neck (Galton and Martin, 2002). The femur of Enaliornis 1325
has been described as lacking a neck between the main shaft and the articular head. In other 1326
hesperornithiforms there is a marked constriction before the head expands out in a ball-like knob. 1327
However, in Pasquiaornis a distinct neck is also unclear. As Pasquiaornis and Enaliornis are 1328
both known solely from highly weathered, completely disarticulated elements found in bone bed- 1329
type deposits (Cumbaa et al., 2006; Galton and Martin, 2002), it is most likely that the seeming 1330
lack of femoral neck is due to the erosion of the edges of the femoral head. Furthermore, the lack 1331
of a femoral neck is plesiomorphic among birds, and so this feature should not be used as 1332
apomoprhic of Enaliornis. 1333
Distal tarsometatarsus arched in distal view (Galton and Martin, 2002). When the 1334
tarsometatarsus is resting in dorsal or plantar view, the overall transverse outline of the distal end 1335
in distal view has been described as ‘c-shaped’ (Galton and Martin, 2002) and plantarly concave, 1336
with the dorsal edge of trochlea III forming the top and trochlea II and IV sloping downward on 1337
either side. This is similar to the state in other hesperornithiforms and basal birds such as 1338
Gansus. Therefore, this feature is not diagnostic. 1339
Centra of preacetabular synsacrum transversely constricted to form a longitudinal ridge 1340
(Galton and Martin, 2002). Little information was given by Galton and Martin (2002) to 1341
differentiate this character from the state in other hesperornithiforms, which also have synsacra 1342
with neural spines that form a longitudinal ridge projecting dorsally up to the level of the 1343
acetabulum. This character should not be considered diagnostic of Enaliornithidae, but it may 1344
perhaps prove to be diagnostic of the Hesperornithiformes. 1345
Cranioproximal process of the tarsometatarsus originates from the proximal end of 1346
metatarsal III (Galton and Martin, 2002). Interpretation of this feature is difficult. The only 1347
61

‘cranioproximal’ feature on the tarsometatarsus of hesperornithiform birds is the intercotylar 1348
eminence. Enaliornis is no different from other hesperornithiforms in having this feature 1349
originate at the proximal end of metatarsal III. Galton and Martin specified that this undefined 1350
process was “not part of hypotarsus” (2002, p. 329), however the hypotarsus is on the dorsal side 1351
of the proximal end (synonymous with the caudal side). As this feature cannot be identified, it is 1352
not supported as diagnostic of Enaliornithidae. 1353
Cranial edge of trochlea IV caudal to the prominent cranial edge of the subequal 1354
trochlea III (Galton and Martin, 2002). This feature is essentially the same as that proposed for 1355
Pasquiaornis hardei, “Neck of the trochlea of metatarsal III higher anteriorly than that of IV” 1356
(Tokaryk et al., 1997). As discussed above, this morphology is observed in both species of 1357
Pasquiaornis as well as in Enaliornis, and so cannot be retained as apomorphic of either group. 1358
Genus Enaliornis 1359
Diagnosis is currently the same as that of the family. 1360
Enaliornis barretti 1361
Large, rugose femoral trochanter (Galton and Martin, 2002). Most specimens of 1362
Enaliornis are poorly preserved and highly eroded, making it difficult to differentiate between 1363
original morphology and weathering. This is true of the preservation of the femoral trochanter, 1364
which presents both rugose and smooth examples within E. barretti. It is impossible to say the 1365
smooth appearance of the trochanter in E. seeleyi and E. sedgewicki is the actual morphology and 1366
not erosion, particularly as thee specimens are so badly weathered in general. Finally, as the 1367
lectotype of E. barretti is a distal tarsometatarsus, features for diagnosing the species should be 1368
restricted to this element in the absence of any additional associated material. 1369
62

Laterally flared, large cnemial crest on proximal tibiotarsus (Galton and Martin, 2002). 1370
While dimensions of the cnemial expansion of the tibiotarsus may prove diagnostic for E. 1371
barretti, the use of this character in other families as well as the very similar appearance of the 1372
tibiotarsus among different members of Enaliornis make the application of this feature difficult 1373
to interpret. Poor preservation also plays a role in this feature, as the cnemial expansion is broken 1374
off completely in many hesperornithiform specimens and eroded to a more rounded, reduced 1375
state in many more. As this feature is highly dependent on good preservation, which does not 1376
usually occur in Enaliornis, it is not useful as a diagnostic feature. Finally, as the lectotype of E. 1377
barretti is a distal tarsometatarsus, features for diagnosing the species should be restricted to this 1378
element in the absence of any additional associated material. 1379
Proximal outline of the medial cotyla almost square in shape (Galton and Martin, 2002). 1380
This feature was reported as unique to E. barretti, however analysis of both the figures provided 1381
(Galton and Martin, 2002) and specimens fails to show any differences in the highly eroded 1382
surfaces between E. barretti and other species. Finally, as the lectotype of E. barretti is a distal 1383
tarsometatarsus, features for diagnosing the species should be restricted to this element in the 1384
absence of any additional associated material. 1385
Enaliornis sedgewicki 1386
Caudally intercondylar angle slightly lateral in position (Galton and Martin, 2002). 1387
Interpreting this feature is difficult, as further explanation was not provided. This feature is here 1388
taken to mean that in caudal view, the negative space between the distal-most condyles (distal to 1389
the shaft) forms a peak proximally, with this peak shifted laterally in E. sedgewicki. This agrees 1390
with observations of the holotype. However, E. barretti was also described as having a laterally 1391
shifted intercondylar sulcus, while that of E. barretti was described as medially shifted (Galton 1392
63

and Martin, 2002). Therefore, this feature is not diagnostic of E. sedgewicki, but the medial 1393
offset of E. barretti may prove diagnostic for that taxon (see discussion below). 1394
Enaliornis seeleyi 1395
Moderately flared lateral cnemial crest on proximal tibiotarsus (Galton and Martin, 1396
2002). Galton and Martin described the cnemial crest of E. barretti as widely flared and that of 1397
E. seeleyi as less flared (2002). Unfortunately the poor preservation and scarce remains of this 1398
species make observations of this level of detail impossible. While one example of E. barretti 1399
(YORYMG 587) does preserve a cnemial crest that appears to flare laterally, it also appears that 1400
the entire cnemial crest has been broken off and glued back on at some point, which brings into 1401
question the accuracy of the reconstruction. Furthermore, one example of E. sedgewicki (SMC 1402
B55314) appears to have a similarly flared cnemial crest, while another (SMC B55314) is much 1403
more even. The cnemial crest of E. seeleyi in one specimen (BMNH A478) does appear to be 1404
more even and less flared than in E. barretti, however it also appears more eroded, as does the 1405
individual of E. sedgewicki that is less flared. Therefore, it appears that the appearance of the 1406
cnemial crest is easily affected by erosion and should not be used as diagnostic. 1407
Medial cotyla on tarsometatarsus subrectangular in outline in proximal view; transverse 1408
width is half of cranio-caudal depth (Galton and Martin, 2002). As discussed earlier, the poor 1409
state of preservation in Enaliornis obscures the details of the proximal articular surface of the 1410
tibiotarsus, thus making features of this region unsupportable. 1411
1412
Family Hesperornithidae 1413
The two features proposed as diagnostic of the Hesperornithidae were supported in this study 1414
(App. I). 1415
64

Genus Asiahesperornis 1416
Tarsometatarsal shaft distinctly compressed (Dyke et al., 2006). This feature was 1417
identified as apomorphic of Asiahesperornis after comparison to North American 1418
hesperornithiforms (Dyke et al., 2006), however the nature of the compression was not fully 1419
described. As a laterally compressed tarsometatarsus is a feature of all hesperornithiforms, it is 1420
unclear how the compression in Asiahesperornis differs, or how this compression would be 1421
separated from the previous character of a slender tarsometatarsus. Given that the overall 1422
dimensions do not differ from those of other hesperornithids, there appears to be no basis for this 1423
character at this time. 1424
Tarsometatarsus lacks a laterally twisted shaft (Dyke et al., 2006). In H. regalis, when 1425
the proximal end of the tarsometatarsus is in dorsal view the distal end is in dorso-medial view, 1426
which is the condition described as twisting by Dyke et al. (2006), who described both 1427
Asiahesperornis and Hesperornis rossicus as lacking this twist. However, Asiahesperornis does 1428
not preserve a complete tarsometatarsus. The holotype is a proximal end with part of the shaft. 1429
While distal tarsometatarsi have been assigned to Asiahesperornis, the observation of a feature 1430
such as this is only possible in complete specimens. While there are differences, the 1431
tarsometatarsus of H. regalis does not appear more twisted. Furthermore, the holotype of H. 1432
rossicus is also a proximal tarsometatarsus for which this character cannot be observed (Rees and 1433
Lindgren, 2005). An additional specimen assigned to H. rossicus (ZIN PO 5464) has a highly 1434
twisted shaft (Panteleyev et al., 2004). 1435
Parallel lateral and medial sides of the tarsometatarsus (Kurochkin, 2000). This 1436
character is unclear, as the sides of the tarsometatarsal shaft appear to intersect distally in dorsal 1437
65

and plantar view, but do seem more parallel in medio-dorsal view. This is true of all 1438
hesperornithiforms though, so it is unclear how this feature varies in Asiahesperornis. 1439
Lateral groove on the tarsometatarsal shaft terminates in a distinct fossa just proximal to 1440
the third trochlea (Dyke et al., 2006). The distal foramen of Asiahesperornis is more obvious 1441
than in some other hesperornithids, however this is most likely due to the tendency for the distal 1442
foramen to be obscured by crushing or filled with sediment that is not removed during 1443
preparation. There does not appear to be any difference between the distal foramina of 1444
Asiahesperornis and those of other well-preserved and prepared specimens of H. regalis. 1445
Medial condyle of tibiotarsus markedly medio-laterally compressed (Nessov and 1446
Prizemlin, 1991). As the holotype of Asiahesperornis is an isolated tarsometatarsus (Dyke et al., 1447
2006), diagnostic features for the genus must be restricted to that element. 1448
Cranial intercondylar furrow of tibiotarsus comparatively deep (Nessov and Prizemlin, 1449
1991). As the holotype of Asiahesperornis is an isolated tarsometatarsus (Dyke et al., 2006), 1450
diagnostic features for the genus must be restricted to that element. 1451
Asiahesperornis bahazanovi 1452
Diagnosis is currently the same as that of the genus. 1453
Genus Canadaga 1454
Large process ventralis occupying anteromedial part of ventral side of centrum (Hou, 1455
1999). All hesperornithiforms possess strong ventral processes on the posterior cervical 1456
vertebrae. While this process is not preserved in Canadaga, the area where it has broken off 1457
extends along nearly the entire surface of the centrum in ventral view, similar to the extent seen 1458
in similarly broken specimens of Hesperornis. As this character is based on a morphology that is 1459
not preserved and does not appear to differ from Hesperornis, it cannot be retained. 1460
66

Dorsal process of the vertebral arch short and robust (Hou, 1999). The neural spines of 1461
Canadaga are not preserved, however from the broken area where they were located it can be 1462
seen that they were restricted antero-posteriorly, as they are in vertebrae 15~18 of Hesperornis. 1463
As the neural spines themselves are not preserved for any specimens of either genus, use of this 1464
feature is inappropriate. 1465
Well-developed area of elastic ligament anterior to arcus vertebrae dorsal process (Hou, 1466
1999). The area for the attachment of the elastic ligament is obvious in most well-preserved 1467
hesperornithiform specimens. As this feature is not unique to Canadaga, it is not supported as 1468
diagnostic. 1469
Centrum wider than width of zygapophysis (Hou, 1999). It is unclear in which view this 1470
feature applies, however in dorsal and cranial views the preserved zygapophyses clearly extend 1471
laterally further than the margins of the constricted centrum. 1472
Large cavity present inside the transverse processes (Hou, 1999 Canadaga is not unique 1473
in this, as Hesperornis regalis shows the same hollow visible in broken specimens. 1474
Canadaga arctica 1475
Diagnosis is currently the same as that of the genus. 1476
Genus Coniornis 1477
See Hesperornis altus. 1478
Genus Hesperornis 1479
The sole diagnostic feature for this genus is discussed in the morphometrics chapter. 1480
Hesperornis altus 1481
Lateral condyle extends relatively further distally than in H. regalis and medial condyle 1482
more in line with the margin of the shaft (Marsh, 1893). This character was invalidated by 1483
67

Shufeldt (1915), who claimed the differences were not more than what might occur between 1484
individuals of the same species (such as within H. regalis). 1485
Hesperronis bairdi 1486
Metatarsal IV trochlea more enlarged and distal to trochlea III than in Parahesperornis 1487
(Martin and Lim, 2002). While this observation is correct, it is true of all species of Hesperornis, 1488
not only H. bairdi. Therefore, this character is not diagnostic of any one species of Hesperornis. 1489
Hesperornis chowi 1490
Anterior metatarsal ridge more slender than in H. regalis (Martin and Lim, 2002). As the 1491
entire tarsometatarsus of H. chowi was described as being more slender than that of H. regalis, it 1492
is to be expected that the individual metatarsals would also be slender. As this feature is simply a 1493
result of the overall shape difference discussed above, it is redundant and is not retained as 1494
diagnostic. 1495
Trochlea of metatarsal IV not as enlarged as in H. regalis (Martin and Lim, 2002). As the 1496
enlarged metatarsal trochlea IV is a derived condition found the Hesperornis, the presence of a 1497
less enlarged trochlea is simply the plesiomorphic state and should not be used as diagnostic. 1498
Inner metatarsal ridge shorter and less prominent than in H. regalis (Martin and Lim, 1499
2002). While the dorsal ridge formed by the surface of metatarsal II may appear fainter in the 1500
holotype of H. chowi (YPM PU 17208), the entire specimen is more weathered and poorly 1501
preserved. If comparison is made to a similarly poorly preserved specimen of H. regalis, the 1502
cranial ridge of metatarsal II is also faint. 1503
Hesperornis crassipes 1504
Lateral cotyla of tarsometatarsus more distally inclined than medial (Rees and Lindgren, 1505
2005). In most species of Hesperornis the articular surface of the medial cotyla is more steeply 1506
68

inclined than the lateral. As illustrated by Marsh, the lateral and medial cotylar surfaces are 1507
similarly angled distally (1880). However, this observation is from drawings that may or may not 1508
accurately represent the holotype specimen. As the holotype is currently unavailable for 1509
observation (in an inaccessible display at the Yale Peabody Museum), this feature cannot be 1510
confirmed. 1511
Large tuberosity on upper half of medial face of tarsometatarsus (Marsh, 1876). This 1512
tuberosity was identified by Marsh on the holotype (YPM 1474), however, Marsh did indicated 1513
that this feature might be due to sexual dimorphism (Marsh, 1880). While the holotype is 1514
unavailable, the assigned specimen, AMNH 5102, does not exhibit anything resembling a large 1515
tuberosity on the medial face of proximal metatarsal II. Rather, there is a noticeable ovoid 1516
indentation present, as is also present on the holotype of H. regalis. Despite Marsh describing 1517
this feature as a ‘striking character’, the illustrations provided do not show this (Marsh, 1880, p. 1518
97 and plate XVII). Instead, the proximal lateral face of metatarsal IV appears very similar to 1519
that of H. regalis, with the same ovoid indentation present. Without analysis of the holotype 1520
further evaluation is difficult, however it seems that this feature is not supported. 1521
Hesperornis gracilis 1522
The diagnostic features for H. gracilis are discussed in Chapter 3. 1523
Hesperornis macdonaldi 1524
The diagnostic features for H. macdonaldi are discussed in Chapter 3. 1525
Hesperornis mengeli 1526
Trochlea of metatarsal II completely behind that of metatarsal III (Martin and Lim, 1527
2002). If this character were accurate, then trochlea II should not be visible in dorsal view. As 1528
figured (Martin and Lim, 2002; Fig. 5), trochlea II is, in fact, partially visible, as it is in other 1529
69

hesperornids (Martin and Lim, 2002; Figs. 2-3). Perhaps this is a question of degree; however the 1530
figures provided are not precise enough for that sort of observation. Furthermore, this feature 1531
was later used by one of the authors as diagnostic of Brodavis baileyi (Martin et al., 2012) and 1532
has also been described for H. rossicus (Rees and Lindgren, 2005). In H. mengeli the state of this 1533
character appears to fall within the range represented by that seen in other hesperornithiform 1534
taxa, invalidating the use of this feature as diagnostic. 1535
Hesperornis regalis 1536
Metatarsal trochlea IV nearly twice the size of trochlea III and extends further distally 1537
(Martin, 1984). This character was first reported as differentiating Parahesperornis from 1538
Hesperornis (Martin, 1984). While this feature is accurate, it is applicable to all analyzed species 1539
of Hesperornis. Therefore, this feature should be considered as diagnostic of the genus. 1540
Hesperornis rossicus 1541
Medial and lateral cotylae of tarsometatarsus flattened (Rees and Lindgren, 2005). 1542
Multiple species of Hesperornis possess a slightly cup-shaped, concave medial cotyla and a 1543
relatively flat lateral cotyla. In H. rossicus both cotylae have been described as flattened (Rees 1544
and Lindgren, 2005). Only drawings have been published of the holotype specimen, in which the 1545
degree of curvature of the cotyla does not appear to differ greatly from that seen in Hesperornis 1546
regalis. In other specimens of H. rossicus the cotylae are too eroded to allow observation of this 1547
feature (Panteleyev et al., 2004). Given the lack of photographic support for this feature, it 1548
cannot be supported at this time. 1549
Medial cotyla slopes distally as compared to lateral, which is nearly horizontal (Rees and 1550
Lindgren, 2005). The angles of the articular surface of the medial and lateral cotylae of H. 1551
rossicus do not differ from those of other species of Hesperornis. In all specimens the medial 1552
70

cotyla slopes dramatically distally while the lateral is more perpendicular to the shaft, however 1553
some distal slope is observed. 1554
Medial cotyla of the tarsometatarsus is distal to the lateral (Nessov and Yarkov, 1993 as 1555
translated in Kurochkin, 2000). This feature is a result of the inclined articular surface also used 1556
as diagnostic of the group. All species of Hesperornis display this feature, as when the shaft of 1557
the tarsometatarsus is oriented vertically, the lateral cotyla is higher than the medial. 1558
Lateral edge of lateral cotyla of the tarsometatarsus exceeds the intercotylar prominence 1559
proximally (Nessov and Yarkov, 1993 as translated in Kurochkin, 2000). This character is caused 1560
by the angle of the proximal articular surface which was discussed earlier, and should not be 1561
used as a separate diagnostic character. 1562
Inner toes strongly reduced (Panteleyev et al., 2004). The toe size of H. rossicus must 1563
have been inferred from the sizes of the tarsometatarsal trochleae (Panteleyev et al., 2004), as no 1564
pedal phalanges have been reported. Unfortunately the distal trochlea are only preserved on two 1565
specimens, and obscured by erosion or breakage on those. While H. rossicus does appear to have 1566
somewhat narrower third trochlea than other hesperonrothiforms, it does not fall outside the 1567
range of the rest of the genus and is therefore probably not reliable as a diagnostic feature. 1568
Condyles of the trochlea for digits II and III undivided (Panteleyev et al., 2004). 1569
Panteleyev et al. (2004) claimed that H. rossicus shared this feature with H. mengeli, however H. 1570
mengeli was not reported to have undivided condyles on trochlea II and III (Martin and Lim, 1571
2002). Examination of the published illustrations does show the trochlear ridges of H. mengeli as 1572
absent, however this is not uncommon in poorly preserved hesperornithiform specimens (Martin 1573
and Lim, 2002 Fig. 5). Additionally, this may simply be due to inaccuracy in the sketch. 1574
Furthermore, the published images of H. rossicus clearly show a divided condyle on trochlea III 1575
71

(Panteleyev et al., 2004 Fig. 3A), while trochlea II appears to be too poorly preserved to 1576
accurately see whether or not the condyle was divided (Panteleyev et al., 2004 Figs. 3-4). 1577
Therefore this feature cannot be considered as valid. 1578
Intercotylar eminence of the tarsometatarsus has acute profile in dorsal view (Rees and 1579
Lindgren, 2005). Rees and Lindgren (2005) claimed that most species of Hesperornis have a 1580
rounded intercotylar eminence in dorsal view while that of H. rossicus is acute and highly 1581
peaked. However, this feature is highly dependent on the preservation of the element. 1582
Observations of multiple tarsometatarsi of H. regalis reveal the appearance of the intercotylar 1583
eminence is highly variable and subject to the weathering of the element. While photographs of 1584
the relevant specimens of H. rossicus are not available, the published drawings show SGU 3442 1585
Ve01 with a highly peaked intercotylar eminence, more so than in H. regalis, but it is also more 1586
peaked than the other H. rossicus specimen, which is most similar to the holotype of H. regalis, 1587
YPM 1200. A highly weathered specimen of H. regalis, FMNH 284, shows the high degree of 1588
rounding possible on the intercotylar eminence which can greatly reduce the peaked appearance. 1589
Therefore, this feature does not seem suitable for a diagnostic feature. 1590
In proximal view, the proximal end of the tarsometatarsus is more dorso-ventrally 1591
flattened in H. rossicus than it is in other species of Hesperornis (Rees and Lindgren, 2005). This 1592
feature overlaps with another, more specific feature, that described the transverse width and 1593
dorso-plantar depth of the proximal tarsometatarsus (Nessov and Yarkov, 1993). Therefore, this 1594
feature will be considered redundant and the other evaluated in the morphometric analysis. 1595
Genus Parahesperornis 1596
Coracoid more elongate than in Hesperornis (Martin, 1984). While the coracoid of 1597
Parahesperornis is much less robust than that of Hesperornis, this state is plesiomorphic in 1598
72

birds, as discussed above. Therefore, while a shortened coracoid might prove diagnostic of 1599
Hesperornis, this feature cannot be considered vaild for Parahesperornis. 1600
Nasal process of lacrimal more extended anteriorly than in Hesperornis (Martin, 1984). 1601
As there are not any examples of Hesperornis that preserve the nasal process of the lacrimal, this 1602
is impossible to say and this feature cannot be supported. 1603
Orbital process of quadrate very elongate (Martin, 1984). Parahesperornis does possess 1604
an elongate orbital process, like Hesperornis. Given the size variation between two, judging 1605
whether or not the orbital process of Parahesperornis is more elongate than that of Hesperornis 1606
cannot be determined. As only one fairly complete quadrate is known for each species, 1607
morphometric analysis is of little use in this case. 1608
Skull mesokinetic (Martin, 1984). Mesokinesis is a state found in some reptiles where the 1609
frontal and parietal bones are not fused, resulting in a transverse hinge which plays a role in 1610
raising the upper jaw (Gingerich, 1973). After study of Hesperornis regalis, Gingerich ruled out 1611
various forms of cranial kinesis and concluded that Hesperornis was akinetic (1976), however 1612
that statement was later amended to say that a novel form of maxillokinesis might have been 1613
possible (1976). Martin claimed that this differs from Parahesperornis and Enaliornis, where the 1614
frontals and parietals are not fused (1984). Unfortunately only a few hesperornithiform 1615
specimens preserve the requisite elements, and those are poorly preserved and disarticulated, 1616
making accurate observations difficult. However, in Enaliornis the only preserved cranium 1617
shows a frontoparietal suture (Galton and Martin, 2002), which makes the ability to infer 1618
movement questionable. In Parahesperornis, the preserved cranial elements are disarticulated 1619
and fragmentary, making reconstruction difficult. As previous work has concluded that 1620
73

mesokinesis was not possible in Hesperornis, it seems unlikely that such a claim is supportable 1621
for the very similar Parahesperornis. 1622
1623
74

FIGURE CAPTIONS 1624
Figure 1. Geographic and temporal distribution of hesperornithiform specimens. A. This map 1625
reflects all reported hesperornithiform discoveries, with the genera known from each area 1626
identified. B. Temporal distribution of hesperornithiform genera. 1627
Figure 2. Comparison of the number of specimens known for each described species of 1628
hesperornithiform bird as well as the portion of specimens consisting of over three associated 1629
elements. 1630
1631
TABLE CAPTIONS 1632
Table 1. Current taxonomic framework of the Hesperornithiformes. All taxa that have been 1633
proposed as part of the order Hesperornithiformes are shown in their current taxonomic position. 1634
Individual taxa that have been questioned or challenged for any reason are highlighted in red. 1635
Table 2. Evaluation of diagnostic features proposed for all taxonomic levels within the 1636
Hesperornithiformes. Features identified as supported are discussed in Appendix I, features 1637
identified as unsupported are discussed in Appendix II, and features in need of morphometric 1638
revision are presented in Chapter 3. 1639
Table 3. All specimens assigned to Baptornis to date. YPM 1465, the holotype of Baptornis 1640
advenus, is highlighted in bold. The holotype of the invalid taxon Parascaniornis stensoi is 1641
indicated with an asterisk (MGUH 1908.214). Citation numbers are as follows: 1 – Marsh, 1877; 1642
2 – Martin and Tate, 1976; 3 – Martin and Bonner, 1977; 4 – Everhart and Bell, 2009; 5 – Rees 1643
and Lindegren, 2005; 6 – Tokaryk and Harrington, 1992. Abbreviations for age are as follows: 1644
CEN – Cenomanian; CMP – Campanian, CON – Coniacian. 1645
75

Table 4. Specimens assigned to Pasquiaornis hardei and Pasquiaornis tankei (designations from 1646
Tokaryk et al., 1997 and Sanchez, 2010). A. Axial skeleton, B. Pectoral girdle and forelimb, C. 1647
Hindlimb. Holotypes and paratypes are indicated in bold. 1648
Table 5. Specimens assigned to the Brodavidae (Martin et al., 2012). Skeletal abbreviations are 1649
as follows: vrt – vertebra, plv – pelvis, fmr – femur, fib – fibula, tbt – tibiotarsus, tmt – 1650
tarsometatarsus. Abbreviations for age are as follows: CMP – Campanian, MAA – 1651
Maastrichtian. All assignments are from Martin et al., 2012. 1652
Table 6. Holotypes, lectotypes, and paralectotypes designated for Enaliornis (from Brodkorb, 1653
1963 and Galton and Martin, 2002). 1654
Table 7. Specimens assigned to Enaliornis (from Brodkorb, 1963 and Galton and Martin, 2002). 1655
Table 8. Specimens assigned to Asiahesperornis. All specimens are from the same quarry in the 1656
Zhuravlovskaya Svita (Maastrichtian) in Kazakhstan (Dyke et al., 2006). The holotype 1657
tarsometatarsus (IZASK 5/287/86a) is highlighted in bold. The tibiotarsus removed from the 1658
holotype is marked with an asterisk. 1659
Table 9. Specimens assigned to Canadaga. All specimens are from one of two locations in the 1660
Canadian High Arctic (Hou, 1999; Wilson et al., 2011). The holotype specimen (NMC 41050) is 1661
highlighted in bold. For the specimens reported by Hou (199), stratigraphic information is not 1662
known. 1663
Table 10. Specimens assigned to Hesperornis. This table presents specimens that have been 1664
assigned to Hesperornis in the literature. This is only a small percentage of the total number of 1665
specimens present in museum collections that have been referred to Hesperornis (most as H. 1666
76

regalis or H. indeterminate), however without published corroboration these assignments require 1667
further analysis. This table is designed to give a brief overview of the distribution and abundance 1668
of reported Hesperornis specimens, and so in some instances localities or taxa that have multiple 1669
specimens are combined and listed as ‘multiple’, with the listed material representing all known 1670
specimens. Taxa marked with an asterisk have been re-assigned from a previous designation, as 1671
discussed in the text. Abbreviations for skeletal elements are as follows: crn – cranial material, 1672
vrt – vertebra, scp – scapula, stn – sternum, plv – pelvic material, fmr – femur, pat – patella, fib – 1673
fibula, tmt – tarsometatarsus, phx – pedal phalanx 1674
Table 11. Specimens assigned to Parahesperornis. All specimens are from the Smoky Hill 1675
Member of the Niobrara Formation (Coniacian) of Kansas (Martin, 1984; Bell and Everhart, 1676
2009). Skeletal abbreviations as for Table 10. 1677
Table 12. Summary of support for the current taxonomic framework, as revealed by the analysis 1678
presented here. For each proposed taxonomic unit, the number of characters identified as 1679
supported, unsupported, and in need of morphometric analysis are listed (for example, 1/0/2 1680
indicates one supported character, zero unsupported characters, and two characters in need of 1681
morphometric analysis). Taxonomic units with no numbers listed do not have a diagnosis 1682
independent from the larger taxonomic group of which it is a part (for example, the diagnosis of 1683
Baptornis advenus is the same as that of the genus Baptornis). 1684
1685
77

Figure 1A. 1686
1687
Figure 1B. 1688
1689
78

Figure 2. 1690
1691
1692
79

Table 1. 1693
Class Family Genus Species
Hesperornithiformes
Enaliornithidae Enaliornis
barretti
sedgewicki
seeleyi
Baptornithidae
Baptornis
advenus
varneri
Judinornis nogontsavensis
Parascaniornis stensoei
Pasquiaornis
hardei
tankei
Brodavidae Brodavis
americanus
baileyi
mongoliensis
varneri
Hesperornithidae
Asiahesperornis bazhanovi
Canadaga arctica
Coniornis altus
Hargeria gracilis
Hesperornis
altus
bairdi
chowi
crassipes
gracilis
macdonaldi
mengeli
montana
regalis
rossicus
Lestornis crassipes
Parahesperornis alexi
1694
1695
80

Table 2. 1696
Hesperornithiformes
Supported characters:
Teeth in grooves (Marsh, 1877)
Pterygoid process of quadrate elongate and appressed onto medial mandibular
condyle (Elzanowski, 2000)
Sternum without keel (Marsh, 1877)
Long bones nonpneumatic (Martin, 1983)
Elongation of post-acetabular pelvis (Martin, 1984)
Large, trihedral patella perforated for ambiens tendon (Martin and Tate, 1976)
Triangular cnemial expansion on tibiotarsus (Martin, 1983)
Tarsometatarsus possesses sharp craniolateral ridge leading to lateral trochlea
(Martin, 1984)
Compressed tarsometatarsus (Martin, 1984)
Unsupported characters:
Thoracic vertebrae heterocoelus (Martin and Tate, 1976)
Reduced quadrate pneumaticity (Martin and Tate, 1976)
Quadrate with undivided head (Marsh, 1877)
Combination of short, broad pterygoids with long, narrow palatines (Martin, 1984)
Unfused mandibular symphysis (Martin and Tate, 1976)
Maxillokinesis in jaw (Martin and Tate, 1976)
Acetabulum partly closed (Marsh, 1880)
Wings reduced (Marsh, 1880)
Coracoid with glenoid facet on tip of scapular end (Martin and Tate, 1976)
Humerus lacking distinct distal condyles (Martin, 1984)
Clavicles unfused (Marsh, 1880)
Posterior extremities of ilium, ischium, and pubis separate (Marsh, 1880)
Tibiotarsus lacking supratendinal bridge (Marsh, 1880)
Tarsometatarsus lacking hypotarsal grooves and proximal foramina (Martin and
Tate, 1976)
Fourth toe longest (Martin, 1984)
Feet lobed (Martin, 1984)
In need of morphometric revision:
Shortened femur (Martin and Tate, 1976)
Baptornithidae
Supported characters:
Small pit directly anterior to diapophysis in thoracic vertebrae (Nessov and Borkin,
1983)
Preacetabular portion of ilium relatively longer than in hesperornithids (Martin and
Tate, 1976)
81

Medial and lateral trochlear ridges more pronounced on trochlea III than trochlea IV
(Everhart and Bell, 2009)
Intercotylar prominence of tarsometatarsus present as a low, rounded bump centered
over metatarsal III (Everhart and Bell, 2009)
Unsupported characters:
Coracoid more slender than in hesperornithids (Martin and Tate, 1976)
Uncinate processes of ribs turned dorsally (Martin and Tate, 1976)
Pyramidal patella (Martin and Tate, 1976)
Pygostyle long and laterally compressed (Martin and Tate, 1976)
Margin of distal foramen of tarsometatarsus delineated by indentations in necks of
trochlea III and IV(Everhart and Bell, 2009)
Circular pits in articular surfaces of centra of thoracic vertebrae (Martin and Bonner,
1977)
Hypapophysis in thoracic vertebrae centered (Tokaryk and Harrington, 1997)
Intracentral bones not fused to caudal vertebrae (Martin and Tate, 1976)
Medial and lateral cotyla of tarsometatarsus tilt dorsally (Everhart and Bell, 2009)
Distal foramen on tarsometatarsus an open groove (Martin and Tate, 1976)
Outer trochlea of tarsometatarsus not enlarged (Martin and Tate, 1976)
Toe-rotation not well developed (Martin and Tate, 1976)
In need of morphometric revision:
Elongate cervical vertebrae (Martin and Cordes-Person, 2007)
Rib articulation and anterior and posterior articular surfaces of centrum of thoracic
vertebrae comparatively smaller than in Hesperornis (Tokaryk and Harrington,
1997)
Baptornis
Unsupported characters:
Short sacrum (Lucas, 1903)
Coracoid lacks procoracoid process (Lucas, 1903)
Peg-and-crescent articulations on distal articular surface of metatarsal IV absent
(Martin, 1984)
Judinornis
Unsupported characters:
Ventral surface of centrum distinctly narrowed at middle but very broad caudally
(Nessov and Borkin, 1983)
Articular surfaces of centra of thoracic vertebrae trapezoidal in shape and extend
transversely (Nessov and Borkin, 1983)
Cranial zygapophyses located low together on midline of thoracic vertebrae (Nessov
and Borkin, 1983)
82

Pasquiaornis
Unsupported characters:
Proximal end of femur relatively less expanded lateromedially than in B. advenus
(Tokaryk et al., 1997)
Trochanteric crest of femur closer to shaft than in Baptornis (Tokaryk et al., 1997)
Intercotylar eminence of tarsometatarsus anteriorly placed, overhangs shaft
(Tokaryk et al., 1997)
In need of morphometric revision:
Metatarsal II trochlea located posterior and close to base of trochlea III (Tokaryk et
al., 1997)
P. hardei
Unsupported characters:
Distal rim of femoral head perpendicular to shaft (Tokaryk et al., 1997)
Medial cotyla of tarsometatarsus deflected toward shaft (Tokaryk et al., 1997)
Neck of metatarsal III trochlea higher anteriorly than that of IV (Tokaryk et al.,
1997)
In need of morphometric revision:
Reduced femur (Tokaryk et al., 1997)
P. tankei
Unsupported characters:
Distal rim of femoral head slanted toward shaft (Tokaryk et al., 1997)
Medial cotyla of tarsometatarsus nearly antero-posteriorly aligned (Tokaryk et al.,
1997)
Neck of metatarsal IV trochlea higher anteriorly than that of III (Tokaryk et al.,
1997)
Brodavidae
In need of morphometric revision:
Tarsometatarsus short and comparatively broad (Martin et al., 2012)
Facet for articulation of metatarsal I displaced proximally on tarsometatarsus shaft
(Martin et al., 2012)
Brodavis americanus
Unsupported characters:
Anterior-distal surface of metatarsal IV trochlea broad and flat (Martin et al., 2012)
In need of morphometric revision:
Facet for metatarsal I placed below tarsomatatarsus midshaft (Martin et al., 2012)
Shaft of tarsometatarsus broader and more robust than B. baileyi but smaller and less
robust than B. varneri (Martin et al., 2012)
Metatarsal IV trochlea swollen proximally and slightly broader than that of
metatarsal III (Martin et al., 2012)
83

Brodavis baileyi
Unsupported characters:
Outer anterior ridge of tarsometatarsus shaft extends further distally (Martin et al.,
2012)
Proximal nutrient foramina reduced practically to absence (Martin et al., 2012)
In need of morphometric revision:
Metatarsal II trochlea more elevated proximally and placed more behind trochlea III
(Martin et al., 2012)
Tarsometatarsus shaft more slender than in B. americanus (Martin et al., 2012)
Metatarsal IV trochlea less expanded at base than in B. americanus (Martin et al.,
2012)
Brodavis mongolienis
Unsupported characters:
Lateral cotyla of proximal tarsometatarsus expanded cranio-caudally, with cranial
and caudal parts inclined distally (Martin et al., 2012)
Proximal nutrient foramina of tarsometatarsus well-developed (Martin et al., 2012)
In need of morphometric revision:
Metatarsal shaft slender (Martin et al., 2012)
Facet for metatarsal I located at midshaft (Martin et al., 2012)
Unable to evaluate:
Nearly quadrangular transverse section near midshaft of tarsometatarsus (Martin et
al., 2012)
Brodavis varneri
Supported characters:
Midpoint of tarsometatarsus shaft narrows to waist (Martin and Cordes-Person,
2007)
Unsupported characters:
Indistinct antitrochanter of pelvis (Martin and Cordes-Person, 2007)
Tibiotarsus with shallow, wide facet for fibular articulation (Martin and Cordes-
Person, 2007)
Head and tuberculum of rib well separated (Martin and Cordes-Person, 2007)
Femur with broad, smooth popliteal fossa (Martin and Cordes-Person, 2007)
Femur with wide and deep intercondylar fossa (Martin and Cordes-Person, 2007)
Tarsometatarsus with proximal foramina (Martin and Cordes-Person, 2007)
Ischium with indistinct anterior fossa (Martin and Cordes-Person, 2007)
Femur with prominent internal condyle (Martin and Cordes-Person, 2007)
Fibular crest of tibiotarsus long and straight, extending at least half length of shaft
(Martin and Cordes-Person, 2007)
In need of morphometric revision:
Intertrochlear notches of distal tarsometatarsus extend to nearly midshaft (Martin
and Cordes-Person, 2007)
84

Tibiotarsus with expanded proximal articulation (Martin and Cordes-Person, 2007)
Enaliornithidae
Unsupported characters:
Reduced antitrochanter, does not project strongly laterally (Galton and Martin,
2002)
Tarsometatarsus with distinct caudomedial ridge leading to metatarsal III
trochlea(Galton and Martin, 2002)
Absence of distinct femoral neck (Galton and Martin, 2002)
Tarsometatarsus arched in distal view (Galton and Martin, 2002)
Centra of preacetabular synsacrum transversely constricted to form a longitudinal
ridge (Galton and Martin, 2002)
Cranioproximal process of tarsometatarsus originates from proximal end of
metatarsal III, in place of hypotarsus (Galton and Martin, 2002)
Cranial edge of metatarsal IV trochlea caudal to prominent cranial edge of subqeual
trochlea III (Galton and Martin, 2002)
Enaliornis barretti
Unsupported characters:
Large, rugose femoral trochanter (Galton and Martin, 2002)
Laterally flared, large cnemial crest on proximal tibiotarsus (Galton and Martin,
2002)
Proximal outline of medial cotyla of tarsometatarsus almost square (Galton and
Martin, 2002)
In need of morphometric revision:
Larger than E. seelyi and E. sedgewicki (Galton and Martin, 2002)
Distal tibiotarsus proportionally deep and narrow with massive lateral and medial
condyles cranially (Galton and Martin, 2002)
Enaliornis sedgewicki
Unsupported characters:
Caudally, intercondylar angle slightly lateral in position (Galton and Martin, 2002)
In need of morphometric revision:
Smaller than E. barretti and E. seeleyi (Galton and Martin, 2002)
Lateral and medial condyles of tibiotarsus reduced and nearly equal in size (Galton
and Martin, 2002)
Broad, shallow intercondylar fossa on distal tibiotarsus (Galton and Martin, 2002)
Enaliornis seeleyi
Unsupported characters:
Moderately flared lateral cnemial crest on proximal tibiotarsus (Galton and Martin,
2002)
Medial cotyla of tarsometatarsus has subrectangular outline in proximal view;
transverse width is half of cranio-caudal depth (Galton and Martin, 2002)
85

In need of morphometric revision:
Medium-sized species of Enaliornis (Galton and Martin, 2002)
Strongly developed, rounded lateral and medial condyles on distal tibiotarsus
(Galton and Martin, 2002)
Intercondylar fossa on distal tibiotarsus narrower and deeper than in E. sedgewicki
(Galton and Martin, 2002)
Hesperornithidae
Supported characters:
Medial ridge of metatarsal IV trochlea enlarged relative to lateral (Bell and Everhart,
2009)
Metatarsal IV trochlea wider than that of III (Tokaryk et al., 1997)
Asiahesperornis
Supported characters:
Prominent medial and lateral grooves present on medial portion of dorsal surface of
tarsometatarsus (Nessov and Prizemlin, 1991)
Unsupported characters:
Tarsometatarsus shaft distinctly compressed (Dyke et al., 2006)
Tarsometatarsus lacks laterally twisted shaft (Dyke et al., 2006)
Parallel lateral and medial sides of tarsometatarsus (Kurochkin, 2000)
Lateral groove on tarsometatarsus terminates in distinct fossa just proximal to
metatarsal III trochlea (Dyke et al., 2006)
Medial condyle of tibiotarsus markedly medio-laterally compressed (Nessov and
Prizemlin, 1991)
Cranial intercondylar furrow of tibiotarsus comparatively deep (Nessov and
Prizemlin, 1991)
Dorsal facet at midshaft of tarsometatarsus deep and narrow, covered by high
dorsolateral crest, with separate medial facet developed on distal shaft
(Kurochkin, 2000)
In need of morphometric revision:
Metatarsal II trochlea deflected medially; clearly separated from trochlea III by
prominent intertrochlear groove (Dyke et al., 2006)
Tarsometatarsus shaft slender and gracile (Dyke et al., 2006)
Scar for attachment of metatarsal I reduced and proximally located on shaft (Nessov
and Prizemlin, 1991)
Canadaga
Supported characters:
Concavitas lateralis large and deep, occupies entire lateral face of centrum of
thoracic vertebrae (Hou, 1999)
Unsupported characters:
Dorsal process of vertebral arch of thoracic vertebrae short and robust (Hou, 1999)
86

Well-developed area of elastic ligament anterior to arcus vertebrae dorsal process of
thoracic vertebrae (Hou, 1999)
Centrum wider than zygapophyses of thoracic vertebrae (Hou, 1999)
Large cavity present inside transverse processes of thoracic vertebrae (Hou, 1999)
Large process ventralis occupying anteromedial part of ventral side of centrum of
thoracic vertebrae (Hou, 1999)
Posterior part of thoracic vertebrae expanded laterally (Hou, 1999)
In need of morphometric revision:
End of centrum of thoracic vertebrae with fan-like expansion and extremely
restricted midline (Hou, 1999)
Hesperornis
In need of morphometric revision:
Distal intercondylar groove of tibiotarsus deeply excavated with smooth, well-
defined walls (Marsh, 1893)
Hesperornis altus
Unsupported characters:
Lateral condyle extends relatively further distally than in H. regalis and medial
condyle more in line with margin of shaft (Marsh, 1893)
In need of morphometric revision:
Size two-thirds that of H. regalis (Marsh, 1893)
Hesperornis bairdi
Unsupported characters:
Metatarsal IV trochlea more enlarged and distal to trochlea III than in
Parahesperornis (Martin and Lim, 2002)
In need of morphometric revision:
Smaller than H. gracilis (Martin and Lim, 2002)
Hesperornis chowi
Unsupported characters:
Inner metatarsal ridge shorter and less prominent than in H. regalis(Martin and Lim,
2002)
Anterior metatarsal ridge more slender than in H. regalis (Martin and Lim, 2002)
Metatarsal IV trochlea not as enlarged as in H. regalis (Martin and Lim, 2002)
In need of morphometric revision:
More elongate tarsometatarsus with more slender shaft than in H. regalis (Martin
and Lim, 2002)
Hesperornis crassipes
Supported characters:
Five sternal pits for rib attachments (Marsh, 1876)
87

In need of morphometric revision:
Larger, more robust than H. regalis (Marsh, 1880)
Unable to evaluate:
Lateral cotyla of tarsometatarsus more distally inclined than medial (Rees and
Lindgren, 2005)
Large tuberosity on upper half of medial face of tarsometatarsus (Marsh, 1876)
Hesperornis gracilis
In need of morphometric revision:
Tarsometatarsus more slender than H. regalis (Marsh, 1876)
Hesperornis macdonaldi
In need of morphometric revision:
Smallest of Hesperornis species (Martin and Lim, 2002)
Hesperornis mengeli
Unsupported characters:
Metatarsal II trochlea completely behind that of metatarsal III (Martin and Lim,
2002)
In need of morphometric revision:
Smaller than H. bairdi (Martin and Lim, 2002)
More slender tarsometatarsus than other Hesperornis species(Martin and Lim, 2002)
Metatarsal III trochlea relatively smaller and distally more compressed than in other
species of Hesperornis (Martin and Lim, 2002)
Hesperornis regalis
Unsupported characters:
Metatarsal IV trochlea nearly twice the size of trochlea III and extends further
distally (Martin, 1984)
Hesperornis rossicus
Supported characters:
Trochlear condyle of metatarsal II completely behind that of digit III (Panteleyev et
al., 2004)
Unsupported characters:
Medial and lateral cotylae of tarsometatarsus flattened (Rees and Lindgren, 2005)
Medial cotyla slopes distally as compared to lateral, which is nearly horizontal (Rees
and Lindgren, 2005)
Medial cotyla of tarsometatarsus distal to lateral (Nessov and Yarkov, 1993)
Lateral edge of lateral cotyla of tarsometatarsus exceeds intercotylar prominence
proximally (Nessov and Yarkov, 1993)
Inner toes strongly reduced (Panteleyev et al., 2004)
Intercotylar eminence of tarsometatarsus has acute profile in dorsal view (Rees and
Lindgren, 2005)
88

Condyles of trochlea for digits II and III undivided (Panteleyev et al., 2004)
Proximal view, surface of tarsometatarsus more dorso-ventrally flattened in H.
rossicus than in other species of Hesperornis (Rees and Lindgren, 2005)
In need of morphometric revision:
Proximal articular surface of tarsometatarsus has strong diagonal slant (Nessov and
Yarkov, 1993)
Proximal articular surface of tarsometatarsus with very large transverse width and
small dorsoplantar depth (Nessov and Yarkov, 1993)
Largest body size, 20% larger than H. regalis (Nessov and Yarkov, 1993)
Parahesperornis
Supported characters:
Lacrimal more elongated dorso-ventrally than in Hesperornis (Martin, 1984)
Unsupported characters:
Coracoid more elongate than in Hesperornis (Martin, 1984)
Nasal process of lacrimal more extended anteriorly than Hesperornis (Martin, 1984)
Orbital process of quadrate very elongate (Martin, 1984)
Skull mesokinetic (Martin, 1984)
In need of morphometric revision:
Metatarsal IV trochlea about one-quarter larger than trochlea III and both with
similar distal extent (Martin, 1984)
Femur more elongate and proximal end less extended laterally than Hesperornis
(Martin, 1984)
Tibiotarsus less compressed than Hesperornis (Martin, 1984)
1697
1698
89

Table 3. 1699
Specimen Age Formation Locality Citation
Baptornis advenus

YPM 1465 CON Niobara Frmtn Kansas (USA) 1
AMNH 5101 CON Niobara Frmtn Kansas (USA) 2
FMNH 395 CON Niobara Frmtn Kansas (USA) 2
KUVP 16112 CON Niobara Frmtn Kansas (USA) 3
KUVP2290 CON Niobara Frmtn Kansas (USA) 2
UNSM 20030 CON Niobara Frmtn Kansas (USA) 3
YPM 5768 CON Niobara Frmtn - -
Baptornis indt.

FHSM VP 6318 CEN Greenhorn Frmtn Kansas (USA) 4
RM PZ R1261 CMP - Kristinistad Basin (Sweden) 5
MGUH 1908.214* CMP - Kristinistad Basin (Sweden) 5
SDSM 5314 CMP Pierre Shale South Dakota (USA) -
SDSM 5893 CMP Pierre Shale South Dakota (USA) -
SMNH P2306.2 CMP Judith River Frmtn Saskatchewan (Canada) 6

1700

1701

1702
90

Table 4.
1703
Table 4.A Pasquiaornis hardei Pasquiaornis tankei
Cranial:

RSM 2831.55, 2989.20, 2989.38,
2995.4, 2997.38, 2997.85 Frontal RSM 2957.12, 2997.36
Angular RSM 2831.18, 2988.19, 2989.37 RSM 2986.2, 2989.36
Splenial RSM 2831.21, 2985.9, 2988.27 RSM 2989.193, 2989.194
Quadrate RSM 2077.120, 2831.52, 2988.25
Maxilla RSM 2988.22, 2995.5
Dentary RSM 2526.4, 2831.6, 2989.19
RSM 2626.42, 2831.5, 2957.23,
2985.10, 2988.10, 2988.11, 2989.21,
2997.35
Vertebrae:
Cervical
RSM 2626.16, 2831.9, 2957.16,
2957.17, 2987.13, 2987.16, 2987.17,
2988.13, 2988.14, 2989.24, 2997.52,
2997.53, 2997.54, 2997.55, 3015.20,
3015.6, 3015.7
Thoracic RSM 2831.7
RSM 2626.39, 2831.10, 2831.11,
2957.15, 2985.1, 2985.3, 2985.4,
2987.18, 2989.23, 2989.22, 2997.50,
2997.51
Unidentified RSM 2831.8 RSM 2988.12
Caudal
RSM 2626.15, 2987.14, 2987.15,
2997.84
Pelvis/Synsacrum
RSM 2626.18, 2957.29, 2831.12,
2989.25, 2988.17, 2989.39,
2997.62, 2997.63
RSM 2077.123, 2077.124, 2077.127,
2626.27, 2626.29, 2626.6, 2831.53,
2987.20, 2988.16, 2997.56, 2997.64

1704

1705
91

Table 4.B Pasquiaornis hardei Pasquiaornis tankei
Scapula RSM 2985.8, 2987.21
RSM 2830.3, 2831.56, 2957.19,
2997.58, 2997.59, 2997.60, 3015.16,
3015.17, 3015.8, 2997.82
Coracoid RSM 2831.54
RSM 2077.113, 2626.10, 2626.11,
2626.30, 2957.10, 2957.20, 2957.9,
2985.7, 2987.5, 2987.6, 2987.7,
2988.9, 2997.31
Humerus
proximal
RSM 2831.2, 2957.7, 2987.4,
2988.18, 2988.6, 2988.5, 2988.7,
2989.11, 2989.14, 2989.13, 2995.1
distal RSM 2487.3, 2989.15
RSM 2467.8, 2487.4, 2957.24,
2957.25, 2957.26, 2987.2, 2987.3,
2989.12, 2989.10
Radius RSM 2997.74
proximal RSM 2987.26, 2997.75, 3015.10
RSM 2526.2, RSM 2526.3, 2831.20,
2831.57, 2987.25, 2988.26, 2989.30,
2997.24, 2997.25
Ulna
proximal RSM 2995.3, 2997.27, 2997.28 RSM 2989.17
distal RSM 2997.29 RSM 2989.16, 2997.30, 3015.5
Carpometacarpus
proximal
RSM 2831.4, 2987.8, 2989.18,
2997.32, 2997.33, 2997.34, 3015.11
distal RSM 2989.34, 2997.78

1706

1707
92

Table 4.C Pasquiaornis hardei Pasquiaornis tankei
Femur
RSM 2077.60 (paratype); RSM
2077.59, 2077.60, 2997.10,
2997.12, 2997.4
RSM 2077.108 (paratype); 2077.109,
2487.2, 2989.1, 2997.13, 2997.2,
2997.3, 2997.6, 2997.7,
proximal
RSM 3015.2, 3015.3, 2409.1,
2831.1, 2988.1, 2997.79, 3015.1 RSM 2626.36, 2957.5
distal
RSM 2077.62, 2626.35, 2626.37,
2997.14, 2997.9
RSM 2077.10, 2077.107, 2077.116,
2409.3, 2467.2, 2467.3, 2626.34,
2988.2, 2988.20, 2997.11, 2997.5
Tibiotarsus

proximal RSM 2830.4, 2997.42, 2997.46
RSM 2487.8, 2957.22, 2987.10,
2997.43, 2997.44, 2997.45, 3015.4
distal
RSM 2626.20, 2626.40, 2957.13,
2957.14, 2995.7, 2997.41, 2997.42,
2997.47, 2997.48 RSM 2957.21
Fibula
(proximal) RSM 2626.41, 2988.21
RSM 2989.35, 2995.9, 2997.77,
2997.8, 3015.12, 2997.80
Tarsometatarsus
RSM 2077.117 (holotype); RSM
2987.1, 2997.15, 2997.18
RSM 2077.63 (holotype); RSM
2957.27, 3015.19
proximal
RSM 2409.11, 2409.9, 2487.7,
2830.1, 2989.2, 2989.8, 2989.9,
2997.16, 2997.19, 2997.22,
2997.81, 2997.83; SMNH 2077.110
RSM 2077.118, 2077.119, 2077.72,
2409.2, 2997.21
distal
RSM 2077.125, 2409.49, 2830.2,
2626.31, 2626.33, 2957.6, 2985.6,
2988.3, 2989.3, 2989.4, 2989.6,
2997.20, 3015.18 RSM 2077.79, 2077.XX, 2997.23
Phalanx
RSM 2831.15, 2831.6, 2985.5,
2989.40, 2997.69, 3015.14
RSM 2626.19, 2626.31, 2831.14,
2987.24, 2988.23, 2988.24, 2989.28,
2989.27, 2989.29, 2995.8, 2989.41,
2997.65, 2997.67, 2997.66, 2997.68,
2997.70, 2997.71, 2997.72, 2997.73,
3015.5, 3015.9

1708

1709
93

Table 5. 1710
Material Age Formation Locality
Brodavis
americanus

RSM P2315.1 tmt (distal+shaft) MAA Frenchman Frmtn Saskatchewan
(Canada)
Brodavis baileyi
UNSM 50665 tmt (shaft) CMP Pierre Shale South Dakota
(USA)
Brodavis
mongoliensis

PIN 4491-8 tmt
(proximal+shaft)
MAA Nemegt Frmtn Aimag (Mongolia)
Brodavis varneri
SDSM 68430 rib, vrt, plv, fmr,
fib, tbt, tmt
CMP Pierre Shale South Dakota
(USA)
1711
Table 6. 1712
E. barretti E. sedgewicki E. seeleyi
Holotype

BGS 87935
(d. tibiotarsus)
Lectotype
BMNH A477
(d. tarsometatarsus)
SMC B55314
(p. tibiotarsus)
Paralectotypes

Braincase SMC B54404
Vertebrae

Cervical YORYMG 507
Thoracic SMC B55277
Pelvis/synsacrum
SMC B55282, BGS
87936
Femur

proximal

BMNH A479,
SMC B55287
distal SMC B55295
Tibiotarsus

proximal SMC B55312
distal SMC B55316 SMC B55315
Tarsometatarsus

proximal BGS 87932
1713
1714
94

Table 7. 1715
E. barretti E. sedgewicki E. seeleyi
Braincase
YORYMG 585,
YORYMG 586
Pelvis/Synsacrum SMC B55284 SMC B55283
Femur SMC B55299
proximal
BMNH A483, BMNH
A5803 SMC B55288
BMNH A5803, SMC
B55290, SMC B55291,
SMC B55929, SMC
B55293
distal
BMNH A163, SMC
55303, SMC 55306,
YORYMG 591, SMC
B55304
SMC B55297, SMC
B55298, SMC
B55300, SMC
B55310, BMNH
A482, BMNH A5802,
YORYMG 583
BMNH A483c, BMNH
A484, BMNH
A485bBMNH A5801,
BMNH 41792, SMC
B55307, SMC B55308,
BGS 87929
Tibiotarsus
proximal SMC B55313
BMNH A480,
YORYMG 581
BMNH A478, BMNH
A481, YORYMG 587,
YORYMG 588
distal SMC B55322 SMC 55317
Tarsometatarsus
proximal SMC B55331
SMC B55318, SMC
B55319
distal
BMNH 41790, SMC
B55320
BMNH A485a,BMNH
A485b, SMC B55321,
YORYMG 560,
YORYMG 589
1716
1717

1718
95

Table 8. 1719
Asiahesperornis bazhanovi Material Citation
IZASK 5/287/86a tarsometatarsus shaft Nessov and Prizemlin, 1991
IZASK 5/287/86b partial tarsometatarsus Nessov and Prizemlin, 1991
IZASK 220/B-2003 partial tarsometatarsus Dyke et al., 2006
IZASK 4/KM 97 partial dentary Dyke et al., 2006
IZASK 22/KM 97 single tooth Dyke et al., 2006
IZASK 2/KM 97 cervical vertebra Dyke et al., 2006
IZASK 5/287/86 thoracic vertebra Nessov and Prizemlin, 1991
IZASK 5/293/87 thoracic vertebra Nessov and Prizemlin, 1991
IZASK 5/KM 97 partial femur Dyke et al., 2006
IZASK 5/287/86B * partial tibiotarsus Nessov and Prizemlin, 1991
IZASK 218/B-2003 partial tibiotarsus Dyke et al., 2006
IZASK 1/KM 97 partial tibiotarsus Dyke et al., 2006
IZASK 3/KM 97 partial tibiotarsus Dyke et al., 2006
1720
Table 9. 1721
Canadaga
arctica Material Age Formation Locality Citation
NMC 41050
cervical vertebrae
15-17 MAA Bylot Island Hou, 1999
NMC 41053 femur MAA Bylot Island Hou, 1999
NMC 41054 femur MAA Bylot Island Hou, 1999
NMC 41064 caudal vertebra MAA Bylot Island Hou, 1999
NUVF 284
cervical vertebrae
16-18 CON Kanguk Frmtn Devon Island Wilson et al., 2011
1722
1723
96

Table 10. 1724
Material Age Formation Locality Citation
Hesperornis altus*

YPM 515 tbt CMP Claggett Shale Montana (USA) Shufeldt, 1915
Hesperornis bairdi

YPM PU 17208A plv, tmt CMP Pierre Shale
South Dakota
(USA)
Martin and Lim,
2002
Hesperornis chowi

YPM PU 17208 tmt CMP Pierre Shale
South Dakota
(USA)
Martin and Lim,
2002
YPM PU 18589
vrt, scp, stn,
fmr, pat, tbt,
fib CMP Pierre Shale
South Dakota
(USA)
Wilson et al.,
2009
YPM PU 17193
fmr, pat, tbt,
fib CMP Pierre Shale
South Dakota
(USA)
Wilson et al.,
2009
Hesperornis
crassipes

YPM 1474
nearly
complete CON Niobrara Frmtn Kansas (USA) Marsh, 1880
Hesperornis gracilis

YPM 1473 tmt, phx CON Niobrara Frmtn Kansas (USA) Marsh, 1880
YPM 1478
vrt, fmr, tbt,
tmt, phx CON Niobrara Frmtn Kansas (USA) Marsh, 1880
YPM 1679
vrt, fmr, pat,
tbt, fib, tmt,
phx CON Niobrara Frmtn Kansas (USA) Marsh, 1880
Hesperornis
macdonaldi

LACM 9728 fmr CMP Pierre Shale
South Dakota
(USA)
Martin and Lim,
2002
Hesperornis mengeli

CFDC
B.78.01.08 tmt CMP Pierre Shale
South Dakota
(USA)
Martin and Lim,
2002
YPM PU 17208 vrt, plv, fmr CMP Pierre Shale
South Dakota
(USA)
Martin, pers.
comm.
Hesperornis
montana*

USNM 8199 vrt CMP Claggett Shale Montana (USA) Shufeldt, 1915.
Hesperornis regalis

YPM 1200
vrt,fmr, tbt,
ttmt,pat, fib,
phx CON Niobrara Frmtn Kansas (USA) Marsh, 1872
97

multiple
vrt, pat, fmr,
tbt, tmt CMP
Smoking Hills
Frmtn
Northwest
Territories
(Canada) Russell, 1967
multiple
crn, vrt, fmr,
tbt, tmt
Vermillion
River Frmtn
Manitoba
(Canada) Bardack, 1968
UCMP V103841 vrt
CON-
CMP Ignek Frmtn Alaska (USA) Bryant, 1983
multiple
nearly
complete CON Niobrara Frmtn Kansas (USA) multiple
Hesperornis rossicus

VPM N 26306/2 tmt CMP
Rybushka
Frmtn
Saratov
Province
(Russia)
Nessov and
Yarkov, 1993
RM PZ R398 tmt CMP -
Kristinistad
Basin (Sweden)
Rees and
Lindgren, 2005
SGU 3442 Ve01 tmt CMP -
Kristinistad
Basin (Sweden)
Rees and
Lindgren, 2005
ZIN PO 5463 tmt CMP
Rybushka
Frmtn
Saratov
Province
(Russia)
Panteleyev et al.,
2004
ZIN PO 5464 tmt CMP
Rybushka
Frmtn
Saratov
Province
(Russia)
Panteleyev et al.,
2004
1725
Table 11. 1726
Material Citation
Parahesperornis alexi
KUVP2287
crn, vrt, plv, cor, stn, hum, fmr, fib,
pat, tbt, tmt, phx Martin, 1984
KUVP 24090
crn, vrt, plv, cor, scp, stn, fmr, fib,
pat, tbt Martin, pers. comm.
Parahesperonis indt.
FHSM VP-17312 tmt Bell and Everhart, 2009
1727
1728
98

1729
Table 12. 1730
Class Family Genus Species
Hesperornithiformes
9/16/1
Enaliornithidae
0/7/0
Enaliornis
barretti 0/3/2
sedgewicki 0/1/3
seeleyi 0/2/3
Baptornithidae
4/12/2
Baptornis
0/3/0
advenus
varneri
Judinornis
0/3/0
nogontsavensis
Parascaniornis stensoei
Pasquiaornis
0/3/1
hardei 0/3/1
tankei 0/3/0
Brodavidae
0/0/2
Brodavis
americanus 0/1/3
baileyi 0/2/3
mongoliensis 0/2/2
varneri 1/9/2
Hesperornithidae
2/0/0
Asiahesperornis
1/7/3
bazhanovi
Canadaga
1/6/1
arctica
Coniornis altus
Hargeria gracilis
Hesperornis
0/0/1
altus 0/2/1
bairdi 0/1/1
chowi 0/3/1
crassipes 1/1/2
gracilis 0/0/1
macdonaldi 0/0/1
mengeli 0/1/3
montana
regalis 0/1/0
rossicus 1/8/3
Lestornis crassipes
Parahesperornis
1/4/3
alexi
1731
1732

99

1
Chapter 2. The Hesperornithiformes: a phylogenetic analysis 2
INTRODUCTION 3
From the late 1800’s to the late 1900’s, research on the hesperornithiforms consisted 4
predominantly of describing new taxa. This focus on descriptive work without synthesizing 5
research resulted in a plethora of little-understood taxa, many based on a single fossil element. 6
Studies looking at the evolutionary context of hesperornithiforms as a group were rare, and most 7
saw these birds as a primitive group of modern birds (e.g. Marsh, 1880; Heilman, 1926; Lucas, 8
1903). It was not until the 1980’s that a few authors began to evaluate relationships within the 9
Hesperornithiformes. Unfortunately, these studies lacked scientific rigor and were often seriously 10
flawed in their methodology (Cracraft, 1982; Martin, 1984; Elzanowski and Galton, 1991). 11
More recent studies with a sound scientific basis have resolved the placement of the 12
hesperornithiforms within the phylogeny of ancient birds (e.g. Chiappe, 2002; Clarke, 2004; Bell 13
et al., 2010). The discovery of abundant avian fossils in China, starting in the 1990’s and 14
continuing to this day, has prompted a dramatic increase in the amount of scientific research into 15
the evolution of early birds (Zhou, 2004; O’Connor et al., 2011). While some groups of birds 16
remain poorly resolved, the hesperornithiforms have been consistently placed as one of the 17
nearest sister groups to modern birds, in a clade referred to as the Ornithurae (Figure 1, e.g. 18
Chiappe, 2002; Chiappe & Dyke, 2002; Zhou, 2004, Bell et al., 2010). These birds are united by 19
a number of synapomorphies, such as a reduced acetabulum relative to the length of the ilium, a 20
prominent patellar groove on the femur, and a cranial cnemial crest on the tibiotarsus (Chiappe, 21
2002). While studies such as these have firmly established the placement of hesperornithiforms 22

100

within the evolutionary tree of Mesozoic birds, the relationships among the more than seven 23
genera and 25 named species of hesperornithiform birds remain poorly understood. 24
In his initial work with Hesperornis and Baptornis, Marsh (1880) believed these birds to 25
be essentially modern, despite the presence of ancestral characters like teeth, and most likely 26
related to the ratites. Marsh based this determination on apparent similarities of the palate of 27
Hesperornis with that of the ostrich (1872; 1880). Despite Marsh’s initial assessment, later 28
researchers never seriously considered the hesperornithiforms to be part of the modern ratite 29
clade, the palaeognathes, but instead thought them to be either closely relative to, but outside of, 30
modern birds (Lucas, 1903; Heilman, 1926) or basal ancestors of modern lineages, such as the 31
grebes (Brodkorb, 1963, 1971), the loons (Simpson, 1980), or a combination of the two 32
(Cracraft, 1982). Despite utilizing morphological similarities to determine relationships between 33
hesperornithiforms and modern birds, none of these early studies employed rigorous cladistic 34
methods. Furthermore, the relationships of various hesperornithiform species to each other have 35
never been investigated. 36
Homology and Convergence within the Hesperornithiformes 37
The first attempt to describe evolutionary relationships within the hesperornithiforms was 38
an analysis that included Hesperornis, Baptornis, and Enaliornis, as well as modern loons, 39
grebes, penguins, and pelicans made by Cracraft (1982). This paper presented a phylogenetic 40
hypothesis (Cracraft, 1982, p.37) that placed Hesperornis and Baptornis as sister taxa to loons 41
and grebes in a monophyletic clade called the Gaviomorphae (Fig. 2). For each node on the 42
proposed phylogenetic tree a list of synapomorphies was provided, with a total of 26 characters 43
presented as support for the proposed hypothesis. The placement of the hesperornithiforms with 44
the modern birds is the result of inappropriate outgroup taxa. By using the modern pelicans, 45

101

certain features were mischaracterized as ancestral, when, in fact, they are derived within modern 46
birds. The inclusion of Archaeopteryx would have resulted in a much different understanding of 47
character polarity in the analysis and helped differentiate convergence from homology. 48
Lucas (1903) was one of the first authors to note that among aquatic birds convergence of 49
morphological features is prevalent and can complicate the interpretation of evolutionary 50
relationships. Stolpe (1935) later undertook an in-depth study of convergence as it relates to the 51
hesperornithiforms, and found that all of the features previously used to unite Hesperornis with 52
modern loons and grebes in a single group were most likely the result of convergence influenced 53
by the similar aquatic lifestyles of these birds (Stolpe, 1935). 54
A close look at the details of the apparently similar morphology shared by these three 55
groups of divers - loons, grebes, and hesperornithiforms – reveals key differences that point to 56
convergence rather than homology. First, all three birds have a greatly expanded region proximal 57
to the tibia for the attachment of the M. gastrocnemius and the M. flexor perforates digiti IV, 58
muscles important for moving the foot during swimming (Wilcox, 1952). However, the structure 59
of this expansion is different in each of these birds. In loons, the muscles attach to a greatly 60
expanded cnemial crest on the proximal tibia with no role for the patella; grebes possess a 61
somewhat reduced cnemial crest (as compared to loons), in conjunction with a robust patella that 62
is incorporated into the muscular attachments (Storer, 1958); and in Hesperornis a moderate 63
cnemial expansion is present on the tibia along with a massive and highly elongate patella that 64
bears numerous muscle scars. Thus each of these lineages has accomplished a superficially 65
similar morphology in different ways. In regard to the modern loons and grebes, molecular 66
analysis has since confirmed that the two groups are only distantly related (Hackett et al., 2008; 67
McCormack et al., 2013). 68

102

Lack of Rigorous Methodology in Hesperornithiform Phylogenetics 69
Another problem influencing studies of hesperornithiform evolution has been an 70
anecdotal approach that lacks the rigor of a comprehensive cladistic analysis. For example, in an 71
effort to test the ratite affinities hypothesized by Marsh (1880), Houde undertook a histological 72
comparison of Hesperornis with multiple palaeognathous and neognathous modern birds (1987). 73
The microstructure of the bone in the hindlimb of Hesperornis was most similar to that of 74
modern neognathous birds, and unlike that of palaeognathous birds (Houde, 1987). While useful, 75
evolutionary studies must be based on more than a single feature. 76
Another study that has attempted to resolve relationships within the Hesperornithiformes 77
and also suffers from a lack of rigorous methodology was presented by Martin to explore 78
relationships among Enaliornis, Baptornis, Parahesperornis, and Hesperornis (1984). Thirty-six 79
characters were identified that united the Hesperornithiformes as a monophyletic group and 80
which placed Parahesperornis and Hesperornis as most closely related within the 81
hesperornithiforms, with Enaliornis as the most primitive member of the group (Fig. 3). This 82
study was also the first to identify hesperornithiforms as more primitive birds, outside the crown 83
group of modern birds (Martin, 1984). However, this analysis was not a cladistic analysis in the 84
sense that a group of characters was developed, coded for all taxa, and analyzed using accepted 85
phylogenetic algorithms. Rather, features which grouped taxa together were identified arbitrarily 86
and then presented as a phylogenetic tree which was more a visualization of an a priori 87
hypothesis than a phylogeny derived from character optimization. This approach is a departure 88
from common practice; however lack of alternative research has led to this study being cited 89
repeatedly as representing the current understanding of hesperornithiform relationships. 90

103

The final and most recent attempt to categorize hesperornithiform relationships was made 91
by Elzanowski and Galton (1991) based on their highly detailed study of the braincase of 92
Enaliornis and Hesperornis as well as Archaeopteryx and several modern birds. Elzanowski and 93
Galton (1991) identified seventeen morphological characters of the braincase that were of 94
evolutionary significance, and then proceeded to discuss them in an evolutionary context without 95
actually coding the features and conducting a cladistic analysis. While such a discussion is useful 96
in determining the polarity of characters, it is in no way a substitute for a rigorous cladistic 97
analysis. 98
The studies presented above highlight a trend in the reliance of hesperornithiform 99
researchers on anecdotal and outdated methods. In order to address these gaps in 100
hesperornithiform research, the study presented here aims to develop a robust morphological 101
dataset for a cladistic analysis investigating the relationships of hesperornithiform birds. 102
Furthermore, as the Hesperornithiformes are in need of a great deal of revisionary taxonomy (see 103
Chapter 1), a goal of this study is to explore the cladistic basis for existing taxonomic units, as 104
well as identify any new clades that may be suitable for taxonomic recognition on the basis of 105
shared features. 106
107
METHODS 108
Morphological Characters 109
Analysis of around 250 hesperornithiform specimens as well as other basal and modern 110
birds resulted in the development of a comprehensive database of 122 previously unidentified 111
cranial and postcranial characters. Additionally, 41 characters were adapted from other large 112
cladistic analyses (Elzanowski and Galton, 1991; Chiappe, 2002; Clarke, 2004), for a total of 163 113

104

characters. While all portions of the skeleton are represented in the matrix, the majority of 114
characters relate to the pelvic limb, as there is considerable bias in the hesperornithiform fossil 115
record for these elements (Table 1). The complete list of characters can be found in Appendix 1, 116
which presents both a written description as well as illustrations explaining each character and 117
character state. The coded matrix for the specimens in this analysis can be found in Appendix 2. 118
Of the characters, 49 were multistate and of these 17 were ordered (see Appendix 1). Ordering 119
was imposed when a morphocline was present (Slowoski, 1993; Thiele, 1993), such as: femoral 120
trochanter and head - nearly continuous (0), separated by a shallow notch (1), separated by a 121
deep notch (2) (Character 59, Appendix 1). Character polarity for binomial and multistate 122
characters was determined by comparison to the basal bird Archaeopteryx or to dromaeosaurid 123
theropods. Whenever possible, characters were erected to reflect the diagnostic features that have 124
been reported in the literature to define each taxonomic grouping. For example, the degree of 125
rounding of the intercotylar eminence of the tarsometatarsus (Martin and Lim, 2002) as well as 126
the relative breadths of the distal metatarsals (Martin et al., 2012) have been suggested as 127
features that vary among species assigned to the genus Hesperornis and have thus been cited as 128
diagnostic features. Therefore, character 131 (Intercotylar eminence - absent (0); reduced to near 129
absence (1); low relief, rounded (2); high relief, peaked (3)) and character 148 (Distal end of 130
tarsometatarsal shaft (just above trochlea), relative widths of metatarsals - II similar width to III 131
and IV (0); III slightly wider than IV, II narrowest (1); II narrower than III and IV, which are of 132
similar widths (2); II, III, and IV progressively wider, with IV twice as wide as III (3)) were 133
erected to incorporate these diagnostic features into the cladistic analysis. It should be noted that 134
during the development of the matrix, a number of characters that had been proposed as 135
diagnostic features were unable to be identified in the specimens, and so were excluded from 136

105

inclusion in the analysis. These instances will be more fully discussed in the revisionary 137
systematics of the various groups. 138
Specimens Included 139
One of the goals of this study is to explore the use of cladistics as a tool for supporting an 140
updated taxonomy of hesperornithiforms. As discussed previously, a number of 141
hesperornithiform taxa are poorly described and not robustly supported, making the assignment 142
of specimens to a given taxon problematic. Therefore, rather than use taxa which may be 143
composed of unrelated specimens as operational taxonomic units (OTUs) in the analysis, in the 144
case of most hesperornithiform material individual specimens were used as OTUs. Specimens 145
were selected on the basis of completeness or potential taxonomic significance, as well as 146
availability for study (Table 1). Eight holotype specimens were available for direct study and 147
were included in the analysis: Baptornis advenus (YPM 1465), Brodavis baileyi (UNSM 50665); 148
Brodavis varneri (SDSM 68430, formerly Baptornis varneri, see Martin et al., 2011), 149
Hesperornis regalis (YPM 1200), H. bairdi (YPM 17208a), H. chowi (YPM 17208), H. gracilis 150
(YPM 1473), H. macdonaldi (LACM 9728), and Parahesperornis alexi (KUVP 2287). Two 151
additional holotypes, Hesperornis mengeli and Hesperornis rossicus, were unavailable for direct 152
study. Hesperornis mengeli (CFDC B780108, mistakenly reported as BO 780106 by Martin and 153
Lim, 2002) was coded for the matrix using publications (Martin and Lim, 2002) and 154
photographs. The holotype of Hesperornis rossicus (VPM N 26306/2, a proximal 155
tarsometatarsus) does not have photographs published, and so instead an assigned 156
tarsometatarsus for which published photos exist was used (ZIN PO 5464; Panteleyev, et al., 157
2004) in addition to other descriptive work on the species (Rees and Lindgren, 2005). While this 158
approach is not ideal, it was necessitated by the inability to access the physical specimens. 159

106

In addition to the holotypes, a fairly loose criterion was adopted for inclusion of other 160
hesperornithiform specimens. In order to fully sample the diversity of hesperornithiforms, all 161
specimens for which over 50% of the matrix could be coded were included (Baptornis advenus: 162
FMNH 395, KUVP 2290, UNSM 20030; Hesperornis gracilis: YPM 1679, YPM 55000; 163
Hesperornis regalis: FMNH 281, NHM 882, YPM 1476; Hesperornis species: SDSM 5312, 164
USNM 13580, YPM 1499; Parahesperornis alexi: KUVP 24090). Additionally, any specimens 165
that were regarded as atypical (i.e., not in conformation with one of the established holotype 166
taxa) were also included in this study (Baptornis advenus: AMNH 5101, FHSM 6318; 167
Hesperornis crassipes: AMNH 5102; Hesperornis chowi: YPM 18589, YPM 17193; 168
Hesperornis mengeli: YPM 17208; Hesperornis regalis: AMNH 2181, FMNH 206, FMNH 219 169
(cast as KUVP 25395), FMNH 321, KUVP 71012, USNM 13581; Hesperornis species: KUVP 170
2280a, SDSM 622, SDSM 53507, UNSM 10148, YPM 1477, YPM 1478, YPM 17208; 171
Parahesperornis species: FHSM 17312; hesperornithid undetermined: UCMP 117605. This 172
approach is justified by the large number of specimens that have never before been studied and 173
the general lack of understanding of hesperornithiform taxonomic units. 174
Additionally, several hesperornithiform groups at various taxonomic levels are highly 175
fragmentary and mostly unavailable for direct study. In these cases, the decision was made to 176
either treat a collection of specimens as a single OTU or to leave the taxa out of the study 177
completely. Asiahesperornis is known from a small number of unrelated and highly fragmentary 178
specimens that were not available for this study. As the holotype (IZASK 5/287/86a, a partial 179
tarsometatarsus) and all other material is very fragmentary, Asiahesperornis was treated as a 180
single OTU in this analysis and coded from the published material (Dyke et al., 2006) and 181
photographs provided by the author (Dyke, pers. comm.). It should be noted that because none of 182

107

the elements are associated, there is some vagueness in the assignment of the vertebrae and 183
tibiotarsi to the same taxon as the holotype tarsometatarsus (Dyke et al., 2006). Both Enaliornis 184
and Pasquiaornis are multi-specific genera known entirely from isolated, fragmentary 185
specimens. While Enaliornis has been reported to consist of three separate species (E. barretti, 186
E. sedgewicki, and E. seeleyi; Galton and Martin, 2002), as discussed in Chapter 1 there is very 187
little basis for these distinctions. Furthermore, of the three species only E. seeleyi is represented 188
by a holotype (distal tibiotarsus BGS 87935). Lectotypes and a number of paralectotypes are 189
assigned to the other species (Galton and Martin, 2002). Because of this taxonomic uncertainty, 190
Enaliornis was treated as a single OTU in this analysis and coded from the specimens listed in 191
Table 2. It should be noted that this compilation did not result in the addition of any 192
polymorphisms to the coded matrix. Pasquiaornis was also mostly unavailable for study and 193
currently highly under-reported in the literature. While a large amount of fossil material is 194
known, only preliminary reports have been made (Tokaryk et al., 1997; Cumbaa et al., 2006), 195
making inclusion of these taxa problematic. Unlike in the case of Enaliornis, combination of 196
specimens assigned to P. hardiei and P. tankei into a single OTU would have resulted in a 197
number of polymorphisms in the matrix, and so these two species were treated separately. Given 198
this morphological diversity it is difficult to be certain of the assignment of disparate elements to 199
one or the other taxa (or even to a different taxon), however at this time the limited available data 200
made a more concrete analysis impossible. The specimens used for coding in the matrix are 201
listed in Table 2. Additional information was collected from the literature (Tokaryk et al., 1997; 202
Cumbaa et al., 2006; Sanchez, 2010). Finally, a number of taxa could not be included because 203
the original specimens were unavailable and the published literature is insufficient for reliable 204

108

coding (Brodavis americanus, Brodavis mongoliensis), or the taxa are too fragmentary 205
(Canadaga arctica, Judinornis nogontsavensis). 206
Archaeopteryx lithographica, Gansus yumenensis, and Ichthyornis dispar were selected 207
as outgroup taxa for the analysis. Unlike the majority of hesperornithiform OTUs, which were 208
coded from individual specimens, Gansus yumenensis and Ichthyornis dispar were coded from 209
multiple specimens assigned to the respective taxa (G. yumenensis: IVPP 04-001, IVPP04-003, 210
IVPP 04-31, IVPP07-004). Archaeopteryx lithographica was treated as a supraspecific taxon 211
(see Chiappe, 2007 for discussion of Archaeopteryx as a single species) and coded from 212
published images and descriptions of the Thermopolis, London, Berlin, and Munich specimens 213
(Ostrom, 1976; Whetstone, 1983; Witmer, 1990; Mayr et al., 2005). The crown clade Aves was 214
represented by two modern birds, Gallus gallus (LACM 87011) and Anas clypeata (LACM 215
102905). 216
Analytical Methods 217
The parsimony program Tree Analysis using New Technology (TNT; Goloboff et al., 218
2008) was used for all analyses. Because the entire matrix of 163 characters coded for 54 OTUs 219
was too large for an exact solution (“branch and bound” methods), multiple approaches were 220
taken using both branch and bound and heuristic searches. First, the goals of the analysis were 221
split into two separate questions: 1) Can the major families of hesperornithiforms (as defined by 222
the current taxonomic framework) be phylogenetically supported? 2) Can the generic 223
assignments Hesperornis and Parahesperornis and the species-rank assignments of specimens in 224
Hesperornis be supported? To address the former question, the original data matrix was pruned 225
to 24 OTU’s (Table 1) by the substitution of a single ‘Hesperornis’ OTU coded to represent all 226
specimens assigned to that genus, with conflicting character states coded as polymorphisms. This 227

109

resulted in a data matrix of 25 OTUs. To address the latter question, the original data matrix was 228
pruned to 30 OTUs through the removal of the majority of non-hesperornithid hesperornithiform 229
taxa (Table 1). The presumed baptornithids FMNH 395 and USNM 20030 were retained due to 230
their fairly complete nature to be used as intermediary outgroups. These matrices were then run 231
separately in TNT using the Implicit Enumeration algorithm, which returns all most 232
parsimonious solutions. In order to analyze the entire data matrix, another analysis was run using 233
a heuristic method starting with Wagner trees that were then resampled using Tree Bisection 234
Reconnection (TBR) over 100 replicates. 235
The program defaults of collapsing all branches with a minimum length of 0 were 236
maintained. When multiple most-parsimonious trees were returned, strict consensus trees were 237
calculated. Jackknife supports, branch lengths, and consistency and retention indices were 238
calculated for all most-parsimonious trees. In order to assess the effect of uninformative 239
characters on the tree statistics, for each analysis uninformative characters were identified and 240
removed from consideration and the analysis rerun, with the consensus, retention, and re-scaled 241
retention index recalculated. 242
243
RESULTS 244
Analysis of each dataset is discussed separately below. For each analysis, the strict consensus of 245
the most parsimonious trees are presented, with branch lengths, mapped synapomorphies, and 246
jack-knife values over 50% shown at all nodes. 247
Evolutionary Relationships within the Hesperornithiformes 248
Analysis of the abbreviated data matrix in order to answer the first research question (Can 249
the major families of hesperornithiforms be phylogenetically supported?) resulted in five most 250

110

parsimonious trees of 317 steps. The strict consensus of these trees is found in Figure 4. The tree 251
is rooted at Archaeopteryx, with Gansus and Ichthyornis progressively more derived. Twenty- 252
one unambiguous synapomorphies unite Gansus with the remaining OTUs (Fig. 4, node A) and 253
four unite Ichthyornis with the remaining OTUs (Fig. 4, node B). A number of 254
hesperornithiforms (YPM 1465, the holotype of Baptornis advenus, Enaliornis, Pasquiaornis 255
hardiei, and Pasquiaornis tankei) form a polytomy (three unambiguous synapomorphies; Fig. 4 256
node C) with the clade Aves + Hesperornithiformes (six unambiguous synapomorphies; Fig. 4, 257
node D). 258
The clade containing the remainder of the hesperornithiforms is united by six 259
unambiguous synapomorphies and is almost fully resolved (Fig. 4, node E). This clade is 260
resolved into two monophyletic clades, one containing the brodavids and some specimens 261
assigned to Baptornis (Fig. 4, node F; united by a single unambiguous synapomorphy) and the 262
second containing Hesperornis + Parahesperornis as well as two specimens assigned to 263
Baptornis (Fig. 4, node J; united by two unambiguous synapomorphies). In the clade stemming 264
from node F, the brodavids are united with the previously unreported UCMP 117605 with a 265
single unambiguous synapomorphy in a polytomy (Fig. 4, node G) that serves as the sister-clade 266
to three specimens assigned to Baptornis advenus (Martin and Tate, 1976; Everhart and Bell, 267
2009) that are united by two unambiguous synapomorphies (Fig. 4, node H). It should be noted 268
that the holotype of Baptornis advenus, YPM 1465, is not present in this clade. In the clade 269
stemming from node J, AMNH 5101 (assigned to Baptornis advenus by Martin and Tate, 1976), 270
the most basal specimen, is the sister-terminal to a monophyletic clade uniting UNSM 20030 271
(assigned to Baptornis advenus by Martin and Tate, 1976) with Hesperornis + Parahesperornis 272

111

with seven unambiguous synapomorphies (Fig. 4, node K). Hesperornis and Parahesperornis 273
alexi are then united by twenty-one unambiguous synapomorphies (Fig. 4, node L). 274
Evolutionary Relationships within Hesperornis 275
Analysis of the second abbreviated data matrix in order to answer the other research 276
question (Can the generic assignments Hesperornis and Parahesperornis and the species-rank 277
assignments of specimens in Hesperornis be supported?) resulted in 1010 most parsimonious 278
trees of 324 steps. The strict consensus of these trees is found in Figure 5. The basal topology of 279
this tree is the same as that in Fig. 4. Here, Pasquiaornis tankei and Enaliornis do not form a 280
polytomy, but are placed outside of the clade Aves + Hesperornithiformes and united with this 281
clade by two (Fig. 5, node C) and one (Fig. 5, node D) unambiguous synapomorphy, 282
respectively. All remaining hesperornithiform terminals form a monophyletic clade united by 283
three unambiguous synapomorphies (Fig 5, node F). 284
Within the hesperornithiform clade, FMNH 395 (assigned to Baptornis advenus by 285
Martin and Tate, 1976) is the most basal specimen, followed by UNSM 20030 (Baptornis 286
advenus - Martin and Tate, 1976) which is united with the remainder of the OTUs by nine 287
unambiguous synapomorphies (Fig. 5, node G).The monophyletic clade containing specimens 288
assigned to Parahesperornis and Hesperornis is united by 25 unambiguous synapomorphies 289
(Fig. 5, node H), and has a basal polytomy between the holotype of Parahesperornis alexi 290
(KUVP 2287), FHSM 17312 (assigned to Parahesperornis by Bell and Everhart, 2009), and the 291
remainder of the OTUs. There is one unambiguous synapomorphy uniting KUVP 24090 292
(assigned to Parahesperornis by Martin, pers. comm.) with the Hesperornis OTUs (Fig. 5, node 293
I). Terminals assigned to Hesperornis as well as the representative of Asiahesperornis form a 294
monophyletic clade with LACM 9728, the holotype of H. macdonaldi, as the basal-most 295

112

member, however there are no unambiguous synapomorphies uniting this clade (Fig. 5, node J). 296
AMNH 2181 is the next basal-most terminal (Fig. 5, node K; 2 unambiguous synapomorphies), 297
followed by FMNH 219 (Fig. 5, node L, 2 unambiguous synapomorphies). Derived from FMNH 298
219 is a polytomy of YPM 55000, FMNH 316, and the remainder of the Hesperornis terminals 299
(Fig. 5, node M; one unambiguous synapomorphy), which are united by 2 unambiguous 300
synapomorphies (Fig. 5, node N). 301
Heuristic Analysis of the Complete Character Matrix 302
Given the inconsistencies between the two exact searches presented above, the heuristic 303
analysis has some importance, despite the potential for not arriving at the absolute most 304
parsimonious trees. Figure 6 shows the strict consensus of the 20 most parsimonious trees 357 305
steps long. Unlike the previous analyses, Pasquiaornis hardiei and Pasquiaornis tankei are 306
found to be monophyletic (united by a single unambiguous synapomorphy) and basal to a 307
polytomy of Enaliornis, YPM 1465, and a clade containing modern birds and the remainder of 308
the hesperornithiforms, which is united by four unambiguous synapomorphies. Within this clade, 309
much of the topography is similar to that in the preceding analyses, with the exception of the 310
placement of some of the species of Baptornis and Hesperornis, as discussed below. 311
312
DISCUSSION 313
Basal Hesperornithiform Phylogeny 314
Basal Placement of Ichthyornis. The analyses presented here differ significantly from 315
other phylogenies of Mesozoic birds in the placement of Ichthyornis as basal to other 316
hesperornithiforms and modern birds. Most studies find Ichthyornis to be derived from 317
Hesperornis and placed closer to modern birds (for example, see Chiappe, 2001; Clarke, 2004; 318

113

O’Connor et al., 2010). While in general it is difficult to compare different phylogenetic studies, 319
an attempt to do so in this case is warranted because of the significant difference in the results of 320
this study. In the early study of Chiappe (2002), Ichthyornis was united with the crown clade 321
Aves on the basis of two synapomorphies of the alular metacarpal and ungual phalanx of digit I, 322
both in the forewing. As neither of these bones are known for any of the birds commonly 323
assigned to the Hesperornithiformes, these characters were not included in the present analysis. 324
Clarke (2004) resolved Ichthyornis with modern birds on the basis of five synapomorphies. Two 325
of these (the number of sacral vertebrae anterior to the acetabulum and the presence of 326
pneumatic foramina on the dorsal surface of the sternum) were not used in the present analysis, 327
while the other three characters (ossified connective tissue present on the dorsal processes of the 328
thoracic vertebrae, lateral process of the coracoid present, and a depressed medial surface of the 329
coracoid in the region of the foramen n. supracoracoideus) were used and coded similarly. The 330
different placement of Ichthyornis and modern birds relative to hesperornithiforms can therefore 331
be regarded as an artifact of the character sampling used in the present study, the goals of which 332
were to investigate the interrelationships of the Hesperornithiformes. This result can also be used 333
as support for the inclusion of the broadest range of characters, regardless of the addition of 334
missing data, to the matrix in any phylogenetic analysis. The addition of any data is valuable, 335
whereas the addition of missing data is not necessarily damaging (Kearney and Clarke, 2003; 336
Wiens, 2003, 2005). 337
Monophyly of the Hesperornithiformes. The most significant result of the phylogenetic 338
analyses presented here is the consistent paraphyly of members of the Hesperornithiformes. 339
Regardless of the taxonomic composition of the database (Hesperornis-dominated versus basal 340
hesperornithid-dominated) and the type of search algorithm used (exact versus heuristic), 341

114

Pasquiaornis and Enaliornis (and YPM 1465, when that specimen is included) are consistently 342
placed outside of the monophyletic clade comprising the crown clade Aves and the remainder of 343
the hesperornithiform terminals. This is in contradiction to a number of descriptive studies that 344
have regarded Enaliornis as the most basal hesperornithiform (Martin and Tate, 1976; Martin, 345
1984; Galton and Martin, 2002; Martin et al., 2012); however, one study has questioned the role 346
of convergence in the morphological similarities often cited between Enaliornis and other 347
hesperornithiforms (Elzanowski and Galton, 1991). The results of the cladistic analyses 348
presented here further questions the placement of Enaliornis within the Hesperornithiformes. 349
Similarly, all previous work has accepted the initial placement of Pasquiaornis within the 350
Baptornithidae (Tokaryk et al., 1997), which is refuted by this analysis. 351
In the present analysis, a number of unambiguous synapomorphies support the clade 352
consisting of Aves + hesperornithiforms as well as the clade of derived hesperornithiforms. The 353
placement of the basal hesperornithid taxa outside of these nodes could be due to one of two 354
reasons, either the characters are coded as missing data or the characters conflict with the coding 355
of the in-group terminals. The clade Aves + hesperornithiforms is united by eight unambiguous 356
synapomorphies, of which one is coded as conflicting for all of the basal taxa, one is unknown in 357
all of the basal taxa, and six are coded as conflicted in some and absent in others of the basal 358
taxa. The in-group hesperornithiform clade is united by twelve unambiguous synapomorphies, of 359
which five are unknown in all of the basal taxa, none is unknown in all of the basal taxa, and 360
seven are coded as conflicting in some and absent in others of the basal taxa. This indicates that 361
while missing data might play a role in the paraphyly of hesperornithiforms (i.e., characters 362
which might unite these taxa within the hesperornithiforms are unknown); there are also a 363

115

number of conflicting codings that do not support placement of these taxa as part of a 364
monophyletic Hesperornithiformes. 365
YPM 1465 as a wildcard taxon. Within the five most parsimonious trees returned in the 366
first analysis of basal hesperornithiform taxa, the placement of YPM 1465, the holotype of 367
Baptornis advenus (Marsh, 1877), was the only terminal that varied in its placement. This is not 368
unsurprising given the extremely fragmentary nature of YPM 1465 today. When Baptornis 369
advenus was first erected, Marsh described the holotype as “a nearly perfect tarso-metatarsal 370
(sic) bone” (1877, p. 86), however YPM 1465 has not survived the intervening years in its 371
original condition – today it only preserves the distal end, resulting in the majority of characters 372
in the database coded as missing (92% missing data). 373
In order to explore a possible jumping effect of YPM 1465 on the overall basal tree 374
topology, the pruned data matrix was run without that specimen. This resulted in a single most- 375
parsimonious tree of 317 steps. This tree is shown in Figure 7, onto which has been mapped the 376
five alternative positions of YPM 1465 that resulted in the polytomy of these terminals in the 377
first analysis. The only one of these five options for which the proposed clade including YPM 378
1465 and other terminals are supported by synapomorphies is the placement of YPM 1465 as the 379
fourth most basal terminal of the tree, between Ichthyornis and Pasquiaornis. In this placement, 380
the subsequent derived position of the Pasquiaornithids is not supported by any unambiguous 381
synapomorphies, and the derived placement of Enaliornis with the remainder of the terminals is 382
supported by one ambiguous and one unambiguous synapomorphy, neither of which could be 383
coded for YPM 1465. 384
Current Taxonomy within the Hesperornithiformes 385

116

Monophyly of the Baptornithidae. Under the current taxonomic framework, 386
Pasquiaornis is united with Baptornis advenus in the Baptornithidae (Tokaryk et al., 1997; see 387
also Cumbaa et al., 2006; Martin and Lim, 2002). The results of the present analysis call this 388
placement into question. Pasquiaornis was originally placed in the Baptornithidae on the basis of 389
not sharing three similarities with Parahesperornis and Hesperornis: a robust femur, prominent 390
intramuscular lines on the femur, and an enlarged trochlea IV on the distal tarsometatarsus 391
(Tokaryk et al., 1997). This is clearly not a valid means of allying taxa, as it is not based on 392
synapomorphies. The lack of all three of those features is synonymous with the plesiomorphic 393
state in birds, as Hesperornis and Parahesperornis are both highly derived. No justification was 394
given for placing Pasquiaornis within the Hesperornithidae (Tokaryk et al., 1997). Other 395
diagnostic studies have used the presence of heteroceolus thoracic vertebrae (Martin and Tate, 396
1976), indistinct distal condyles on the humerus (Martin, 1984), and the proximal placement of 397
the glenoid facet on the coracoid (Martin, 1984) as diagnostic features of the 398
Hesperornithiformes, none of which are present in the Pasquiaornis material available for this 399
study. There are a few features that have been proposed as diagnostic of hesperornithiforms that 400
Pasquiaornis appears to possess, such as a triangular cnemial expansion on the proximal 401
tibiotarsus. However, this has been repeatedly demonstrated to be found convergently in modern 402
loons, grebes, and hesperornithiforms (Lucas, 1903; Stolpe, 1935), and so should not be 403
immediately accepted as a synapomorphy of the hesperornithiforms. When taken into 404
consideration with the results of the present analysis, the placement of Pasquiaornis within the 405
Hesperornithiformes is questionable, and within the Baptornithidae is entirely unsupported. 406
Monophyly of Baptornis. Of the specimens included in the analysis of basal 407
hesperornithiform birds, four had previously been assigned to Baptornis advenus (YPM 1465 - 408

117

Marsh, 1877; AMNH 5101, FMNH 395, UNSM 20030, KUVP 2290 - Martin and Tate, 1976) 409
and one to an undefined species of Baptornis due to its fragmentary nature (FHSM 6318 - Bell 410
and Everhart, 2009). Of these specimens, only three comprise a monophyletic clade in the 411
present analysis (FHSM 6318, FMNH 395, KUVP 2290), while the holotype (YPM 1465) falls 412
outside the clade inclusive of most hesperornithiforms (as discussed above), and the remaining 413
two specimens are placed progressively basal to the clade of Hesperornis + Parahesperornis 414
alexi. This has significant ramifications for the previous use of Baptornis as a generic terminal 415
taxon in earlier cladistic analyses (such as Clarke, 2004; Chiappe, 2002; etc.). In previous 416
analyses, a composite scoring was created for the terminal taxon ‘Baptornis’ using multiple 417
specimens, usually AMNH 5101, USNM 20030, and FMNH 395 in addition to YPM 1465 418
(Clarke, 2004). The use of multiple apparently disparate specimens to code a single terminal 419
taxon might explain the monophyly of Hesperornis + ‘Baptornis’ in previous phylogenetic 420
analyses. 421
The heuristic analysis performed on the complete dataset (Fig. 6) confirms the disparate 422
placement of the different Baptornis specimens. Like in the exact analyses, YPM 1465 is placed 423
as basal to Aves + most hesperornithiforms, KUVP 2290 is placed in a monophyletic clade with 424
FMNH 395, and UNSM 20030 is placed as just basal to Hesperornis + Parahesperornis. Unlike 425
the exact analysis, AMNH 5101 is placed in a clade with FHSM 6318; however no unambiguous 426
synapomorphies are identified for this clade. Despite this small difference, the analyses presented 427
here demonstrate that Baptornis is in need of morphologic revision. It seems clear that 428
morphological features should be identified to unite UNSM 20030 in a group with the 429
Hesperornis-like specimens, while KUVP 2290 and FMNH 395 should form their own group, 430

118

perhaps with FHSM 6318 or AMNH 5101, pending the clarification of appropriate diagnostic 431
features. 432
Monophyly of the Brodavidae. The current taxonomic framework of the 433
Hesperornithiformes recognizes four species of Brodavis, three of which are represented by 434
isolated, partial tarsometatarsi (B. americanus, B. baileyi, B. mongoliensis) and the fourth of 435
which is represented by a poorly preserved partial skeleton originally described as a second 436
species of Baptornis (B. varneri, originally described by Martin and Cordes-Person, 2007, 437
reassigned by Martin et al., 2012). While two of the four taxa were unavailable for inclusion in 438
this analysis, B. varneri and B. baileyi were supported as a monophyletic clade, along with a new 439
specimen, UCMP 117605. A single synapomorphy, the presence of a waist in the midshaft 440
region of the tarsometatarsus, supports this clade. This morphology was not identified in the 441
descriptive work of the group (Martin et al., 2012), however it was identified as diagnostic of the 442
species B. varneri in the original descriptive work (Martin and Cordes-Person, 2007). Other 443
diagnostic features were either not addressed by this analysis and will instead be addressed by 444
the morphometric analysis (for example, the proposed differences in the length: breadth of the 445
tarsometatarsus). At this time, the taxonomic grouping of B. varneri and B. baileyi into a single 446
group is supported cladistically, however the other brodavid taxa are not considered at this time 447
and the description of the group should be revised. 448
Monophyly of the Hesperornithidae. Currently, the family Hesperornithidae contains 449
four genera, three of which were included in this study (Hesperornis, Parahesperornis, and 450
Asiahesperornis). Both cladistic analyses found Hesperornis + Parahesperornis to form a 451
monophyletic clade supported by 26 unambiguous synapomorphies, making this one of the best- 452
supported hesperornithiform clades. However, currently only two features have been proposed as 453

119

diagnostic of the family: reduction of the lateral trochlear ridge of metatarsal IV relative to the 454
medial (Bell and Everhart, 2009) and enlargement of the trochlea of metatarsal IV relative to III 455
(Tokaryk et al., 1997). Clearly additional diagnostic features should be identified. 456
Monophyly of Parahesperornis. To date, Parahesperornis is represented by two 457
published specimens: the nearly complete holotype of P. alexi, KUVP 2287 (Martin, 1984), and 458
a second isolated tarsometatarsus designated as P. indeterminate (FHSM 17312; Bell and 459
Everhart, 2009). Neither a complete description of the holotype nor of a third, also nearly 460
complete, specimen (KUVP 24090, Martin, pers. comm.) of P. alexi has been published. The 461
majority of the features identified as diagnostic of the species are related to the skull, which is 462
unknown in any but the holotype specimen, making assignment of less complete specimens 463
problematic (Martin, 1984; Bell and Everhart, 2009). In the present analysis, a monophyletic 464
Parahesperornis is not supported. A single synapomorphy places KUVP 24090 as more derived 465
than KUVP 2287 and FHSM 17312, which form a more basal polytomy. That synapomorphy 466
relates to the proximal surface of the tibiotarsus, which cannot be coded for in KUVP 2287 and 467
FHSM 17312. 468
Monophyly of Hesperornis. All specimens assigned to Hesperornis were resolved into a 469
monophyletic clade united by three synapomorphies. The most striking feature of this clade is 470
the massive polytomy among all but five of the included specimens. None of the current species- 471
level distinctions of Hesperornis specimens are reflected in the cladogram calculated in this 472
analysis, despite the inclusion of features reported as diagnostic within the data matrix. This 473
strongly contradicts the current taxonomic framework of hesperornithiforms. While the strict 474
consensus of the initial analysis places five terminals as progressively basal to the main 475
polytomic clade (YPM 55000 and FMNH 316, followed by FMNH 219, AMNH 2181, and 476

120

LACM 9728, respectively), this arrangement is based on few unambiguous synapomorphies and 477
is not upheld in subsequent analyses when the included hesperornithiform specimens are 478
changed. Figure 8 shows a number of alternative topologies that are generated when other 479
hesperornithiform specimens not included in the initial analysis are substituted in. 480
The inability of TNT to incorporate the complete dataset into a single analysis exact 481
analysis necessitated the use of a heuristic search with all taxa included (Figure 6). While there 482
does appear to be some structure among a few of the Hesperornis specimens in this analysis, 483
only two of these placements correspond with the results of the exact searches of limited 484
specimens, and none of these clades are supported by unambiguous synapomorphies. 485
Furthermore, of the holotypes included, four are placed together in the main polytomy (H. 486
gracilis, H. regalis, H. mengeli, and H. macdonaldi, in contradiction to its more basal placement 487
in some of the exact analyses) while two are placed together in a smaller polytomy (H. bairdi 488
and H. chowi). Finally, the specimens assigned to various species (H. mengeli – YPM 17208, H. 489
chowi – YPM 18589, H. gracilis – YPM 55000) usually are not placed with their respective 490
holotypes. The single exception to this is YPM 17193, which has been assigned to H. chowi 491
(Martin, pers. comm.) and which is placed in a polytomy with the H. chowi holotype, along with 492
the H. bairdi holotype. These arguments provide strong support for the reassignment of 493
specimens of all analyzed species of Hesperornis in this study (H. gracilis, H. bairdi, H. chowi, 494
H. mengeli, and H. macdonaldi as well as the genus Asiahesperornis) to the original species, 495
Hesperornis regalis. However, a number of diagnostic features proposed for some of these 496
species relate to the morphometric proportions of various elements, and so will be further 497
explored in the subsequent chapter. Two specimens are consistently retained as basal to the 498

121

remainder of the Hesperornis terminals – FMNH 219 and YPM 55000. Further evaluation of 499
these specimens as unique taxa is therefore warranted. 500
501
CONCLUSIONS 502
It is clear from the above analysis that the current taxonomic framework for the 503
Hesperornithiformes is in need of serious revision. In particular, the status of Pasquiaornis and 504
Enaliornis as hesperornithiforms should be re-evaluated, while the genus Baptornis needs to be 505
carefully revised. Only features of the distal tarsometatarsus, the sole portion of the skeleton 506
present in the holotype YPM 1465, should be considered diagnostic of Baptornis advenus. 507
Furthermore, AMNH 5101, FMNH 395, FHSM 6318, and KUVP 2290 need to be re-evaluated 508
as potential new taxonomic groups. Finally, while the family Hesperornithidae has strong 509
support, subdivision of the genus Hesperornis is mostly unfounded. 510
511

122

APPENDIX I. 512
0. Frontal - parietal suture in adult: open (0); fused (1). (Clarke, 2004) 513
1. Basilar tubercle prominent, in marginal position: present (0); reduced or absent (1). 514
(Elzanowski and Galton, 1991) 515
2. Glossopharyngeal foramen: absent (0); present (1). Elzanowski and Galton, 1991 516
3. Nuchal crest and zygomatic process: zygomatic process absent (0); nuchal crest extends 517
onto zygomatic process far rostral to the extent of the nuchal crest (2). Ordered (Modified 518
from Elzanowski and Galton, 1991) 519
4. Orbital process of quadrate, pterygoid articulation: pterygoid broadly overlapping medial 520
surface of orbital process (i.e., ‘‘pterygoid ramus’’) (0); restricted to anteromedial edge of 521
process (1). (from Clarke, 2004) 522
5. Quadrate/pterygoid contact: as a facet, variably with slight anteromedial projection cradling 523
base (0); condylar, with a well-projected tubercle on the quadrate (1). (from Clarke, 2004) 524
6. Well-developed tubercule on anterior surface of dorsal process: absent (0); present (1). 525
(from Clarke, 2004) 526
7. Quadratojugal articulation: overlapping (0); peg-and-socket (1). (from Clarke, 2004) 527
8. Quadrate, dorsal process articulation: with squamosal only (0); with squamosal and prootic 528
(1). (Clarke, 2004) 529
9. Quadrate, mandibular articulation: biconcave (0); tricondylar articulation, additional 530
posterior condyle or broad surface (1). (Clarke, 2004) 531
10. Cervical vertebrae: amphicoelus (0); heterocoelus (1) [Fig. 1]. 532
11. Thoracic vertebrae, one or more with prominent hypapophyses: absent (0); present (1). 533
(from Clarke, 2004) 534

123

12. Thoracic vertebra, centra length and midpoint width: approximately equal (0); length 535
markedly greater (1). (from Clarke, 2004) 536
13. Thoracic vertebrae, lateral surfaces of centra: flat to slightly depressed (0); deep, emarginate 537
fossa (1); central ovoid foramina (2). (from Clarke, 2004) [Fig. 2] 538
14. Thoracic vertebrae, count: 12 or more (0); 10 or fewer (1). (Clarke, 2004) 539
15. Thoracic vertebrae with ossified connective tissue bridging transverse processes: absent (0); 540
present (1). (Clarke, 2004) 541
16. Thoracic vertebrae: amphicoelus (0); heterocoelus (1). [Fig. 3] 542
17. Sacral vertebrae, number ankylosed: less than 7 (0); 10-14 (1); 15 or more (2). Ordered 543
(modified from Clarke, 2004) 544
18. Synsacrum shape (in lateral view): straight (0); slight convex curve (dorsal edge at peak of 545
curve) (1). [Fig. 4] 546
19. Ribs - ossified uncinate processes: absent (0); present and unfused to rib (1); present and 547
fused to rib (2). Ordered (Clarke, 2004). 548
20. Glenoid and scapular facets: located near the proximal end of the coracoid (0); displaced 549
distally (1). [Fig. 5] 550
21. Glenoid facet reduced: absent (0); present (1). [Fig. 5] 551
22. Pro-coracoid process: present (0); absent (1). [Fig. 6] 552
23. Shape: thin triangle with long neck (0); blunt triangle with short neck (1). [Fig. 6] 553
24. Sterno-coracoidal facet: thin, arched ridge (0); thickened ridge along the sternal margin (1). 554
[Fig. 7] 555
25. Coracoid, medial surface, area of the foramen n. supracoracoideus (when developed): 556
strongly depressed (0); flat to convex (1). (Clarke, 2004) 557

124

26. Coracoid: trapezoidal (0); strut-like (1). (Chiappe, 2002) [Fig. 6] 558
27. Coracoid, dorsal surface: strongly concave (0); flat to convex (1). (from Clarke, 2004) [Fig. 559
6] 560
28. Coracoid, lateral process: absent (0); present (1). (from Clarke, 2004) 561
29. Humerus shape: robust (0); elongate or slender (1). [Fig. 8] 562
30. Humerus shape: strongly (0); slightly (1) curved. [Fig. 8] 563
31. Bicipital crest: present (0); reduced or absent (1). [Fig. 8] 564
32. Distal condyles: dorsal condyle larger than ventral condyle (0); condyles indistinct or absent 565
(1). [Fig. 9] 566
33. Cranial view, ventral epicondyle extends: to a similar extent as condyles (0); epicondyles 567
absent (1). [Fig. 9] 568
34. Humerus, distal end, posterior surface - groove for passage of m. scapulotriceps: absent (0); 569
present (1). (Clarke, 2004) [Fig. 9] 570
35. Humerus, distal condyles: developed distally (0); developed on anterior surface of humerus 571
(1). (from Clarke, 2004) [Fig. 9] 572
36. Ulna, brachial scar: absent (0), present (1). (from Clarke, 2004). [Fig. 10] 573
37. Ulna shape: elongate, slender (0); shortened (1). [Fig. 10] 574
38. Radius, muscle impression on ventro-posterior surface: absent (0); present (1). (modified 575
from Clarke, 2004) 576
39. Length of preacetabular ilium approximately: equal to (0); 60% (1); less than 40% (2) the 577
length of the post-acetabular ilium. [Fig. 11] 578
40. Preacetabular ilium: approach on anterodorsal midline, open (0); coossified dorsal closure to 579
form iliosyncral canals (1). (Clarke, 2004) [Fig. 12] 580

125

41. Ilium/ischium, distal coossification to completely enclose the ilioischiadic fenestra: absent 581
(0); present (1). (Clarke, 2004) [Fig. 14] 582
42. Ilium, ischium, pubis - proximal contact in adult: unfused (0); completely fused (1). 583
(modified from Clarke, 2004) [Fig. 14] 584
43. Ischium, obturator foramen closed distally: absent (0); present (1). (modified from Clarke, 585
2004) [Fig. 14] 586
44. Acetabulum: open (0); closed (1). [Fig. 14] 587
45. Antitrochanter on pelvis: directly posterior to acetabulum (0); posterodorsal to acetabulum 588
(1). (from Clarke, 2004) [Fig. 14] 589
46. Antitrochanter: absent (0); width is approximately equal to that of the acetabulum (1); width 590
is less than that of the acetabulum (2). [Fig. 14] 591
47. Antitrochanter: absent (0); poorly developed with minimal lateral protrusion (0); well- 592
developed, with marked lateral protrusion (1). [Fig. 13] 593
48. Pubis, cross section: suboval (0); mediolateraly compressed (1). (Clarke, 2004) [Fig. 14] 594
49. Pubis: directed nearly ventrally (0); subparallel with ischium, posteriorly directed (1). 595
(modified from Clarke, 2004) [Fig. 14] 596
50. Femur shape in caudal view - proximal width is 20% of length (0); caudal view, proximal 597
width is 30% of length (1); caudal view, proximal width is 50% of length (2). Ordered [Fig. 598
20] 599
51. Shaft in cranial or caudal view: relatively even width along length, narrows only very 600
slightly (0); waist distal to midshaft (1); waist at or proximal to midshaft (2). [Fig. 20] 601
52. Groove running along long axis of bone on cranio-medial side: absent (0); present (1). [Fig. 602
20] 603

126

53. Lateral margin of shaft (between trochanter and fibular condyle): relatively straight (0); 604
slightly concave (1); highly concave (2) angled (3). [Fig. 20] 605
54. Caudal view, medial margin of shaft distal to head: concave (0); slight s-shape (1). [Fig. 20] 606
55. Medial or lateral view: "hunched", thickened appearance along shaft: absent (0); minimal 607
(1); pronounced (2). [Fig. 16] 608
56. Intramuscular lines: absent or faint (0); well-developed, with high relief (1). [Fig. 20] 609
57. Proximal and distal ends: similar medio-lateral widths (0); distal end wider (1); proximal 610
end wider (2). [Fig. 20] 611
58. Shaft: cranially convex (in medial view, proximal and distal ends inflected caudally): 612
present (0); absent (1). [Fig. 16] 613
59. Trochanter and head: nearly continuous (0); separated by a shallow notch (1); separated by a 614
deep notch (2). Ordered [Fig. 18] 615
60. Head: neck absent (0); merges into neck without sharp margin (1); distinct knob separated 616
from neck (2). Ordered [Fig. 18] 617
61. Caudal view, flattened surface of head (where the fovea for the capital ligament is located) 618
directed: medially (0); proximo-medially (1) [Fig. 18] 619
62. Caudal view, flattened surface of head (where the fovea for the capital ligament is located) 620
inflected slightly caudally: absent (0); present (1). [Fig. 18] 621
63. Proximal-most margin of trochanter forms a shelf which crosses the proximal surface of the 622
bone and joins the lateral edge of the head: absent (0); present (1). [Fig. 18] 623
64. Lateral margin of trochanter: forms a smooth curve (0); forms a sharp angle directed 624
laterally (1). [Fig. 18] 625

127

65. Trochanter: reduced (0); lateral margin lies close to femoral shaft (1); greatly expanded into 626
a bulbous, laterally-projecting edge (2). Ordered [Fig. 18] 627
66. Proximal surface of trochanter: smooth, flattened surface (0); flattened with a peak on the 628
lateral edge (1); highly rounded proximal projection (2). [Fig. 18] 629
67. Proximal view, shape of the trochanter: circular (0); elongate or ovoid (1). [Fig. 15] 630
68. Cranial view, trochanter extends proximally: similar extent (0); further (1) than head. *when 631
femur aligned along axis between notch between head and trochanter and intercondylar 632
sulcus [Fig. 22] 633
69. Lateral view, margin of trochanter: smooth, flattened surface (0); smooth ridge (1); sharp 634
ridge with highly irregular margin (2). [Fig. 17] 635
70. Popliteal fossa: minimally excavated (0); deeply excavated (1). [Fig. 21] 636
71. Caudal view, distal condyles widely spaced, such that the popliteal fossa is bounded distally 637
by intercondylar bridge: absent, popliteal fossa open (0); absent, condyles adjacent (1); 638
present (2). Ordered [Fig. 21] 639
72. Pocket behind intercondylar bridge: absent (0); present (1). [Fig. 21] 640
73. Popliteal fossa: centered on shaft (0); laterally offset (1). [Fig. 21] 641
74. Patellar groove: absent (0); present, elongate (1); present, circular (2). [Fig. 22] 642
75. Patellar groove: absent (0); extends smoothly onto the distal shaft as a broad shallow groove 643
(1); divided from the remainder of the shaft by a slight ridge, giving it a ‘pocketed’ 644
appearance (2). Ordered [Fig. 22] 645
76. Cranio-caudal widths of medial and lateral condyles in distal view: lateral 20-30% larger 646
(0); lateral over 24090% larger (1); medial longer (2). [Fig. 23] 647

128

77. Distal view, cranio-caudal axis of medial condyle: perpendicular (0); at a slight angle (1) to 648
medio-lateral axis of distal end. [Fig. 23] 649
78. Medial condyle in caudal view: sub-circular or oval (0); wedge-shaped (1); kidney bean 650
shaped (2). [Fig. 21] 651
79. Lateral condyle: similar distal extent as medial (0); extends distally past medial (1). [Fig. 652
21] 653
80. Caudal view, medial condyle: parallel (0); angled (1) relative to the proximo-distal axis of 654
the shaft. [Fig. 21] 655
81. Medial view, caudal margin of medial condyle possesses a lip: absent (0), present (1). [Fig. 656
19] 657
82. Laterally projected fibular trochlea: absent (0); shelf-like projection (1). (modified from 658
Clarke, 2004) [Fig. 16] 659
83. Face of fibular condyle oriented: caudally or slightly laterally (0); laterally (1). [Fig. 16] 660
84. Lateral view, lateral condyle: merges smoothly into shaft (0); constricts into neck before 661
widening at shaft (1). [Fig. 25] 662
85. Distal surface of lateral condyle: featureless curve (0); small bump or rounded prominence 663
(1); projected, flattened prominence (2). Ordered [Fig. 21] 664
86. Caudal view, fibular condyle bears a small depression on distal-most portion: absent (0); 665
present (1). [Fig. 21] 666
87. Cranial surface of lateral condyle: ovoid (0); bulbous (1). [Fig. 26] 667
88. Femur, posterior trochanter: present (0); absent (1). (modified from Clarke, 2004) 668
89. Femur, greater and lesser trochanters: separated by a notch (0); developed as a single 669
trochanteric crest (1). (from Clarke, 2004) 670

129

90. Femur, ectocondylar tubercle and lateral condyle: separated by a deep notch (0); developed 671
as a single trochanteric crest (1). (from Clarke, 2004) 672
91. Patella shape: roughly pyramidal (0); laterally compressed, nearly flat and highly elongate 673
(1). [Fig. 24] 674
92. Patella - foramen for the ambiens tendon: located centrally (0); displaced to the caudal edge 675
of the medial face (1). [Fig. 24] 676
93. Tibiotarsus - proximal articular surface: perpendicular to longitudinal axis of shaft (0); 677
inclined latero-medially slightly (1); inclined latero-medially dramatically (2). [Fig. 27] 678
94. Proximal articular surface expanded: minimally, medio-lateral width of articular surface is 679
similar to that of the shaft in caudal view (0); moderately, articular surface flares out past the 680
margins of the shaft (1); extremely, entire proximal end wider than proximal shaft (2). [Fig. 681
27] 682
95. Proximal articular surface, central depression between cotylae: absent (0); present (1). [Fig. 683
30] 684
96. Proximal articular surface, grooves define the medial and cranial margins of the medial 685
cotyla: absent (0); present (1). [Fig. 30] 686
97. Lateral view, proximal end of lateral shaft deeply excavated, forms groove between cnemial 687
expansion and lateral cotyla that wraps up onto articular surface: absent or slight (0); 688
present, deeply excavated (1). [Fig. 29] 689
98. Medial view, roughened triangular surface covers proximo-cranial portion of shaft and 690
cnemial expansion: faint (0); well-developed and bounded by a diagonal ridge (1). [Fig. 31] 691
99. Cranial view, lateral cnemial crest angles sharply outward from the shaft (lateral projection): 692
absent (0); present (1). [Fig. 32] 693

130

100. Medial view, cnemial expansion: relatively straight (0); convex cranially (1). [Fig. 31] 694
101. Cranial view, intercondylar sulcus: symmetric (0); asymmetric, medial margin steeper than 695
lateral margin (1). [Fig. 32] 696
102. Cranial view, intercondylar sulcus: slightly (0); deeply (1) proximally indented. [Fig. 32] 697
103. Distal extent of medial cnemial crest: medial crest absent (0), proximal to midpoint of 698
shaft/or at midpoint of shaft (1); past midpoint of shaft (2). [Fig. 28] 699
104. Caudal view, lateral border of lateral cotylae: smooth bulge (0); abrupt lip (1) outward from 700
shaft. [Fig. 27] 701
105. Medial view, shaft: is fairly straight (0), slight s-shape (1); is bowed (2). [Fig. 28] 702
106. Fibular crest extends approximately: over half-way (0); half-way (1) down shaft; restricted 703
to upper 1/3 of shaft (2). [Fig. 34] 704
107. Distal fibular crest is interrupted distally by a broad longitudinal gap: absent (0); present (1). 705
[Fig. 34] 706
108. Shaft twists such that when the proximal end is in caudal view the distal end is in: caudal 707
view (no twisting) (0); caudo-lateral view (1). [Fig. 34] 708
109. Groove runs along caudal surface of shaft adjacent to fibular crest: faint (0); well-developed 709
(1) 710
110. Distal shaft, medio-lateral expansion: symmetric, medial and lateral condyles evenly 711
expanded on either side of shaft (0); medial condyle more expanded than lateral (1); 712
expansion entirely medial, with lateral condyle in-line with margin of shaft (2); entire distal 713
end shifted medially, both lateral and medial condyles (3). [Fig. 37] 714
111. In cranial view, distal end medio-laterally: similar to (0); wider than (1) midshaft. [Fig. 37] 715
112. Crista sulci extend onto the distal caudal surface: low relief (0); high relief (1). [Fig. 37] 716

131

113. Cranial view, lateral side of distal shaft leading to lateral condyle: widens and flattens 717
slightly (0); widens and flattens with a bulge extending over extensor sulcus (1); fully 718
developed bridge enclosing extensor sulcus (2). Ordered [Fig. 39] 719
114. Extensor sulcus: faint or shallow (0); deep (1). [Fig. 39] 720
115. Cranial view, proximo-distal length of medial condyle: similar to (0); shorter than (1) lateral 721
condyle. [Fig. 39] 722
116. Cranial view, distal extent of medial condyle: similar to (0); further distally than (1) lateral 723
condyle. [Fig. 39] 724
117. Medial view, medial condyle: roughly circular (0); strongly ‘J’-shaped, cranial margin of 725
condyle projects cranially (1); slightly J-shaped, less cranial projection (2), elongate (3); 726
subcircular, hook on proximo-cranial margin (4). [Fig. 36] 727
118. Lateral view, lateral condyle: roughly circular (0); j-shaped (1); half-circle (2); irregular 728
half-circle with caudal margin flattened (3). [Fig. 38] 729
119. Distal view, intercondylar incision: broad and shallow (0); deep with sloped sides (1); deep 730
with near-vertical sides (2). [Fig. 35] 731
120. Distal view, lateral condyle cranio-caudally: similar length to (0); longer than (1); shorter 732
than (2) medial condyle. [Fig. 35] 733
121. Distal view, medial and lateral condyles: at a slight angle toward cranio-caudal midline (0); 734
parallel (1). [Fig. 35] 735
122. Large rugosity on lateral side of distal shaft, just above condyle: absent (0); present (1). 736
[Fig. 39] 737
123. Calcaneum and astragalus completely fused to each other and tibia in adult, creating a 738
tibiotarsus: absent (0); present (1). (modified from Chiappe, 2002) 739

132

124. Fibula - lateral view, proximal articular surface: flat (0); slightly concave (1); deeply 740
concave, saddle-shaped (2). [Fig. 40] 741
125. Cranial margin bulges near proximal end: absent (0); present (1). [Fig. 40] 742
126. Lateral view, concave articular surface: absent (0); centered over shaft (1); cranially offset 743
(2). [Fig. 40] 744
127. Tarsometatarsus - proximal view, medial and lateral cotylae: approximately equal in size 745
(0); medial cotyla smaller (1). [Fig. 41] 746
128. Proximal view, articular surface: square (0); rectangular (1). [Fig. 41] 747
129. Lateral view, proximal surface of lateral cotyla: flat and perpendicular to long axis of shaft 748
(0); slopes distally at dorsal margin (1). [Fig. 43] 749
130. Lateral view, proximo-dorsal edge of metatarsal IV: rounded (0); flattened or angular (1). 750
[Fig. 43] 751
131. Intercotylar eminence: absent (0); reduced to near absence (1); low relief, rounded (2); high 752
relief, peaked (3). [Fig. 44] 753
132. Intercotylar eminence: absent (0); projects symmetrically from articular surface (1); projects 754
asymmetrically, such that it tilts laterally in dorsal view (2). [Fig. 44] 755
133. Proximal vascular foramina: absent (0); present (1). [Fig. 44] 756
134. Dorsal view, articular surface forms a transverse plane positioned, relative to the long axis 757
of the shaft: roughly perpendicular (0), slight angle (medial cotyla slightly higher than 758
lateral) (1), steep angle (medial cotylae much higher than lateral) (2). Ordered [Fig. 44] 759
135. Medial and plantar views: proximal plantar margin of metatarsal II possesses a bulbous 760
flange: absent or slight (0); enlarged or bulbous (1). [Fig. 42] 761

133

136. Medial view, shaft of metatarsal II: nearly straight (0); dramatically curved into an s-shape 762
(1). [Fig. 47] 763
137. Medial view, metatarsal II: uniform cranio-caudal width (0); thins distally (1). [Fig. 47] 764
138. Lateral view, round depression on proximal-most lateral face of metatarsal IV: weakly 765
developed (0); well-developed (1). [Fig. 43] 766
139. Plantar-medial view, medial margin of metatarsal II: rounded edge (0); sharp ridge (1). 767
140. Dorsal view, midshaft region: maintains width (0); narrows slightly (1); narrows 768
dramatically to a waist (2). [Fig. 51] 769
141. Dorsal view, relative position of metatarsals at midshaft: coplanar (0); IV and III coplanar, 770
II shifted plantarly (1); IV shifted dorsally, II and III coplanar (2). [Fig. 51] 771
142. Trochlea of metatarsal II oriented: nearly vertical (0); angles slightly medially (1); angles 772
dramatically medially, such that the distal-most end of the trochlea does not overlap 773
metatarsal III in plantar view (2). [Fig. 53] 774
143. Shaft twisted laterally - when the distal end is in dorsal view, the proximal end is in: dorsal 775
view (no or minimal twisting) (0); dorso-lateral view (1). [Fig. 49] 776
144. Dorsal view, grooves separating metatarsals: absent - metatarsals unfused along shaft (0); 777
absent - seam or crack only (1); faint depressions along a portion of shaft (2); prominent - 778
deep groove between metatarsals III and IV along entire length of shaft (3). [Fig. 49] 779
145. Dorsal surface of metatarsal IV forms a ridge: absent (0); ridge extends to midshaft (1); 780
ridge extends to trochlea (2). [Fig. 51] 781
146. Lateral view, metatarsal IV shaft: tapers evenly to distal end (0); widest at midshaft, tapers 782
at both proximal and distal end (1). [Fig. 52] 783
147. Lateral view, shaft: bowed (0); straight (1). [Fig. 52] [Fig. 52] 784

134

148. Dorsal view, distal end of shaft (just above trochlea), relative widths of metatarsals: II 785
similar width to III and IV (0); III slightly wider than IV, II narrowest (1); II narrower than 786
III and IV, which are of similar widths (2); II, III, and IV progressively wider, with IV twice 787
as wide as III (3). [Fig. 50] 788
149. Scar for attachment of metatarsal I reduced: absent (0); present (1). [Fig. 46] 789
150. Plantar view, intertrochlear incision between III and IV: absent, metatarsals unfused (0); 790
wedge-shaped with pointed proximal end (1); enclosed oval or tear-drop shape with rounded 791
proximal end (2); u-shaped, with rounded proximal end (3). [Fig. 45] 792
151. Distal foramen between trochlea III and IV: absent (0); open to (1); closed and separate 793
from (2) intertrochlear incision. [Fig. 45] 794
152. Plantar view, intertrochlear incision between III and IV: absent (0); begins proximal to (1); 795
even with or distal to (2) distal-most end of trochlea II. [Fig. 53] 796
153. Distal extent of trochlea IV: not as far as (0); to a similar level as or slightly further than (1); 797
markedly further than (2) trochlea III. Ordered [Fig. 50] 798
154. Distal view, articular ridges on trochlea III: aligned and roughly parallel (0); angled toward 799
each other dorsally, forming a triangular shape for the trochlea (1). [Fig. 58] 800
155. Metatarsal V: present (0); absent (1). (from Clarke, 2004) 801
156. Proximal metatarsal III: in plane with II and IV (0); proximally displaced plantarly from II 802
and IV (1). (from Clarke, 2004) 803
157. Hypotarsus (projected surface on proximoplantar surface of tarsometatarsus): absent (0); 804
developed as a flattened posterior surface (1); developed as a projection with distinct crests 805
and/or grooves, (2); at least one groove enclosed by bone (3). Ordered (from Clarke, 2004) 806

135

158. Metatarsal II tubercle (associated with the insertion of the tendon of the m. tibialis cranialis 807
in Aves): absent (0); present, developed on lateral surface of metatarsal II or III (1). 808
(modified from Clarke, 2004) 809
159. Pedal phalanges of digit IV, medial articular ridge relative to lateral: subequal (0); lateral 810
moderately enlarged (1), lateral greatly enlarged with medial ridge reduced to a rounded peg 811
(2). Ordered [Fig. 54] 812
160. Phalanx 1 of digit IV: elongate (0), robust (1). [Fig. 25] 813
161. Phalanx I of Digit IV, ventral groove: absent (0), present (1). [Fig. 54] 814
162. Phalanx I of Digit IV, medial edge of ventral surface of phalanx: smooth (0); possesses a 815
high, ventrally-projecting flange (1); possesses a high flange curved laterally as cross the 816
ventral groove (2). [Fig. 54] 817
818

136

819
820

137

821
822

138

823
824

139

825
826

140

827

141

828

142

829
830

143

831

144

832

145

833
834

146

835
836

147

837

148

838

149

839

150

840
841
APPENDIX II 0 1 2 3 4 5 6 7 8
Archaeopteryx lithographica 0 0 0 0
0 ? 0 0 0
Gansus yumenensis ? ? ? ?
? ? ? ? ?
Ichthyornis dispar ? 0 ? ?
1 1 0 1 1
Asiahesperornis ? ? ? ?
? ? ? ? ?
B. advenus
YPM1465
? ? ? ?
? ? ? ? ?
B. advenus
AMNH5101
? ? ? ?
? ? ? ? ?
B. advenus
FMNH395
? ? ? ?
? ? ? ? ?
B. advenus
KUVP16112
? ? ? ?
? ? ? ? ?
B. advenus
KUVP2290
? ? ? ?
? ? ? ? ?
B. advenus
UNSM20030
? ? ? ?
? ? ? ? ?
B. indet.
FHSM6318
? ? ? ?
? ? ? ? ?
B. varneri
SDSM68430
? ? ? ?
? ? ? ? ?
B. baileyi
UNSM25665
? ? ? ?
? ? ? ? ?
Enaliornis 0 0 0 1
? ? ? ? 1
Hesperornis 0 0 0 1
1 0 0 1 1
H. bairdi
YPM17208A
? ? ? ?
? ? ? ? ?
H. chowi
YPM17208
? ? ? ?
? ? ? ? ?
H. chowi
YPM18589
? ? ? ?
? ? ? ? ?
H. chowi
YPM17193
? ? ? ?
? ? ? ? ?
H. gracilis
YPM1473
? ? ? ?
? ? ? ? ?
H. gracilis
YPM1478
? ? ? ?
? ? ? ? ?
H. gracilis
YPM1679
? ? ? ?
? ? ? ? ?
H. macdonaldi
LA9728
? ? ? ?
? ? ? ? ?
H. mengeli
CFDC78.01.08
? ? ? ?
? ? ? ? ?
H. mengeli
YPM17208
? ? ? ?
? ? ? ? ?
H. regalis
YPM1200
0 0 0 1
1 0 0 1 1
H. regalis
KUVP71012
0 0 0 1
1 0 0 1 1
H. regalis
YPM1476
? ? ? ?
? ? ? ? ?
H. regalis
YPM1477
? ? ? ?
? ? ? ? ?
H. regalis
YPM1499
? ? ? ?
? ? ? ? ?
H. rossicus
ZIN5463
? ? ? ?
? ? ? ? ?
H.indet.
AMNH2181
? ? ? ?
? ? ? ? ?
H.indet.
AMNH5102
? ? ? ?
? ? ? ? ?
H.indet.
BMNH882
? ? ? ?
? ? ? ? ?
H.indet.
FMNH206
? ? ? ?
? ? ? ? ?
H.indet.
FMNH219
? ? ? ?
? ? ? ? ?
H.indet.
FMNH281
? ? ? ?
? ? ? ? ?
H.indet.
FMNH316
? ? ? ?
? ? ? ? ?
H.indet.
FMNH321
? ? ? ?
? ? ? ? ?
151
H.indet.
KUVP2280a
? ? ? ?
? ? ? ? ?
H.indet.
SDSM5312
? ? ? ?
? ? ? ? ?
H.indet.
SDSM53507
? ? ? ?
? ? ? ? ?
H.indet.
SDSM622
? ? ? ?
? ? ? ? ?
H.indet.
UNSM10148
? ? ? ?
3 ? ? ? ?
H.indet.
USNM13580
? ? ? ?
? ? ? ? ?
H.indet.
USNM13581
? ? ? ?
? ? ? ? ?
H.indet.
YPM17208
? ? ? ?
? ? ? ? ?
H.indet.
YPM55000
? ? ? ?
? ? ? ? ?
P. alexi
KUVP2287
0 0 0 1
1 0 0 1 1
P. alexi
KUVP24090
? ? ? ?
? ? ? ? ?
P.indet.
FHSM17312
? ? ? ?
? ? ? ? ?
Pasquiaornis hardiei ? ? ? ?
1 1 0 1 1
Pasquiaornis tankei ? ? ? ?
? ? ? ? ?
H. indeterminate
UCMP117605
? ? ? ?
? ? ? ? ?
Gallus 1 1 1 2
1 1 1 1 1
Anas 1 1 1 2
1 1 1 1 1
percent coded 9% 9% 9% 9% 11% 9% 9% 9% 11%
152
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
9 10 11 12 13 14 15 16 17
0 0 0 0 0
0 1
0
0
? ? 1 1 1
1 ? 1 1
0 ? ? 1 1
1 1
0
1
? ? ? 1 2
? 0 1 ?
? ? ? ? ?
? ?
?
?
? 1 1 1 2
? 0 1 1
? ? ? 1 2
? 0
1
1
? 1 1 1 2
? 0 1 ?
? 1 1 1 2
? 0
1
?
? 1 1 1 2
1 0 1 1
? ? ? ? ?
? ?
?
?
? 1 1 1 2
? 0 1 1
? ? ? ? ?
? ?
?
?
? 1 ? 1 2
? ? 1 ?
0 1 1 1 2
1 0
1
1
? ? ? ? ?
? ? ? 1
? ? ? ? ?
? ?
?
?
? 1 1 1 2
1 0 1 ?
? ? ? ? ?
? ?
?
?
? ? ? ? ?
? ? ? ?
? ? ? 1 2
? 0
1
?
? 1 1 1 2
? 0 1 ?
? ? ? ? ?
? ?
?
?
? ? ? ? ?
? ? ? ?
? ? ? 1 2
? 0
1
1
0 1 1 1 2
? 0 1 ?
0 1 1 ? ?
? ?
?
?
? 1 1 1 2
1 0 1 ?
? 1 1 1 2
? 0
1
?
? 1 1 1 2
1 0 1 ?
? ? ? ? ?
? ?
?
?
? ? ? ? ?
? ? ? ?
? ? ? ? ?
? ?
?
?
? ? ? 1 2
? 0 1 ?
? ? ? ? ?
? ?
?
?
? ? ? ? ?
? ? ? ?
? ? ? ? ?
? ?
?
?
? 1 1 1 2
1 0 1 ?
? ? ? 1 2
? 0
1
1
153
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ? ? ? ?
? ? ? ?
? 1 1 1 2
? 0
1
?
? ? ? ? ?
? ? ? ?
? ? ? ? ?
? ?
?
?
? ? ? 1 2
1 0 1 ?
? 1 1 1 2
1 0
1
?
? ? ? ? ?
? ? ? ?
? ? ? ? ?
? ?
?
?
? ? ? 1 2
? 0 1 1
0 1 1 1 2
1 0
1
1
? 1 1 ? ?
1 0 1 1
? ? ? ? ?
? ?
?
?
0 1 1 1 2
? 0 0 ?
? 1 ? ? ?
? ?
?
?
? ? ? ? ?
? ? ? ?
1 1 1 1 0
1 1
1
2
1 1 1 1 0
1 1 1 2
9% 40% 36% 49% 49% 19% 49% 51% 21%
154
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
18 19 20 21 22 23 24 25 26
0 0 0
0
1 0 0 0 0
?
0
1 0 0 0 0 1
1
0 ? 1 0 0 0 0 1 1
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
1
?
? ? ? ? ? ?
?
1 ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
1 ? 0 1 1 0 1 0 0
?
1
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
1
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
1 1 ? ? ? 1 1 0 0
1
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
1 ? ? ? ? ? ? ? ?
?
1
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
1 ? ? ? ? ? ? ? ?
155
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
0
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
1
?
? ? ? ? ? ?
?
1 ? 0 1 1 1 1 0 0
?
1
0 1 1 1 1 0
0
? ? ? ? ? ? ? ? ?
0
?
1 0 0 0 ? ?
1
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ?
?
1 2 1 0 0 0 0 1 1
1
2
1 0 0 0 0 1
1
19% 9% 8% 8% 8% 9% 8% 8% 11%
156
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
27 28 29 30 31 32 33 34 35
0 0 0 0 0 0 0 0
0
1 1
0 0 0 0 0 1
1
1 1 0 0 0 0 0 [01] 1
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
1 ? 1 1 1 1 1 0 1
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
1 0 1 1 ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
157
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
1 0
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
? ?
? ? ? ? ? ?
?
1 0 1 1 1 1 1 0 1
1 0
? ? ? ? ? ?
?
? ? ? ? ? ? ? ? ?
1 0
? ? ? 1 1 0
1
? ? 0 0 0 0 0 1 1
? ?
? ? ? ? ? ?
?
1 1 0 1 0 0 0 1 1
1 1
0 1 0 0 0 1
1
11% 9% 8% 8% 6% 8% 8% 8% 8%
158
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
36 37 38 39 40 41 42 43 44
0 0 0 0 0 0 0 0 0
1
0
1
1 0 1
1 0
0
1 0 1 0 0 0 1 ? 0
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? 0 0
1 ?
1
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
1 1 1 ? ? ? ? ? ?
?
?
?
1 0 0
1 0
1
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
1 0
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
1 ?
?
? ? ? 2 0 0 1 0 1
?
?
?
? 0 0
? ?
1
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
1 ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? 0 0 ? ? 1
?
?
?
? ? ?
1 0
?
? ? ? ? ? ? ? ? ?
?
?
?
2 0 0
1 0
1
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
1 ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
2 0 0
1 ?
1
? ? ? ? 0 0 1 ? 1
159
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
?
?
?
? 0 0
? ?
1
? ? ? 2 0 0 1 ? 1
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? ? ?
? ?
?
? ? ? ? 0 0 1 ? 1
?
?
?
? ? ?
? ?
?
? ? ? ? ? ? ? ? ?
?
?
?
? 0 0
? ?
1
? ? ? 2 0 0 1 0 1
?
?
?
2 0 0
1 0
1
? ? ? ? ? ? ? ? ?
?
?
1
? 0 0
1 0
0
1 0 1 ? ? ? 1 ? ?
?
?
?
? ? ?
? ?
?
1 0 1 1 1 1 1 1 0
1
0
1
1 1 1
1 1
0
4% 4% 6% 13% 28% 28% 32% 15% 28%
160
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
45 46 47 48 49 50 51 52 53
0 0 0 0 0 0 0
0
0
1
1 ? 1
1
0 0 0 0
1
1
1 1 1 0 0
0
0
?
? ? ?
?
? ? ? ?
?
?
? ? ? ? ?
?
?
1
1 2 1
?
? ? ? ?
?
?
? ? ? 1 2
1
1
?
? ? ?
?
? ? ? ?
?
?
? ? ? 1 2
1
1
1
1 1 1
1
1 2 0 1
?
?
? ? ? ? ?
?
?
1
? ? ?
1
? ? ? ?
?
?
? ? ? ? ?
?
?
1
? ? ?
?
1 2 0 0
1
?
? ? 1 2 2
0
3
?
1 1 1
?
? ? ? ?
?
?
? ? ? ? ?
?
?
?
? ? ?
?
2 2 0 2
?
?
? ? ? 2 2
0
2
?
? ? ?
?
? ? ? ?
?
?
? ? ? ? ?
?
?
1
? ? ?
?
? 2 0 3
?
?
? ? ? 2 2
0
2
?
? ? ?
?
? ? ? ?
?
?
? ? ? 2 2
0
2
1
? ? ?
1
2 2 0 3
?
?
? ? ? ? ?
?
?
1
2 2 1
1
2 2 0 2
?
?
? ? ? 2 2
0
3
?
? ? ?
?
2 2 0 3
?
?
? ? ? ? ?
?
?
?
? ? ?
?
2 2 0 2
?
?
? ? ? ? ?
?
?
1
? 2 1
?
2 2 0 3
?
?
? ? ? ? ?
?
?
?
? ? ?
?
2 2 0 3
?
?
? ? ? 2 2
0
2
1
2 1 1
1
? ? ? ?
1
2
? ? ? 2 2
0
2
161
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
?
2 ? ?
?
? 2 0 3
1
?
? ? ? 2 2
0
2
?
? ? ?
?
2 2 0 2
?
?
? ? ? 2 2
0
3
?
? ? ?
?
2 2 0 2
1
2
2 1 ? 2 2
0
3
?
? ? ?
?
2 2 0 [23]
?
?
? ? ? ? ?
?
?
?
2 ? ?
?
2 2 ? 3
1
2
1 1 1 2 2
0
2
1
2 ? ?
1
2 2 0 2
?
?
? ? ? ? ?
?
?
1
2 ? 1
1
1 1 0 0
1
?
? ? ? 1 1
0
0
?
? ? ?
?
? ? ? ?
1
2
1 1 1 0 0
0
0
1
2 1 1
1
0 0 0 0
32% 23% 15% 17% 17% 55% 58% 57% 57%
162
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
54 55 56 57 58 59 60 61 62
0
0 0 0 0 ? 0
0 0
0 0 0 0 0 1 1 0 0
0
0 0 1 1 1
2 0 0
? ? ? ? ? ? ? ? ?
?
? ? ? ? ?
? ? ?
? ? ? ? ? ? ? ? ?
0
0 0 0 1 1
2 1 0
? ? 0 ? ? 1 2 1 ?
0
0 0 0 1 1
2 1 0
0 1 1 0 0 2 2 1 0
?
? ? ? ? ?
? ? ?
? ? ? ? ? ? ? ? ?
?
? ? ? ? ?
? ? ?
0 1 0 ? 0 0 [12] 1 0
1
1 1 0 0 2
2 1 0
? ? ? ? ? ? ? ? ?
?
? ? ? ? ?
? ? ?
1 1 1 0 0 2 2 1 ?
1
1 1 0 0 2
2 1 1
? ? ? ? ? ? ? ? ?
?
? ? ? ? ?
? ? ?
1 1 1 0 0 2 2 1 1
0
? ? 0 0 2
2 1 ?
? ? ? ? ? ? ? ? ?
1
1 1 0 0 2
2 1 0
1 1 1 0 0 2 2 1 0
1
? ? ? ? ?
? ? ?
1 1 1 0 0 2 2 1 0
1
1 1 0 0 2
2 1 0
1 1 1 0 0 2 2 1 1
1
? ? ? ? ?
? ? ?
1 1 1 0 0 2 2 1 1
?
? ? ? ? ?
? ? ?
1 1 1 0 0 2 2 1 1
?
? ? ? ? ?
? ? ?
1 1 1 0 0 ? 2 1 1
1
1 ? 0 0 2
2 1 0
? ? ? ? ? ? ? ? ?
1
1 ? 2 0 2
2 1 1
163
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
1 1 1 ? 0 ? ? ? ?
1
1 ? 0 0 2
2 1 1
1 1 ? ? 0 2 2 1 ?
1
1 1 0 0 2
2 1 1
1 1 1 0 0 2 2 1 1
1
1 1 0 0 2
2 1 1
1 1 1 0 0 2 2 1 0
1
? ? ? ? ?
? ? ?
1 1 ? 0 0 2 2 1 1
0
1 1 0 0 2
2 1 0
0 1 1 0 0 2 2 1 0
?
? ? ? ? ?
? ? ?
0 0 0 1 0 2 2 1 0
0
0 0 1 0 2
2 1 0
? ? ? ? ? ? ? ? ?
0
0 0 2 0 1
2 1 0
0 0 0 1 1 1 2 1 0
64% 57% 49% 53% 58% 57% 57% 58% 51%
164
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
63 64 65 66 67 68 69 70 71
0 ?
0
? ? ? ? ? 0
0 0 0 0 ? 0 ? 0 1
0
0 0 2
?
0 ?
0 1
? ? ? ? ? ? ? ? ?
?
? ? ?
?
? ?
? ?
? ? ? ? ? ? ? ? ?
1
0 1 0
?
1 1
0 2
? ? ? ? ? ? ? ? ?
1
0 1 0
0
1 2
0 2
1 0 1 2 0 0 1 0 2
?
? ? ?
?
? ?
? ?
? ? ? ? ? ? ? ? 2
?
? ? ?
?
? ?
? ?
0 0 1 0 ? 1 0 0 2
1
1 2 1
1
1 2
1 2
? ? ? ? ? ? ? ? ?
?
? ? ?
?
? ?
? ?
1 0 2 0 1 ? 2 1 2
1
1 2 2
1
0 2
1 2
? ? ? ? ? ? ? ? ?
?
? ? ?
?
? ?
? ?
1 0 2 2 0 0 2 1 2
1
? 2 2
?
1 ?
1 2
? ? ? ? ? ? ? ? ?
1
1 2 1
1
1 2
? 2
1 1 2 1 1 1 2 1 2
1
? ? ?
?
? ?
? ?
1 1 2 2 0 1 2 1 2
1
1 2 2
1
0 2
1 2
1 1 2 2 0 0 2 1 2
1
? ? ?
?
? ?
? ?
1 0 2 1 0 0 2 1 2
?
? ? ?
?
? ?
? ?
1 1 2 2 1 0 2 1 2
?
? ? ?
?
? ?
? ?
1 1 2 ? ? ? 2 1 2
1
0 2 2
0
0 2
? 2
? ? ? ? ? ? ? ? ?
1
1 2 1
0
0 2
1 2
165
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ? ? ? ? ? ? 1 2
1
0 2 2
1
0 2
1 2
1 0 2 ? ? 1 2 ? 2
1
1 2 1
1
0 2
1 2
1 1 2 2 0 0 2 1 2
1
1 2 2
1
1 2
1 2
1 1 2 1 1 1 2 1 2
1
? ? ?
?
? ?
? ?
1 1 2 1 1 0 2 1 2
1
0 2 1
?
1 2
1 2
1 0 2 2 0 1 2 1 2
1
? ? ?
?
? ?
? ?
0 0 1 2 ? 0 0 0 2
0
0 1 2
?
0 0
0 2
? ? ? ? ? ? ? ? ?
1
0 1 0
?
1 1
0 2
1 0 1 0 1 1 1 0 2
64% 55% 57% 53% 42% 53% 55% 53% 60%
166
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
72 73 74 75 76 77 78 79 80
? ? 0 0 ? ? ? ? 0
? ? ? 1 ? ? 0 0 0
?
0
1 1 ? ? 1 0 0
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
?
?
1 1 1 1 1 1 1
? ? ? ? ? ? ? ? ?
0
1
[12] 1 1 1 1 1 1
? ? 2 1 1 1 1 1 1
?
?
? ? ? ? ? ? ?
0 0 1 ? 0 1 ? 1 1
?
?
? ? ? ? ? ? ?
0 1 1 1 0 0 ? 1 1
1
0
2 2 0 0 2 0 1
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ? ?
1 1 2 2 0 1 2 0 1
?
?
2 2 0 1 2 0 1
? ? ? ? ? ? ? ? ?
?
?
? ? ? ? ? ? ?
1 1 2 2 0 1 2 0 1
1
1
2 2 0 0 ? 0 ?
? ? ? ? ? ? ? ? ?
?
1
2 2 ? ? 2 ? 1
1 0 ? 2 0 0 2 0 1
?
?
? ? ? ? ? ? ?
? ? 2 2 0 0 2 0 1
?
?
2 2 0 1 2 0 1
? ? 2 2 0 0 2 0 1
?
?
? ? ? ? ? ? ?
? ? ? 2 0 0 2 0 1
?
?
? ? ? ? ? ? ?
? ? 2 2 0 0 2 0 1
?
?
? ? ? ? ? ? ?
0 1 2 2 0 0 2 1 1
?
?
2 2 0 1 2 0 1
? ? ? ? ? ? ? ? ?
1
1
? 2 0 1 2 0 1
167
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
0 1 2 2 0 0 2 0 1
1
1
? 2 0 1 2 0 1
? ? ? 2 ? ? ? ? ?
0
0
2 2 0 1 2 0 1
? 1 2 2 0 0 2 0 1
1
?
2 2 0 0 2 0 1
1 1 2 2 0 1 2 0 1
?
?
? ? ? ? ? ? ?
? ? ? ? 1 0 2 0 1
0
1
2 2 0 1 2 1 1
0 1 2 2 0 0 2 1 1
?
?
? ? ? ? ? ? ?
0 0 1 1 ? 0 ? 1 0
0
0
1 1 0 ? 1 1 0
? ? ? ? ? ? ? ? ?
0
0
1 1 2 0 0 1 1
0 0 1 1 0 1 0 1 1
36% 38% 47% 57% 55% 55% 51% 57% 57%
168
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
81 82 83 84 85 86 87 88 89
0 0
?
0 0 ? ?
0 0
0 1 0 0 0 0 0 1 1
? 1 0 ?
0
0
0 1 1
? ? ? ? ? ? ? ? ?
? ? ? ?
?
?
? ? ?
? ? ? ? ? ? ? ? ?
1 1 0 0
1
0
0 1 1
? ? ? 0 1 0 0 ? ?
? ? 0 ?
1
0
0 1 1
1 1 0 1 2 1 0 1 1
? ? ? ?
?
?
? ? ?
1 1 0 0 1 ? 0 ? ?
? ? ? ?
?
?
? ? ?
0 1 0 0 1 ? 0 ? ?
1 1 1 1
2
1
1 1 1
? ? ? ? ? ? ? ? ?
? ? ? ?
?
?
? ? ?
1 1 1 1 2 1 1 1 1
1 1 1 ?
2
1
1 1 1
? ? ? ? ? ? ? ? ?
? ? ? ?
?
?
? ? ?
1 1 1 1 2 1 1 1 1
? 1 1 ?
?
?
1 ? ?
? ? ? ? ? ? ? ? ?
1 1 ? ?
2
?
1 1 1
1 1 1 1 2 1 1 1 1
? ? 1 ?
?
?
? ? ?
1 1 1 ? 2 ? 1 1 1
1 1 1 1
2
1
1 1 1
1 1 1 ? 2 1 1 1 1
? ? 1 ?
?
?
? ? ?
1 1 1 1 2 1 1 1 1
? ? ? ?
?
?
? ? ?
1 1 1 1 2 1 1 1 1
? ? ? ?
?
?
? ? ?
1 1 1 ? ? 1 1 1 1
1 1 1 1
?
?
1 1 1
? ? ? ? ? ? ? ? ?
1 1 1 1
2
1
1 1 1
169
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
1 1 1 1 2 1 1 1 1
1 1 1 ?
2
?
1 1 1
? ? ? ? 2 ? 1 ? ?
1 1 1 1
2
1
1 1 1
1 1 1 1 2 1 1 1 1
1 1 1 ?
2
1
1 1 1
1 1 1 1 2 1 1 1 1
? ? ? ?
?
?
? ? ?
1 1 1 1 2 1 1 1 1
1 1 1 1
2
?
1 1 1
1 1 1 ? 2 0 1 1 1
? ? ? ?
?
?
? ? ?
? 1 0 0 1 ? 0 1 1
? 1 0 0
1
?
0 1 1
? ? ? ? ? ? ? ? ?
1 1 0 ?
0
0
0 1 1
1 1 0 0 1 0 0 1 1
51% 57% 60% 42% 57% 42% 62% 53% 53%
170
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
90 91 92 93 94 95 96 97 98
0
? ? 0 2 ? ? 0 ?
1 ? ? 0 ? ? ? 0 1
1 ? ? 1 0 0 0 0 ?
? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
1 ? ? 2 0 ? ? 0 1
? ? ? ? ? ? ? ? ?
1 0 0 2 0 0 0 0 1
1 0 0 2 0 0 0 0 1
? ? ? ? ? ? ? ? ?
? ? ? [12] 0 ? ? ? ?
? ? ? ? ? ? ? ? ?
? ? ? 1 0 0 0 0 ?
1 1 1 2 1 1 1 1 0
? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
1 1 1 ? 1 ? ? 1 ?
1 1 1 2 1 1 1 1 0
? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
1 1 1 2 1 ? ? 1 0
? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
1 ? ? ? ? ? ? ? ?
1 1 1 2 1 1 1 1 0
? ? ? ? ? ? ? ? ?
1 1 1 2 1 ? ? 1 0
1 1 1 2 1 1 1 1 0
1 ? ? 2 1 1 ? 1 0
? ? ? ? ? ? ? ? ?
1 1 1 2 1 1 1 1 0
? ? ? ? ? ? ? ? ?
1 1 1 2 1 1 1 1 0
? ? ? 2 1 1 1 1 0
1 ? ? ? ? ? ? ? ?
1 ? ? 2 1 ? ? 1 0
? ? ? ? ? ? ? ? ?
1 ? ? ? 1 ? ? 1 ?
171
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
1 ? ? 2 1 1 0 1 0
1 1 1 2 1 1 ? 1 0
? ? ? ? ? ? ? ? ?
1 ? ? 2 1 ? 1 1 0
1 ? ? ? ? ? ? ? ?
1 1 1 2 1 ? ? 1 0
1 ? ? ? ? ? ? ? ?
? ? ? 2 1 ? ? 1 0
1 1 1 2 ? 1 1 1 ?
1 1 1 2 1 ? ? 1 0
1 1 1 2 1 1 0 1 0
? ? ? ? ? ? ? ? ?
1 ? ? ? ? ? ? ? ?
1 ? ? 1 0 ? ? 0 ?
? ? ? ? ? ? ? ? ?
1 ? ? 0 1 0 0 0 1
1 ? ? 1 1 0 0 0 1
53% 30% 30% 45% 49% 28% 26% 49% 40%
172
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
99 100 101 102 103 104 105 106 107
?
0
0 0 0
0
0
0 0
? 0 0 0 0 ? 0 1 0
1 ? 0 0
?
1 0 ? 0
? ? ? ? ? ? ? ? ?
? ? ? ?
?
? ? ? ?
? ? 1 1 2 ? 0 0 0
1 0 0 0
?
0 ? 1 0
? ? ? ? ? ? ? ? ?
1 0 0 0
1
0 ? 1 ?
1 0 1 1 1 1 0 1 0
? ? ? ?
?
? ? ? ?
? 0 0 ? 2 ? 0 0 ?
? ? ? ?
?
? ? ? ?
1 0 0 0 ? ? ? ? ?
0 1 1 1
1
1 0 0 1
? ? ? ? ? ? ? ? ?
? ? ? ?
?
? ? ? ?
0 1 ? ? ? ? 0 ? ?
? 1 1 1
?
? 0 0 1
? ? ? ? ? ? ? ? ?
? ? 1 1
?
? ? ? ?
? ? 1 1 1 1 0 0 1
? ? ? ?
?
? ? ? ?
? ? ? ? ? ? ? ? ?
? ? ? ?
?
? ? ? ?
0 1 1 1 [12] 1 0 0 1
? ? ? ?
?
? ? ? ?
? 1 1 1 2 ? [12] 0 1
0 1 1 1
?
1 ? ? ?
? 1 1 1 ? 1 ? 0 1
? ? ? ?
?
? ? ? ?
0 1 1 1 1 1 0 0 1
? ? ? ?
?
? ? ? ?
? 1 1 1 2 1 1 0 1
0 1 1 1
?
1 [12] 0 1
? ? ? ? ? ? ? ? ?
? 1 1 1
2
1 1 0 1
? ? ? 1 ? ? ? ? ?
? 1 ? ?
?
1 ? ? ?
173
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ? 1 1 1 1 0 0 1
0 1 1 1
?
1 0 0 ?
? ? ? ? ? ? ? ? ?
0 1 1 1
[12]
1 2 0 1
? ? ? ? ? ? ? ? ?
? 1 1 1
1
1 0 0 1
? ? ? ? ? ? ? ? ?
0 1 ? ?
?
1 0 0 1
0 1 1 1 2 1 0 0 1
1 0 1 1
1
1 0 0 1
1 1 1 1 1 1 0 0 1
? ? ? ?
?
? ? ? ?
? ? ? ? ? ? ? ? ?
? ? 1 0
?
? ? ? ?
? ? ? ? ? ? ? ? ?
1 1 0 0
0
1 2 2 0
1 1 0 0 0 0 2 2 0
30% 45% 49% 49% 28% 40% 36% 43% 38%
174
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
108 109 110 111 112 113 114 115 116
0 0 [10] 0 ? 0 ? 0 0
0 ? 0 1 ? 1 0 0 0
?
? 1 1 1 1 1 0 1
? ? [23] ? 1 1 1 ? ?
?
? ? ? ? ? ? ? ?
? 1 2 1 1 1 1 0 0
0
? 1 1 0 1 0 0 0
? ? ? ? ? ? ? ? ?
?
? 1 1 0 1 1 ? ?
0 1 1 1 1 1 1 1 1
?
? ? ? ? ? ? ? ?
0 ? 1 1 ? 1 1 0 0
?
? ? ? ? ? ? ? ?
? ? 0 1 ? ? 0 0 0
0
0 [23] 1 1 1 1 1 1
? ? ? ? ? ? ? ? ?
?
? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
0
0 3 1 1 1 1 1 0
? ? ? ? ? ? ? ? ?
?
? ? ? 1 ? ? 1 1
? 0 2 1 1 1 1 1 0
?
? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
?
? ? ? ? ? ? ? ?
0 0 3 ? 1 1 ? 1 1
?
? ? ? ? ? ? ? ?
1 1 3 0 ? 1 1 ? 1
?
? ? ? ? ? ? ? ?
? 0 [23] 0.01 ? 1 ? 1 0
?
? ? ? ? ? ? ? ?
0 0 2 1 1 1 1 1 1
?
? ? ? ? ? ? ? ?
? 1 2 0 1 1 1 1 0
?
? 3 ? 1 1 1 1 0
? ? ? ? ? ? ? ? ?
1
? 2 1 ? ? 1 ? 0
? ? ? ? ? ? 1 ? 0
?
? ? ? ? ? ? ? ?
175
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
1 0 2 1 1 1 1 1 1
1
? 3 ? ? 1 1 1 0
? ? ? ? ? ? ? ? ?
1
1 2 1 1 1 ? 1 0
? ? ? ? ? ? ? ? ?
1
0 3 ? ? 1 1 1 1
? ? ? ? ? ? ? ? ?
?
1 ? ? ? ? ? ? ?
0 ? 2 ? ? 1 ? ? 0
1
1 2 1 1 1 1 1 0
0 1 2 1 1 1 1 ? ?
?
? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?
?
? 0 ? 0 1 0 0 1
? ? ? ? ? ? ? ? ?
0
0 1 1 1 2 1 0 0
0 0 2 1 1 2 1 1 0
30% 30% 42% 34% 34% 43% 42% 38% 45%
176
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
117 118 119 120 121 122 123 124 125
0 0 ? ? ?
0
0
?
?
1 2 ? ? ?
0
1 ? ?
? ? 0 0 0
0
1
?
?
? ? ? ? ?
0
? ? ?
? ? ? ? ?
?
?
?
?
2 1 1 0 0
?
? ? ?
2 1 1 0 ?
0
1
1
?
? ? ? ? ?
?
? 1 0
? 1 1 ? ?
0
1
2
1
2 1 1 0 0
0
1 1 0
? ? ? ? ?
?
?
?
?
2 1 1 ? ?
1
1 1 1
? ? ? ? ?
?
?
?
?
1 1 0 0 0
0
1 ? ?
[34] [23] 2 1 1
0
1
2
1
? ? ? ? ?
?
? ? ?
? ? ? ? ?
?
?
?
?
? ? ? ? ?
?
? ? ?
3 3 2 1 1
1
1
2
1
? ? ? ? ?
?
? ? ?
3 2 2 1 ?
0
?
?
?
4 3 2 1 1
0
1 2 1
? ? ? ? ?
?
?
?
?
? ? ? ? ?
?
? ? ?
? ? ? ? ?
?
?
?
?
3 2 2 1 1
0
1 2 1
? ? ? ? ?
?
?
?
?
3 2 2 1 0
1
1 ? ?
[34] 2 2 1 1
?
1
2
1
3 2 ? 1 ?
0
1 2 1
? ? ? ? ?
?
?
?
?
[34] 2 2 1 1
0
1 2 1
? ? ? ? ?
?
?
?
?
3 2 2 1 1
0
1 2 1
4 3 2 1 0
1
1
?
?
? ? ? ? ?
?
? ? ?
3 2 2 1 1
0
1
?
?
? 3 ? ? 1
0
? ? ?
? ? ? ? ?
?
?
?
?
177
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
3 2 2 1 1
1
1 ? ?
4 3 2 1 1
0
1
?
?
? ? ? ? ?
?
? ? ?
4 3 2 1 1
1
1
?
?
? ? ? ? ?
?
? ? ?
3 2 2 ? ?
?
1
?
?
? ? ? ? ?
?
? 2 1
? ? ? ? ?
?
1
?
?
[34] 3 2 1 0
0
1 2 1
3 1 2 1 0
0
1
2
0
? 1 2 ? ?
0
1 2 ?
? ? ? ? ?
?
?
?
?
? ? ? ? ?
0
1 ? ?
? ? ? ? ?
0
1
?
?
? ? ? ? ?
?
? ? ?
3 0 2 2 1
0
1
0
0
2 1 1 2 0
0
1 0 0
36% 47% 45% 40% 36% 49% 49% 32% 28%
178
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
126 127 128 129 130 131 132 133 134
? ? ? ? ? 0 0 0
0
? ? 1 0 1 2 1 1 0
?
0 1 0 1
2
1 1 0
? ? ? ? ? 3 ? ? ?
?
? ? ? ?
?
? 1 ?
? 0 0 0 0 2 2 1 1
1
0 0 0 0
2
2 1 0
2 ? ? 0 ? ? ? 1 ?
2
0 0 0 ?
2
2 1 0
1 1 1 0 0 1 2 1 1
?
0 0 0 ?
2
2 1 0
2 ? ? 0 ? 1 ? 1 ?
?
? ? 0 ?
?
? 1 ?
? 0 0 0 0 ? ? 1 0
2
1 1 1 1
3
2 1 1
? 1 1 1 ? 3 2 1 2
?
1 1 1 1
3
2 1 2
? ? ? ? ? ? ? ? ?
2
? ? ? ?
?
? ? ?
? 1 1 1 1 3 2 1 2
?
1 1 1 1
3
2 1 1
2 1 1 1 1 3 2 1 1
?
? ? ? ?
?
? ? ?
? 1 1 1 1 3 2 1 1
?
? ? ? ?
?
? ? ?
2 1 1 1 1 3 2 1 1
?
1 1 1 1
3
2 1 2
? 1 1 1 1 3 2 1 1
1
? ? ? ?
?
? ? ?
? 1 1 1 1 3 2 1 2
?
? ? ? ?
3
2 1 ?
1 ? ? ? ? ? ? ? ?
?
1 1 1 1
3
2 1 1
1 ? ? ? ? ? ? ? ?
?
1 1 1 ?
3
2 1 2
? ? 1 1 1 2 2 1 0
?
1 1 1 1
3
2 1 2
? 1 1 1 1 3 2 1 1
?
? ? ? ?
?
? ? ?
179
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ? ? ? ? ? ? ? ?
?
1 1 1 1
3
2 1 1
? ? ? ? ? ? ? ? ?
?
? ? ? ?
?
? ? ?
? ? ? ? ? ? ? ? ?
?
1 1 1 ?
3
2 1 1
? ? ? 1 1 3 2 1 ?
?
1 1 1 ?
3
2 1 2
1 1 1 1 1 3 2 1 1
2
1 1 0 0
2
2 1 0
? ? ? ? ? ? ? ? ?
?
1 1 0 0
2
2 1 0
? 0 0 ? ? ? ? 1 0
?
0 ? 0 0
?
? 1 0
? ? ? ? ? ? ? ? ?
0
0 1 1 1
2
1 1 1
0 0 1 1 1 2 1 1 1
26% 55% 55% 62% 45% 58% 55% 68% 57%
180
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
135 136 137 138 139 140 141 142 143
0 0 1 ? 0
0 0
0 0
0 0 1 0 0 1 1 ? 0
0
? ? 0
?
? 1 2
0
? ? ? ? 1 1 2 ? ?
?
? ? ?
?
? ? 1
?
0 0 0 0 1 1 1 2 0
0
0 0 0
1
1 1 2
1
? ? ? ? ? ? ? ? ?
0
? ? 0
?
? ? 1
?
0 1 0 0 1 1 2 2 1
0
? ? 0
?
? ? ?
?
? 1 0 ? ? 2 1 2 1
?
1 0 0
1
2 1 2
1
0 ? ? ? ? ? ? 1 ?
1
0 1 1
1
1 2 0
1
1 0 1 1 1 1 2 1 1
1
0 1 1
1
1 2 1
1
? ? ? ? ? ? ? ? ?
?
? ? ?
?
? ? ?
?
0 ? 1 1 1 1 2 ? 1
0
0 1 1
1
1 2 1
1
0 0 1 1 1 1 2 0 1
?
? ? ?
?
? ? ?
?
0 0 1 ? 1 1 2 1 1
?
? ? ?
?
? ? ?
?
1 0 1 1 1 1 2 0 1
1
0 1 1
1
1 2 1
1
0 0 1 1 1 1 2 0 1
?
? ? ?
?
? ? ?
?
0 0 1 1 1 1 2 ? 1
0
0 1 ?
?
1 ? ?
1
? ? ? ? ? ? ? ? ?
0
0 1 1
1
1 2 1
1
? ? ? ? ? ? ? ? ?
1
0 1 ?
1
1 2 1
1
0 0 1 ? 1 1 2 ? 1
0
0 1 1
1
1 2 0
1
0 ? ? ? ? ? ? ? ?
?
? ? ?
?
? ? 0
?
181
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ? ? ? ? ? ? ? ?
1
0 1 1
1
1 2 1
1
? ? ? ? ? ? ? ? ?
?
? ? ?
?
? ? ?
?
? ? ? ? ? ? ? ? ?
0
0 1 ?
1
? 2 ?
1
0 0 1 1 1 1 2 1 ?
1
0 1 1
1
1 2 1
1
0 0 1 ? 1 1 2 1 1
0
0 1 1
1
1 2 1
1
? ? ? ? ? ? ? ? ?
0
0 1 1
1
1 2 1
1
0 0 1 ? ? 1 1 1 1
0
? ? ?
?
1 1 1
1
? 1 ? ? ? ? 1 2 ?
0
0 1 0
1
1 1 2
0
0 0 1 0 1 1 1 2 0
60% 53% 55% 43% 51% 57% 57% 55% 55%
182
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
144 145 146 147 148 149 150 151 152
0 0
? 1
0
? 0 0 0
1 1 ? 1 1 ? 1 2 2
3
1
?
?
1
0 1 2 2
3 2 ? ? 3 ? 2 2 ?
1
?
?
?
1
? 1 1 1
1 0 0 0 2 1 1 1 2
1
1
0
0
2
1 1 1 2
1 ? ? 0 ? ? ? ? ?
1
?
?
?
2
? 1 2 2
3 2 0 0 2 0 1 ? 2
1
0
?
?
2
? 1 1 ?
1 1 0 0 2 0 1 ? 2
1
0
?
0
2
0 ? 2 2
1 ? 0 ? 1 0 1 ? ?
3
1
1
1
3
0 2 2 2
3 2 1 1 3 0 2 2 2
3
2
1
1
3
0 2 2 2
? ? ? ? ? ? ? ? ?
?
?
?
?
?
? ? ? ?
3 [12] 1 1 ? ? ? ? ?
3
2
1
1
3
0 2 2 2
3 2 1 1 3 0 2 2 2
?
?
?
?
?
? ? ? ?
3 2 1 1 3 0 2 ? ?
?
?
?
?
?
? ? ? ?
3 2 1 1 3 0 2 2 2
3
2
1
1
3
0 2 2 2
3 2 1 1 3 0 2 2 2
?
?
?
?
?
? ? ? ?
3 2 1 1 3 0 ? 2 ?
?
2
?
?
3
0 ? 2 ?
? ? ? ? ? ? ? ? ?
3
2
1
1
?
0 ? ? ?
? ? ? ? ? ? ? ? ?
3
2
?
?
3
0 2 ? 2
3 2 1 1 3 0 ? ? ?
3
2
1
1
3
0 2 2 2
? ? ? ? ? ? ? ? ?
?
?
?
?
?
? ? ? ?
183
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ? ? ? ? ? ? ? 1
3
2
1
1
3
0 2 2 2
? ? ? ? ? ? ? ? ?
?
?
?
?
?
? ? ? ?
? ? ? ? ? ? ? ? ?
3
2
1
1
3
? ? 2 ?
3 2 1 1 ? 0 ? ? ?
3
2
1
1
3
0 2 2 2
3 2 1 1 3 ? ? ? ?
3
2
1
1
3
0 2 2 1
? ? ? ? ? ? ? ? ?
3
2
1
1
3
0 2 2 2
1 1 0 ? 1 1 ? 2 1
1
1
?
0
1
? ? ? ?
1 ? ? ? [21] 0 ? ? ?
1
0
1
1
2
0 3 2 1
1 0 0 0 2 1 3 2 1
66% 58% 51% 53% 60% 51% 45% 45% 43%
184
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
153 154 155 156 157 158 159 160 161
0 ?
0 0 0 0
?
0 0
0 0
1 1 1 1
? 0 1
0
0
1 1 [12] 1 ? ? ?
2 ?
1 1 1 1
? ? ?
1
1
1 1 1 1 ? ? ?
1 1
1 1 1 1
? ? ?
1
1
1 1 1 1 1 0 0
? ?
1 1 1 1
? ? ?
1
1
1 1 1 1 ? ? ?
1 1
1 1 1 1
1 0 0
1
?
1 1 1 1 ? ? ?
1 1
1 1 1 1
? ? ?
?
1
1 1 1 1 ? ? ?
1 1
1 1 1 1
? ? ?
2
1
1 1 1 1 2 1 1
2 1
1 1 1 1
? ? ?
2
1
1 1 1 1 ? ? ?
? ?
? ? ? ?
? ? ?
?
?
? ? ? ? ? ? ?
? ?
1 1 1 1
? ? ?
2
1
1 1 1 1 2 1 1
2 1
1 1 1 1
2 1 1
?
?
? ? ? ? ? ? ?
2 1
1 1 1 1
? ? ?
?
?
? ? ? ? ? ? ?
2 1
1 1 1 1
2 1 1
2
1
1 1 1 1 2 ? ?
2 1
1 1 1 1
2 1 1
?
?
? ? ? ? ? ? ?
2 1
1 1 1 1
2 1 1
2
?
? ? ? ? ? ? ?
? ?
? ? ? ?
? ? ?
?
?
1 1 1 1 ? ? ?
? ?
? ? ? ?
? ? ?
2
1
? ? ? ? ? ? ?
? ?
1 1 1 1
? ? ?
2
1
1 1 1 1 ? ? ?
? ?
? ? ? ?
1 1 1
?
?
? ? ? ? ? ? ?
185
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
? ?
? ? ? ?
? ? ?
2
1
1 1 1 1 2 1 1
? ?
? ? ? ?
? ? ?
?
?
? ? ? ? ? ? ?
? ?
? ? ? ?
? ? ?
2
1
1 1 1 1 ? ? ?
? ?
1 1 1 1
? ? ?
2
1
1 1 1 1 2 1 1
1 1
1 1 1 1
2 1 1
1
1
1 1 1 1 2 1 1
? ?
? ? ? ?
? ? ?
1
1
1 1 1 1 ? ? ?
1 1
? 1 1 1
? ? ?
1
?
? 1 1 1 ? ? ?
? ?
1 1 1 1
? ? ?
0
1
1 1 3 1 0 0 0
0 0
1 1 3 1
0 0 0
57% 51% 60% 64% 64% 64% 26% 25% 25%
186
APPENDIX II
Archaeopteryx lithographica
Gansus yumenensis
Ichthyornis dispar
Asiahesperornis
B. advenus
YPM1465
B. advenus
AMNH5101
B. advenus
FMNH395
B. advenus
KUVP16112
B. advenus
KUVP2290
B. advenus
UNSM20030
B. indet.
FHSM6318
B. varneri
SDSM68430
B. baileyi
UNSM25665
Enaliornis
Hesperornis
H. bairdi
YPM17208A
H. chowi
YPM17208
H. chowi
YPM18589
H. chowi
YPM17193
H. gracilis
YPM1473
H. gracilis
YPM1478
H. gracilis
YPM1679
H. macdonaldi
LA9728
H. mengeli
CFDC78.01.08
H. mengeli
YPM17208
H. regalis
YPM1200
H. regalis
KUVP71012
H. regalis
YPM1476
H. regalis
YPM1477
H. regalis
YPM1499
H. rossicus
ZIN5463
H.indet.
AMNH2181
H.indet.
AMNH5102
H.indet.
BMNH882
H.indet.
FMNH206
H.indet.
FMNH219
H.indet.
FMNH281
H.indet.
FMNH316
H.indet.
FMNH321
162
0
0
?
?
?
?
0
?
?
0
?
?
?
?
2
?
?
?
?
?
2
1
?
?
?
2
?
2
?
1
?
?
?
?
?
?
?
[12]
?
187
H.indet.
KUVP2280a
H.indet.
SDSM5312
H.indet.
SDSM53507
H.indet.
SDSM622
H.indet.
UNSM10148
H.indet.
USNM13580
H.indet.
USNM13581
H.indet.
YPM17208
H.indet.
YPM55000
P. alexi
KUVP2287
P. alexi
KUVP24090
P.indet.
FHSM17312
Pasquiaornis hardiei
Pasquiaornis tankei
H. indeterminate
UCMP117605
Gallus
Anas
percent coded
?
1
?
?
?
?
?
2
1
1
?
?
?
?
?
0
0
23%
188

189

1
FIGURE CAPTIONS 2
Figure 1. Consensus cladogram of basal birds, adapted from O’Connor et al., 2010. While 3
taxonomic placement in the more basal birds may fluctuate, Hesperornis has been consistently 4
placed as the basal member of the clade Ornithurae. 5
6
Figure 2. Hypothesis of the evolutionary relationships of the ‘Gaviomorphae’ by Cracraft (1982). 7
Note this is not a cladogram, but rather a visualization of a hypothesis of relationships based on a 8
limited number of unanalyzed characters. 9
10
Figure 3. Hypothesis of the evolutionary relationships of the Hesperornithiformes by Martin 11
(1984). Note this is not a cladogram, but a visualization of a hypothesis of relationships based on 12
a limited number of unanalyzed characters. To date, this is the only such hypothesis published, 13
and so represents the current state of hesperornithiform ‘cladistics’. 14
15
Figure 4. Strict consensus of five most parsimonious trees returned from analysis of the pruned 16
data matrix representing the basal hesperornithiform taxa (Table 1). In this analysis all specimens 17
of Hesperornis were represented by a single compiled terminal taxon. Labels at the nodes 18
correspond to the following groups of unambiguous synapomorphies: A) 11,12, 14, 27, 35, 36, 19
38, 42, 45, 48, 49, 82, 88, 89, 90, 111, 123, 133, 155, 156, 158; B) 4, 7, 8, 60; C) 10, 61, 71; D) 20
18, 30, 63, 81, 139; E) 21, 24, 29, 31, 93, 132, F) 110; G) 140; H) 52, 76; I) none; J) 44, 102; K) 21
56, 74, 83, 84, 86, 127, 141; L) 5, 23, 39, 50, 65, 70, 75, 78, 83, 87, 91, 92, 97, 98, 107, 120, 138, 22
148, 150, 159, 160. Branch lengths are labeled on the tree. Consistency Index = 0.675, Retention 23

190

Index = 0.792. After removal of 26 uninformative characters: Consistency Index = 0.648, 24
Retention Index = 0.792. 25
26
Figure 5. Strict consensus of 1010 most parsimonious tress returned from analysis of the pruned 27
data matrix representing the derived hesperornithiforms (Table 1). Labels at the nodes 28
correspond to the following groups of unambiguous synapomorphies: A) 4, 7, 8, 12, 14, 27, 35, 29
36, 38, 42, 45, 48, 49, 82, 88, 89, 90, 123, 133, 155, 156, 158; B) 48, 89; C) 10, 71; D) 80; E) 18, 30
30, 63, 81, 139; F) 15, 93, 132; G) 19, 44, 56, 74, 84, 85, 127, 141, 145; H) 5, 21, 23, 24, 39, 50, 31
64, 69, 70, 75, 78, 83, 87, 91, 92, 97, 98, 102, 107, 120, 124, 138, 148, 150, 160; I) 95; J) none; 32
K) 54, 99; L) 64, 96; M) 131; N) 126, 153. Branch lengths are labeled on the tree. Consistency 33
Index = 0.742, Retention Index= 0.854. After removal of 25 uninformative characters: 34
Consistency Index = 0.702, Retention Index= 0.854. 35
36
Figure 6. Strict consensus of the 20 most parsimonious trees returned after heuristic analysis of 37
the complete data matrix. Labels at the nodes correspond to the following groups of 38
unambiguous synapomorphies: A) 11, 12, 14, 27, 35, 36, 38, 42, 45, 48, 49, 82, 88, 89, 90, 123, 39
133, 155, 156, 158; B) 4, 7, 8, 60; C) 10, 17, 61, 71; D) 59; E) 80; F) 18, 30, 63, 81; H) 19; I) 40
140; J) 21, 24, 29, 31, 93, 132; K) 52; L) 44, 102; M) none; N) 56, 74, 84, 85, 127, 141, 145; O) 41
5, 23, 39, 50, 65, 70, 75, 78, 83, 87, 91, 92, 97, 98, 107, 120, 138, 146, 150, 160; P) 95; Q) none; 42
R) none; S) 99, 131; T) 153; U) none; V) none; W) none; X) none. Branch lengths are labeled on 43
the tree. Consistency Index = 0.547, Retention Index= 0.802. After removal of 25 uninformative 44
characters: Consistency Index = 0.524, Retention Index=0.802. 45
46

191

Figure 7. Single most parsimonious tree recovered from analysis of the basal data matrix without 47
YPM 1465 included. The four alternate equally parsimonious placements of YPM 1465 are 48
indicated. 49
50
Figure 8. Comparison of the effect of taxonomic sampling on analysis of the second pruned data 51
matrix. Only the more derived hesperornithiform taxa are shown, as the basal portion of the tree 52
did not change with the composition of the Hesperornis taxa. A. Strict consensus of 1000 equally 53
most parsimonious trees of 323 steps each. Consistency Index= 0.732, Retention Index=0.835. 54
Unambiguous synapomorphies, by node: A - none; B - 131, 96; C - 108. B. Strict consensus of 55
1010 equally most parsimonious trees of 329 steps each. Consistency Index= 0.604, Retention 56
Index=0.778. Unambiguous synapomorphies, by node: A - 72, 79; B - 54; C - 108, 121. C. Strict 57
consensus of 1000 equally most parsimonious trees of 331 steps each. Consistency Index= 0.790, 58
Retention Index=0.853. Unambiguous synapomorphies, by node: A - 96, 99; B - 153. 59
60
TABLE CAPTIONS 61
Table 1. Hesperornithiform specimens included in the current analysis. Specimens are listed with 62
their currently recognized taxonomic status, with holotypes in bold. Notes: 1. FMNH 219 was 63
coded from a cast at the University of Kansas. The original specimen was on loan to KU and can 64
no longer be located. 2. ‘YPM 17208’ is a single specimen number that represents at least four 65
different individuals. One portion of this was designated “YPM 17208A” as the holotype for H. 66
bairdi while another portion was retained as “YPM 17208” as the holotype for H. chowi (Martin 67
and Lim, 2002). The remainder of the material was divided into a specimen assigned to H. 68
mengeli and a separate specimen assigned as H. species, however new collection numbers were 69

192

not assigned (Martin, pers. comm.). 3. YPM 55000 – coded from a cast of an un-numbered 70
specimen at the University of Wisconsin. 4. KUVP 2280a – identified as the larger, Hesperornis- 71
like elements located with a series of baptornithid vertebrae and a single specimen card for 72
KUVP 2280. As the vertebrae articulated and clearly went together, while the remainder of the 73
material was consistent in size, elements, and appearance, the hesperornithid elements were 74
separated as KUVP 2280a, in coordination with the collection manager at KU. 75
76
Table 2. Taxonomic framework for the Hesperornithiformes showing the range of taxa coded in 77
the present analysis. Taxa that have been invalidated by previous work are shaded. 78

193

Figure 1. 79
80
81
82

194

Figure 2. 83
84
85

195

Figure 3 86
87
88

196

Figure 4. 89
90
91

197

Figure 5. 92
93
94

198

Figure 6. 95
96

199

Figure 7. 97
98
99

200

Figure 8. 100
101

201

Table 1. 102
Family Genus species
specimen
number
% coded
data
basal
analysis
derived
analysis
heuristic
analysis
Enaliornithidae Enaliornis sp. compiled 48% x x x
Baptornithidae Baptornis advenus YPM 1465 8% x

x

Baptornis advenus AMNH 5101 41% x

x

Baptornis advenus FMNH 395 66% x x x

Baptornis advenus KUVP 2290 67% x

x

Baptornis advenus
UNSM
20030 80% x x x

Baptornis sp. FHSM 6318 12% x

x

Pasquiaornis hardiei compiled 55% x

x

Pasquiaornis tankei compiled 50% x

x
Brodavidae Brodavis varneri SDSM 68430 43% x

x

Brodavis baileyi
UNSM
50665 13% x

x
Hesperornithidae Asiahesperornis bazhanovi compiled 13%

x x

Hesperornis regalis YPM 1200 74%

x x

Hesperornis regalis YPM 1476 72%

x x

Hesperornis regalis FMNH 281 56%

x x

Hesperornis regalis NHM 882 50%

x

Hesperornis regalis AMNH 2181 45%

x x

Hesperornis regalis FMNH 219
1
36%

x x

Hesperornis regalis FMNH 321 34%

x x

Hesperornis regalis FMNH 206 30%

x x

Hesperornis regalis KUVP 71012 23%

x

Hesperornis bairdi
YPM
17208a
2
24%

x x

Hesperornis chowi YPM 17208
2
20%

x x

Hesperornis chowi YPM 17193 44%

x x

Hesperornis chowi YPM 18589 33%

x x

Hesperornis crassipes AMNH 5102 16%

x x

Hesperornis gracilis YPM 1473 13%

x x

Hesperornis gracilis YPM 55000
3
64%

x x

Hesperornis macdonaldi LACM 9728 16%

x x

Hesperornis mengeli
CFDC
780108 18%

x x

Hesperornis mengeli YPM 17208 26%

x x

Hesperornis rossicus ZIN PO 5463 10%

x x

Hesperornis sp. YPM 1679 71%

x x

Hesperornis sp. SDSM 5312 69%

x

202

Hesperornis sp.
USNM
13580 64%

x x

Hesperornis sp. YPM 1499 62%

x x

Hesperornis sp. SDSM 622 42%

x x

Hesperornis sp. YPM 1477 40%

x x

Hesperornis sp.
USNM
13581 39%

x

Hesperornis sp.
KUVP
2280a
4
38%

x x

Hesperornis sp. YPM 1478 31%

x

Hesperornis sp. YPM 17208
2
30%

x

Hesperornis sp.
UNSM
10148 29%

x

Hesperornis sp. KUVP 71012 23%

x x

Hesperornis sp. FMNH 316 21%

x x

Hesperornis sp. SDSM 53507 12%

x

Parahesperornis alexi KUVP 2287 89%

x x

Parahesperornis alexi KUVP 24090 58%

x x

Parahesperornis sp. FHSM 17312 20%

x x
undetermined

UCMP
117605 6% x x x
103
104

203

Table 2. 105
Family Genus Species included specimens
Enaliornithidae
Enaliornis
barretti
sedgwicki compiled into a single OTU
seeleyi
Baptornithidae
Baptornis advenus holotype + 5 specimens
Judinornis nogontsavensis unavailable
Parascaniornis stensoei invalid
Pasquiaornis
hardei compiled into a single OTU
tankei compiled into a single OTU
Brodavidae
Brodavis
americanus unavailable
baileyi holotype
mongoliensis unavailable
varneri holotype
Hesperoronithidae
Asiahesperornis bazhanovi compiled into a single OTU
Canadaga arctica unavailable
Coniornis altus invalid
Hargeria gracilis invalid
Hesperornis
altus invalid
bairdi holotype
chowi holotype + 3 specimens
crassipes holotype unavailable, 1 specimen
gracilis holotype + 1 specimen
macdonaldi holotype
mengeli holotype + 1 specimen
montana invalid
regalis holotype + 8 specimens
rossicus holotype unavailable, 1 specimen
Lestornis crassipes invalid
Parahesperornis alexi holotype + 1 specimen
?
Potamornis skutchi not included
106
Femur
measurement: 1 2 3 4 5
Baptornis advenus (FMNH 395) 7.84 5.25 10.11 23.97 9.61
Baptornis advenus (KUVP 2290) 12.21 7.14 11.04 27.03 9.81
Baptornis advenus (UNSM 20030) 10.55 5.33 10.3 24.16 9.19
Baptornis indet. (Bonner collection) 8.7 5.1 9.47 21.54 7.96
Hesperornis chowi (YPM PU 17193) 12.25 5.64 15.69 46.3 15.31
Hesperornis chowi (YPM PU 18589) 12.22 5.69 15.39 46.79 14.73
Hesperornis gracilis (YPM 55000) 8.21 5.81 15.93 45.24 14.45
Hesperornis macdonaldi (LACM 9728) 7.82 3.51 8.61 23.19 6.15
Hesperornis macdonaldi (CFDC B.81.03.16) 8.99 3.88 8.38 24.73 8.13
Hesperornis mengeli (YPM PU 17208) 9.41 4.34 9.11 28.04 9.32
Hesperornis regalis (AMNH 2181) 15.74 7.71 16.91 49.55 14.82
Hesperornis regalis (FMNH 281) 15.07 5.19 14.33 43.48 15.6
Hesperornis regalis (FMNH 321) 13.3 6.36 14.75 42.39 15.43
Hesperornis regalis (FMNH 348) 16.68 6.78 16.35 50.1 17.09
Hesperornis regalis (KUVP 2289) 18.03 8.56 19.09 49.97 16.79
Hesperornis regalis (NHM 13581) 11.82 5.18 16.64 49.25 14.99
Hesperornis regalis (NHM 721) 14.64 6.58 16.74 50.23 17.39
Hesperornis regalis (UNSM 10148) 15.52 7.91 15.89 45.05 14.46
Hesperornis regalis (YPM 1200) 13.93 7.08 18.32 51.81 16.5
Hesperornis regalis (YPM 1207) 16.33 8.79 18.97 52.14 17.35
Hesperornis regalis (YPM 1476) 16.11 7.55 19.08 54.01 16.99
Hesperornis indet. (FMNH 32) 12.08 5.01 14.32 42.71 14.55
Hesperornis indet. (NHM 882) 14.03 6.84 17.67 49.68 18.16
Hesperornis indet. (SDSM 5312) 11.49 5.58 16.95 48.32 14.5
Hesperornis indet. (SDSM 54347 ) 11.77 3.82 11.96 38.41 14.24
Hesperornis indet. (SDSM 56127 ) 13.43 7.48 16.58 47.06 14.26
Hesperornis indet. (SDSM 622) 12.96 6.39 16.25 48.21 14.1
Hesperornis indet. (UNSM 20029) 15.52 6.96 14.69 44.2 16.15
Hesperornis indet. (USNM 13580) 15.41 7.84 15.3 42.08 15.25
Hesperornis indet. (YPM 1477) 13.37 6.03 18.69 53.02 16.02
Hesperornis indet. (YPM 1499) 15.32 7.96 17.3 44.87 14.96
Parahesperornis alexi (KUVP 2287) 11.29 5.61 13.19 33.44 9.27
Parahesperornis alexi (KUVP 24090) 12.09 7.97 14.41 34.88 11.07
Aechmophorus occidentalis (LACM 100231) 4.83 3.21 5.05 12.76 5.61
Aechmophorus occidentalis (LACM 101066) 5.47 3.18 5.26 12.85 5.8
Aechmophorus occidentalis (LACM 101372) 4.41 3.24 5.25 13.15 5.53
Aechmophorus occidentalis (LACM 103093) 4.9 3.6 5.37 13.34 6.09
Aechmophorus occidentalis (LACM 107430) 5.1 3.42 5.49 13.21 5.7
Aechmophorus occidentalis (LACM 111490) 5.04 3.34 5.74 14.5 6.17
Aechmophorus occidentalis (LACM 111491) 4.72 3.47 5.42 14.63 5.84
Aechmophorus occidentalis (LACM 111843) 4.72 3.39 5.54 14.71 6.55
Aechmophorus occidentalis (LACM 113768) 4.99 3.19 5.68 14.63 6.16
Aechmophorus occidentalis (LACM 113871) 5.45 3.43 5.98 14.05 6.54
238
Femur
measurement: 1 2 3 4 5
Gavia immer (LACM 86318) 7.67 4.53 6.9 16.08 6.32
Gavia immer (LACM 101063) 9.03 5.13 7.79 17.28 7.29
Gavia immer (LACM 103466) 8.04 4.45 7.34 16.57 6.61
Gavia immer (LACM 114714) 9.16 4.64 7.53 17.45 6.83
Gavia immer (LACM 114856) 8.35 4.93 7.43 16.92 6.11
Gavia immer (LACM 86319) 10.01 5.27 7.94 18.43 6.76
Gavia immer (LACM 99905) 9.51 5.49 7.85 18.04 7.45
Gavia immer (LACM 100432) 8.89 4.45 7.46 17.16 6.85
Gavia immer (LACM 100726) 8.62 4.74 7.99 18.42 6.41
Gavia immer (LACM 115000) 8.63 4.41 6.96 16.18 6.56

239
Femur
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (KUVP 2290)
Baptornis advenus (UNSM 20030)
Baptornis indet. (Bonner collection)
Hesperornis chowi (YPM PU 17193)
Hesperornis chowi (YPM PU 18589)
Hesperornis gracilis (YPM 55000)
Hesperornis macdonaldi (LACM 9728)
Hesperornis macdonaldi (CFDC B.81.03.16)
Hesperornis mengeli (YPM PU 17208)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (FMNH 321)
Hesperornis regalis (FMNH 348)
Hesperornis regalis (KUVP 2289)
Hesperornis regalis (NHM 13581)
Hesperornis regalis (NHM 721)
Hesperornis regalis (UNSM 10148)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (FMNH 32)
Hesperornis indet. (NHM 882)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (SDSM 54347 )
Hesperornis indet. (SDSM 56127 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (UNSM 20029)
Hesperornis indet. (USNM 13580)
Hesperornis indet. (YPM 1477)
Hesperornis indet. (YPM 1499)
Parahesperornis alexi (KUVP 2287)
Parahesperornis alexi (KUVP 24090)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
6 9 10 11 12
13.52 23.71 71.91 15 11.99
17.79 24.49 74.9 17.19 12.36
16.88 24.88 71.8 15.45 10.93
13.05 22.63 69.19 13.2 9.66
25.48 46.14 87.71 22.83 16.21
27.33 46.38 80.79 23.88 12.12
27.67 34.6 90.59 24.95 15.79
10.5 21.85 44.63 11.26 6.8
13.62 23.91 47.56 12.69 9.19
13.04 28.85 57.91 12.45 10.36
26.91 46.9 97.7 28.28 18.96
21.69 44.57 84.61 18.49 15.27
23.9 49.18 80.19 22.12 17.86
27.87 51.41 87.16 22.65 16.98
32.08 51.6 102.4 28.43 18.27
27.79 48.15 98.66 29.97 12.91
33.76 54.94 113.75 31.71 11.05
27.87 45.6 91.77 24.64 14.82
32.16 52.19 97.92 27.18 15.76
31.16 52.48 98.51 26.88 20.42
31.62 52.56 103.79 26.51 20.21
24.97 43.24 80.33 20.76 13.74
31.94 52.18 105.3 30.64 19.13
25.6 51.05 91.56 25.54 16.06
21.18 40.68 69.56 18.71 13.21
25.98 46.01 87.35 25.29 13.89
24.42 46.68 84.76 21.96 15.05
24.18 46.26 83.78 21.78 15.88
29.49 44.46 90.68 26.89 19.49
31.08 51.82 97.75 24.57 19.84
31.92 44.16 94.33 24.72 21.95
19.27 31.9 68.8 18.64 12.53
19.4 33.21 74.53 19.55 12.56
9.6 13.08 41.49 9.04 6.96
9.63 13.24 40.39 9.45 8.1
10.14 13.87 42.03 8.96 7.88
10.17 13.6 43.06 9.21 8.54
10.34 14.13 44.18 10.04 8.52
10.8 15 47.07 10.23 9.01
10.45 15.16 45.83 10.27 8.98
11 15.34 47.58 9.58 9.54
10.82 14.91 45.88 11.24 9.36
11.23 14.79 45.61 10.64 9.63
240
Femur
measurement:
Baptornis advenus (FMNH 395) Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)

6 9 10 11 12
11.2 16.23 50.36 11.24 7.75
11.84 18.23 52.5 12.88 10.13
11.59 15.96 52.18 11.66 8.38
12.54 18.57 51.52 12.44 9.1
12.26 17.75 53.49 12.21 8.77
12.07 17.8 53.79 11.9 10.18
12.61 17.27 53.17 12.66 9.51
12.68 17.92 53.58 12.72 9.32
12.84 18.49 56.57 12.8 9.15
11.91 16.83 53.62 12.01 9.53
241
Femur
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (KUVP 2290)
Baptornis advenus (UNSM 20030)
Baptornis indet. (Bonner collection)
Hesperornis chowi (YPM PU 17193)
Hesperornis chowi (YPM PU 18589)
Hesperornis gracilis (YPM 55000)
Hesperornis macdonaldi (LACM 9728)
Hesperornis macdonaldi (CFDC B.81.03.16)
Hesperornis mengeli (YPM PU 17208)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (FMNH 321)
Hesperornis regalis (FMNH 348)
Hesperornis regalis (KUVP 2289)
Hesperornis regalis (NHM 13581)
Hesperornis regalis (NHM 721)
Hesperornis regalis (UNSM 10148)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (FMNH 32)
Hesperornis indet. (NHM 882)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (SDSM 54347 )
Hesperornis indet. (SDSM 56127 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (UNSM 20029)
Hesperornis indet. (USNM 13580)
Hesperornis indet. (YPM 1477)
Hesperornis indet. (YPM 1499)
Parahesperornis alexi (KUVP 2287)
Parahesperornis alexi (KUVP 24090)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
13 14 15 17 18
11.34 67.96 7.59 9.51 11.77
13.07 70.24 10.5 11.45 11.42
11.73 70.38 6.66 10.3 10.04
11.21 67.22 6.93 9.8 9.87
15.3 83.73 14.6 16.04 15.54
14.98 80.63 13.59 17.85 15.33
15.53 87.84 14.23 18.39 16.69
6.62 43.39 6.26 7.15 8.68
8.61 46.84 6.08 9.45 8.53
9.81 55.47 4.87 9.97 9.54
19.72 98.42 17.11 22.66 17.57
15.58 80.78 13.61 17.69 14.47
16.35 75.3 12.6 16.16 14.76
17.15 88.29 13.96 18.95 16.43
20.3 100.98 18.42 20.01 18.98
17.3 100.15 18.18 20.45 23.28
21.14 111.35 9.81 21 19.22
16.01 90.18 13.3 18.99 15.9
17.98 98.64 16.23 21.33 18.21
16.79 99.17 17.58 20.78 18.93
17.42 105.38 18.14 21.84 19.3
17.2 81.67 13.89 16.52 14.3
20.91 110.92 12.26 25.36 20.1
18.19 92.18 17.62 20.47 16.65
15.26 67.85 10.65 14.78 11.59
15.35 84.78 13.81 17.84 16.06
15.47 84.39 13.77 17.96 16.46
16.32 81.13 16.52 17.53 15.36
16.09 91.86 12.44 18.97 16.85
19.46 96.35 14.25 21.05 18.74
15.72 94.23 14.47 20.97 17.05
12.53 69.36 13.6 13.83 13.23
13.89 73.87 12.79 13.86 14.28
6.11 38.25 5.19 5.46 5.04
6.03 37.4 5.5 5.59 5.32
6.27 39.11 5.45 5.76 5.19
6.59 40.55 5.52 5.53 5.52
6.4 40.87 5.26 5.47 5.48
6.53 43.9 5.81 6.5 5.75
6.58 42.27 5.46 6.52 5.56
7.03 44.77 5.49 6.79 5.54
6.43 41.88 5.75 6.31 5.64
6.48 42.28 6.09 6.46 6.14
242
Femur
measurement:
Baptornis advenus (FMNH 395) Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)

13 14 15 17 18
8.74 46.84 6.59 6.99 6.85
9.75 48.12 7.87 7.1 7.95
8.09 48.64 6.99 6.78 7.38
8.73 47.48 7.22 6.88 7.6
9.37 49.23 7.24 7.46 7.35
10.09 50.16 7.43 6.83 7.93
9.87 49.31 7.69 7.73 8.03
8.54 49.98 7.45 7.35 7.34
9.09 52 8.15 7.57 7.96
9.62 49.06 7 6.83 7.07
243
Femur
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (KUVP 2290)
Baptornis advenus (UNSM 20030)
Baptornis indet. (Bonner collection)
Hesperornis chowi (YPM PU 17193)
Hesperornis chowi (YPM PU 18589)
Hesperornis gracilis (YPM 55000)
Hesperornis macdonaldi (LACM 9728)
Hesperornis macdonaldi (CFDC B.81.03.16)
Hesperornis mengeli (YPM PU 17208)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (FMNH 321)
Hesperornis regalis (FMNH 348)
Hesperornis regalis (KUVP 2289)
Hesperornis regalis (NHM 13581)
Hesperornis regalis (NHM 721)
Hesperornis regalis (UNSM 10148)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (FMNH 32)
Hesperornis indet. (NHM 882)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (SDSM 54347 )
Hesperornis indet. (SDSM 56127 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (UNSM 20029)
Hesperornis indet. (USNM 13580)
Hesperornis indet. (YPM 1477)
Hesperornis indet. (YPM 1499)
Parahesperornis alexi (KUVP 2287)
Parahesperornis alexi (KUVP 24090)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
19 20 21 22 23
5.09 11.53 66.63 8.76 5.74
7.33 12.87 69.21 7.45 6.63
6.18 11.25 63.31 10.62 6.52
5.96 11.44 61.91 7.79 5.76
8.62 16.17 78.1 19.09 15.88
8.58 16.02 74.52 20.38 14.4
9.04 16.02 82.79 16.31 13.98
4.72 7.2 41.13 7.47 7.23
4.13 9.31 41.12 8.89 7.9
4.6 9.27 50.38 12.3 10.7
10.41 19.72 92.08 19.99 20.69
9.21 15.35 76 20.76 15.96
9.01 15.83 75.59 22.78 15.08
9.43 16.95 81.99 25.85 17.74
11.23 21.17 93.34 22.11 15.08
12.69 22.43 84.72 21.4 18.63
10.45 18.3 97.94 25.21 19.42
9.25 16.16 85.71 17.59 15.83
10.41 17.16 91.7 20.36 18.98
10.61 18.14 91.67 21.94 18.34
11.79 17.64 97.88 26.33 20.76
8.5 13.97 75.65 20.67 15.84
9.96 20.01 92.85 24.61 21.53
9.49 15.16 86.12 21.58 17.58
6.49 13.54 63.43 14.36 13.59
8.28 15.14 80.95 18.22 16.19
8.55 15.54 78.94 17.08 16.7
9.29 15.53 77.1 20.06 18.71
8.5 17.38 80.4 29.14 24.43
9.73 17.88 91.04 24.23 21.01
9.4 15.72 88.26 24.65 20.85
8.21 12.06 64.81 14.52 12.23
9.29 14.26 68.79 14.18 10.68
2.66 6.11 38.27 3.94 2.72
2.21 6.01 37.24 4.39 2.75
2.19 6.18 38.95 4.51 2.73
2.39 6.55 40.28 4.3 2.39
2.12 6.46 40.67 5 2.45
3.63 6.52 43.57 4.23 3.27
2.99 6.5 42.16 5.04 3.15
3.27 7.01 44.56 4.18 3.47
3.42 6.69 41.18 4.31 3.4
2.84 6.38 41.9 5.21 3.18
244
Femur
measurement:
Baptornis advenus (FMNH 395) Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)

19 20 21 22 23
4.22 8.59 45.46 6.22 3.01
5.35 9.1 46.68 5.88 3.75
4.38 7.95 47.44 7.54 3.56
4.59 8.78 46.48 6.77 3.35
4.81 8.67 48.13 3.43 2.42
5.33 9.62 48.58 4.81 3.02
5.57 9.05 48.38 6.54 3.88
5.47 8.45 48.84 5.33 3.08
5.01 8.82 51.21 6.29 3.89
4.43 9.28 48.82 5.19 2.53
245
Femur
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (KUVP 2290)
Baptornis advenus (UNSM 20030)
Baptornis indet. (Bonner collection)
Hesperornis chowi (YPM PU 17193)
Hesperornis chowi (YPM PU 18589)
Hesperornis gracilis (YPM 55000)
Hesperornis macdonaldi (LACM 9728)
Hesperornis macdonaldi (CFDC B.81.03.16)
Hesperornis mengeli (YPM PU 17208)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (FMNH 321)
Hesperornis regalis (FMNH 348)
Hesperornis regalis (KUVP 2289)
Hesperornis regalis (NHM 13581)
Hesperornis regalis (NHM 721)
Hesperornis regalis (UNSM 10148)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (FMNH 32)
Hesperornis indet. (NHM 882)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (SDSM 54347 )
Hesperornis indet. (SDSM 56127 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (UNSM 20029)
Hesperornis indet. (USNM 13580)
Hesperornis indet. (YPM 1477)
Hesperornis indet. (YPM 1499)
Parahesperornis alexi (KUVP 2287)
Parahesperornis alexi (KUVP 24090)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
26 28 29 30 31
8.39 9.28 24.34 22.9 33.6
9.41 11.04 26.56 24.57 35.84
9.88 9.03 23.58 25.24 35.34
9 8.76 21.33 22.99 33.65
18.27 18.73 46.12 46.48 37.01
17.81 21.01 45.29 47.6 35.68
18.35 19.87 44.52 45.43 41.17
7.16 9.48 23.32 21.88 22.68
8.91 9.44 23.15 22.86 21.05
9.87 10.96 28.19 29.19 24.5
21.66 20.43 51.73 49.6 45.24
16.18 18.6 44.24 44.91 36.98
16.92 17.04 42.34 48.81 36.15
19.97 21.81 49.76 50.27 37.43
21.19 20.98 36.5 37.01 49.76
18.1 21.57 48.477 45.46 40.7
21.9 22.76 53.25 54.15 45.44
18.09 19.14 45.1 44.98 36.84
21.31 23.23 52.98 52.81 45.04
20.61 22.96 52.36 52.37 46.04
20.62 22.26 53.96 51.74 48.11
16.53 18.56 41.62 43.97 31.48
23.48 23.64 49.88 51.9 45.22
20.11 19.86 50.47 52.18 39.38
13.88 13.12 39.55 40.88 25.55
17.64 18.06 47.39 46.12 31.59
18.17 18.86 48.02 48.28 36.74
15.64 16.64 43.96 44.65 34.63
20.26 22.25 43.31 43.17 35.31
21.29 22.81 52.86 51.92 47.21
18.99 20.31 45.22 48.21 47.22
12.56 13.65 33.59 32.26 39.02
13.78 14.76 34.02 33.64 33.63
4.95 4.75 12.78 13.08 19.31
5.87 4.94 12.98 13.39 18.19
4.82 4.84 12.99 13.91 22.22
5.17 5.35 13.27 13.58 21.12
6.02 4.83 13.23 14.14 18.73
5.05 5.59 14.42 14.88 22.36
5.58 5.2 14.56 15.26 20.3
4.92 5.36 13.5 15.01 22.57
5.54 5.58 13.7 14.95 21.25
5.93 5.23 13.05 14.2 23.16
246
Femur
measurement:
Baptornis advenus (FMNH 395) Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)

26 28 29 30 31
5.97 5.84 15.74 16.37 23.26
7.57 6.54 16.29 17.58 23.57
6.04 5.97 16.41 17.13 23.64
7.11 6.45 16.82 18.66 23.34
7.59 6.33 16.47 16.26 25.77
6.68 7.22 18.2 17.91 26.95
6.75 6.52 17.91 17.76 24.49
6.5 6.02 17.23 18.73 24.92
7.24 7.01 18.43 18.56 26.01
6.69 6.47 16.21 17.7 24.9
247
Femur
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (KUVP 2290)
Baptornis advenus (UNSM 20030)
Baptornis indet. (Bonner collection)
Hesperornis chowi (YPM PU 17193)
Hesperornis chowi (YPM PU 18589)
Hesperornis gracilis (YPM 55000)
Hesperornis macdonaldi (LACM 9728)
Hesperornis macdonaldi (CFDC B.81.03.16)
Hesperornis mengeli (YPM PU 17208)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (FMNH 321)
Hesperornis regalis (FMNH 348)
Hesperornis regalis (KUVP 2289)
Hesperornis regalis (NHM 13581)
Hesperornis regalis (NHM 721)
Hesperornis regalis (UNSM 10148)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (FMNH 32)
Hesperornis indet. (NHM 882)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (SDSM 54347 )
Hesperornis indet. (SDSM 56127 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (UNSM 20029)
Hesperornis indet. (USNM 13580)
Hesperornis indet. (YPM 1477)
Hesperornis indet. (YPM 1499)
Parahesperornis alexi (KUVP 2287)
Parahesperornis alexi (KUVP 24090)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
32 33
39.42 9.39
33.47 10.49
34.36 9.24
33.86 7.41
44.85 14.69
43.18 16.27
51.41 14.37
22.41 7.74
25.45 6.02
31.24 8.1
51.13 16.15
49.94 16.05
46.46 16.07
47.03 17.94
53.08 4.23
51.27 9.44
54.26 14.28
53.56 15.37
54.06 16.55
53.98 17.46
62.17 17.43
44.3 13.9
57.64 12.87
52.88 16.29
39.67 12.3
44.73 13.63
47.31 12.72
47.16 15.04
45.68 12.39
55.37 16.1
51.52 15.67
34.47 10.44
43.03 10.67
19.53 5.63
19.95 6.04
19.53 5.93
20.96 6.17
22.32 5.63
22.97 5.89
23.02 6.17
21.57 6.4
22.63 6.77
22.28 6.62
248
Femur
measurement:
Baptornis advenus (FMNH 395) Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)

32 33
27.56 5.67
27.82 7.53
28.29 6.94
26.78 6.97
28.71 6.81
25.81 7.05
27.74 6.17
27.08 7.55
28.49 6.44
27.62 6.26
249
Tibiotarsus
measurement: 36 37 38 40
Baptornis advenus (FMNH 395) 6.97 8.63 9.63 12.78
Baptornis advenus (UNSM 20030) 6.97 7.88 9.38 15.57
Brodavis varneri (SDSM 68430) 10.02 11.81 10.29 16.9
Hesperornis gracilis (YPM 55000) 12.93 13.95 16.35 18.94
Hesperornis regalis (AMNH 2181) 13.41 20.9 16.71 20.48
Hesperornis regalis (FMNH 206) 15.52 20.99 17.39 22.98
Hesperornis indet. (SDSM 5312) 14.56 22.74 16.67 23.61
Hesperornis regalis (SDSM 56X7 ) 11.4 15.18 18.33 22.98
Hesperornis indet. (SDSM 622) 12.4 19.36 17.42 23.44
Hesperornis indet. (SDSM 82695) 12.66 17.6 18.23 20.76
Hesperornis regalis (YPM 1200) 12.9 18.75 18.15 20.43
Hesperornis regalis (YPM 1476) 12.69 16.13 17.19 21.66
Parahesperornis alexi (KUVP 2287) 10.01 9.03 12.11 13.18
Aechmophorus occidentalis (LACM101066) 4.53 5.23 3.47 9.44
Aechmophorus occidentalis (LACM100231) 4.42 5.79 4.21 10.03
Aechmophorus occidentalis (LACM103093) 4.11 5.39 4.03 9.51
Aechmophorus occidentalis (LACM107430) 4.41 6.47 3.99 9.89
Aechmophorus occidentalis (LACM101372) 4.08 5.71 4.59 9.84
Aechmophorus occidentalis (LACM113768) 5.26 6.18 4.54 10.04
Aechmophorus occidentalis (LACM113871) 4.61 6.21 4 10.51
Aechmophorus occidentalis (LACM11843) 5.12 6.05 4.01 10.46
Aechmophorus occidentalis (LACM111490) 5.03 5.97 4.44 10.79
Aechmophorus occidentalis (LACM111491) 4.33 6.46 4.31 10.8
Gavia immer (LACM101063) 5.86 7.56 5.35 12.8
Gavia immer (LACM114856) 5.86 7.46 5.26 12.27
Gavia immer (LACM114714) 5.65 6.78 5.31 13.26
Gavia immer (LACM103466) 5.75 5.71 5.23 12.31
Gavia immer (LACM86318) 6.82 6.94 5.02 12.6
Gavia immer (LACM86319) 5.83 6.86 5.54 12.53
Gavia immer (LACM11500) 5.29 7.07 5.06 12.57
Gavia immer (LACM99905) 5.94 7.72 5.89 13.98
Gavia immer (LACM100726) 6.94 6.31 5.9 12.81
Gavia immer (LACM100432) 6.45 6.53 5.34 13.99
250
Tibiotarsus
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 206)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (SDSM 56X7 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (SDSM 82695)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
41 42 44 45
13.44 3.67 3.55 194.76
15.28 4.78 8.79 193.82
20.95 10.37 11.24 226.64
27.43 10.25 12.99 291.25
32.09 14.61 14.68 304.3
29.37 9.02 15.91 329.87
33.45 13.3 22.67 319.42
27.78 11.4 10.52 301.6
29.65 11.35 15.03 298.9
26.72 9.7 13.35 283.48
31.34 9.29 12.61 321.11
32.28 9.89 13.39 332.8
17.63 7.5 7.96 212.97
10.3 3.55 5.06 131.57
10.29 3.33 5.54 124.97
10.4 3.23 6.92 129.02
10.47 3.3 6.76 139.16
10.54 3.7 6.81 135.09
10.86 3.29 5.93 142.52
11.23 3.73 7 149.46
11.55 3.69 6.18 142.02
11.07 3.98 7.63 139.08
11.12 3.74 6.23 147.52
15.19 5.1 6.77 182.38
14.47 4.9 6.4 183.14
14.61 5.01 6.31 184.67
14.2 4.98 6.35 176.21
13.94 4.97 6.74 180.3
13.62 5.45 6.86 182.02
14.16 5.6 7.24 183.13
14.91 4.98 7.35 182.31
14.86 5.45 7.19 189.43
15.67 5.47 7.12 193.31
251
Tibiotarsus
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 206)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (SDSM 56X7 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (SDSM 82695)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
46 47 48 49
8.25 13.45 98 14.34
8.93 14.75 120.4 12.57
15.72 20.64 62.6 16.6
16.43 27.33 125.3 23.21
18.14 31.8 125.9 24.57
11.11 29.56 123.1 26.94
14.27 33.67 112 27.81
15.68 28.3 112.5 23.36
14.94 30.63 110.4 24.98
16.53 26.44 112.9 22.76
19.64 31.19 106.5 26.43
17.49 31.61 108.4 25.54
11.42 19.26 101.8 17.31
5.56 10.3 107.4 6.88
5.02 10.37 114.5 6.49
5.08 10.32 110 7.52
5.3 10.64 108.2 7.3
5.64 10.63 109.3 7.33
5.68 10.74 108.4 7.9
6.14 11.48 107.9 8.21
5.3 11.57 108.5 7.47
5.68 10.96 107.1 7.4
5.97 11.46 109.3 7.7
9.03 15.22 98.8 8.65
8.3 14.61 105.7 9.04
8.76 14.54 102.8 8.65
7.55 14.24 108.9 8.51
7.53 13.66 102.6 8.45
7.66 13.7 100.2 8.39
7.79 14.23 105.5 9.11
7.63 14.92 103.6 9.02
7.19 14.72 107.2 9.65
7.9 15.7 101.9 8.26
252
Tibiotarsus
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 206)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (SDSM 56X7 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (SDSM 82695)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
50 51 52 53
12.67 10.25 14.48 17.23
13.02 11.66 18.76 18.14
17.12 14.09 23.38 19.98
20.03 16.76 25.13 27.59
19.97 18.96 19.23 21.56
22.35 19.54 27.15 26.28
25.06 20.79 24.12 21.46
23.78 18.76 19.86 22.22
23.85 20.86 22.73 29.33
22.01 18.53 19.67 22.67
22.17 19.4 22.29 29.62
24.47 20.46 24.66 28.09
15.15 12.14 16.6 12.57
9.5 8.2 6.64 5.02
9.6 8.5 7.23 4.65
9.53 8.12 6.97 4.76
10.01 9.56 7.62 4.09
9.89 9.14 7.74 5.18
10.23 10.01 7.31 4.99
10.71 10.25 7.4 4.62
10.49 10.03 8.16 5.02
10.78 9.98 7.66 4.96
10.79 10.26 8.05 4.37
13.23 12.56 7.05 7.75
13.19 12.89 7.01 8.16
13.46 12.54 6.49 5.92
12.52 12.01 8.32 7.25
12.52 11.89 7.91 6.4
12.49 12 8.14 7.95
12.28 11.88 8.28 7.72
13.68 12.99 10.84 6.77
13.7 12.75 9.31 8.78
13.74 13.01 7.81 7.11
253
Tibiotarsus
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 206)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (SDSM 56X7 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (SDSM 82695)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
54 56 57 59
30.67 197.08 10.23 7.06
33.8 193.11 12.49 8.43
40.57 222.83 17.2 11.3
37.19 281.3 21.31 14.44
56.33 312.01 25.35 15.8
50.32 334.33 14.58 17.92
55.92 328.49 29.35 16.71
54.28 304.42 16.05 16.17
46.28 296.61 21.91 16.25
52.01 281.22 21.49 14.55
52.03 324.11 24.64 13.82
52.96 332.43 25.03 15.56
30.32 212.34 14.86 10.91
32.53 131.56 6.19 5.3
32.51 124.79 6.62 4.77
31.37 128.94 7.56 5.41
33.99 139.35 6.91 5.02
33.25 135.14 6.87 5.12
32.35 142.65 7.55 4.69
37.01 149.49 6.82 4.95
34.99 141.96 8.1 5.55
33.53 138.6 8.17 5.08
36.93 147.44 6.98 5.769
62.73 178.75 9.02 6.87
64.62 182.95 8.48 7.26
61.27 185.07 8.62 7.51
59.57 174.11 8.41 6.81
61.54 180.55 8.8 5.71
65 181.51 8.21 5.94
64.39 183.25 8.34 6.27
63.1 181.25 9.63 6.62
66.52 187.74 8.69 6.3
66.19 192.52 8.36 7.2
254
Tibiotarsus
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 206)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (SDSM 56X7 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (SDSM 82695)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
60 61 62 63
4.3 7.38 6.11 16.78
4.74 11.01 6.69 18.87
6.54 14.85 8.33 25.34
8.3 13.06 8.8 29.27
7.09 12.53 6.5 24.37
9.8 18.84 10.08 35.64
9.93 14.35 10.16 34.42
9.21 12.41 8.72 29.88
9.32 14.68 9.25 31.52
7.57 12.58 8.06 28.52
9.42 12.27 5.93 26.63
8.72 13.08 8 28.2
6.35 10.51 7.21 23.38
3.17 5.51 4.88 10.94
2.81 5.31 3.97 10.73
3.14 4.59 4.57 10.73
2.99 5.19 4.91 10.83
3.54 6.1 4.79 11.78
3.28 6.03 5.23 11.45
3.35 6.05 4.71 12.39
3.34 5.95 4.89 12.04
3.3 5.59 5.1 12.26
3.53 6.3 4.45 11.93
3.56 5.38 4.89 14.7
3.55 6.27 4.55 14.31
3.51 6.6 4.6 14.51
4.31 5.15 3.9 13.41
3.61 5.46 4.16 13.31
3.54 5.82 4.26 14.23
3.71 5.38 4.59 14.83
4.15 6.44 5.27 15.41
4.05 6.18 5.02 15.52
4.11 6.51 5.06 15.58
255
Tibiotarsus
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis regalis (FMNH 206)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (SDSM 56X7 )
Hesperornis indet. (SDSM 622)
Hesperornis indet. (SDSM 82695)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
64 65
197.2 16.07
190.82 18.7
224.07 24.92
287.5 27.23
318.81 36.25
325.04 36.06
319.11 38.99
300.51 31.46
299.06 36.18
284.01 34.92
322.65 34.83
324.74 32.96
212.4 23.69
131.67 9.52
124.95 10.02
128.92 10.22
139.27 10.25
135.08 10.45
142.42 10.86
149.3 10.9
141.96 10.04
138.66 10.61
147.58 11.18
180.74 12.32
182.21 12.76
184.6 11.79
174.48 11.36
183.96 12.06
182.63 11.47
182.38 12.14
183.92 12.28
189.34 12.5
196.51 11.78
256
Tarsometatarsus
measurement: 66 69 70 71
Baptornis advenus (AMNH 5101) 19.11 10.15 7.04 10.46
Baptornis advenus (FMNH 395) 17.45 8.14 6.41 9.51
Baptornis advenus (UNSM 20030) 18.79 9.37 9.44 10.52
Brodavis varneri (SDSM 68430) 25.59 11.24 7.33 9.74
Hesperornis bairdi (YPM 17208A) 22.32 15.33 9.5 12.94
Hesperornis chowi (YPM 17208) 31.54 22.1 13.77 17.04
Hesperornis gracilis (YPM 1478) 28.89 20.59 12.45 16.73
Hesperornis gracilis (YPM 1679) 29.28 18.85 13.64 17.22
Hesperornis gracilis (YPM 55000) 27.54 18.35 12.43 16.66
Hesperornis mengeli (CFDC B78.01.08) 16.33 8.58 8.24 10.45
Hesperornis regalis (CFDC B.84.04.18) 30.42 19.42 13.62 19.39
Hesperornis regalis (FMNH 206) 34.18 21.37 13.75 18.49
Hesperornis regalis (FMNH 218) 27.72 17.12 11.79 15.63
Hesperornis regalis (FMNH 281) 30.81 22.16 12.75 17.05
Hesperornis regalis (SDSM 5313) 35.05 17.83 13.99 17.9
Hesperornis regalis (YPM 1200) 35.65 20.3 14.28 19.35
Hesperornis regalis (YPM 1207) 32.47 21.21 13.92 18.11
Hesperornis regalis (YPM 1476) 31.39 20.52 12.95 17.2
Hesperornis indet. (SDSM 5312) 35.28 23.77 12.55 19.37
Hesperornis indet. (YPM 1499) 30.2 20.38 13.03 17.79
Hesperornis indet. (YPM 17208) 34.02 20.46 13.41 18.77
Parahesperornis alexi (KUVP 2287) 21.35 13.85 9.2 11.57
Parahesperornis indet. (FHSM 17312) 17.35 10.81 7.25 10.34
Aechmophorus occidentalis (LACM101066) 11.57 5.17 3.85 5.34
Aechmophorus occidentalis (LACM100231) 11.48 4.21 4.24 5.27
Aechmophorus occidentalis (LACM103093) 12.66 5.39 4.87 5.6
Aechmophorus occidentalis (LACM107430) 11.66 4.55 4.56 5.7
Aechmophorus occidentalis (LACM101372) 11.4 4.74 4.27 5.45
Aechmophorus occidentalis (LACM113768) 12.09 4.8 4.46 6.01
Aechmophorus occidentalis (LACM113871) 12.86 5.16 4.57 5.88
Aechmophorus occidentalis (LACM11843) 13.28 5.36 4.78 6.09
Aechmophorus occidentalis (LACM111490) 12.83 5.1 5.46 6.14
Aechmophorus occidentalis (LACM111491) 12.98 4.92 4.9 5.35
Gavia immer (LACM101063) 15.65 8.06 5.19 8.16
Gavia immer (LACM114856) 15.33 6.72 5.41 7.36
Gavia immer (LACM114714) 15.51 6.92 5.73 7.3
Gavia immer (LACM103466) 14.47 7.4 4.97 7.57
Gavia immer (LACM86318) 14.38 6.54 4.95 8.21
Gavia immer (LACM86319) 15.84 7.07 5.92 7.77
Gavia immer (LACM11500) 14.87 6.69 5.62 7.28
Gavia immer (LACM99905) 15.31 7.37 4.99 7.7
Gavia immer (LACM100726) 15.69 7.56 5.37 7.63
Gavia immer (LACM100432) 15.05 7.48 5.39 7.66
257
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
72 73 74 75
12.24 6.77 6.52 123.3
11.23 6.19 6.39 139.4
10.7 5.13 5.12 137
12.22 6.67 7.48 160.6
10.9 10.34 5.41 114.3
14.87 13.43 6.37 116.9
12.51 13.9 7.08 119.5
13.46 13.17 6.58 120.1
13.75 13.84 6.67 131
8.33 7.69 4.17 110.7
18.83 16.85 8.22 129.1
16.09 15.83 7.23 129.2
12.1 13.27 6.95 120.7
13 13.88 6.53 124.7
14.15 15 6.31 125.9
15.55 14.72 8.41 117.4
15.22 14.87 7.62 121.7
16.3 14.43 8.54 120.2
15.4 14.96 8.3 116.6
14.77 13.47 6.97 119.5
12.98 15.24 6.59 119.6
11.33 10.1 6.51 116.3
10.25 8.66 5.33 119.7
6.16 2.8 3.6 117.9
6.01 3.02 3 122.9
6.18 3.18 3.16 118.5
6.48 3.08 2.95 114.8
5.99 2.72 3.14 114.2
5.55 3.22 3.07 116.1
6.22 3.43 3.38 119.4
6.33 3.41 2.71 117
6.23 3.18 3.2 113.4
6.52 3.24 3.31 110.8
9.79 4.27 4.56 132.7
8.95 4.01 4.62 128
8.77 4.3 4.44 117.5
9.58 3.79 4.16 126.5
8.7 3.36 4.25 126.7
9.63 3.98 4.96 123.8
9.21 4 4.32 133
8.79 3.91 4.65 123.2
9.15 4.24 4.81 120.2
8.99 3.91 4.25 124.7
258
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
76 77 79 80
103.1 6.71 14.42 9.65
130.2 5.47 10.18 8.94
135.2 6.19 10.66 9.74
159.4 8.3 11.34 11.08
109.8 7.17 12.55 14.45
117.5 8.85 19.96 16.9
110.4 8.22 17.27 16.72
100.6 6.61 15.65 17.48
116.8 7.22 19.33 17.22
102.4 3.98 12.89 12.44
107.4 6.08 19.01 21.95
106.9 9.24 23.6 20.23
110.5 8.07 18.46 16.35
105.8 9.06 17.59 18.98
107.1 7.73 20.91 19.09
104 9.17 18.39 17.01
112.5 8.25 17.65 20.55
110.3 8.92 17.27 17.29
113.2 8.92 20.01 26.25
100.8 8.21 17.16 18.54
112 8.72 19.03 24.27
114.2 7.45 14.63 11.02
119.7 4.49 17.95 12.81
113.7 1.98 5.72 5.64
115 2.09 5.58 5.58
115 1.83 6.08 5.71
110.8 2.18 6.61 5.37
108.1 2.29 5.36 5.09
115.6 1.92 6.17 5.76
112.8 1.81 5.83 5.84
109.9 1.89 6.95 5.83
110.7 2.35 6.97 6.37
105.3 2.23 7.12 6.16
110.7 3.8 11.39 7.79
123.1 3.72 10.51 7.06
116.9 3.58 10.48 7.14
112.9 3.07 10.47 6.11
123.9 3.71 10.3 6.86
118.9 3.67 11.43 7.44
124.4 3.35 10.47 6.64
119.6 3.56 10.36 7.25
113.8 3.54 10.57 7.6
121.1 3.27 10.69 6.24
259
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
81 83 84 87
8.34 9.95 9.04 55.46
5.71 7.99 8.33 53.01
10.28 7.9 9.42 46.52
9.56 12.31 11.26 49.44
9.35 15.33 13.6 59.72
13.72 21.34 17.79 80.62
12.46 18.6 16.7 80.17
13.47 18.56 16.74 75.29
12.29 17.67 16.35 68.67
7.45 8.52 10.7 58.82
12.46 21.46 20.51 76.35
13.93 22.73 18.75 84.16
12.1 19.05 16.07 67.04
12.89 19.68 17.51 82.36
14.69 19.84 15.94 81.86
14.22 20.58 18.55 81.89
13.76 20.39 17.82 78.97
13.1 20.43 18.29 79.68
13.9 23.26 18.8 80.7
13.05 18.99 17.23 79.64
13.72 18.6 21.08 77.31
9.09 13.34 10.76 61.35
7.77 10.41 13.3 58.33
5.18 4.94 5.72 48.31
5.26 4.04 5.22 49
4.87 4.57 5.57 54.41
4.77 4.53 5.13 53.71
4.18 4.49 4.84 48.96
4.91 4.14 5.48 54.49
5.22 4.42 5.61 53.13
5.04 5.09 5.37 55.94
5.02 4.51 5.93 55.41
4.76 5.01 5.92 52.27
7.59 7.52 6.6 54.6
7.35 6.98 6.34 55.59
7.34 6.68 6.58 51.03
7.46 6.94 6.05 52.01
6.58 6.4 6.53 52.11
7.05 6.96 7.41 56.84
6.72 6.05 6.41 53.88
7.07 7.21 6.98 52.17
6.92 7.4 7.3 58.21
7.19 6.99 6.18 52.25
260
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
88 89 90 91
10.35 12.19 2.76 11.32
8.84 10.86 3.27 10.27
9.2 9.45 2.78 8.52
12.84 9.41 4.02 12.83
10.29 10.82 2.51 11.07
13.22 14.59 5.71 26.52
12.11 15.01 4.71 15.78
10.9 13.45 3.65 12.48
11.89 14.07 3.22 14.31
7.91 6.04 2.48 6.81
13.67 14.62 2.91 13.04
14.3 13.12 4.31 15.22
12.27 12.51 2.97 10.08
11.1 13.61 3 15.3
13.3 15.75 4.11 13.6
13.96 15.99 3.4 13.3
14.66 15.27 5.62 13.05
14.71 14.76 4.37 13.27
15.72 15.85 5.81 14.21
13.4 15.08 4.09 12.39
14.23 14.97 2.71 11.98
11.56 11.38 3.15 10.12
8.5 9.87 2.42 10.4
5.98 5.92 2.5 4.45
5.86 5.94 2.08 4.44
6.96 6.05 2.65 4.34
6.49 6.37 2.8 4.82
5.88 5.98 2.23 4.75
6.56 6.7 2.27 4.35
6.02 6.21 2.22 5.36
7.11 6.53 2.27 4.98
6.82 6.45 2.25 4.28
6.8 6.89 2.56 5.19
8.71 8.48 2.06 7.9
9.22 9.16 2.41 7.25
9.12 9.48 2.3 7.08
8.15 9.36 2.02 7.57
8.85 8.19 1.89 7.08
9.51 9.36 1.93 7.92
8.76 9.63 1.94 7.58
9.07 9.18 2.2 8.2
9.29 9.06 2.33 7.82
9.33 9.47 2.38 7.46
261
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
92 93 94 95
89.9 88.95 76.43 77
82.84 82.3 71.99 74.2
83.72 83.17 73.91 77.5
96.03 95.6 84 83.5
102.76 100.39 89.69 70.5
136.24 131.35 119.15 78.1
137.18 132.16 116.97 80.6
126.44 120.79 108.7 78.4
122.91 120.43 105.98 76.4
86.25 82.5 75.18 80.1
131.4 125.64 114.06 73.9
143.46 138.37 123.22 72.6
106.43 101.69 92.07 75.8
134.1 128.41 114.21 70.8
142.4 136.65 121.53 74.9
136.88 130.86 115.94 81.7
136.93 132.35 118.49 80.3
136.76 134.09 119.77 76.9
143.23 139.56 125.46 79.7
133.18 127.59 115.55 76
132.15 125.92 114.57 76.3
100.73 99.66 88.54 88.7
99.93 96.43 86.9 83.8
67.63 67.42 63.03 14.4
69.36 68.32 63.93 12.6
75.39 75.01 70.14 10.2
74.6 74.16 69.37 12.9
68.4 67.68 62.43 13.1
76.39 76.21 71.06 11.5
73.62 73.27 68.22 11.1
79.37 78.07 73.61 10.2
78.74 77.77 72.86 13.2
77.1 76.96 71.64 10.9
81.56 83.03 74.86 7
83.68 83.76 76.28 6.9
83.33 84.39 76.38 8.6
83.93 84.88 73.57 8.4
82.5 83.14 75.22 9.5
89.72 91.25 82.12 6.1
83.33 84.5 75.86 6
84.56 85.61 77.12 5.9
90.91 91.75 83.01 7.1
84.57 85.54 77.21 8.1
262
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
96 97 100 104
7.94 32.1 7.78 6.5
7.14 23.5 6.96 4.38
8.37 22.2 10.25 3.76
15.84 28.1 16.79 8.67
13.6 18.1 14.67 6.93
17.75 21 17.55 9.47
18.04 13.7 18.79 11.23
19.17 11.5 16.76 11.6
17.68 22.8 15.94 10.9
11.8 36.4 10.02 5.44
17.66 33.02 19.12 10.76
18.54 22.5 18.39 15.13
16.08 24.4 15.77 9.1
18.5 23.2 18.69 11.1
18.86 19 19.33 11.24
19.29 22.7 19.85 10.63
21.33 23.2 20.02 12.79
17.61 20.9 16.79 11.44
20.75 24.2 22.25 12.11
21.44 23.9 17.84 12.52
20.06 22.8 18.15 10.28
8.04 20.9 9.45 8.18
10.25 23.7 10.08 6.22
3.32 20.3 3.22 2.2
3.28 21.2 3.48 1.3
3.61 21.5 3.78 1.6
3.36 23.4 3.5 1.5
3.33 19.8 3.36 1.15
4.18 22.5 3.88 1.79
3.92 19.4 3.68 1.73
3.8 21.1 3.57 1.83
3.8 19.3 3.73 1.34
3.49 22.8 3.47 1.62
5.93 9.2 5.32 1.83
5.13 10.3 4.99 2.09
4.89 11.5 4.7 1.9
4.98 8.2 4.71 1.8
4.68 8.3 4.4 1.83
4.95 11.7 4.65 2.3
4.77 12.5 4.47 2.11
5.2 12 4.66 2.09
4.95 10.5 4.58 1.65
4.61 10.7 4.74 1.92
263
Tarsometatarsus
measurement:
Baptornis advenus (AMNH 5101)
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Brodavis varneri (SDSM 68430)
Hesperornis bairdi (YPM 17208A)
Hesperornis chowi (YPM 17208)
Hesperornis gracilis (YPM 1478)
Hesperornis gracilis (YPM 1679)
Hesperornis gracilis (YPM 55000)
Hesperornis mengeli (CFDC B78.01.08)
Hesperornis regalis (CFDC B.84.04.18)
Hesperornis regalis (FMNH 206)
Hesperornis regalis (FMNH 218)
Hesperornis regalis (FMNH 281)
Hesperornis regalis (SDSM 5313)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1207)
Hesperornis regalis (YPM 1476)
Hesperornis indet. (SDSM 5312)
Hesperornis indet. (YPM 1499)
Hesperornis indet. (YPM 17208)
Parahesperornis alexi (KUVP 2287)
Parahesperornis indet. (FHSM 17312)
Aechmophorus occidentalis (LACM101066)
Aechmophorus occidentalis (LACM100231)
Aechmophorus occidentalis (LACM103093)
Aechmophorus occidentalis (LACM107430)
Aechmophorus occidentalis (LACM101372)
Aechmophorus occidentalis (LACM113768)
Aechmophorus occidentalis (LACM113871)
Aechmophorus occidentalis (LACM11843)
Aechmophorus occidentalis (LACM111490)
Aechmophorus occidentalis (LACM111491)
Gavia immer (LACM101063)
Gavia immer (LACM114856)
Gavia immer (LACM114714)
Gavia immer (LACM103466)
Gavia immer (LACM86318)
Gavia immer (LACM86319)
Gavia immer (LACM11500)
Gavia immer (LACM99905)
Gavia immer (LACM100726)
Gavia immer (LACM100432)
105
4.27
4.02
3.96
6.6
3.5
5.88
7.63
6.3
7.3
3.25
4.68
8.82
6.35
8.85
7.52
8.91
9.07
8.42
7.98
6.72
9.63
4.91
4.79
5.17
4.12
4.58
3.87
4.71
4.49
4.51
4.81
4.33
4.27
3.32
3.63
3.56
3.89
4.45
4.35
3.84
4.47
3.94
3.28
264
All Measurements
measurement: 1 2 3 4 5
Baptornis advenus (FMNH 395) 7.84 5.25 10.11 23.97 9.61
Baptornis advenus (UNSM 20030) 10.55 5.33 10.3 24.16 9.19
Hesperornis gracilis (YPM 55000) 8.21 5.81 15.93 45.24 14.45
Hesperornis regalis (AMNH 2181) 15.74 7.71 16.91 49.55 14.82
Hesperornis indet. (SDSM 5312) 11.49 5.58 16.95 48.32 14.5
Hesperornis regalis (YPM 1200) 13.93 7.08 18.32 51.81 16.5
Hesperornis regalis (YPM 1476) 16.11 7.55 19.08 54.01 16.99
Parahesperornis alexi (KUVP 2287) 11.29 5.61 13.19 33.44 9.27
Aechmophorus occidentalis (LACM 100231) 4.83 3.21 5.05 12.76 5.61
Aechmophorus occidentalis (LACM 101066) 5.47 3.18 5.26 12.85 5.8
Aechmophorus occidentalis (LACM 101372) 4.41 3.24 5.25 13.15 5.53
Aechmophorus occidentalis (LACM 103093) 4.9 3.6 5.37 13.34 6.09
Aechmophorus occidentalis (LACM 107430) 5.1 3.42 5.49 13.21 5.7
Aechmophorus occidentalis (LACM 111490) 5.04 3.34 5.74 14.5 6.17
Aechmophorus occidentalis (LACM 111491) 4.72 3.47 5.42 14.63 5.84
Aechmophorus occidentalis (LACM 111843) 4.72 3.39 5.54 14.71 6.55
Aechmophorus occidentalis (LACM 113768) 4.99 3.19 5.68 14.63 6.16
Aechmophorus occidentalis (LACM 113871) 5.45 3.43 5.98 14.05 6.54
Gavia immer (LACM 86318) 7.67 4.53 6.9 16.08 6.32
Gavia immer (LACM 101063) 9.03 5.13 7.79 17.28 7.29
Gavia immer (LACM 103466) 8.04 4.45 7.34 16.57 6.61
Gavia immer (LACM 114714) 9.16 4.64 7.53 17.45 6.83
Gavia immer (LACM 114856) 8.35 4.93 7.43 16.92 6.11
Gavia immer (LACM 86319) 10.01 5.27 7.94 18.43 6.76
Gavia immer (LACM 99905) 9.51 5.49 7.85 18.04 7.45
Gavia immer (LACM 100432) 8.89 4.45 7.46 17.16 6.85
Gavia immer (LACM 100726) 8.62 4.74 7.99 18.42 6.41
Gavia immer (LACM 115000) 8.63 4.41 6.96 16.18 6.56
265
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
6 9 10 11 12
13.52 23.71 71.91 15 11.99
16.88 24.88 71.8 15.45 10.93
27.67 34.6 90.59 24.95 15.79
26.91 46.9 97.7 28.28 18.96
25.6 51.05 91.56 25.54 16.06
32.16 52.19 97.92 27.18 15.76
31.62 52.56 103.79 26.51 20.21
19.27 31.9 68.8 18.64 12.53
9.6 13.08 41.49 9.04 6.96
9.63 13.24 40.39 9.45 8.1
10.14 13.87 42.03 8.96 7.88
10.17 13.6 43.06 9.21 8.54
10.34 14.13 44.18 10.04 8.52
10.8 15 47.07 10.23 9.01
10.45 15.16 45.83 10.27 8.98
11 15.34 47.58 9.58 9.54
10.82 14.91 45.88 11.24 9.36
11.23 14.79 45.61 10.64 9.63
11.2 16.23 50.36 11.24 7.75
11.84 18.23 52.5 12.88 10.13
11.59 15.96 52.18 11.66 8.38
12.54 18.57 51.52 12.44 9.1
12.26 17.75 53.49 12.21 8.77
12.07 17.8 53.79 11.9 10.18
12.61 17.27 53.17 12.66 9.51
12.68 17.92 53.58 12.72 9.32
12.84 18.49 56.57 12.8 9.15
11.91 16.83 53.62 12.01 9.53
266
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
13 14 15 17 18
11.34 67.96 7.59 9.51 11.77
11.73 70.38 6.66 10.3 10.04
15.53 87.84 14.23 18.39 16.69
19.72 98.42 17.11 22.66 17.57
18.19 92.18 17.62 20.47 16.65
17.98 98.64 16.23 21.33 18.21
17.42 105.38 18.14 21.84 19.3
12.53 69.36 13.6 13.83 13.23
6.11 38.25 5.19 5.46 5.04
6.03 37.4 5.5 5.59 5.32
6.27 39.11 5.45 5.76 5.19
6.59 40.55 5.52 5.53 5.52
6.4 40.87 5.26 5.47 5.48
6.53 43.9 5.81 6.5 5.75
6.58 42.27 5.46 6.52 5.56
7.03 44.77 5.49 6.79 5.54
6.43 41.88 5.75 6.31 5.64
6.48 42.28 6.09 6.46 6.14
8.74 46.84 6.59 6.99 6.85
9.75 48.12 7.87 7.1 7.95
8.09 48.64 6.99 6.78 7.38
8.73 47.48 7.22 6.88 7.6
9.37 49.23 7.24 7.46 7.35
10.09 50.16 7.43 6.83 7.93
9.87 49.31 7.69 7.73 8.03
8.54 49.98 7.45 7.35 7.34
9.09 52 8.15 7.57 7.96
9.62 49.06 7 6.83 7.07
267
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
19 20 21 22 23
5.09 11.53 66.63 8.76 5.74
6.18 11.25 63.31 10.62 6.52
9.04 16.02 82.79 16.31 13.98
10.41 19.72 92.08 19.99 20.69
9.49 15.16 86.12 21.58 17.58
10.41 17.16 91.7 20.36 18.98
11.79 17.64 97.88 26.33 20.76
8.21 12.06 64.81 14.52 12.23
2.66 6.11 38.27 3.94 2.72
2.21 6.01 37.24 4.39 2.75
2.19 6.18 38.95 4.51 2.73
2.39 6.55 40.28 4.3 2.39
2.12 6.46 40.67 5 2.45
3.63 6.52 43.57 4.23 3.27
2.99 6.5 42.16 5.04 3.15
3.27 7.01 44.56 4.18 3.47
3.42 6.69 41.18 4.31 3.4
2.84 6.38 41.9 5.21 3.18
4.22 8.59 45.46 6.22 3.01
5.35 9.1 46.68 5.88 3.75
4.38 7.95 47.44 7.54 3.56
4.59 8.78 46.48 6.77 3.35
4.81 8.67 48.13 3.43 2.42
5.33 9.62 48.58 4.81 3.02
5.57 9.05 48.38 6.54 3.88
5.47 8.45 48.84 5.33 3.08
5.01 8.82 51.21 6.29 3.89
4.43 9.28 48.82 5.19 2.53
268
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
26 28 29 30 31
8.39 9.28 24.34 22.9 33.6
9.88 9.03 23.58 25.24 35.34
18.35 19.87 44.52 45.43 41.17
21.66 20.43 51.73 49.6 45.24
20.11 19.86 50.47 52.18 39.38
21.31 23.23 52.98 52.81 45.04
20.62 22.26 53.96 51.74 48.11
12.56 13.65 33.59 32.26 39.02
4.95 4.75 12.78 13.08 19.31
5.87 4.94 12.98 13.39 18.19
4.82 4.84 12.99 13.91 22.22
5.17 5.35 13.27 13.58 21.12
6.02 4.83 13.23 14.14 18.73
5.05 5.59 14.42 14.88 22.36
5.58 5.2 14.56 15.26 20.3
4.92 5.36 13.5 15.01 22.57
5.54 5.58 13.7 14.95 21.25
5.93 5.23 13.05 14.2 23.16
5.97 5.84 15.74 16.37 23.26
7.57 6.54 16.29 17.58 23.57
6.04 5.97 16.41 17.13 23.64
7.11 6.45 16.82 18.66 23.34
7.59 6.33 16.47 16.26 25.77
6.68 7.22 18.2 17.91 26.95
6.75 6.52 17.91 17.76 24.49
6.5 6.02 17.23 18.73 24.92
7.24 7.01 18.43 18.56 26.01
6.69 6.47 16.21 17.7 24.9
269
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
32 33 36 37 38
39.42 9.39 6.97 8.63 9.63
34.36 9.24 6.97 7.88 9.38
51.41 14.37 12.93 13.95 16.35
51.13 16.15 13.41 20.9 16.71
52.88 16.29 14.56 22.74 16.67
54.06 16.55 12.9 18.75 18.15
62.17 17.43 12.69 16.13 17.19
34.47 10.44 10.01 9.03 12.11
19.53 5.63 4.53 5.23 3.47
19.95 6.04 4.42 5.79 4.21
19.53 5.93 4.11 5.39 4.03
20.96 6.17 4.41 6.47 3.99
22.32 5.63 4.08 5.71 4.59
22.97 5.89 5.26 6.18 4.54
23.02 6.17 4.61 6.21 4
21.57 6.4 5.12 6.05 4.01
22.63 6.77 5.03 5.97 4.44
22.28 6.62 4.33 6.46 4.31
27.56 5.67 5.86 7.56 5.35
27.82 7.53 5.86 7.46 5.26
28.29 6.94 5.65 6.78 5.31
26.78 6.97 5.75 5.71 5.23
28.71 6.81 6.82 6.94 5.02
25.81 7.05 5.83 6.86 5.54
27.74 6.17 5.29 7.07 5.06
27.08 7.55 5.94 7.72 5.89
28.49 6.44 6.94 6.31 5.9
27.62 6.26 6.45 6.53 5.34
270
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
40 41 42 44 45
12.78 13.44 3.67 3.55 194.76
15.57 15.28 4.78 8.79 193.82
18.94 27.43 10.25 12.99 291.25
20.48 32.09 14.61 14.68 304.3
23.61 33.45 13.3 22.67 319.42
20.43 31.34 9.29 12.61 321.11
21.66 32.28 9.89 13.39 332.8
13.18 17.63 7.5 7.96 212.97
9.44 10.3 3.55 5.06 131.57
10.03 10.29 3.33 5.54 124.97
9.51 10.4 3.23 6.92 129.02
9.89 10.47 3.3 6.76 139.16
9.84 10.54 3.7 6.81 135.09
10.04 10.86 3.29 5.93 142.52
10.51 11.23 3.73 7 149.46
10.46 11.55 3.69 6.18 142.02
10.79 11.07 3.98 7.63 139.08
10.8 11.12 3.74 6.23 147.52
12.8 15.19 5.1 6.77 182.38
12.27 14.47 4.9 6.4 183.14
13.26 14.61 5.01 6.31 184.67
12.31 14.2 4.98 6.35 176.21
12.6 13.94 4.97 6.74 180.3
12.53 13.62 5.45 6.86 182.02
12.57 14.16 5.6 7.24 183.13
13.98 14.91 4.98 7.35 182.31
12.81 14.86 5.45 7.19 189.43
13.99 15.67 5.47 7.12 193.31
271
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
46 47 48 49 50
8.25 13.45 98 14.34 12.67
8.93 14.75 120.4 12.57 13.02
16.43 27.33 125.3 23.21 20.03
18.14 31.8 125.9 24.57 19.97
14.27 33.67 112 27.81 25.06
19.64 31.19 106.5 26.43 22.17
17.49 31.61 108.4 25.54 24.47
11.42 19.26 101.8 17.31 15.15
5.56 10.3 107.4 6.88 9.5
5.02 10.37 114.5 6.49 9.6
5.08 10.32 110 7.52 9.53
5.3 10.64 108.2 7.3 10.01
5.64 10.63 109.3 7.33 9.89
5.68 10.74 108.4 7.9 10.23
6.14 11.48 107.9 8.21 10.71
5.3 11.57 108.5 7.47 10.49
5.68 10.96 107.1 7.4 10.78
5.97 11.46 109.3 7.7 10.79
9.03 15.22 98.8 8.65 13.23
8.3 14.61 105.7 9.04 13.19
8.76 14.54 102.8 8.65 13.46
7.55 14.24 108.9 8.51 12.52
7.53 13.66 102.6 8.45 12.52
7.66 13.7 100.2 8.39 12.49
7.79 14.23 105.5 9.11 12.28
7.63 14.92 103.6 9.02 13.68
7.19 14.72 107.2 9.65 13.7
7.9 15.7 101.9 8.26 13.74
272
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
51 52 53 54 56
10.25 14.48 17.23 30.67 197.08
11.66 18.76 18.14 33.8 193.11
16.76 25.13 27.59 37.19 281.3
18.96 19.23 21.56 56.33 312.01
20.79 24.12 21.46 55.92 328.49
19.4 22.29 29.62 52.03 324.11
20.46 24.66 28.09 52.96 332.43
12.14 16.6 12.57 30.32 212.34
8.2 6.64 5.02 32.53 131.56
8.5 7.23 4.65 32.51 124.79
8.12 6.97 4.76 31.37 128.94
9.56 7.62 4.09 33.99 139.35
9.14 7.74 5.18 33.25 135.14
10.01 7.31 4.99 32.35 142.65
10.25 7.4 4.62 37.01 149.49
10.03 8.16 5.02 34.99 141.96
9.98 7.66 4.96 33.53 138.6
10.26 8.05 4.37 36.93 147.44
12.56 7.05 7.75 62.73 178.75
12.89 7.01 8.16 64.62 182.95
12.54 6.49 5.92 61.27 185.07
12.01 8.32 7.25 59.57 174.11
11.89 7.91 6.4 61.54 180.55
12 8.14 7.95 65 181.51
11.88 8.28 7.72 64.39 183.25
12.99 10.84 6.77 63.1 181.25
12.75 9.31 8.78 66.52 187.74
13.01 7.81 7.11 66.19 192.52
273
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
57 59 60 61 62
10.23 7.06 4.3 7.38 6.11
12.49 8.43 4.74 11.01 6.69
21.31 14.44 8.3 13.06 8.8
25.35 15.8 7.09 12.53 6.5
29.35 16.71 9.93 14.35 10.16
24.64 13.82 9.42 12.27 5.93
25.03 15.56 8.72 13.08 8
14.86 10.91 6.35 10.51 7.21
6.19 5.3 3.17 5.51 4.88
6.62 4.77 2.81 5.31 3.97
7.56 5.41 3.14 4.59 4.57
6.91 5.02 2.99 5.19 4.91
6.87 5.12 3.54 6.1 4.79
7.55 4.69 3.28 6.03 5.23
6.82 4.95 3.35 6.05 4.71
8.1 5.55 3.34 5.95 4.89
8.17 5.08 3.3 5.59 5.1
6.98 5.769 3.53 6.3 4.45
9.02 6.87 3.56 5.38 4.89
8.48 7.26 3.55 6.27 4.55
8.62 7.51 3.51 6.6 4.6
8.41 6.81 4.31 5.15 3.9
8.8 5.71 3.61 5.46 4.16
8.21 5.94 3.54 5.82 4.26
8.34 6.27 3.71 5.38 4.59
9.63 6.62 4.15 6.44 5.27
8.69 6.3 4.05 6.18 5.02
8.36 7.2 4.11 6.51 5.06
274
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
63 64 65 66 69
16.78 197.2 16.07 17.45 8.14
18.87 190.82 18.7 18.79 9.37
29.27 287.5 27.23 27.54 18.35
24.37 318.81 36.25 30.81 22.16
34.42 319.11 38.99 35.28 23.77
26.63 322.65 34.83 35.65 20.3
28.2 324.74 32.96 31.39 20.52
23.38 212.4 23.69 21.35 13.85
10.94 131.67 9.52 11.57 5.17
10.73 124.95 10.02 11.48 4.21
10.73 128.92 10.22 12.66 5.39
10.83 139.27 10.25 11.66 4.55
11.78 135.08 10.45 11.4 4.74
11.45 142.42 10.86 12.09 4.8
12.39 149.3 10.9 12.86 5.16
12.04 141.96 10.04 13.28 5.36
12.26 138.66 10.61 12.83 5.1
11.93 147.58 11.18 12.98 4.92
14.7 180.74 12.32 15.65 8.06
14.31 182.21 12.76 15.33 6.72
14.51 184.6 11.79 15.51 6.92
13.41 174.48 11.36 14.47 7.4
13.31 183.96 12.06 14.38 6.54
14.23 182.63 11.47 15.84 7.07
14.83 182.38 12.14 14.87 6.69
15.41 183.92 12.28 15.31 7.37
15.52 189.34 12.5 15.69 7.56
15.58 196.51 11.78 15.05 7.48
275
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
70 71 72 73 74
6.41 9.51 11.23 6.19 6.39
9.44 10.52 10.7 5.13 5.12
12.43 16.66 13.75 13.84 6.67
12.75 17.05 13 13.88 6.53
12.55 19.37 15.4 14.96 8.3
14.28 19.35 15.55 14.72 8.41
12.95 17.2 16.3 14.43 8.54
9.2 11.57 11.33 10.1 6.51
3.85 5.34 6.16 2.8 3.6
4.24 5.27 6.01 3.02 3
4.87 5.6 6.18 3.18 3.16
4.56 5.7 6.48 3.08 2.95
4.27 5.45 5.99 2.72 3.14
4.46 6.01 5.55 3.22 3.07
4.57 5.88 6.22 3.43 3.38
4.78 6.09 6.33 3.41 2.71
5.46 6.14 6.23 3.18 3.2
4.9 5.35 6.52 3.24 3.31
5.19 8.16 9.79 4.27 4.56
5.41 7.36 8.95 4.01 4.62
5.73 7.3 8.77 4.3 4.44
4.97 7.57 9.58 3.79 4.16
4.95 8.21 8.7 3.36 4.25
5.92 7.77 9.63 3.98 4.96
5.62 7.28 9.21 4 4.32
4.99 7.7 8.79 3.91 4.65
5.37 7.63 9.15 4.24 4.81
5.39 7.66 8.99 3.91 4.25
276
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
75 76 77 79 80
139.4 130.2 5.47 10.18 8.94
137 135.2 6.19 10.66 9.74
131 116.8 7.22 19.33 17.22
124.7 105.8 9.06 17.59 18.98
116.6 113.2 8.92 20.01 26.25
117.4 104 9.17 18.39 17.01
120.2 110.3 8.92 17.27 17.29
116.3 114.2 7.45 14.63 11.02
117.9 113.7 1.98 5.72 5.64
122.9 115 2.09 5.58 5.58
118.5 115 1.83 6.08 5.71
114.8 110.8 2.18 6.61 5.37
114.2 108.1 2.29 5.36 5.09
116.1 115.6 1.92 6.17 5.76
119.4 112.8 1.81 5.83 5.84
117 109.9 1.89 6.95 5.83
113.4 110.7 2.35 6.97 6.37
110.8 105.3 2.23 7.12 6.16
132.7 110.7 3.8 11.39 7.79
128 123.1 3.72 10.51 7.06
117.5 116.9 3.58 10.48 7.14
126.5 112.9 3.07 10.47 6.11
126.7 123.9 3.71 10.3 6.86
123.8 118.9 3.67 11.43 7.44
133 124.4 3.35 10.47 6.64
123.2 119.6 3.56 10.36 7.25
120.2 113.8 3.54 10.57 7.6
124.7 121.1 3.27 10.69 6.24
277
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
81 83 84 87 88
5.71 7.99 8.33 53.01 8.84
10.28 7.9 9.42 46.52 9.2
12.29 17.67 16.35 68.67 11.89
12.89 19.68 17.51 82.36 11.1
13.9 23.26 18.8 80.7 15.72
14.22 20.58 18.55 81.89 13.96
13.1 20.43 18.29 79.68 14.71
9.09 13.34 10.76 61.35 11.56
5.18 4.94 5.72 48.31 5.98
5.26 4.04 5.22 49 5.86
4.87 4.57 5.57 54.41 6.96
4.77 4.53 5.13 53.71 6.49
4.18 4.49 4.84 48.96 5.88
4.91 4.14 5.48 54.49 6.56
5.22 4.42 5.61 53.13 6.02
5.04 5.09 5.37 55.94 7.11
5.02 4.51 5.93 55.41 6.82
4.76 5.01 5.92 52.27 6.8
7.59 7.52 6.6 54.6 8.71
7.35 6.98 6.34 55.59 9.22
7.34 6.68 6.58 51.03 9.12
7.46 6.94 6.05 52.01 8.15
6.58 6.4 6.53 52.11 8.85
7.05 6.96 7.41 56.84 9.51
6.72 6.05 6.41 53.88 8.76
7.07 7.21 6.98 52.17 9.07
6.92 7.4 7.3 58.21 9.29
7.19 6.99 6.18 52.25 9.33
278
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
89 90 91 92 93
10.86 3.27 10.27 82.84 82.3
9.45 2.78 8.52 83.72 83.17
14.07 3.22 14.31 122.91 120.43
13.61 3 15.3 134.1 128.41
15.85 5.81 14.21 143.23 139.56
15.99 3.4 13.3 136.88 130.86
14.76 4.37 13.27 136.76 134.09
11.38 3.15 10.12 100.73 99.66
5.92 2.5 4.45 67.63 67.42
5.94 2.08 4.44 69.36 68.32
6.05 2.65 4.34 75.39 75.01
6.37 2.8 4.82 74.6 74.16
5.98 2.23 4.75 68.4 67.68
6.7 2.27 4.35 76.39 76.21
6.21 2.22 5.36 73.62 73.27
6.53 2.27 4.98 79.37 78.07
6.45 2.25 4.28 78.74 77.77
6.89 2.56 5.19 77.1 76.96
8.48 2.06 7.9 81.56 83.03
9.16 2.41 7.25 83.68 83.76
9.48 2.3 7.08 83.33 84.39
9.36 2.02 7.57 83.93 84.88
8.19 1.89 7.08 82.5 83.14
9.36 1.93 7.92 89.72 91.25
9.63 1.94 7.58 83.33 84.5
9.18 2.2 8.2 84.56 85.61
9.06 2.33 7.82 90.91 91.75
9.47 2.38 7.46 84.57 85.54
279
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
94 95 96 97 100
71.99 74.2 7.14 23.5 6.96
73.91 77.5 8.37 22.2 10.25
105.98 76.4 17.68 22.8 15.94
114.21 70.8 18.5 23.2 18.69
125.46 79.7 20.75 24.2 22.25
115.94 81.7 19.29 22.7 19.85
119.77 76.9 17.61 20.9 16.79
88.54 88.7 8.04 20.9 9.45
63.03 14.4 3.32 20.3 3.22
63.93 12.6 3.28 21.2 3.48
70.14 10.2 3.61 21.5 3.78
69.37 12.9 3.36 23.4 3.5
62.43 13.1 3.33 19.8 3.36
71.06 11.5 4.18 22.5 3.88
68.22 11.1 3.92 19.4 3.68
73.61 10.2 3.8 21.1 3.57
72.86 13.2 3.8 19.3 3.73
71.64 10.9 3.49 22.8 3.47
74.86 7 5.93 9.2 5.32
76.28 6.9 5.13 10.3 4.99
76.38 8.6 4.89 11.5 4.7
73.57 8.4 4.98 8.2 4.71
75.22 9.5 4.68 8.3 4.4
82.12 6.1 4.95 11.7 4.65
75.86 6 4.77 12.5 4.47
77.12 5.9 5.2 12 4.66
83.01 7.1 4.95 10.5 4.58
77.21 8.1 4.61 10.7 4.74
280
All Measurements
measurement:
Baptornis advenus (FMNH 395)
Baptornis advenus (UNSM 20030)
Hesperornis gracilis (YPM 55000)
Hesperornis regalis (AMNH 2181)
Hesperornis indet. (SDSM 5312)
Hesperornis regalis (YPM 1200)
Hesperornis regalis (YPM 1476)
Parahesperornis alexi (KUVP 2287)
Aechmophorus occidentalis (LACM 100231)
Aechmophorus occidentalis (LACM 101066)
Aechmophorus occidentalis (LACM 101372)
Aechmophorus occidentalis (LACM 103093)
Aechmophorus occidentalis (LACM 107430)
Aechmophorus occidentalis (LACM 111490)
Aechmophorus occidentalis (LACM 111491)
Aechmophorus occidentalis (LACM 111843)
Aechmophorus occidentalis (LACM 113768)
Aechmophorus occidentalis (LACM 113871)
Gavia immer (LACM 86318)
Gavia immer (LACM 101063)
Gavia immer (LACM 103466)
Gavia immer (LACM 114714)
Gavia immer (LACM 114856)
Gavia immer (LACM 86319)
Gavia immer (LACM 99905)
Gavia immer (LACM 100432)
Gavia immer (LACM 100726)
Gavia immer (LACM 115000)
104 105
4.38 4.02
3.76 3.96
10.9 7.3
11.1 8.85
12.11 7.98
10.63 8.91
11.44 8.42
8.18 4.91
2.2 5.17
1.3 4.12
1.6 4.58
1.5 3.87
1.15 4.71
1.79 4.49
1.73 4.51
1.83 4.81
1.34 4.33
1.62 4.27
1.83 3.32
2.09 3.63
1.9 3.56
1.8 3.89
1.83 4.45
2.3 4.35
2.11 3.84
2.09 4.47
1.65 3.94
1.92 3.28
281

282

FIGURE CAPTIONS 1
Figure 1. Principle component analysis of measurements collected from the femur. A. Upper: 2
PC1 plotted against PC2, lower: loadings for each measurement B. Upper: PC2 plotted against 3
PC3, lower: loadings for each measurement. The following outlying specimens are designated by 4
letter: a – SDSM 54347, an undescribed hesperornithid, b – CFDC B.81.03.16, assigned to 5
Hesperornis macdonaldi, c – KUVP 2289, an undescribed hesperornithid. In Figures 1-4, color 6
codes for the PC plots of specimens are as follows: blue – Baptornis, green – Hesperornis, 7
orange – Brodavis, purple - Parahesperornis. Stars indicate holotype specimens. In Figures 1-4 8
all specimens included in the cladistic analysis presented in the previous chapter are repented by 9
number, as follows: 1 – AMNH 5101; 2 – FMNH 395; 3 - KU2290; 4 – UNSM 20030; 5 – 10
SDSM 68430; 6 – YPM 17208A; 7 – YPM 17208; 8 – LACM 9728; 9 – CFDC BO 780106; 10 11
– YPM 1200; 11 – YPM 17193; 12 – YPM 18589; 13 – YPM 17208; 14 – YPM 55000; 15 – 12
AMNH 2181; 16 – NHM 882; 17 – FMNH 206; 18 – FMNH 281; 19 – FMNH 321; 20 - 13
Neb10148; 21 – SDSM 5312; 22 – SDSM 622; 23 – USNM 13580; 24 – USNM 13581; 25 – 14
YPM 17208; 26 – YPM 1476; 27 – YPM 1478; 28 – YPM 1499; 29 – YPM 1679; 30 – KUVP 15
2287; 31 – FHSM 17312; 32 – KUVP 24090; 33 – YPM 1477. 16
17
Figure 2. Principle component analysis of measurements collected from the tibiotarsus. A. 18
Upper: PC1 plotted against PC2, lower: loadings for each measurement B. Upper: PC2 plotted 19
against PC3, lower: loadings for each measurement. Color and numeric codes are as for Figure 1. 20
21

283

Figure 3. Principle component analysis of measurements collected from the tarsometatarsus. A. 22
Upper: PC1 plotted against PC2, lower: loadings for each measurement B. Upper: PC2 plotted 23
against PC3, lower: loadings for each measurement. Color and numeric codes are as for Figure 1. 24
25
Figure 4. Principle component analysis of all measurements. A. Upper: PC1 plotted against PC2, 26
lower: loadings for each measurement B. Upper: PC2 plotted against PC3, lower: loadings for 27
each measurement. Color and numeric codes are as for Figure 1. 28
29
Figure 5. Comparison of the length of the centrum and width of the ventral surface of the 30
posterior cervical vertebrae (16-17) of hesperornithiforms. 31
32
Figure 6. Comparison of size differences in the articular facets of the thoracic vertebrae of 33
hesperornithiform birds. A. Width of the cranial articular facet versus centrum length. B. Width 34
of the caudal articular facet versus centrum length. 35
36
Figure 7. Width of the rib articulation on thoracic vertebrae 20-23 compared to centrum length. 37
38
Figure 8. Comparisons of the proportions of hindlimb elements in modern and fossil birds. A. 39
Ternary diagram comparing the contribution of femur, tibiotarsus, and tarsometatarsus to total 40
leg length (1 – Parahesperornis, 2 – Hesperornis, 3 – Baptornis). B. Femur and tarsometatarsal 41
length compared in hesperornithiforms (blue), Mesozoic birds (red), and modern birds (green) 42
with linear trends extrapolated for each group. C. Femur and tibiotarsal length compared in 43

284

hesperornithiforms (blue), Mesozoic birds (red), and modern birds (green) with linear trends 44
extrapolated for each group. 45
46
Figure 9. Comparison of the length: width ratio of the femur of hesperornithiform birds. 47
48
Figure 10. Comparison of the femoral trochanteric crest across hesperornithiforms. A. 49
Comparison of the lateral extent of the trochanteric crest past the shaft of the femur to the total 50
proximal width of the femur. B. Comparison of the lateral extent of the trochanteric crest past the 51
shaft of the femur to the midshaft width of the femur. C. Comparison of the total proximal width 52
to the midshaft width of the femur. 53
54
Figure 11. Comparison of the tibiotarsal shaft cross-section in hesperornithiforms. 55
56
Figure 12. Analysis of the medio-lateral expansion of the proximal articular surface of the 57
tibiotarsus in hesperornithiform birds. 58
59
Figure 13. Comparison of the degree of excavation of the intercondylar sulcus of the distal 60
tibiotarsus of hesperornithiforms. 61
62
Figure 14. Dimensions of the distal tibiotarsus of Enaliornis and Baptornis. A. Cranio-caudal 63
width of the lateral condyle compared to the medio-lateral width of distal end. B. Cranio-caudal 64
width of the medial condyle compared to the medio-lateral width of the distal end. C. 65
Comparison of the cranio-caudal widths of the medial and lateral condyles. 66

285

67
FIgrue 15. Elongation of the tarsometatarsus of hesperornithiforms. A. Midshaft width compared 68
to total length. B. Shaft width at the scar for metatarsal I compared with the distance from the 69
proximal end of metatarsal II to the scar for metatarsal I. 70
71
Figure 16. Shape of the proximal tarsometatarsus of hesperornithiform birds. A. Variance in the 72
angle of the proximal articular surface. For taxa with more than one individual, the mean is 73
shown with the sample size indicated in the label. Taxa for which measurements were taken from 74
drawings are indicated with an asterisk. B. Medio-lateral width compared with dorso-plantar 75
width of the proximal articular surface of the tibiotarsus. 76
77
Figure 17. Comparison of the sizes of metatarsal trochlea III and IV in hesperornithiforms. A. 78
Comparison of the medio-lateral width of trochlea III and IV. B. Comparison of the cranio- 79
caudal depth of trochlea III and IV. 80
81
Figure 18. Comparison of the distal extents of the metatarsal trochlea of hesperornithiforms. A. 82
Length of tarsometatarsus compared to distal extent of trochlea IV past that of trochlea III. B. 83
Length of tarsometatarsus compared to distal extent of trochlea III past that of trochlea II. 84
Figure 19. Angle at which metatarsal II trochlea projects medially from the long axis of the shaft. 85
For taxa with more than one individual, the mean is shown with the sample size indicated in the 86
label. 87
Figure 20. Comparison of the length of the intertrochlear incision between metatarsal IV and III 88
trochleae with the length of the tarsometatarsus in hesperornithiform birds. 89

286

TABLE CAPTIONS 90
Table 1. Qualitative morphometric diagnoses. This table presents the range of morphologic 91
features best described quantitatively but that have previously been used only qualitatively in the 92
taxonomy of the Hesperornithiformes. Column 2 (Proposed Utility. describes the context in 93
which the feature has been proposed as diagnostic. This has been in the form of either 94
apomorphies of certain taxonomic levels or as comparisons differentiating one taxon from 95
another taxon or higher grouping (presented as taxon vs. taxon.. 96
97
Table 2. Specimens of hesperornithiform birds included in the four multivariate analyses. 98
Holotype specimens are highlighted in bold. 99
100
Table 3. Comparison of photographic and direct measurement methods. For each measurement 101
the number of individuals measured is shown (n) as well as the p value from comparing the 102
photographic and direct values for each measurement. 103
104
Table 4. Results of single-factor ANOVA for measurements present in three or more of each 105
species of Enaliornis. Measurement numbers correspond with those from the original database 106
(App. 1.. 107
108
Table 4. Results of single-factor ANOVA for measurements present in three or more of each 109
species of Pasquiaornis. Measurement numbers correspond with those from the original database 110
(App. 1.. 111
112

287

Table 6. Measurements with strongly positive or negative loadings in each of the principle 113
component analyses presented here. Measurement numbers correspond to those used in 114
Appendix I. 115
116

288

Figure 1. 117
118
119
120

289

Figure 2. 121
122
123
124

290

Figure 3 125
126
127

291

Figure 4. 128
129
130
131

292

Figure 5. 132
A. 133
B.

134
C. 135
0
2
4
6
8
10
12
14
0 10 20 30 40
Length of Centrum (mm)
Height of Centrum (mm)
B. varneri
H. gracilis
H. regalis
P. alexi
0
5
10
15
20
25
0 5 10 15 20 25
Length of Centrum (mm)
Height of Centrum (mm)
B. varneri
H. regalis
P. alexi
B. advenus
0
5
10
15
20
25
0 5 10 15 20 25
Length of Centrum (mm)
Height of Centrum (mm)
B. varneri
H. regalis
P. alexi
B. advenus

293

Figure 6 136
A. 137
B. 138
139
0
2
4
6
8
10
12
0 10 20 30 40
Width of Caudal Articular Surface (mm)
Width of Waist in Ventral Surface (mm)
B. advenus
C. arctica
H. regalis
P. alexi
0
1
2
3
4
5
6
7
8
9
0 10 20 30 40
Width of Caudal Articular Surface (mm)
Width of Waist in Ventral Surface (mm)
B. advenus
C. arctica
H. regalis
P. alexi

294

Figure 7. 140
141
142
0
2
4
6
8
10
12
0 5 10 15 20 25 30 35 40
height of rib articulation, mm
centrum length, mm
B. advenus
B. varneri
enaliornis
P. alexi
H. regalis

295

Figure 8. 143
144

296

Figure 9. 145
146
147
148
149
0
5
10
15
20
25
0 20 40 60 80 100 120 140
Midshaft Width (mm)
Length (mm)
P. tankei
P. hardiei
Baptornis
H. chowi
H. macdonaldi
H. macdonaldi
H. regalis
Parahesperornis

297

150
Figure 10. 151
A. 152
153
154
B. 155
156
157
0
5
10
15
20
25
0 10 20 30 40 50 60
Lateral extent of Trochanter (mm)
Medio-lateral Width of Proximal End (mm)
E. barretti
E. sedgewicki
E. seeleyi
P. hardei
P. tankei
B. advenus
H. chowi
H. gracilis
H. macdonaldi
H. mengeli
H. regalis
P. alexi
0
5
10
15
20
25
0 5 10 15 20 25
Lateral Extent of Trochanter (mm)
Midshaft Width (mm)
E. barretti
E. sedgewicki
E. seeleyi
P. hardiei
P. tankei
B. advenus
H. regalis
H. chowi
H gracilis
H. macdonaldi
H. mengeli
P. alexi

298

C. 158
159
160
0
10
20
30
40
50
60
0 5 10 15 20 25
Medio-lateral width of Proximal End (mm)
Midshaft Width (mm)
E. barretti
E. sedgewicki
E. seeleyi
P. hardiei
P. tankei
B. advenus
H. regalis
H. chowi
H. gracilis
H. macdonaldi
P. alexi

299

Figure 11. 161
162
163
Figure 12 164
165
166
167
0
5
10
15
20
25
30
0 5 10 15 20 25
Cranio-caudal Midshaft Width (mm)
Medio-lateral Midshaft Width (mm)
Hesperornis
Baptornis
Brodavis
Parahesperornis
0
5
10
15
20
25
30
35
40
45
0 5 10 15 20 25
Width of Proximal End (mm)
Midshaft Width (mm)
Baptornis
B. varneri
Hesperornis
Parahesperornis

300

Figure 13. 168
169
170
171
0
5
10
15
20
25
30
35
40
0 2 4 6 8 10 12 14 16
Depth of Intercondylar Sulcus (mm)
Distal Width (mm)
E. sedgewicki
E. seeleyi
E. barretti
B. advenus
B. varneri
H. altus
H.chowi
H. gracilis
H. regalis
P. alexi

301

Figure 14. 172
A. 173
B. 174
C. 175
0
5
10
15
20
25
30
35
40
0 10 20 30 40
Medio-lateral Width of Distal Shaft
(mm)
Cranio-caudal Width of Lateral Condyle (mm)
B. advenus
B. varneri
H. gracilis
H. chowi
H. altus
H. regalis
P. alexi
E. barretti
E. sedgewicki
E. seeleyi
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25
Medio-lateral Width of Distal Sshaft
(mm)
Cranio-caudal Length of Medial Condyle (mm)
B. advenus
B. varneri
H. gracilis
H. chowi
H. altus
H. regalis
P. alexi
E. seeleyi
E. sedgewicki
E. barretti
0
5
10
15
20
25
30
0 10 20 30 40
Cranio-caudal Width of Lateral
Condyle (mm)
Cranio-caudal Width of Medial Condyle (mm)
E. barretti
E. sedgewicki
E. seeleyi
B. advenus
B. varneri
H. altus
H. chowi
H. gracilis
H. regalis

302

Figure 15. 176
A. 177
B. 178
179
180
181
182
0
20
40
60
80
100
120
140
160
180
0 10 20 30 40 50
Medio-lateral Width of Proximal End (mm)
Length of Tarsometatarsus (mm)
P. tankei
B. advenus
B. varneri
H. bairdi
H. chowi
H. gracilis
H. mengeli
H. regalis
H. rossicus
P. alexi
0
20
40
60
80
100
120
0 20 40 60 80 100 120 140 160
Distance to Sscar for Metatarsal I (mm)
Total Length (mm)
P. tankei
P. hardiei
B. advenus
H. regalis
A. bazhanovi
H. gracilis
H. bairdi
H. chowi
H. indet
P. alexi
B. mongoliensis
B. baileyi
B. americanus
B. varneri

303

Figure 16. 183
A. . 184
B. 185
186
187
68
70
72
74
76
78
80
82
84
86
angle of proximal articular surface
0
5
10
15
20
25
30
0 10 20 30 40 50
Dorso-plantar Depth of Proximal Surface
(mm)
Medio-lateral Width of Proximal Surface (mm)
Baptornis
H. bairdi
H. chowi
H. crassipes
H conlini
H gracilis
H. regalis
H. rossicus
Parahesperornis

304

Figure 17. 188
A. 189
B. 190
191
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15 20
Medio-ateral Width of Trochlea III (mm)
Medio-lateral Width of Trochlea IV (mm)
E. barretti
E. seeleyi
P. tankei
B. advenus
B. varneri
H. bairdi
H. chowi
H. crassipes
H. mengeli
H. gracilis
H. regalis
P. alexi
0
2
4
6
8
10
12
14
16
18
20
0 5 10 15 20
Cranio-caudal Depth of TrochleaIII (mm)
Cranio-caudal Depth of Trochlea IV (mm)
P. tankei
B. advenus
B. varneri
H. bairdi
H. chowi
H. gracilis
H. mengeli
H. regalis
P. alexi

305

Figure 18. 192
A. 193
B. 194
195
0
2
4
6
8
10
0 50 100 150 200
Extent of Trochlea IV past Trochlea III (mm)
Length of Tarsometatarsus (mm)
B. advenus
B. varneri
H. bairdi
H. chowi
H. crassipes
H. gracilis
H. mengeli
H. regalis
H. rossicus
P. alexi
0
4
8
12
16
20
0 50 100 150 200
Extent of Trochlea III past Trochlea II
(mm)
Length of Tarsometatarsus (mm)
B. advenus
B. varneri
H. bairdi
H. chowi
H. crassipes
H. gracilis
H. mengeli
H. regalis
H. rossicus
P. alexi

306

Figure 19. 196
197
198
Figure 20. 199
200
201
202
0
5
10
15
20
25
30
35
Angle of Metatasal II
0
20
40
60
80
100
120
140
160
180
0 5 10 15 20 25 30
Length of Intertrochlear Incision (mm)
Length of Tarsometatarsus (mm)
P. tankei
B. advenus
B. varneri
H. bairdi
H. chowi
H. gracilis
H. mengeli
H. regalis
H. rossicus
P. alexi

307

Table 1. 203
Genus Species Collection Code femur
tibio-
tarsus
tarso-
metatarsus all
Baptornis advenus AMNH 5101 x
Baptornis advenus FMNH 395 x x x x
Baptornis advenus KUVP 2290 x
Baptornis advenus UNSM 20030 x x x x
Baptornis indeterminate Bonner A x
Brodavis varneri SDSM 64830 x x
Hesperornis bairdi YPM 17208-A x
Hesperornis chowi YPM PU 17208

x
Hesperornis chowi YPM PU 17193 x
Hesperornis chowi YPM PU 18589 x

Hesperornis gracilis ypm 1478 x
Hesperornis gracilis YPM 1679

x
Hesperornis gracilis YPM 55000 x x x x
Hesperornis macdonaldi LACM 9728 x

Hesperornis macdonaldi CFDC B.81.03.16 x
Hesperornis mengeli CFDC B78.01.08

x
Hesperornis mengeli YPM PU 17208 x
Hesperornis regalis YPM 1200 x x x x
Hesperornis regalis AMNH 2181 x x x
Hesperornis regalis CFDC B.84.04.18

x
Hesperornis regalis FMNH 206 x x
Hesperornis regalis fmnh 218

x
Hesperornis regalis FMNH 281 x x
Hesperornis regalis FMNH 321 x

Hesperornis regalis FMNH 348 x
Hesperornis regalis KUVP 2289 x

Hesperornis regalis NHM 13581 x
Hesperornis regalis NHM 721 x

Hesperornis regalis SDSM 5313 x
Hesperornis regalis SDSM 5619 x
Hesperornis regalis UNSM 10148 x
Hesperornis regalis YPM 1207 x

x
Hesperornis regalis YPM 1476 x x x x
Hesperornis indeterminate FMNH 32 x

Hesperornis indeterminate NHM 882 x
Hesperornis indeterminate SDSM 5312 x x x x
Hesperornis indeterminate SDSM 54347 x
Hesperornis indeterminate SDSM 56127 x

Hesperornis indeterminate SDSM 622 x x

308

Hesperornis indeterminate SDSM 82695 x
Hesperornis indeterminate UNSM 20029 x
Hesperornis indeterminate USNM 13580 x

Hesperornis indeterminate YPM 1477 x
Hesperornis indeterminate YPM 1499 x

x
Hesperornis indeterminate YPM PU 17208 x
Parahesperornis alexi KUVP 2287 x x x x
Parahesperornis alexi KUVP 24090 x
Parahesperornis indeterminate SMNH 17312 x
204
205

309

Table 2. 206
Differences in Morphology Proposed Utility Proposed by:
Overall variation in body
size species within Enaliornis Galton and Martin, 2002
species within Pasquiaornis Tokaryk et al., 1997
H. macdonaldi vs. Hesperornis Martin and Lim, 2002
H. bairdi vs. H. gracilis Martin and Lim, 2002
H. mengeli vs. H. bairdi Martin and Lim, 2002
H. rossicus vs. Hesperornis
Nessov and Yarkov,
1993
Elongate cervical vertebrae apomorphic of Baptornithidae
Martin and Cordes-
Person, 2007
Laterally expanded caudal
articular surface of posterior
cervical vertebrae apomorphic of Canadaga Hou, 1999
Relative size of rib
articulations on thoracic
vertebrae Baptornithidae vs. Hesperornis
Tokaryk and Harrington,
1997
Relative sizes of cranial and
caudal articular surfaces of
thoracic vertebrae Baptornithidae vs. Hesperornis
Tokaryk and Harrington,
1997
Shortened or reduced femur apomorphic of Hesperornithiformes Martin and Tate, 1976
Parahesperornis vs. Hesperornis Martin, 1984
Tibiotarsus with expanded
proximal articulation apomorphic of B. varneri
Martin and Cordes-
Person, 2007
Compression of tibiotarsus Parahesperornis vs. Hesperornis Martin, 1984
Relative cranio-caudal
expansion of distal condyles
on tibiotarsus species within Enaliornis Galton and Martin, 2002
Relative dimensions of
intercondylar fosssa on
tobiotarsus species within Enaliornis Galton and Martin, 2002
Hesperornis vs. Baptornis Marsh, 1893
Broad, shallow
intercondylar fossa on distal
tibiotarsus apomoprhic of E. sedgewicki Galton and Martin, 2002
Intercondylar fossa on distal
tibiotarsus narrower and
deeper than in E. sedgewicki apomorphic of E. seeleyi Galton and Martin, 2002
Distal intercondylar groove
of tibiotarsus deeply
excavated with smooth,
well-defined walls apomorphic of Hesperornis Marsh, 1893

310

Shape of tarsometatarsus:
slender species within Brodavis Martin et al., 2012
H. gracilis vs. H. regalis Marsh, 1880
H. chowi vs. H. regalis Martin and Lim, 2002
H. mengeli vs. Hesperornis Martin and Lim, 2002
Asiahesperornis vs. Hesperornis Dyke et al., 2006
Shape of tarsometatarsus:
broad or robust apomorphic of Brodavidae Martin et al., 2012
B. americanus vs. Brodavis Martin et al., 2012
H. crassipes vs. H. regalis Marsh, 1880
Angle of proximal surface
of tarsometatarsus apomorphic of H. rossicus
Nessov and Yarkov,
1993
Shape of proximal surface
of tarsometatarsus apomorphic of H. rossicus
Nessov and Yarkov,
1993
Relative placement of facet
for metatarsal I on
tarsometatarsus apomorphic of Brodavidae Martin et al., 2012
B. americanus vs. Brodavis Martin et al., 2012
B. mongoliensis Martin et al., 2012
apomorphic of Asiahesperornis
Nessov and Prixemlin,
1991
Relative lenth of
intertrochlea incsions on
tarsometatarsus apomoprhic of B. varneri
Martin and Cordes-
Person, 2007
Relative distal extent of
trochlea II and III apomoprhic of Pasquiaornis Tokaryk et al., 1997
H. bairdi vs. Parahesperornis Martin and Lim, 2002
Angle of medial projectuon
of metatarsal II apomorphic of Asiahesperornis Dyke et al., 2006
Relative sizes of metatarsals
III and IV B. americanus vs. Brodavis Martin et al., 2012
B. baileyi vs. B. americanus Martin et al., 2012
H. bairdi vs. Parahesperornis Martin and Lim, 2002
H. bairdi vs. Hesperornis Martin and Lim, 2002
apomorphic of Parahesperornis Martin, 1984
207
208

311

Table 3. 209
# measurement n p value
Femur:
1 Cranio-caudal length of trochanter: through widest portion 7 3.27E-08
2 Narrowest cranio-caudal length 7 1.07E-06
3 Cranio-caudal length of head 8 1.20E-07
4
total medio-lateral width of proximal end
2 8.96E-07
5 width across articular surface of lateral condyle 8 1.97E-03
6 length of lateral condyle: from articular surface to caudal end 7 1.99E-06
7 Angle of fibular condyle - photo only na na
8 cranio-caudal length of medial condyle 8 1.70E-02
9 total width 2 7.00E-06
10 femur length, lateral view 14 1.28E-07
11 width of lateral condyle 14 2.70E-06
12 width of trochanter 14 3.86E-02
13 midshaft width 14 9.03E-07
14 femur length, medial view 13 2.05E-07
15 proximo-distal length of head 15 6.77E-07
16 width of medial condyle 17 2.26E-01
17 height of med condyle 2 1.16E-04
18 medial width of head 15 2.80E-07
19 width of neck 15 5.37E-07
20 midshaft width 13 2.33E-07
21 minimum length, taken along midshaft 16 6.12E-07
22 lateral extent of lateral condyle - photo only na na
23 lateral extent of trochanter - photo only na na
24 depth of intercondylar sulcus - photo only na na
25 midshaft width, caudal view 8.23E-06
26 prox-dist length of medial condyle 18 6.28E-07
27 medio-lateral width of medial condyle 1.54E-02
28 femur width at narrowest point of shaft 17 2.47E-07
29 medio-lateral width of proximal end 17 3.05E-09
30 medio-lateral width of distal end 19 3.82E-07
31 length from proximal end to narrowest point of shaft, on lateral side 15 9.85E-05
32 length from distal end to narrowest point of shaft, on lateral side 14 1.56E-06
34
width of trochanter (measured from center of gap between head and trochanter
to lateral margin) 6 6.35E-05
35
width of head (measured from center of gap between head and trochanter to
medial margin) 4 5.21E-07
Tibiotarsus:
36 Width of lateral cotyla 5 8.18E-05
37 Length of lateral cotyla 6 9.03E-03

312

38 Width of medial cotyla 6 8.77E-06
39 Length of medial cotyla 5 1.63E-07
40 length of medial condyle in distal view 4 2.16E-07
41 length of lateral condyle in distal view 4 3.12E-08
42 depth of intercondylar sulcus -photo only na na
43 Length of fibular groove - photo only na na
44 Width of lateral cotyla 15 1.23E-01
46 mid-shaft width 5 4.97E-06
47 cranio-caudal width of lateral condyle 14 4.17E-08
48 surface angle of medial cotyla - photo only na na
49 width of medial cotyla 13 1.41E-04
50 cranio-caudal width of medial condyle in medial view 10 4.43E-07
51 proximo-caudal width of medial cotyla (angled across articular face) 9 3.25E-05
52 width of cnemial crest: taken at base 13 8.90E-03
53 length of cnemial crest, from point at which width was taken to proximal end 3 2.56E-04
54 neck displacement from proximal end 8 2.50E-11
55 width of neck 5 9.23E-06
56 total length 2 3.68E-08
57 midshaft width in cranial view 5 7.00E-05
58 displacement of gap in fibular crest from proximal end 4 5.84E-09
59 width of medial condyle in cranial view 11 1.13E-05
60 width of cranial-most extending (upper) part of medial condyle in cranial view 1 7.27E-04
61 width of lateral condyle in cranial view 12 3.40E-03
62 width of cranial-most extending (upper) part of lateral condyle in cranial view 7 5.30E-03
63 distal width 2 3.17E-07
64 total tibiotarsal length in caudal view 6 2.50E-08
65 width of articular surface, measured along articular plane (angled) 13 5.15E-03
Tarsometatarsus:
66 medio-lateral width of proximal surface 16 7.13E-07
67 dorso-plantar width of proximal surface 4 2.49E-06
68 maximum dorso-plantar length of lateral cotyla 15 3.44E-03
69 maximum dorso-plantar length of medial cotyla 16 1.29E-05
70 length of lateral trochlear ridge of metatarsal IV 14 5.72E-07
71 length of medial trochlear ridge of metatarsal IV 14 4.35E-05
72 length of metatarsal III trochlea 12 2.01E-08
73 width of trochlea IV 10 9.69E-06
74 width of trochlea III 12 1.59E-06
75 angle of inclination of trochlea IV - photo only na
na
76 angle of inclination of trochlea III - photo only na
na
77 width of trochlea II at plantar edge 6 1.71E-07
78 width of trochlea II at middle of distal surface 2 7.64E-01
79 width of lateral face of lateral cotyla 18 2.54E-08
80 midshaft width 13 8.14E-05

313

81 dorso-plantar width of lateral face of trochlea IV 15 8.35E-09
82 angle of proximal margin of medial cotyla - photo only na na
83 width of medial cotyla - along plane across proximal margin 20 5.76E-08
84 midshaft width 15 4.19E-05
85 length of scar for metatarsal I 19 8.52E-05
86 width of scar of metatarsal I 4 2.43E-01
87 length from proximal end of metatarsal II to scar for metatarsal I 17 3.19E-08
88 width of metatarsal II at distal end 20 7.32E-08
89 width of metatarsal III at distal end 17 4.18E-09
90 trochlear ridge displacement of metatarsal II - photo only na na
91 proximal displacement of trochlea II - photo only na na
92 length from proximal to distal end of metatarsal IV 15 2.50E-06
93 length from proximal to distal end of metatarsal III 14 6.19E-07
94 length from proximal to distal end of metatarsal III 17 3.09E-05
95 angle of proximal articular surface - photo only 10 4.62E-07
96 shaft width, at proximal end of scar for metatarsal I 18 4.62E-07
97 angle of metatarsal II from proximo-distal midline - photo only na na
98 portion of trochlea II plantar to trochlea III - dorsal view 5 3.23E-02
99 portion of trochlea II medial to trochlea III - dorsal view 5 2.56E-01
100 shaft width at prox end of scar for met I 15 8.54E-06
101 extent of trochlea IV past III trochlea 3 7.00E-04
102 extent of trochlea IV past trochlea II 4 3.19E-08
103 extent of trochlea III past trochlea II 3 3.26E-09
104 depth of lateral cotyla - photo only na
na
105 depth of medial cotyla - photo only na
na
210
211

314

Table 4. 212
n Mean Variance F ratio p-value Ƞ
2

12. Femur: width of trochanter
E. barretti 3 9.64 7.62 2.909 0.112 0.421
E. sedgewicki 4 6.82 1.21

E. seeleyi 4 8.97 1.00
15. Femur: proximo-distal length of head
E. barretti 3 5.10 0.35 4.039 0.061 0.502
E. sedgewicki 4 4.31 1.22

E. seeleyi 4 6.83 2.90

18. Femur: medial with of head
E. barretti 3 9.47 37.54 2.389 0.154 0.374
E. sedgewicki 4 4.33 0.42

E. seeleyi 4 7.14 0.41
19. Femur: width of neck
E. barretti 3 3.44 0.06 43.580 1.1E-4 0.926
E. sedgewicki 4 3.70 0.04

E. seeleyi 3 4.86 0.01

20. Tibiotarsus: midshaft width
E. barretti 5 6.56 3.48 1.403 0.287 0.203
E. sedgewicki 4 5.45 0.65

E. seeleyi 5 7.18 2.66
213
214

315

Table 5. 215
n Mean Variance F ratio
p-
value Ƞ
2

5. Femur: width of lateral condyle
P. tankei 3 9.31 0.19 14.361 0.019 0.782
P. hardiei 3 6.63 1.31

9. Femur: distal width
P. tankei 3 22.12 1.98 50.619 0.002 0.927
P. hardiei 3 14.69 1.30
24. Femur: depth of intercondylar sulcus

P. tankei 3 3.99 0.77 1.960 0.220 0.282
P. hardiei 4 2.69 1.94
216
Table 6. 217
Individual Element Analyses Analysis of All Elements
PC 2 PC 3 PC 2 PC 3
+ - + - + - + -
54 22 44 1 54 95 44 77
23 97 19 97 97 95
95 105 53 105
97 77
218
219

316

Table 7. 220
Hesperornithiformes
Invalid characters:
Shortened femur (Martin and Tate, 1976)
Baptornithidae
Invalid characters:
Elongate cervical vertebrae (Martin and Cordes-Person, 2007)
Rib articulation and anterior and posterior articular surfaces of centrum of thoracic
vertebrae comparatively smaller than in Hesperornis (Tokaryk and Harrington, 1997)
Pasquiaornis
Invalid characters:
Metatarsal II trochlea located posterior and close to base of trochlea III (Tokaryk et al.,
1997)
P. hardei
Invalid characters:
Reduced femur (Tokaryk et al., 1997)
Brodavidae
Unsupported at current taxonomic level:
Tarsometatarsus short and comparatively broad (Martin et al., 2012)
Invalid characters:
Facet for articulation of metatarsal I displaced proximally on tarsometatarsus shaft (Martin
et al., 2012)
Brodavis americanus
Invalid characters:
Facet for metatarsal I placed below tarsometatarsus midshaft (Martin et al., 2012)
Shaft of tarsometatarsus broader and more robust than B. baileyi but smaller and less robust
than B. varneri (Martin et al., 2012)
Metatarsal IV trochlea swollen proximally and slightly broader than that of metatarsal III
(Martin et al., 2012)
Brodavis baileyi
Invalid characters:
Metatarsal II trochlea more elevated proximally and placed more behind trochlea III (Martin
et al., 2012)
Tarsometatarsus shaft more slender than in B. americanus (Martin et al., 2012)
Metatarsal IV trochlea less expanded at base than in B. americanus (Martin et al., 2012)

317

Brodavis mongolienis
Invalid characters:
Metatarsal shaft slender (Martin et al., 2012)
Facet for metatarsal I located at midshaft (Martin et al., 2012)
Brodavis varneri
Supported characters:
Intertrochlear notches of distal tarsometatarsus extend to nearly midshaft (Martin and
Cordes-Person, 2007)
Invalid characters:
Tibiotarsus with expanded proximal articulation (Martin and Cordes-Person, 2007)
Enaliornis barretti
Invalid characters:
Larger than E. seelyi and E. sedgewicki (Galton and Martin, 2002)
Distal tibiotarsus proportionally deep and narrow with massive lateral and medial condyles
cranially (Galton and Martin, 2002)
Enaliornis sedgewicki
Invalid characters:
Smaller than E. barretti and E. seeleyi (Galton and Martin, 2002)
Lateral and medial condyles of tibiotarsus reduced and nearly equal in size (Galton and
Martin, 2002)
Broad, shallow intercondylar fossa on distal tibiotarsus (Galton and Martin, 2002)
Enaliornis seeleyi
Invalid characters:
Medium-sized species of Enaliornis (Galton and Martin, 2002)
Strongly developed, rounded lateral and medial condyles on distal tibiotarsus (Galton and
Martin, 2002)
Intercondylar fossa on distal tibiotarsus narrower and deeper than in E. sedgewicki (Galton
and Martin, 2002)
Asiahesperornis
Invalid characters:
Metatarsal II trochlea deflected medially; clearly separated from trochlea III by prominent
intertrochlear groove (Dyke et al., 2006)
Tarsometatarsus shaft slender and gracile (Dyke et al., 2006)
Scar for attachment of metatarsal I reduced and proximally located on shaft (Nessov and
Prixemlin, 1991)

318

Canadaga
Invalid characters:
End of centrum of thoracic vertebrae with fan-like expansion and extremely restricted
midline (Hou, 1999)
Hesperornis
Invalid characters:
Distal intercondylar groove of tibiotarsus deeply excavated with smooth, well-defined walls
(Marsh, 1893)
Hesperornis altus
Invalid characters:
Size two-thirds that of H. regalis (Marsh, 1893)
Hesperornis bairdi
Unsupported at current taxonomic level:
Metatarsal IV trochlea more enlarged and distal to trochlea III than in Parahesperornis
(Martin and Lim, 2002)
Invalid characters:
Smaller than H. gracilis (Martin and Lim, 2002)
Hesperornis chowi
Invalid characters:
More elongate tarsometatarsus with more slender shaft than in H. regalis (Martin and Lim,
2002)
Hesperornis crassipes
Invalid characters:
Larger, more robust than H. regalis (Marsh, 1880)
Hesperornis gracilis
Invalid characters:
Tarsometatarsus more slender than H. regalis (Marsh, 1876)
Hesperornis macdonaldi
Valid characters:
Among the smallest of Hesperornis species (modified from Martin and Lim, 2002)
Hesperornis mengeli
Valid characters:
Among the smallest of Hesperornis species (modified from Martin and Lim, 2002)
Invalid characters:
More slender tarsometatarsus than other Hesperornis species(Martin and Lim, 2002)
Metatarsal III trochlea relatively smaller and distally more compressed than in other species
of Hesperornis (Martin and Lim, 2002)

319

Hesperornis rossicus
Supported characters:
Largest body size, 20% larger than H. regalis (Nessov and Yarkov, 1993)
Unsupported at current taxonomic level:
Proximal articular surface of tarsometatarsus with very large transverse width and small
dorsoplantar depth (Nessov and Yarkov, 1993)
Invalid characters:
Proximal articular surface of tarsometatarsus has strong diagonal slant (Nessov and Yarkov,
1993)
Parahesperornis
Unsupported at proposed taxonomic level:
Femur more elongate and proximal end less extended laterally than Hesperornis (Martin,
1984)
Invalid characters:
Metatarsal IV trochlea about one-quarter larger than trochlea III and both with similar distal
extent (Martin, 1984)
Tibiotarsus less compressed than Hesperornis (Martin, 1984)
221
222
223
320

Chapter 4. The Hesperornithiformes: a taxonomic revision 1
INTRODUCTION 2
Integrating the cladistic analysis with the taxonomic framework 3
One of the more interesting results of the cladistic analysis was the placement of 4
Pasquiaornis, Enaliornis, and the holotype of Baptornis advenus in a polytomy basal to the clade 5
Aves + Hesperornithiformes. This implies that including these taxa in the Hesperornithiformes is 6
inappropriate, as it would create a polyphyletic group rather than a monophyletic clade. 7
However, this interpretation is complicated by a number of factors. First, YPM 1465 was a 8
wildcard taxon that could be placed in one of four equally parsimonious places; however none of 9
these were within the Hesperornithiformes (Fig. 1). If YPM 1465 was removed from 10
consideration, then the basal topography of the tree was fully resolved, with Pasquiaornis 11
hardiei and P. tankei forming a monophyletic clade basal to Enaliornis, which was in turn basal 12
to Aves + Hesperornithiformes. Thus the behavior of YPM 1465 as a wildcard obscured an 13
otherwise consistent placement of Enaliornis and Pasquiaornis (Fig.1). The clade Aves + 14
Hesperornithiformes was united with five unambiguous synapomorphies, none of which could be 15
coded for YPM 1465, and only two of which were coded for the remaining taxa. Thus placement 16
outside of the Aves + Hesperornithiformes clade was primarily due to missing, rather than 17
conflicting, character data. For Enaliornis and YPM 1465 the high quantity of missing data was 18
due to the fragmentary nature of the specimens, however Pasquiaornis has been reported to have 19
an extensive collection of specimens (Sanchez et al., 2010). The majority of specimens are 20
unavailable for collection consultation and have not been illustrated in the literature, resulting in 21
a large quantity of missing data for both species of Pasquiaornis. Therefore, it is likely that more 22
complete analysis of Pasquiaornis will resolve its placement. Given the paucity of available data 23
321

at this time, both Enaliornis and Pasquiaornis will be regarded as Hesperornithiformes 24
indeterminate (further discussion below). The holotype of Baptornis advenus, YPM 1465, will 25
also be retained as a hesperornithiform at this time, as its exclusion from the clade was entirely 26
due to missing data. 27
Another result of the cladistic analysis was the disparate placement of a number of 28
specimens previously assigned to Baptornis advenus, none of which were grouped with the 29
holotype (YPM 1465). Instead, three formed a monophyletic cluster supported by two 30
unambiguous synapomorphies (KUVP 2290, FHSM 6318, FMNH 395). UNSM 20030 and 31
AMNH 5101 were placed closer to Hesperornis + Parahesperornis in a monophyletic clade 32
supported by two unambiguous synapomorphies. Despite these disparate placements, the 33
removal of all specimens from Baptornis advenus may not be justified because of the highly 34
fragmentary nature of the holotype (see above). Additionally, of the unambiguous 35
synapomorphies uniting the relevant nodes, the majority of characters were coded as missing 36
data for several specimens (Table 1, Fig. 1). This is particularly relevant to YPM 1465, the 37
holotype of B. advenus, which was coded as missing for every relevant unambiguous 38
synapomorphy. The only two specimens that could be coded for a majority of the relevant 39
characters were FMNH 395 and UNSM 20030. Therefore, to resolve the taxonomy of AMNH 40
5101, FMNH 395, KUVP 2290, FHSM 6318, and YPM 1465 new diagnostic features must be 41
identified. Specimens that cannot be allied with the holotype YPM 1465 should be removed from 42
Baptornis advenus and re-examined taxonomically. 43
Finally, the cladistic analyses returned little support for the vast majority of Hesperornis 44
species. This was somewhat complicated by the inability to include all taxonomic units in a 45
single branch and bound analysis (see Chapter 2 for details); however placement of taxa was 46
322

generally consistent between analyses. Of the eighteen Hesperornis-like taxonomic units 47
included in the analysis, the majority formed a polytomy. YPM 55000 was always placed as the 48
nearest outgroup of the Hesperornis polytomy, separated by 3-4 unambiguous synapomorphies, 49
of which only one conflicted Hesperornis. All of the other unambiguous synapomorphies were 50
missing in YPM 55000 as well as other specimens that were excluded from the main polytomy in 51
the various analyses. This casts serious doubt on the validity of the majority of Hesperornis-like 52
taxa. Species of Hesperornis for which robust diagnostic features cannot be identified should 53
therefore be regarded as invalid. 54
Integrating the morphometric analyses with the taxonomic framework 55
The morphometric analyses presented in Chapter 3 evaluated previous diagnostic features 56
proposed without appropriate quantitative support, and found that the vast majority (forty-two of 57
forty-five) were unsupported. Of these, four did vary across specimens, but were not applied at 58
the appropriate taxonomic level. These characters, as well as the supported diagnostic features, 59
can be integrated into a revised taxonomic framework. 60
61
SYSTEMATIC TAXONOMY 62
Aves Linnaeus 1758 63
Ornithurae Haeckel 1866 64
Hesperornithiformes Furbringer 1888 65
66
Type species – Hesperornis regalis Marsh 1872 67
Included families – Hesperornithidae (Hesperornis, Parahesperornis), Baptornithidae 68
(Baptornis), Enaliornis, Pasquiaornis 69
Occurrence – Global distribution in marine, estuarine, and alluvial strata from the Late 70
Cretaceous. 71
323

Retained diagnostic features – Foot-propelled diving birds with teeth in grooves (Marsh, 72
1877), pterygoid process of the quadrate elongate (Elzanowski, 2000), quadrate with undivided 73
head (Marsh, 1877), sternum without keel (Marsh, 1877), long bones nonpneumatic (Martin, 74
1983), trihedral patella perforated for ambiens tendon (modified from Martin and Tate, 1976), 75
triangular cnemial expansion on the tibiotarsus (modified from Martin and Tate, 1976), 76
tarsometatarsus possesses sharp craniolateral ridge along portion of shaft (modified from Martin, 77
1984), metatarsals dorso-plantarly elongated with ovoid cross-sections (modified from Martin, 78
1984). 79
New diagnostic features – Posterior cervical vertebrae 15-18 with large ventral processes with 80
expanded ventral-most ends, femur arched in medial or lateral view, lengthened tibiotarsus that 81
comprises 50-60% of the total leg length, metatarsal II shifted plantarly behind metatarsal III (the 82
degree to which this is true varies among taxa), distal foramen between trochleae of metatarsals 83
III and IV partially or completely closed and separate from the intertrochlear incision. 84
Discussion – The revised diagnostic features presented here accurately describe all specimens 85
assigned to the Hesperornithiformes for which the relevant elements are preserved, however 86
most specimens only preserve hindlimb material, thus limiting the utility of cranial, axial, and 87
forelimb features. While Enaliornis and Pasquiaornis are currently retained in the 88
Hesperornithiformes, they should be regarded as Hesperornithiformes indeterminate. This is a 89
very tentative placement as the available material is either scarce or largely unavailable for 90
study. The generic and family-level composition of most groups has changed considerably, as 91
discussed below. 92
93
94
324

Aves Linnaeus 1758 95
Ornithurae Haeckel 1866 96
Hesperornithiformes Furbringer 1888 97
Baptornithidae American Ornithologists Union 1910 98
99
Type species – Baptornis advenus Marsh 1877 100
Included genera – Baptornis 101
Occurrence – Global occurrence in marine, estuarine, and alluvial strata from the Late 102
Cretaceous. 103
Retained diagnostic features –Small round pit anterior to the diapophysis of thoracic vertebrae 104
(modified from Nessov and Borkin, 1983), medial and lateral trochlear ridges more pronounced 105
on trochlea III than on trochlea IV (Everhart and Bell, 2009). 106
New diagnostic features – Metatarsals III and IV align medio-laterally distal to midshaft, 107
trochlea of metatarsals III and IV subequal in size; distal foramen between trochlea III and IV 108
partially closed, such that the trochlea of metatarsals III and IV only contact along a small ridge 109
within the intertrochlear incision. 110
Discussion – The main revision to the Baptornithidae is the removal of Pasquiaornis, which was 111
originally included by Tokaryk et al. (1997), and will be further discussed below. Judinornis, 112
known from an isolated thoracic vertebra, is largely unavailable for direct study. As all previous 113
diagnostic features of Judinornis are unsupported and no new diagnostic features could be 114
identified, it is considered Baptornithidae indeterminate at this time. Therefore, the generic and 115
species composition of the Baptornithidae reverts to the original make-up described by Marsh 116
(1880), of Baptornis advenus. 117
118
119
120
325

Aves Linnaeus 1758 121
Ornithurae Haeckel 1866 122
Hesperornithiformes Furbringer 1888 123
Baptornithidae American Ornithologists Union 1910 124
Baptornis Marsh 1877 125
126
Type species – Baptornis advenus Marsh 1877 127
Included species – Baptornis advenus 128
Occurrence – Currently known from Late Cretaceous strata in the United States (Marsh, 1877; 129
Martin and Tate, 1976; Everhart and Bell, 2009), Canada (Tokaryk and Harrington, 1992), and 130
Sweden (Rees and Lindgren, 2005). 131
Retained diagnostic features – Uncinate processes of ribs turned dorsally (Martin and Tate, 132
1976), preacetabular portion of ilium elongate, making up approximately 60% of total ilium 133
length (modified from Martin and Tate, 1976), pyramidal patella (Martin and Tate, 1976), medial 134
and lateral cotyla of tarsometatarsus tilt dorsally (modified from Everhart and Bell, 2009). 135
New diagnostic features – Cranial ridge on the surface of metatarsal IV extends to midshaft. 136
Discussion – While the diagnosis of the genus has been revised considerably, the composition of 137
Baptornis has not changed as a result of this study, with Baptornis advenus remaining the sole 138
species. The list of diagnostic features has been shortened considerably as a result of this study – 139
of the sixteen proposed features four were retained and an additional two were identified. 140
Aves Linnaeus 1758 141
Ornithurae Haeckel 1866 142
Hesperornithiformes Furbringer 1888 143
Baptornithidae American Ornithologists Union 1910 144
Baptornis advenus Marsh 1877 145
146
Holotype specimen – YPM 1465, an isolated distal tarsometatarsus 147
Referred specimens – AMNH 5101, FMNH 395, KUVP 2290, KUVP 16112 148
326

Occurrence – B. advenus has been reported from the Niobrara Formation (Coniacian) of 149
Kansas. 150
Retained diagnostic features – As for genus. 151
New diagnostic features – As for genus. 152
Discussion – In the cladistic analysis, AMH 5101 (assigned to Baptornis advenus – Martin and 153
Tate, 1976) was placed as a basal terminal to the clade containing Hesperornis + 154
Parahesperornis on the basis of two unambiguous synapomorphies, neither of which could be 155
coded for most of the other Baptornis specimens. AMNH 5101 can be included in Baptornis 156
advenus on the basis of all the features presented above in the amended diagnosis of the genus. 157
UNSM 20030 is the other specimen previously assigned to Baptornis advenus that was placed 158
adjacent to Hesperornis + Parahesperornis in the cladistic analysis; however this placement is 159
supported by numerous characters. Furthermore, UNSM 20030 does not exhibit all the 160
diagnostic features of the genus. At this time, it is recommended that UNSM 20030 be removed 161
from Baptornis advenus and re-assigned to the Hesperornithidae (see below). 162
Aves Linnaeus 1758 163
Ornithurae Haeckel 1866 164
Hesperornithiformes Furbringer 1888 165
Hesperornithidae Marsh 1872 166
167
Type species – Hesperornis regalis Marsh 1872 168
Included genera – Hesperornis, Parahesperornis 169
Occurrence – Global distribution in marine and estuarine strata from the Late Cretaceous. 170
Retained diagnostic features – Robust, trapezoidal coracoid (modified from Martin et al., 171
1984), reduced quadrate pneumaticity (modified from Martin and Tate, 1976), combination of 172
short, broad pterygoids with long, narrow palatines (modified from Martin, 1984), entire anterior 173
surface of metatarsal IV developed as a dorsally projecting ridge, from just below the articular 174
327

surface to the trochlea (modified from Martin and Tate, 1976), lateral ridge of metatarsal IV 175
trochlea enlarged relative to the medial (Everhart and Bell, 2009), phalanges one, two, and three 176
of digit IV with distal peg-and-crescent articulations (Bell and Everhart, 2009). 177
New diagnostic features – Femur narrows to a waist proximal to midshaft, highly medio- 178
laterally flattened patella shaped like an isosceles triangle, deeply excavated groove along the 179
proximal lateral shaft of the proximal tibiotarsus that wraps up onto the proximal articular 180
surface cranial to the lateral cotyla, prominent lateral groove separating metatarsals III and IV 181
along length. 182
Discussion – Both the diagnosis and generic composition of the Hesperornithidae is heavily 183
revised in this study. Two genera, Asiahesperornis and Canadaga, are considered 184
Hesperornithidae indeterminate. An additional specimen, UNSM 20030, is added to the 185
Hesperornithidae on the basis of the following diagnostic features: entire anterior surface of 186
metatarsal IV developed as a dorsally projecting ridge, from just below the articular surface to 187
the trochlea; greatly enlarged metatarsal IV trochlea (nearly twice the width of trochlea III); 188
lateral ridge of metatarsal IV trochlea enlarged relative to the medial; prominent lateral groove 189
separating metatarsals III and IV along length. 190
Aves Linnaeus 1758 191
Ornithurae Haeckel 1866 192
Hesperornithiformes Furbringer 1888 193
Hesperornithidae Marsh 1872 194
Hesperornis Marsh 1872 195
196
Type species – Hesperornis regalis Marsh 1872 197
Included species –Hesperornis bairdi, Hesperornis regalis, Hesperornis rossicus, Hesperornis 198
macdonaldi, Hesperornis mengeli 199
Occurrence – Global distribution in marine and estuarine strata from the Late Cretaceous. 200
328

Retained diagnostic features – Lacrimal with shortened nasal process (modified from Martin, 201
1984), greatly enlarged metatarsal IV trochlea (nearly twice the width of trochlea III) (modified 202
from Tokaryk et al., 1997). 203
New diagnostic features – Distal tibiotarsus dramatically inflected medially, peaked intercotylar 204
eminence on the proximal tarsometatarsus, proximo-lateral face of metatarsal IV marked with a 205
large round depression, articular surface of metatarsal II highly angled distally at dorsal margin, 206
medial trochlear ridge over twice the dorso-plantar length of the lateral, . 207
Discussion – The majority of features adopted here as diagnostic of Hesperornis were originally 208
proposed for a more inclusive taxonomic group, usually at the species level. Evaluation during 209
the course of this study recognized these features as common to all members of Hesperornis, 210
thus invalidating some taxa (H. altus, H. chowi, H. crassipes, H. gracilis). As mentioned above, 211
Canadaga and Asiahesperornis are here reassigned to Hesperornithidae indeterminate. 212
Aves Linnaeus 1758 213
Ornithurae Haeckel 1866 214
Hesperornithiformes Furbringer 1888 215
Hesperornithidae Marsh 1872 216
Hesperornis bairdi Martin and Lim 2002 217
218
Type specimen – YPM PU 17208A, left tarsometatarsus. 219
Referred specimens – None. 220
Occurrence – Known from the Sharon Springs Member of the Pierre Shale (Campanian) of 221
South Dakota, USA. 222
Retained diagnostic features – Among the smallest of Hesperornis species (modified from 223
Martin and Lim, 2002). 224
New diagnostic features – Enlarged distal foramen between metatarsals III and IV with a slight 225
intertrochlear groove present distal to the foramen. 226
329

Discussion – Determining speciation almost solely on the basis of size with few diagnostic 227
features seems to be warranted in this case by the extreme size discrepancy between the holotype 228
and most other species of Hesperornis (Fig. 3), with the possible exception of H. macdonaldi and 229
H. mengeli. H. mengeli, also known from a small tarsometatarsus, is smaller than H. bairdi. H. 230
macdonaldi is known from an isolated femur and cannot be compared to the isolated 231
tarsometatarsi of H. bairdi and H. mengeli. It is possible that all or some of these specimens 232
represent juvenile individuals or are con-specific; however in the absence of histologic data this 233
remains unclear. While H. bairdi is nearly identical in size and dimension to Parahesperornis, it 234
displays all of the features of the tarsometatarsus found in other species of Hesperornis and so is 235
not allied with Parahesperornis. 236
Aves Linnaeus 1758 237
Ornithurae Haeckel 1866 238
Hesperornithiformes Furbringer 1888 239
Hesperornithidae Marsh 1872 240
Hesperornis macdonaldi Martin and Lim 2002 241
242
Type specimen – LACM 9728, an isolated femur. 243
Referred specimens – None. 244
Occurrence – Known from the Sharon Springs Member of the Pierre Shale (Campanian) of 245
South Dakota, USA. 246
Retained diagnostic features – Among the smallest of Hesperornis species (modified from 247
Martin and Lim, 2002). 248
New diagnostic features – Femur more elongate than in other hesperornithiforms. 249
Discussion – As discussed for H. bairdi above, the assignment of H. macdonaldi is supported on 250
the basis of its extremely small size as compared to other species of Hesperornis, however 251
histological analysis may reveal the holotype to be a juvenile. Morphometric analysis shows that 252
330

the length: width proportions of H. macdonaldi do not fall along the trend seen in other 253
hesperornithiforms (Fig. 4), thus providing an additional diagnostic feature of the species. 254
Aves Linnaeus 1758 255
Ornithurae Haeckel 1866 256
Hesperornithiformes Furbringer 1888 257
Hesperornithidae Marsh 1872 258
Hesperornis mengeli Martin and Lim 2002 259
260
Type specimen – CFDC B78.01.08, an isolated tarsometatarsus (erroneously reported as BO 261
780106 in Martin and Lim, 2002). 262
Referred specimens – None. 263
Occurrence – Known from the Sharon Springs Member of the Pierre Shale (Campanian) of 264
southern Manitoba, Canada. 265
Retained diagnostic features – Among the smallest Hesperornis species (modified from Martin 266
and Lim, 2002). 267
New diagnostic features – None. 268
Discussion – H. mengeli is retained as a valid taxon at this time due to the extreme difference in 269
size between the holotype and other species of Hesperornis for which a tarsometatarsus is 270
preserved. See the above discussion regarding the use of the small size as diagnostic of H. bairdi, 271
H. macdonaldi, and H. mengeli. 272
Aves Linnaeus 1758 273
Ornithurae Haeckel 1866 274
Hesperornithiformes Furbringer 1888 275
Hesperornithidae Marsh 1872 276
Hesperornis regalis Marsh 1872 277
278
Type specimen – YPM 1200, a partial skeleton preserving vertebrae (two anterior cervical; one 279
posterior thoracic, possibly number 20-21; ten free caudal; pygostyle with two fused centra), 280
331

ribs, femora, patellae, fibulae, tibiotarsi, tarsometatarsi, and pedal phalanges (left & right – digit 281
IV:1, 2, digit III:2; right – digit III:1, 3) 282
Referred specimens - Numerous, with the addition at this time of all specimens previously 283
referred to H. altus (YPM 515), H. chowi (YPM 17193, YPM 18589), H. crassipes (YPM 1474), 284
and H. gracilis (YPM 1473, YPM 1478, YPM 1679) 285
Occurrence – Late Cretaceous marine and estuarine strata of the central United States, with 286
most specimens found in the Niobrara Formation (Cenomanian) of Kansas or the Pierre Shale 287
(Campanian) of South Dakota. 288
Retained diagnostic features – None. 289
New diagnostic features – As for the genus. 290
Discussion – The present study failed to find support for all of the proposed features for a 291
number of Hesperornis species (H. altus, H. chowi, H. crassipes, and H. gracilis). As no new 292
features were identified to distinguish these invalid taxa, the majority of assigned specimens 293
revert to H. regalis. The following species are considered junior synonyms of H. regalis: H. 294
altus, H. chowi, H. crassipes, and H. gracilis. 295
Aves Linnaeus 1758 296
Ornithurae Haeckel 1866 297
Hesperornithiformes Furbringer 1888 298
Hesperornithidae Marsh 1872 299
Hesperornis rossicus Nessov and Yarkov 1993 300
301
Type specimen – VPM N 26306/2, a proximal tarsometatarsus 302
Referred specimens – RM PZ R398 (Nessov and Yarkov, 1993); ZIN PO 5463, ZIN PO 5464 303
(Panteleyev et al., 2004); SGU 3442 Ve01 (Rees and Lindgren, 2005) 304
Occurrence – Known from the Kristianistad Basin (Campanian) of Sweden and the Rybushka 305
Formation (Campanian) of Russia. 306
332

Retained diagnostic features – Size 20% larger than other species of Hesperornis (Nessov and 307
Yarkov, 1993), proximal articular surface of tarsometatarsus with very large transverse width 308
and small dorso-plantar depth (Nessov and Yarkov, 1993), trochlear condyle of metatarsal II 309
completely behind that of digit III (Panteleyev et al., 2004). 310
New diagnostic features – Shaft highly twisted, with the proximal end in near lateral view when 311
the distal end of metatarsal IV is in dorsal view. 312
Discussion – While some features diagnostic of H. rossicus were invalidated in the present 313
study, several were retained and an additional feature identified. Quantitative support of the 314
comparatively large size of H. rossicus was identified (Fig. 3B). Even with the inclusion of all 315
available hesperornithiform specimens, H. rossicus is significantly larger than other specimens 316
of Hesperornis. While several vertebrae and a fragmentary tibiotarsus are also known from the 317
same localities as H. rossicus (Nessov and Yarkov, 1993; Rees and Lindgren, 2005), assignment 318
of the unassociated elements to H. rossicus is not justified at this time. 319
Aves Linnaeus 1758 320
Ornithurae Haeckel 1866 321
Hesperornithiformes Furbringer 1888 322
Hesperornithidae Marsh 1872 323
Hesperornis indeterminate 324
325
A number of specimens have been identified during the course of this study that appear to differ 326
substantially from the typical form found in H. regalis. These specimens are briefly described 327
here. 328
YPM 55000 – Cast of a partial skeleton in the collection of the University of Wisconsin at 329
Madison (partial scapula, two posterior cervical, six dorsal, and two free caudal vertebrae, 330
synsacrum and partial pelvis, partial pygostyle, femora, patellae, fibula, tibiotarsi, tarsometatarsi, 331
phalanges). Geographic or geologic information is not known at this time. 332
333

Discussion. YPM 55000 was consistently placed in a basal location as compared to other 333
specimens of Hesperornis in the cladistic analysis and usually fell outside of the morphospace 334
occupied by Hesperornis. Additionally, the specimen appears to suffer some form of deformity 335
of the right hindlimb, which is certainly worth additional study. 336
KUVP 2280a – Nearly complete femur and tarsometatarsus from the Vermillion River 337
Formation of Manitoba, Canada. 338
Discussion. This material differs in several morphological details of the distal femur 339
from Hesperornis regalis, while maintaining an overall similarity with Hesperornis in the form 340
of the femur and tarsometatarsus. This individual is also fairly small, and may prove to be related 341
to H. bairdi, as both are of a similar size. 342
SDSM 80452 – Isolated femur from the Pierre Shale of South Dakota, USA. 343
Discussion – Analysis of this specimen is complicated by extremely poor preservation, 344
however the specimen appears to be incredibly thickened around the midshaft region in cranial 345
or caudal view and possessing a distinct hump on the caudal surface of the midshaft. Further 346
analysis may identify these proportions as sufficiently different from other Hesperornis 347
specimens to warrant species distinction. 348
YPM 1476 – Partial skeleton consisting of sternum fragments, eleven vertebrae (five posterior 349
cervical and six thoracic), a fairly complete pelvis, and a femur, patella, tibiotarsus, 350
tarsometatarsus, and phalanges IV:1 and III:1,2 from the Niobrara Formation of Kansas, USA. 351
Discussion – One of the bivariate analyses identified an unusual discrepancy in the extent 352
of metatarsal IV trochlea past that of III in YPM 1476. In this feature, YPM 1476 very closely 353
resembles Parahesperornis, despite being of average Hesperornis size and possessing other 354
features common to Hesperornis (e.g. peaked intercotylar eminence, trochlea IV twice as wide as 355
334

trochlea III, etc.). The morphology of the femur also differs slightly from that of H. regalis, 356
primarily in details of the proximal end. 357
Aves Linnaeus 1758 358
Ornithurae Haeckel 1866 359
Hesperornithiformes Furbringer 1888 360
Hesperornithidae Marsh 1872 361
Parahesperornis Martin 1984 362
363
Type species – Parahesperornis alexi Martin 1984 364
Included species – Parahesperornis alexi 365
Occurrence – Known exclusively from the Smoky Hill Chalk of the Niobrara Formation 366
(Cenomanian) in central Kansas, USA. 367
Retained diagnostic features – Metatarsal IV trochlea about one-quarter larger than trochlea III 368
(Martin, 1984) 369
New diagnostic features – Groove separating metatarsals III and IV along tarsometatarsus shaft 370
very deeply excavated distally. 371
Discussion – Despite being similar in size to the small Hesperornis species H. bairdi, 372
Parahesperornis is morphologically distinct, lacking the apomorphies found in Hesperornis and 373
possessing features not seen in Hesperornis. 374
Aves Linnaeus 1758 375
Ornithurae Haeckel 1866 376
Hesperornithiformes Furbringer 1888 377
Hesperornithidae Marsh 1872 378
Parahesperornis alexi Martin 1984 379
380
Type specimen – KUVP 2287, a nearly complete disarticulated specimen preserving most of the 381
skull (braincase, dentary, frontal, lacrimal, mandibles, maxillae, nasals, palatine, premaxilla, 382
pterygoid, quadrates, vomers), vertebrae (axis, 17 cervical, and 4 thoracic), synsacrum and pelvis 383
with nearly complete pubes and ilia and partial ischia; ribs; forelimb (coracoid, humerus, 384
335

sternum); and hindlimb (femora, tibiotarsi, patellae, fibula, tarsometatarsi, the majority of pedal 385
phalanges, missing only the ungual phalanx from the left digit I and the third phalanx from the 386
left digit III). 387
Referred specimens - KUVP 24090 388
Occurrence, retained diagnostic features, and new diagnostic features – As for genus. 389
Aves Linnaeus 1758 390
Ornithurae Haeckel 1866 391
Hesperornithiformes Furbringer 1888 392
Hesperornithidae Marsh 1872 393
Hesperornithidae indeterminate 394
395
Type specimen – NMC 41050, two partial and one complete articulated posterior cervical 396
vertebrae, most likely numbers 15-17, previously assigned to Canadaga arctica. 397
Referred specimens – NUVF 284 (Wilson et al., 2011) 398
Occurrence – The stratigraphic occurrence of the holotype has not been reported, however the 399
age was described as Maastrichtian (Hou, 1999). An additional referred specimen was reported 400
from the Kanguk Formation (Campanian) (Wilson et al., 2011). Both specimens are from islands 401
in the Canadian High Arctic. 402
Retained diagnostic features – Concavitas lateralis large and deep, occupies entire lateral face 403
of centrum of thoracic vertebrae (Hou, 1999). 404
New diagnostic features – Among the largest Hesperornis species. 405
Discussion –Originally separated from Hesperornis in its own genus, Canadaga, the analysis 406
presented here has invalidated the majority of features used for to justify this placement. At this 407
time, there are insufficient distinctions to warrant a separate genus, and so the specimens should 408
be considered as Hesperornithidae indeterminate. NMC 41050 is characterized by the extremely 409
large size of the vertebrae as compared to those of other hesperornithiforms. Fig. 2 compares the 410
336

ratio of centrum length to height in hesperornithiforms, where H. arctica plots well beyond the 411
morphospace occupied by the other specimens. This size discrepancy is similar to that seen in H. 412
rossicus, however as only tarsometatarsi are known from H. rossicus a direct size comparison 413
cannot be made. Hou originally referred two femora and a caudal vertebra to Canadaga arctica 414
(1999), but the lack of association among any of the elements and the holotype makes this 415
assignment invalid. As a final note, the temporal distribution of this group (Coniacian – 416
Maastrichtian) is only weakly supported, as the geological occurrence of the holotype has not 417
been reported and so the Maastrichtian age cannot be verified. In summary, as some diagnostic 418
features do exist for NMC 41050, referral of this material as a species of Hesperornis in future 419
work appears to be justified. 420
Aves Linnaeus 1758 421
Ornithurae Haeckel 1866 422
Hesperornithiformes Furbringer 1888 423
Hesperornithidae Marsh 1872 424
Hesperornithidae indeterminate 425
426
Type specimen – IZASK 5/287/86a, an isolated partial tarsometatarsus lacking the proximal and 427
distal-most ends, originally described as Asiahesperornis bazhanovi (Nessov and Prizemlin, 428
1991). 429
Referred specimens – IZASK 5/287/86b, IZASK 1/KM 97, IZASK 3/KM 97 all tarsometatarsi 430
Occurrence – Known exclusively from the Zhuravlovskaya Svita (Maastrichtian) of northern 431
Kazakhstan. 432
Retained diagnostic features – Medial as well as lateral grooves separating metatarsals along 433
shaft deeply excavated, metatarsal II trochlea separated from trochlea III by an intercondylar 434
incision (modified from Dyke et al., 2006). 435
New diagnostic features – None. 436
337

Discussion – Originally described as Asiahesperornis (Nessov and Prizemlin, 1991) and later 437
upheld (Dyke et al., 2006), no justification has been provided for the generic distinction of this 438
material. This study finds little difference between specimens assigned to Asiahesperornis and 439
those of Hesperornis regalis, particularly given the wide diversity of hesperornithiforms and the 440
fairly narrow diversity of Hesperornis. Therefore, Asiahesperornis is designated as 441
Hesperornithidae indeterminate. As some diagnostic features do exist, referral of this material as 442
a species of Hesperornis appears to be justified. Additionally, while several other elements have 443
been described as con-specific with the holotype and assigned tarsometatarsi (Nessov and 444
Prizemlin, 1992; Dyke et al., 2006), the lack of association among elements makes these 445
assignments unsupported. 446
Aves Linnaeus 1758 447
Ornithurae Haeckel 1866 448
Hesperornithiformes Furbringer 1888 449
Hesperornithiformes indeterminate 450
Brodavis Martin et al. 2012 451
452
Type species – Brodavis varneri Martin et al. 2012 453
Included species –Brodavis varneri 454
Occurrence – Known from the Sharon Springs Member of the Pierre Shale (Campanian) of 455
South Dakota, USA. 456
Retained diagnostic features – Tarsometatarsus shaft narrows to a waist distal to midshaft in 457
dorsal view (modified from Martin and Cordes-Person, 2007). 458
New diagnostic features – None at this time. 459
Discussion – The extremely poor preservation off all specimens previously assigned to Brodavis 460
makes the identification of diagnostic features difficult. B. americanus and B. mongoliensis have 461
never been adequately illustrated in the literature and are unavailable for study. Because of this, 462
338

the holotypes are here reassigned to Brodavis indeterminate. Furthermore, none of the features 463
originally proposed for these species, or for B. baileyi, which was available for study, were 464
supported. As no new features could be identified as apomorphic of B. baileyi, it is also 465
considered as Brodavis indeterminate at this time. 466
Aves Linnaeus 1758 467
Ornithurae Haeckel 1866 468
Hesperornithiformes Furbringer 1888 469
Hesperornithiformes indeterminate 470
Brodavis Martin et al., 2012 471
Brodavis varneri Martin et al. 2012 472
473
Type specimen – SDSM 98430, a partial skeleton preserving vertebrae (one anterior and three 474
posterior cervical, two thoracic), ribs, fragmentary pelvis and synsacrum (small portion of both 475
ilia and about half of both ischia and one pubis), distal femur, fibula, tibiotarsus, and 476
tarsometatarsus. 477
Referred specimens – None. 478
Occurrence – As for the genus. 479
Retained diagnostic features – Intertrochlear incision between trochlea III and IV extends to 480
nearly midshaft (modified from Martin and Cordes-Person, 2007). 481
New diagnostic features – None. 482
Discussion – The apparent elongation of the intertrochlear incision is here retained as diagnostic, 483
however this should be treated with care given the poor preservation of the specimen. This 484
feature is only diagnostic of B. varneri and not B. baileyi, hence its use at the species level and 485
not that of the genus. It should be noted that during trips to the SDSM collections, three small 486
thoracic vertebrae (most likely belonging to Baptornis) were present mixed with the material of 487
the holotype. There is a clear size discrepancy between these vertebrae and the holotype material 488
and it is therefore recommended that these elements be removed from SDSM 98430. Despite the 489
339

original referral of this specimen to Baptornis as B. varneri (Martin and Cordes-Person, 2007), 490
no features could be identified at this time to support such a grouping, and so the removal of the 491
specimen and re-assignment to Brodavis by Martin et al. (2012) is retained at this time. 492
Aves Linnaeus 1758 493
Ornithurae Haeckel 1866 494
Hesperornithiformes Furbringer 1888 495
Hesperornithiformes indeterminate 496
Enaliornis Seeley 1876 497
Type species – None. 498
Included species – None (previously E. barretti, E. sedgewicki, and E. seeleyi). 499
Occurrence – Known exclusively from the Cambridge Greensand (Late Albian) of southern 500
England. 501
Retained diagnostic features – None. 502
New diagnostic features – None. 503
Discussion – The inability to identify any diagnostic features unique to Enaliornis necessitates 504
the treatment of this taxa with considerable caution. Enaliornis can be grouped within the 505
Hesperornithiformes on the basis of the following diagnostic features: long bones nonpneumatic 506
(Martin, 1983), triangular cnemial expansion on the tibiotarsus (modified from Martin and Tate, 507
1976), tarsometatarsus possesses sharp craniolateral ridge along portion of shaft (modified from 508
Martin, 1984, metatarsal II shifted plantarly behind metatarsal III (the degree to which this is true 509
varies among taxa), distal foramen between trochleae of metatarsals III and IV partially or 510
completely closed and separate from the intertrochlear incision. However, the preserved material 511
is too sparse for the identification of apomorphic features. Furthermore, statistical analyses 512
presented in Chapter 3 failed to identify any significant measures that could be supported as 513
differing among the three previously proposed species. As none of the proposed diagnostic 514
340

features for the species could be supported, there is no justification for dividing the highly 515
fragmented and entirely un-associated remains into three species. 516
Aves Linnaeus 1758 517
Ornithurae Haeckel 1866 518
Hesperornithiformes Furbringer 1888 519
Hesperornithidae indeterminate 520
Pasquiaornis Tokaryk et al. 1997 521
Type species – None. 522
Included species – Pasquiaornis hardiei, Pasquiaornis tankei 523
Occurrence – Known exclusively from the Belle Fourche Formation (middle Cenomanian) of 524
Saskatchewan, Canada. 525
Retained diagnostic features – None. 526
New diagnostic features – None. 527
Discussion – Pasquiaornis taxonomy and relationships are complicated by the sparse amount of 528
fossil material that has been adequately described and published. At this time, two main 529
difficulties exist in validating Pasquiaornis. First, the assignment of what has been reported as 530
hundreds of small avian fossils (Tokaryk et al., 1997; Sanchez, 2010) to one of two species 531
solely on the basis of size is problematic. As the material is entirely disarticulated and 532
unassociated, as is the case in Enaliornis, there is no way to definitively state that different 533
elements belong to the same species. While all can be sorted into large and small groups (as 534
shown in the ANOVA in Chapter 3), judging whether the large femora are con-specific or even 535
con-generic with the large humeri is impossible. 536
Second, the elements assigned to Pasquiaornis differ from those of all other 537
hesperornithiforms in several ways. For example, the acetabulum of Pasquiaornis is clearly 538
completely open with very short, steep sides, much unlike the acetabula of Baptornis, 539
Hesperornis, and Parahesperornis which have sloped walls that might have formed a closed 540
341

acetabulum, however none are preserved well enough to say this with any certainty. 541
Additionally, the forelimb material assigned to Pasquiaornis is completely unlike that of any 542
hesperornithiform. While few wing elements are known, Baptornis, Hesperornis, and 543
Parahesperornis all have extremely thin, elongate humeri that curve only slightly and do not 544
show any development of the deltopectoral crest or distal condyles. The humeri known from 545
Pasquiaornis are much more similar to those of flighted birds, with a deltopectoral crest, globose 546
humeral head, and distinct distal condyles. The ulna and radius also appear to be similarly well- 547
developed. In other hesperornithiforms, the ulna and radius are only known from KUVP 2290, 548
assigned to Baptornis advenus, and they are incredibly rudimentary. 549
Therefore, while Pasquiaornis does share features in common with the 550
Hesperornithiformes [pterygoid process of the quadrate elongate (Elzanowski, 2000), quadrate 551
with undivided head (Marsh, 1877), long bones nonpneumatic (Martin, 1983), triangular cnemial 552
expansion on the tibiotarsus (modified from Martin and Tate, 1976), tarsometatarsus possesses 553
sharp craniolateral ridge along portion of shaft (modified from Martin, 1984), metatarsals dorso- 554
plantarly elongated with ovoid cross-sections (modified from Martin, 1984), metatarsal II shifted 555
plantarly behind metatarsal III (the degree to which this is true varies among taxa), distal 556
foramen between trochleae of metatarsals III and IV partially or completely closed and separate 557
from the intertrochlear incision], these features are all found in various modern diving taxa and 558
so could well be homologous for a similar aquatic lifestyle of Pasquiaornis and other 559
hesperornithiforms. 560
561
562
563
342

CONCLUSIONS 564
This study has demonstrated that much of the taxonomic distinctions previously recognized in 565
the Hesperornithiformes cannot be supported. While there does appear to be fairly strong support 566
for a monophyletic Hesperornithiformes containing Hesperornis, Parahesperornis, Baptornis, 567
and Brodavis, the relationships of these genera to each other is more complicated. The 568
Hesperornithidae is the only sub-grouping which retains a large number of diagnostic features. 569
Within this family, there is strong support for Hesperornis regalis, H. rossicus, and 570
Parahesperornis alexi. Other retained species of Hesperornis, H. bairdi, H. macdonaldi, and H. 571
mengeli, all share the distinction of being incredibly small as compared to all other known 572
specimens, however the relation of these taxa to each other is unclear. Histologic work may 573
reveal these specimens to be juveniles of the larger Hesperornis species. Among the other taxa, 574
recognition of the Baptornithidae may be unwarranted, as the family remains mono –generic and 575
–specific. Future work may recognize a closer relationship between Brodavis and one of the 576
other groups of hesperornithiforms, possibly Baptornis, however at this time such a grouping is 577
not supported. Finally, the disassociated elements assigned to Enaliornis and Pasquiaornis 578
should be treated cautiously, as little evidence exists to unite specimens into the respective 579
genera to which they have been assigned. 580
581
343

FIGURE CAPTIONS 582
Figure 1. Most equally parsimonious trees returned from the cladistic analysis presented in 583
Chapter 2, with the alternate positions of YPM 1465 indicated. Labeled nodes correspond to the 584
columns listed in Table 1, which shows the distribution of missing data across taxa for each 585
node. Branches are labeled with branch lengths. 586
587
Figure 2. Size distribution of hesperornithiform posterior cervical vertebrae 16 (A) and 17 (B). 588
589
Figure 3. Size distribution of femora (A) and tarsometatarsi (B) in hesperornithiform taxa. 590
591
TABLE CAPTIONS 592
Table 1. Distribution of missing data across Enaliornis, Pasquiaornis, and specimens assigned 593
to Baptornis. Column labels correspond to the node labels used in Figure 1. For each taxa the 594
number of coded unambiguous synapomorphies is indicated before the number of unambiguous 595
synapomorphies identified for each node (for example, 1/2 indicates that out of two 596
unambiguous synapomorphies at that node, one could be coded). Dashes indicate the specimen 597
or taxon is included in that node. 598
599
344

Figure 1. 600
601
602
345

Figure 2. 603
A. 604
B. 605
606
0
2
4
6
8
10
12
0 10 20 30 40
Width of Caudal Articular Surface (mm)
Width of Waist in Ventral Surface (mm)
B. advenus
C. arctica
H. regalis
P. alexi
0
1
2
3
4
5
6
7
8
9
0 10 20 30 40
Width of Caudal Articular Surface (mm0
Width of Waist in Ventral Surface (mm)
B. advenus
C. arctica
H. regalis
P. alexi
346

Figure 3. 607
A. 608
B. 609
610
0
5
10
15
20
25
0 20 40 60 80 100 120 140
Midshaft Width (mm)
Length of Femur (mm)
P. tankei
P. hardiei
Baptornis
H. chowi
H. macdonaldi
H. macdonaldi
H. regalis
Parahesperornis
0
20
40
60
80
100
120
140
160
180
0 10 20 30 40 50
Medio-lateral Width of Proximal End
(mm)
Length of Tarsometatarsus (mm)
P. tankei
B. advenus
B. varneri
H. bairdi
H. chowi
H. gracilis
H. mengeli
H. regalis
H. rossicus
P. alexi
347

Table 1. 611
A B C D E
YPM 1465 0/5 0/1 0/2 0/2 0/6
FMNH 395 - - - 1/2 6/6
KUVP 2290 - - - 1/2 3/6
FHSM 6318 - - - 0/2 1/6
AMNH 5101 - 1/1 0/2 - 2/6
UNSM 20030 - 1/1 1/2 - -
Pasquiaornis hardiei 2/5 1/1 1/2 1/2 5/6
Pasquiaornis tankei 2/5 1/1 2/2 1/2 5/6
Enaliornis 2/5 1/1 2/2 1/2 5/6
612
348

Chapter 5. Description and ecologic analysis of a Late Cretaceous bird 1
from the Gobi Desert (Mongolia). 2
3
INTRODUCTION 4
Despite the abundance of nonavian dinosaur fossils that have been discovered from the Late 5
Cretaceous of the Mongolian Gobi Desert (reviewed in Novacek, 1996; Benton et al., 2000), those 6
belonging to birds have remained elusive. While diverse Mesozoic avifaunas from coastal environments 7
have been discovered in several localities around the globe (Sanz et al., 1988; Zhou et al., 2003), very few 8
continental deposits with bird fossils are known. Of these sites, the Gobi is becoming the most productive in 9
terms of avian diversity, with a wide range of taxa representing multiple stages of avian evolution 10
(Kurochkin et al., 2002; Clarke and Norell, 2004; Chiappe et al., 2007). However, just a handful of fossils 11
comprise this diversity. Therefore the discovery of additional avian taxa from the Late Cretaceous Gobi 12
Desert is of significance to our understanding of early avian evolution and in particular of terrestrial avian 13
diversity during the Mesozoic. 14
This paper presents a new taxon of primitive ornithuromorph bird from the Late Cretaceous Barun 15
Goyot Formation at Khermeen Tsav in the southern Mongolia Gobi Desert (Fig. 1). The specimens 16
presented here were collected by a joint expedition of the Hayashibara Museum of Natural Science and the 17
Mongolian Academy of Science-Mongolian Paleontological Center in 1997 and are part of the collections of 18
the Mongolian Paleontological Center (MPC) in Ulaanbaatar. The specimens described here were found 19
closely associated with the bones of at least one additional bird as well as a small non-avian theropod 20
dinosaur. The specimens were allied with each other as those belonging to a new taxon—and most likely to 21
a single individual—on the basis of element size and morphology. 22
23
24
349

Geographic and Stratigraphic Setting 25
The Late Cretaceous of the Mongolian Gobi Desert contains three main fossil-bearing stratigraphic 26
units: the Nemegt Formation and the older Djadohkta and Barun Goyot formations (Dingus et al., 2008; 27
Dashzeveg et al., 2005). The Barun Goyot is very similar to the better-studied Djadohkta Formation in 28
terms of lithology and faunal assemblage. While it is not possible to determine the exact stratigraphic 29
relationships or correlations between these formations at this time, the Barun Goyot has been considered as 30
“Djadohkta-like” (Gao and Norell, 2000) and is commonly dealt with as a close contemporary of the 31
Djadohkta (Clarke and Norell, 2004), and so these formations will be dealt with synonymously here. Both 32
are characterized by thick, cross-bedded red sandstone units with some fluvial sediments, indicating a semi- 33
arid environment with intermittent ponds and streams in the interdunes (Gao and Norell, 2000). Extensive 34
lithological analysis of the cross-bedded and unbedded sandstones of the Djadohkta Formation supports an 35
eolian origin for the cross-stratified units (Fastovsky et al., 1997); however analysis of the unbedded 36
sandstones indicates a different depositional environment. The unbedded units are characterized by caliche 37
deposits, mud lenses, rhizoliths, and mud-filled Ophiomorpha burrows which imply they represent sandslide 38
deposits generated by periodic heavy rain events on stabilized dune slopes (Loope et al., 1999; Dashzeveg et 39
al., 2005). The Barun Goyot and Djadohkta formations are laterally extensive. Fossiliferous outcrops of the 40
Djadohkta occur at the Mongolian sites Bayn Dzak and Tögrögiin Shiree (also known as Tugrugin Shireh), 41
approximately 150 km east of the Barun Goyot outcrops at Khermeen Tsav, as well as at Bayan Mandahu in 42
northern China (Jerzykiewicz et al., 1993; Dashzeveg et al., 2005). While the age of the Barun Goyot 43
Formation has yet to be determined, biostratigraphy and paleomagnetostratigraphy correlate the Djadohkta 44
Formation at Bayn Dzak to a series of short normal and reversed intervals in the late Campanian, between 45
71 and 75 mya (Dashzeveg et al., 2005). 46
47
48
350

SYSTEMATIC PALEONTOLOGY 49
Aves Linnaeus, 1758 50
Ornithurae Haeckel, 1986 51
Hollanda luceria, new taxon. 52
Derivation of name. “Hollanda” is in honor of the Holland family (Janice, Charles, Carl, and J.-P.), whose 53
generous donations have supported a great deal of paleontological research and field work at the Dinosaur 54
Institute of the Natural History Museum of Los Angeles County. The species name was inspired by the band 55
Lucero of Memphis, Tennessee, and comes from the Latin “to shine.” 56
Holotype. MPC-b100/202, a distal right tibiotarsus preserved in articulation with the complete right 57
tarsometatarsus, the proximal portion of the first phalanx of digit II, the complete first and second, and the 58
proximal third phalanges of digit III (Fig. 2). 59
Type locality & stratigraphic horizon. Khermeen Tsav, southern Gobi Desert, Mongolia. Latitude 60
43°29’36” N, Longitude 99°49’41” E., Middle Red Bed (sensu Gradzinski et al., 1977), Barun Goyot 61
Formation. 62
Referred Specimens. On the basis of similarities in preservation, morphology and size, five additional 63
specimens found in association with the holotype are also referred to Hollanda luceria: a left tibiotarsus 64
missing its distal end [MPC-b100/204], a right proximal tibiotarsus [MPC-b100/205] (Fig. 3), a left fibula 65
[MPC-b100/206], a right fibula [MPC-b100/207], and a left distal femur [MPC-b100/203] (Fig. 4). The 66
shaft of the left tibiotarsus [MPC-b100/204] can be aligned with the distal right tibiotarsus of the holotype 67
because a small tuberosity (see Description) is present on both and allows the reconstruction of the original 68
tibiotarsal length. These elements were found associated with the holotype material, and so may belong to 69
the same individual. 70
Description of Holotype. 71
Tibiotarsus. The tibia, astragalus, and calcaneum are fully fused, although the outline of the 72
ascending process of the astragalus is visible (Fig. 2). Distally, the medial condyle is larger than the lateral. 73
351

In caudal view the medial and lateral crests of the tibial cartilaginous trochlea extend onto the caudal surface 74
of the bone; however the medial crest is largely weathered. In cranial view, just proximal to the condyles, 75
there is a shallow fossa flanked by well-developed medial and lateral crests. On the proximal margin of this 76
fossa a small, elongated tuberosity is present in the center of the shaft (Fig. 2a). This may represent the 77
Tuberositas proximalis retinaculi mm. extensorum, which serves as the skeletal attachment for the 78
retinaculum extensorum tibiotarsi, a transverse fibrous ligament which secures the tendon of the m. tibialis 79
anticus before it passes over the ankle joint. In medial view the sulcus for the m. fibularis is present, 80
however the medial epicondylar depression is only weakly developed. The lateral epicondylar depression is 81
more strongly developed. 82
Tarsometatarsus. The right tarsometatarsus is thin and elongate, curving slightly laterally (Fig. 2). 83
The mediolateral widths of the proximal and distal ends of the tarsometatarsus are similar, with only a slight 84
narrowing of the element along the shaft. The proximal and distal ends of the tarsometatarsus are not 85
oriented in the same mediolateral plane. Rather, the mediolateral axis of the proximal end is rotated 30° 86
about the longitudinal axis of the bone from the mediolateral axis of the distal end. Metatarsals I – IV are 87
preserved proximally and distally, however most of the shaft of metatarsal IV is absent. 88
The proximal articulation is formed by the fully fused ends of metatarsals II, III, and IV aligned 89
mediolaterally. Some weathering of the cotylae has occurred on the plantar proximal surface; however it is 90
evident that a hypotarsus was developed to some degree. The medial cotyla is slightly larger and more 91
circular than the lateral cotyla, which slopes slightly dorsally. Both cotylae are strongly concave. The lateral 92
margin of the lateral cotyla extends further proximally than the remainder of the articular surface, causing 93
the lateral cotyla to be more excavated than the medial in dorsal view. The cotylae are separated by a 94
distinct intercotylar eminence which is in line with the medial margin of metatarsal III in dorsal view. 95
Metatarsal III is the longest of the metatarsals. The shaft is fairly uniform in width and straight along 96
its length, narrowing slightly from the proximal end and flaring slightly at the trochlea. The shaft of 97
352

metatarsal III is ovate in cross-section and tightly sutured to the shaft of metatarsal II. Distally, the medial 98
and lateral trochlear ridges of metatarsal III are well defined and subequal in size. In distal view, the dorsal 99
surface of this trochlea is wider than the plantar surface. On both the medial and lateral faces of the trochlea 100
a fovea is developed. 101
Metatarsal II is narrower and extends less far distally than metatarsal III. In dorsal view, metatarsal 102
II narrows gradually along the shaft to the distal end, where the trochlea then widens to approximately the 103
same width as at the proximal end of the shaft. Computed tomography (CT) scanning revealed no evidence 104
of a proximal vascular foramina; however a very deep proximal indentation is present in cranial view 105
between metatarsals II and III (Fig. 2). Proximally, a poorly developed tubercle for the attachment of the m. 106
tibialis cranialis is present on the lateral margin of metatarsal II. In plantar view, metatarsal II narrows to a 107
very thin crest, the cristae plantaris medialis of Baumel and Witmer (1993), that runs along most of the 108
midshaft of the metatarsal before this bone flattens and widens out to its original proportions just above the 109
trochlea. This ridge is most striking in medial view, where it widens significantly from the proximal and 110
distal portions of the metatarsal shaft. The trochlea of metatarsal II diverges approximately 14.3° medially 111
from that of metatarsal III. A well-developed lateral fovea is present on the trochlea, while the medial one is 112
significantly shallower. The plantar surface of the trochlea slopes sharply laterally, however further 113
observations of this trochlea are obscured by the distal portion of the phalange, which is still in articulation 114
with the metatarsal and has been somewhat crushed onto its distal surface. 115
The proximal and distal ends of metatarsal IV are well preserved, however a majority of the shaft is 116
missing. In lateral view the proximal end of metatarsal IV is rounded and almost as broad dorsoplantarly as 117
the other metatarsals, however it narrows very quickly, becoming extremely reduced as compared to 118
metatarsals II and III (Fig. 2d). Distally, metatarsal IV widens into a bulbous trochlea that is tightly sutured 119
to the distal shaft of metatarsal III—the former ends further proximal to the latter, making metatarsal IV 120
353

much shorter than both metatarsal III and II. In distal view, the trochlea of metatarsal IV has a depression in 121
its plantar surface and is displaced plantarly from the trochlea of metatarsal III. 122
Metatarsal I is preserved on the medial margin of metatarsal II slightly displaced and plantarly 123
rotated from its original position. In medial view, metatarsal I is j-shaped and widens as it curves plantarly 124
and slightly laterally. This results in the articular surface of metatarsal I facing plantarly, indicating an 125
anisodactyl foot with a rotated hallux. The medial surface of the trochlea of this metatarsal is excavated by a 126
deep fovea. 127
Phalanges. The three distal phalanges of the third digit were preserved in articulation. Phalanges I 128
and II are complete, while the trochlea of phalange III is broken off (Fig. 2). However, the preserved portion 129
of this phalange shows the widening at the base of the trochlea, allowing the full length of the phalange to 130
be reconstructed. Phalange I is the longest, while phalange II and III are progressively shorter. The fovea on 131
the medial and lateral margins of the trochlea are deeply excavated in both phalange II and III. 132
Description of Referred Specimens. 133
Tibiotarsus. The tibiotarsus (MPC-b100/204) is long, slender, and concave medially. The proximal 134
articular surface is irregular in outline and caudally inclined. While the surface is mostly flat, the lateral 135
articular facet forms a distinctive, conical peak (Fig. 3b,d). The medial articular facet is oval in shape and 136
larger than the lateral, which is more circular. In proximal view the cranial margins of the articular facets are 137
not clearly defined; however they are separated from the portion corresponding to the cnemial crests by 138
notches, the lateral of which is very prominent (Fig. 3e). The caudal margin of the proximal articular 139
surface, between the lateral and medial articular facets is also marked by a notch. 140
Just distal to the articular surface, two distinctive cnemial crests are present on the proximal portion 141
of the craniolateral side of the shaft. The lateral crest is more prominent than the cranial. Proximolaterally 142
the lateral crest ends in an arched, lateroproximal-oriented surface. The cranial crest is straighter and 143
extends further distally than the lateral. Distal to the articular surface in caudal view, the articular facets are 144
354

strongly excavated by a fossa which is deepest at the apex of the notch separating the medial and lateral 145
articular facets. 146
The proximal shaft has a roughly circular cross section which flattens distally to a subovate cross 147
section. The fibular crest is located along the lateral face of the proximal third of the shaft. Distally, the 148
outline of the ascending process is visible on the distal fourth of the cranial surface of the shaft (Fig. 3d). 149
The ascending process is roughly triangular in shape and very slightly raised from the cranial surface of the 150
shaft. A small tuberosity is present on the lateral margin of the distal shaft (Fig. 3c). This tuberosity is 151
present on both the left and right tibiotarsal fragments, and thus allows reconstruction of the tibiotarsus in its 152
entirety. 153
Fibulae. Both the left [MPC-b100/206] and right [MPC-b100/207] fibulae are preserved. While the 154
distal-most ends are not preserved, the highly tapered nature of the fibulae and the lack of an articular scar 155
on the tibiotarsus indicate that the majority of the length of the fibula is preserved. The preserved left and 156
right fibulae are nearly identical in terms of size, proportions, and features. The fibula is compressed 157
lateromedially, with the proximal-most portion of the bone showing a higher degree of compression than the 158
more distal portions. The medial face of the fibula is concave proximally. The depressed area of the medial 159
face is bounded on the proximal margin by a ridge which forms the articular surface of the fibula. The 160
lateral face of the fibula is more flat and a proximal ridge is not developed as on the medial face. Distally, 161
the fibula tapers to a very thin shaft with an oval cross section. Along the proximal portion of the caudal 162
margin of the shaft the tubercle for the m. tibiofibularis is present and oriented caudolaterally. 163
Femur. The distal two-thirds of the femoral shaft and the distal articulation of the right femur are 164
preserved (MPC-b100/203, Fig. 4). The preserved shaft is roughly triangular in cross section and tilts 165
medially from the distal articulation to the proximal-most preserved end. At the distal articulation, the 166
lateral condyle extends much further distally than the medial condyle when the longitudinal axis of the shaft 167
is held in the vertical plane, however if the element is oriented such that the transverse axis of the distal end 168
355

is in the horizontal plane the distal extent of the condyles is equal. The medial and lateral condyles extend 169
onto the cranial surface of the shaft as narrow crests which define the margins of the patellar sulcus. A 170
distinct, round impression for the m. tibialis cranialis is present on the edge of the lateral crest which 171
extends from the lateral condyle and defines the patellar sulcus. In caudal view, the popliteal fossa is deep 172
and distally well-demarcated by the bridge connecting the condyles (Fig. 4d). This bridge originates on the 173
medial margin of the lateral condyle and widens before terminating against the lateral margin of the medial 174
condyle. 175
The tibiofibular crest of the femur is well developed with a distinctive fibular condyle facing 176
caudally. In distal view, medial to the fibular condyle the articular surface possesses a deep, round 177
depression bordered by the lateral condyle and the bridge connecting the condyles. This depression may 178
allow for the articulation of the high peak of the lateral articular surface of the tibiotarsus (see below). 179
Phylogenetic Relationships of Hollanda luceria 180
In order to investigate the evolutionary relationships of Hollanda luceria, a cladistic analysis of 242 cranial 181
and postcranial characters—based on the analysis of Gao et al. (2008)—was performed for a dataset of H. 182
luceria and 17 other taxa. Character scoring of H. luceria was taken from observations of the holotype and 183
referred specimens. The analysis was performed with 100 random addition sequence replications, each 184
followed by tree bisection-reconnection branch swapping (TBR) implemented in TNT (Goloboff et al., 185
2008). 186
Four most parsimonious trees of 413 steps were generated by the cladistic analysis, hypothesizing 187
two alternative placements for H. luceria (Fig. 5). Both alternatives agree in the placement of H. luceria 188
within the Ornithuromorpha and more derived than Patagopteryx. Hollanda is placed within the 189
Ornithuromorpha on the basis of two derived characters of the tarsometatarsus, the development of a 190
hypotarsal-like structure and the enclosure of the distal vascular foramen by metatarsals III and IV (Fig. 2). 191
The competing hypotheses for the relationships of H. luceria within the Ornithuromorpha result from the 192
356

fragmentary nature of the specimens. Missing data on the sacral vertebrae, humerus, and proximal femur 193
precludes inclusion of H. luceria in the clade containing the common ancestor of Yixianornis + more 194
derived birds (Fig. 5). Alternatively, the well-developed intercotylar eminence on the proximal 195
tarsometatarsus of H. luceria supports its placement as the sister taxon to Hesperornis regalis + modern 196
birds (Neornithes). Other variations in the four most parsimonious trees obtained from the cladistic analysis 197
include the relationship of the enantiornithines Eoenantiornis buhleri and Longipteryx chaoyangensis and so 198
will not be further considered here. The addition of Hollanda luceria to the character matrix used here does 199
not greatly alter the topology of the tree presented from a previous analysis of this character matrix (Gao et 200
al., 2008). 201
Histological Description 202
A histological section was made from the fractured proximal right tibiotarsus [MPC-b100/205]. A 203
diaphyseal transverse plane section was made approximately 43 mm from the proximal end of the element. 204
The slide was examined using polarized microscopy and the histological patterning was described based on 205
the criteria of Francillon-Vieilliot et al. (1990). The majority of the cortical matrix is composed of 206
longitudinally vascularized, parallel-fibered bone (Fig. 6a). However, the inner 1/3
rd
of the cortex shows 207
woven bone matrix with centripetally infilled (with lamellar bone) circumferentially and longitudinally 208
oriented primary vascular canals to form a fibro-lamellar complex. The thickest regions of the cortex shows 209
a blend of parallel fibered and woven bone with both longitudinal and reticular primary vascular canals in 210
the inner 2/3
rd
of the bone (Fig. 6b). These primary vascular canals are infilled with centripetally deposited 211
parallel-fibered and occasionally lamellar bone (i.e. there is local expression of fibro-lamellar bone). The 212
outer 1/3rd of such regions are composed solely of parallel-fibered bone with longitudinal vascularization 213
that diminishes towards the periosteal surface. A single line of arrested growth (LAG) is found in the outer 214
cortex and an annulus occurs just short of the periosteal surface. The bone is avascular peripheral to the 215
357

LAG. The endosteal border of the cortex is highly scalloped from osteoclastic erosion associated with 216
medullar cavity expansion. 217
The absence of endosteal bone and tightly packed peripheral growth lines (i.e. External Fundamental 218
System) and Haversian remodeling points to MPC-b100/205 as still experiencing growth at the time of 219
death. The development shifts to parallel-fibered and avascular matrices just prior to death suggest that the 220
animal’s growth was slowing when it met its demise. From these considerations we tentatively conclude that 221
the animal was in the transition to the stationary phase of development. 222
Phylogenetically the histology of Hollanda is consistent with a taxon more crownward than 223
Enantiornithes but earlier diverging than Ornithurae (Hesperornithiformes + Neornithes and all its 224
descendents). This conclusion is supported by the presence of fibro-lamellar bone and fewer LAGs than 225
seen among Enantiornithines, with an overall similarity to the bone microstructure of the basal 226
ornithuromorph Patatgopteryx (Chinsamy et al., 1995). The presence of a LAG and an annulus in H. luceria 227
supports its more basal phylogenetic placement than Ornithurae, which typically display uninterrupted 228
growth and lack LAGs (Houde, 1987; Chinsamy, 2005). 229
230
DISCUSSION 231
Ecological Considerations 232
A number of recent studies have investigated the correlation between various hindlimb 233
measurements of modern birds and the ecospace inhabited by these birds (Hopson, 2001; Zeffer et al., 2003; 234
Habib and Ruff, 2008). Previous studies have demonstrated the ability to distinguish terrestrial and arboreal 235
birds using the proportions of the phalanges of the third digit (Hopson, 2001) and the use of mass- 236
independent trends in element lengths to sort birds into more specialized ecospace bins (Zeffer et al., 2003). 237
Despite the incomplete nature of the skeletal material of Hollanda luceria, the preservation of the 238
majority of the hindlimb enables the evaluation of hypotheses regarding the ecology of this bird. The 239
358

elongation of H. luceria’s slender tarsometatarsus and tibiotarsus resembles specializations seen in modern 240
wading birds, such as storks or flamingos, as well as modern ground-dwelling runners, such as the Secretary 241
bird. These qualitative observations were quantitatively evaluated using a morphometric analysis comparing 242
H. luceria to modern birds of known ecology. The study presented here focuses on the use of the pedal 243
phalanges as correlates to ecospace because the femur of H. luceria was incomplete. As phalanges 1 and 2 244
of digit III were complete, and phalanx 3 was lacking only the distal-most end, these elements were chosen 245
as more appropriate. 246
The database of relative phalangeal lengths of digit III presented here is enlarged from that of 247
Hopson (2001) in terms of both the number of taxa included as well as the specificity of habitats considered. 248
Hopson’s (2001) database consisted of 191 Holocene birds, which was expanded to 350 in this study. 249
Additionally, while Hopson (2001) classed taxa as either terrestrial or arboreal, our study sought to identify 250
the habit of birds more precisely. In the present study birds were classified into one of the following 251
ecologic bins on the basis of the primary feeding mechanism of the birds: ground foragers, arboreal 252
foragers, and on-the-wing hunters. Classification was made on the basis of the ecologic description provided 253
in del Hoyo et al. (2007). A more specific analysis of the ground foraging birds was undertaken which 254
further divided this group on the basis of secondary ecologies, such as nesting habitat, into one of the 255
following bins: obligatory ground nesting (flightless), preferential ground nesting, waders, or arboreal 256
nesting. The preliminary analyses presented here were carried out by plotting the proportion of the length of 257
digit III (not including the claw) of each of the phalanges on ternary diagrams. 258
The results of the morphometric analysis conducted here are shown in Figure 7. Figure 7 shows taxa 259
divided into three basic feeding strategies, while the inset shows only the terrestrial feeders further divided 260
by secondary ecology. As seen in Figure 7, taxa included in this analysis separate well in morphospace at 261
the level of ground vs. arboreal foraging strategies, however distinctions between arboreal foragers and on- 262
the-wing hunters are not reflected in the data. Within the ground foragers, a small subset overlap with 263
359

arboreal foragers (indicated by the dashed circle in Fig. 7). However, these overlapping ground foragers are 264
those that fall into the subcategory of arboreal nesters, indicting they utilize flight for nesting. The overlap 265
of birds which all utilize flight may indicate a uniform selective pressure on phalangeal length in flying 266
birds, regardless of the feeding strategy. In looking at the subdivisions of ground foragers, the bins 267
designated in this analysis are fairly well-categorized by phalangeal proportions, with obligatory ground 268
nesting birds falling at the highest end of the data, and preferential ground nesters occupying a wide scatter 269
clustered around the low end of the data (Fig. 7). Hollanda luceria plots well within the ground foragers in 270
this data set, plotting most closely to the preferential ground nesters. In particular, H. luceria plots most 271
closely to the Chucao (Scelorchilus rubecula) and Roadrunner (Geococcyx californianus) as well as the 272
Wild Turkey (Meleagris gallopavo) and Mountain Caracara (Polyborus plancus). These are birds that are 273
capable of flight to a varying degree, however nest on the ground, where they actively pursue prey. While 274
the Chucao is an insectivore, the Roadrunner and Wild Turkey are opportunistic and will eat insects or kill 275
small animals. 276
The results of this analysis indicate that Hollanda luceria has similar limb proportions to modern 277
terrestrial hunters. This shared morphology may imply that Hollanda luceria also shared the ecology of 278
terrestrial hunters, somewhat like the modern Roadrunner. This conclusion is congruent with recent 279
interpretations of the Late Cretaceous Mongolian Gobi Desert as a stabilized dune field with periodic 280
flooding events given that such an environment would provide a range of habitats suitable to active cursorial 281
predators. The presence of this type of environment is further reflected in the abundance of cursorial non- 282
avian theropod dinosaurs, such as Avimimus, Velociraptor, and Oviraptor (Currie, 2000). 283
Avian Fauna of the Djadohkta/Barun Goyot Formation 284
While both ornithuromorphs and enantiornithines have been described from the Djadohkta/Barun 285
Goyot and Nemegt formations (Table 1), the three fossiliferous units of the Mongolian Gobi’s Late 286
Cretaceous, these formations do not share any less inclusive taxonomic levels, and so this review will focus 287
360

on the avifauna of which Hollanda luceria was a member (for a review of the Late Cretaceous avifauna of 288
the Nemegt Formation, see Clarke and Norell, 2004). 289
With the addition of Hollanda luceria, three avian taxa and several indeterminate avian fossils have 290
been described from the Djadohkta/Barun Goyot formations of the Gobi desert. Khermeen Tsav, the locality 291
at which Hollanda was discovered, has yielded the remains of enantiornithine eggs and embryos 292
(Elzanowski, 1981) as well as a specimen originally described as Nanantius valifanovi (Kurochkin, 1996) 293
but later reassigned to Gobipteryx minuta (Chiappe et al., 2001). Gobipteryx was first described from cranial 294
material found at Khulsan and initially allied with palaeognathous neornithine birds (Elzanowski, 1974). 295
Further research on new and existing specimens by Martin (1983) revealed synapomorphies allying 296
Gobipteryx with the Enantiornithes. Gobipteryx was about the size of a pheasant, with strong second and 297
third toes and a weak first toe (Kurochkin, 1996). These observations combined with the highly robust jaws 298
suitable for feeding on tough materials such as seeds or fruit in trees, led Kurochkin (1996) to interpret this 299
bird as a climber, however this conclusion is highly subjective and not quantitatively supported. 300
At Tögrögiin Shiree the Djadohkta Formation has yielded a partial post-cranial skeleton of a second 301
enantiornithine, the chicken-sized Elsornis keni (Chiappe et al., 2007). Elsornis represents a novel ecology 302
for the Late Cretaceous Gobi Desert, as it was most likely a very poor flier, or perhaps even flightless, as 303
indicated by the ratio of the length of the humerus to that of the ulna (Chiappe et al., 2007). 304
In addition to enantiornithines, a single ornithuromorph, the jacana-sized Apsaravis ukhaana, has 305
been described from the Djadohkta Formation at Ukhaa Tolgod (Norell and Clarke, 2001). Phylogenetic 306
analysis indicates Apsaravis forms a sister-group with Hesperornithiformes + Neornithes (Clarke and 307
Norell, 2004). Preservation of skeletal features related to the modern attachment of the extensor carpi 308
radialis muscle indicate that Apsaravis was potentially a highly capable flier, as this type of muscle 309
attachment is used to reposition the distal-most part of the wing during the transition from upstroke to 310
downstroke in modern birds (Norell and Clarke, 2001). 311
361

The review of the Djadohkta/Barun Goyot avifauna presented here reveals that while the quantity of 312
avian remains may be sparse, the ecological diversity represented by these remains is not. With the addition 313
of H. luceria, birds with flighted, ground-dwelling, and cursorial lifestyles are all represented. This variety 314
of lifestyles represents a complex avifauna with a well-developed partitioning of ecospace. When 315
considering the effects of competition, adaptations for niche specialization such as these would reduce 316
competitive pressures. The implications presented here for a diverse, albeit incomplete, avifauna also have 317
important ramifications for the environment and ecosystem of the Late Cretaceous Gobi as a whole. The 318
species-energy theory of ecosystem functioning predicts that biodiversity is a reflection of the amount of 319
energy available in an ecosystem, with greater energy resources allowing higher biodiversity and requiring 320
more ecosystem partitioning (Wright, 1983; Srivastava and Lawton, 1998). Studies of modern avifaunas 321
have provided support for this theory as well as illustrating the importance of habitat diversity to avian 322
diversity (Hurlbert, 2004). As deserts typically have lower habitat diversity, they also typically display 323
lower bird diversity (Hurlbert, 2004). Following these principles, it emerges that the previously proposed 324
environment of an active dune field (Eberth, 1993) may not provide the best explanation for the observed 325
ecological diversity of the Late Cretaceous Gobi Desert. The more recent interpretations of Loope et al. 326
(1999) and Dashzeveg et al. (2005) of an environment that altered over thousands of years from an active 327
dune field with some fluvial deposits to a stabilized dune field with periodic water flow may have provided 328
greater habitat complexity for ancient birds. The existence of cursorial birds, ground-dwelling birds, and 329
multiple species of flighted birds in the Djadohkta/Barun Goyot formation is much better explained by these 330
more recent interpretations of a complex environment. 331
332
CONCLUSIONS 333
Despite the rich non-avian dinosaur fauna that has been described from Late Cretaceous rocks 334
throughout the Mongolian Gobi Desert, avian fossils have remained elusive. The inland setting of the 335
362

modern Gobi during the Late Cretaceous provides a window into a type of avifauna much more poorly 336
known in the Mesozoic record than the near shore avifaunas, making fossils from this area even more 337
significant. The new specimens presented here document the existence of a new basal lineage of 338
ornithuromorph birds and increase the avian as well as the ecological diversity of the Late Cretaceous 339
Djadohkta/Barun Goyot formations. 340
Hollanda luceria is only the second ornithuromorph to be described from the Djadohkta/Barun 341
Goyot formations. As such, the ecological specializations for a cursorial lifestyle implied by the 342
morphometric analysis are significant; as they support that during the Late Cretaceous basal 343
ornithuromorphs occupied a diverse ecospace alongside enantiornithines. 344
345
363

APPENDIX 1 346
Coding of applicable characters from Gao et al., 2008 in the phylogenetic analysis. While the Gao et al 347
(2008) matrix consists of 242 cranial and post-cranial characters, the fragmentary nature of H. luceria 348
resulted in “?” codings for all but the pelvic limb characters, and so these unknown characters have not been 349
included here. Codings are provided for Hollanda luceria. Coding of the other taxa is unchanged from Gao 350
et al., 2008. 351
character: 200 201 202 203 204 205 206 207
Hollanda luceria ? ? ? ? 1 2 1 ?1

character: 208 209 210 211 212 213 214 215
Hollanda luceria 0 ?1 0 1 0 0 0 0

character: 216 217 218 219 220 221 222 223
Hollanda luceria 0 0 0 0 1 0 0 ?

character: 224 225 226 227 228 229 230 231
Hollanda luceria 0 0 1 37623 1 1 1 2

character: 232 233 234 235 236 237 238
Hollanda luceria 0 0 3 0 0 0 ?
352
353
364

APPENDIX II 354
Synapomorphies mapped onto the two most parsimonius trees (A and B) returned by the cladistic analysis 355
of Hollanda luceria and 17 other taxa. Nodes labelled with numbers represent synapomorphies present in 356
both trees, while nodes labelled with a letter and number represent synapomorphies unique to that tree. 357
365

358
366

359
FIGURE CAPTIONS 360
Figure 1. Late Cretaceous localities of the Mongolian Gobi Desert. Numbers indicate locality names as 361
follows: 1. Khermeen Tsav, 2. Ukhaa Tolgod, 3. Khulsan, 4. Togrogiin Shiree, 5. Bayn Dzak (Flaming 362
Cliffs), 6. Tsaagan Khushu, 7. Gurilyn Tsav, 8. Burgeen Tsav. 363
364
Figure 2. Specimens identified as the holotype [MPC-b100/202] of Hollanda luceria. Right distal 365
tibiotarsus, tarsometatarsus, and digit III in A. plantar, B. medial, C. dorsal, and D. lateral views. Also 366
shown are the tibiotarsus in distal view (E), and the proximal (F) and distal (G) views of the tarsometatarsus. 367
Insets from tarsometatarsus (C) show computed tomography (CT) cross-sections of the bone along the line 368
indicated. Abbreviations are as follows: I – metatarsal I, II – metatarsal II, III – metatarsal III, IV – 369
metatarsal IV, III.1 – Phalanx 1, Digit III, III.2 – Phalanx 2, Digit III, III.3 – Phalanx 3, Digit III, df – distal 370
foramen, pf – proximal foramen, lc – lateral condyle, lct – lateral cotyla, mc – medial condyle, mct – medial 371
cotyla, t1 – tubercule 1, t2 – tubercule 2, p – phalanx. 372
373
Figure 3. Left (upper, MPC-b100/204) and right (lower, MPC-b100/205) referred tibiotarsi of Hollanda 374
luceria in A. caudal, B. lateral, C. cranio-medial, D. medio-cranial, and E. proximal views. Inset (F) shows 375
reconstructed tibiotarsal length determined by aligning tubercules on proximal and distal fragments. The 376
asterisk in (A) indicates the approximate location of the section taken for histological analysis. 377
Abbreviations are as follows: ap – ascending process of astragalus, cc – cranial cnemial crest, laf – lateral 378
articular facet, lc – lateral cnemial crest, maf – medial articular facet, t – tubercule. 379
380
367

Figure 4. Left referred femur [[MPC-b100/203] of Hollanda luceria in A. medial, B. cranial, C. lateral, D. 381
caudal, and E. distal views. Abbreviations are as follows: lc – lateral condyle, mc – medial condyle, pf – 382
popliteal fossa. 383
384
Figure 5. Two most parsimonious trees returned by TNT (Goloboff et al., 2008) from an analysis of 242 385
characters for 18 taxa. The clade containing Concornis, Longipteryx, and Eoenantiornis has been collapsed 386
due to ambiguities in the returned trees. See discussion for the methods of the cladistic analysis. 387
388
Figure 6. A) Predominant tibiotarsal cortical histology for Hollanda viewed with polarized light 389
microscopy. Note: the scalloped endosteal border (lower left), woven-fibered circumferentially vascularized 390
inner cortex, and parallel-fibered longitudinally vascularized outer cortex. The outermost cortex of the bone 391
is not present. It delaminated at the LAG interface. B) Hollanda tibiotarsal cortical histology showing 392
localized fibro-lamellar bone and a line of arrested growth. Note: The woven-fibered nature of the inner 393
most cortices with a mix of reticular and longitudinally vascularized canals, and a pattern of reduced 394
vascularization and switch to parallel-fibered bone towards the periosteal surface of the bone. 395
396
Figure 7. Morphometric analysis correlating phalangeal length of digit three with primary foraging ecology, 397
expanded from Hopson (2001). The new taxon presented here, Hollanda luceria, is represented by a star, 398
while the dashed circle surrounds birds which are classified as terrestrial foragers but which overlap in 399
morphospace with the arboreal birds. See text for further discussion. Inset: Morphometric analysis 400
correlating phalangeal length of digit three with secondary ecologies for the ground foragers shown in the 401
ternary diagram. 402
403
368

TABLE CAPTIONS 404
Table 1. Review of the known avifauna of the Nemegt and Djadohkta/Barun Goyot avifauna from the Gobi 405
Desert of Mongolia with implications for ecological adaptations. Adapted from Clarke and Norell, 2004. 406
References are as follows: 1 – Elzanowski, 1981; 2 – Nessov and Borkin, 1983; 3 – Kurochkin, 1988; 4 – 407
Kurochkin, 1999; 5 – Kurochkin, 2000; 6 – Kurochkin et al., 2002; 7 – Chiappe et al., 2001; 8 – Clarke and 408
Norrell, 2004; 9 – Norrell, 2004; 10 – Elzanowski, 1974; 11 – Elzanowski, 1977; 12 – Clarke and Norrell, 409
2002. 410
411
369

Figure 1. 412
413
414
370

Figure 2. 415
416
371

Figure 3. 417
418
Figure 4. 419
420
372

Figure 5. 421
422
Figure 6. 423
424
373

Figure 7. 425
426
427
374

Table 1
Group Taxonomic
Assignment
Formation Ecological
Interpretation
Evidence for Interpretation Ref
Avialae incertae
sedis
unassigned Barun Goyot insufficient material (fragmentary)

1,5,8
Enantiornithines Enantiornithines
incertae sedis
Barun Goyot insufficient material (embryonic) 1,8
Enantiornithines Elsornis keni Djadohkta flightless / poor flyer high brachial index 7
Enantiornithines Gobipteryx minuta Barun Goyot
& Djadohkta
arboreal, possibly
climber
Robust front toes, reduced back toe 7,
8,10,
11
Enantiornithines Gurilynia nesovi Nemegt Insufficient material (fragmentary) 4,8
Ornithurae Ornithurae incertae
sedis
Nemegt insufficient material (fragmentary) 8
Ornithurae Apsaravis ukhanna Djadohkta modern aerial
(flighted) ecology
presence of extensor process on metacarpal I 8,12
Ornithurae Hollanda luceria Barun Goyot cursorial predator relative lengths of pedal phalanges this
paper
Ornithurae
(Hesperornithes)
Hesperornithes
incertae sedis
Nemegt aquatic diving similarity to hesperornithiforms 5
Ornithurae
(Hesperornithes)
Judinornis
nogontsavensis
Nemegt aquatic diving similarity to hesperornithiforms 2, 8
Ornithurae unassigned Nemegt aquatic similarity to charadriiforms, and
presbyornithid anseriforms
5,8
Ornithurae Teviornis gobiensis Nemegt aquatic similarity to presbyornithid anseriforms 6,8

375

Chapter 6. Statistical approach for inferring ecology of 1
Mesozoic birds. 2
3
INTRODUCTION 4
A cornerstone of modern ecology is the identification of niches and trophic guilds in an 5
ecosystem and the organisms which belong to them. The identification of niche specialization 6
allows comparisons of ecosystems with similar environments but disparate animal communities 7
as well as the identification of different roles closely related species may play in their ecosystem 8
(Hurlburt, 2004). Within paleontology, the association of a fossil animal with a particular 9
ecologic niche requires the identification of morphological features uniquely specialized for the 10
occupation of that niche. When dealing solely with the fossil record, often it is difficult or 11
impossible to identify appropriate features. However, by using modern analogues or close 12
relatives of the fossil taxa in question, identifying these features may become easier, thus 13
enhancing our ability to make plausible ecological correlations. While it may not always be 14
possible to correlate form with function, in many cases such correlation has been shown to be an 15
important tool for inferring ecological attributes in birds (Fielder, 2005; Keast & Saunders, 16
2008). 17
Since the 1990’s the abundance of early bird fossils has increased dramatically, such that 18
our understanding of avian evolution during the Mesozoic is much improved (Chiappe, 2007; 19
Zhou, 2004; Zhou and Zhang, 2007). These discoveries have documented an enormous 20
morphological diversity that likely reflects an equally large ecological diversity. Determining 21
what ecological roles were played by these novel avians, however, has often been problematic. A 22
number of studies have extrapolated ecological roles for Mesozoic birds based upon various 23
morphological features; early papers most commonly sought to better understand the lifestyle of 24
376

Archaeopteryx (Peters & Gorgner, 1992; Glen & Bennett, 2007). Additionally, a number of 25
studies have categorized many other early birds (e.g., Kurochkin, 1996; Clarke & Norell, 2004; 26
Hou et al., 2004; Chiappe et al., 2007) within specific ecological niches or trophic guilds on the 27
basis of qualitative observations. However, using modern birds, paleoecological studies have the 28
potential to contain large datasets of morphological measurements that are correlated with known 29
ecological behaviors. This quantitative approach for inferring ecological attributes in Mesozoic 30
birds can supplement, and perhaps prove more reliable than, inferences based on particular 31
isolated morphologies. 32
A number of recent studies have investigated the correlation between various hindlimb 33
measurements and the ecology of modern birds (Hopson, 2001; Zeffer et al., 2003a; Hertel & 34
Campbell, 2007; Tickle et al., 2007). These studies have explored the use of a variety of skeletal 35
measurements and statistical techniques to identify a range of ecological habits within modern 36
birds. For example, Hopson was able to distinguish terrestrial and arboreal birds using the 37
proportions of the phalanges of the third digit (Hopson, 2001). Zeffer and colleagues used mass- 38
independent trends in hindlimb element lengths to sort birds into specialized ecospace bins 39
(Zeffer et al., 2003a). By developing a comprehensive database of modern birds which includes 40
skeletal measurements and observed ecological niches, trends in these data may be identified and 41
applied to fossil taxa which share morphological similarities. 42
There have been few studies carried out to date which utilize a modern avian dataset to 43
better understand early avian ecological diversification. Some recent studies have focused on 44
using morphometrics to characterize functional adaptations for different foraging strategies. The 45
first such studies compared the contribution of individual hindlimb (Gatesy & Middleton, 1997) 46
and forelimb (Middleton & Gatesy, 2000) element lengths to total limb length across modern 47
birds, Mesozoic birds, and non-avian theropod dinosaurs. Both of these studies showed that 48
377

modern and ancient birds occupy a much wider morphospace than non-avian theropods (Gatesy 49
& Middleton, 1997; Middleton & Gatesy, 2000), while also highlighting repeated trends in 50
element length proportions associated with flightlessness (Middleton & Gatesy, 2000). Most 51
recently, Dyke and Nudds used the comparative element proportions of the forelimb and 52
hindlimb separately to look at the distribution of enantiornithine birds in modern avian 53
morphospace (Dyke & Nudds, 2007). Researchers found that while the majority of 54
enantiornithine birds fell within modern morphospace, there were some exceptions in wing 55
proportions, indicating that some early birds may have utilized flight strategies not seen in 56
modern birds (Dyke & Nudds, 2007). While the focus of the research was to examine flight 57
strategies in Mesozoic birds (Dyke & Nudds, 2007), as opposed to ecological strategies, the 58
general approach is very similar to that presented here. A more similar study was presented by 59
Bell and colleagues in an attempt to characterize the ecology of the early ornithurine bird 60
Hollanda luceria (Bell et al., 2010). This study compared the proportional lengths of the pedal 61
phalanges of digit III of 250 modern birds with varying ecologies, a database expanded from that 62
of Hopson et al. (1991). 63
The goal of the research presented here is to investigate correlations between morphology 64
and ecology for a number of extinct avian taxa utilizing a broad modern morphopmetric 65
database. While previous studies have only explored morphospace for a single functional unit 66
(i.e., the forelimb or hindlimb separately) using non-statistical techniques, this study seeks to 67
include both the forelimb and hindlimb in a single statistical analysis in order to characterize taxa 68
as completely as possible. To fulfill this goal, two main objectives were met. First, correlations 69
between morphometrics and niche occupation of modern birds were established using a database 70
of 10 skeletal dimensions for nearly 500 modern bird species. The second objective was to match 71
fossil avian taxa with specific ecological niches using the morphological measurements and 72
378

ecological characterizations of the modern database. The hypotheses which were tested were as 73
follows - first, element proportions within the limbs correlate to ecological habits of modern 74
birds; second, when considered with modern birds, fossil taxa will share the modern 75
morphospace and fall into a specific ecological category based upon comparative element 76
lengths. 77
78
METHODS 79
Data Collection 80
Modern avian morphometrics. Measurements were collected from the skeletal 81
collections of the Ornithology Department at the Natural History Museum of Los Angeles 82
County. Length measurements were collected from the forelimb (humerus, ulna, radius, and 83
carpometacarpus) and hindlimb (femur, tibiotarsus, tarsometatarsus phalanges I-III of pedal digit 84
III, excluding the ungual phalanx). All measurements were taken from the same side of a single 85
individual for each species. Measurements were collected from 477 bird species representing 16 86
orders and 38 families. Species were selected for inclusion based on availability in the collection 87
with the intent of sampling as broad a range of modern avian diversity (both phylogenetic and 88
ecologic) as possible. While a number of previous studies have amassed similar datasets, 89
combining them for use in this study was not possible, as no previous study has presented data 90
for both the forelimb and hindlimb of a single individual. 91
Mesozoic avian morphometrics. Mesozoic taxa were added to the modern database 92
from measurements reported in the primary literature or gathered from photographs or casts. Due 93
to the incomplete nature of most avian fossils, very few specimens were able to provide all 94
measurements. Seventeen Mesozoic taxa from across the phylogenetic tree of pre-modern birds 95
were included in the final analysis: Archaeopteryx lithographica (BMNH 37001; JM 2257; 96
379

Wellnhofer, 2008), Cathayornis yandica (IVPP V9769; Zhou et al., 1992), Changchengornis 97
hengdaoziensis (GMV 2129b; Qiang et al., 1999), Confuciusornis sanctus (BM.Av.1168-1171, 98
GMV-2130, D2454; Hou et al., 1995), Eoconfuciusornis zhengi (IVPP V11977; Zhang et al., 99
2008), Hongshanornis longicresta (IVPP v 14533; Zhou and Zhang, 2005), Jeholornis prima 100
(IVPP v 13274; Zhou and Zhang, 2002), Longipteryx chaoyangensis (HKScM DG-002-1-1-1a; 101
Zhou and Zhong, 2002), Neuquenornis volans (MUCPv-142; Chiappe and Calvo, 1994), 102
Patagopteryx deferrisi (MACN-N-11; Chiappe et al., 1992), Pengornis houi (IVPP v 11844; 103
Zhou et al., 2008), Protopteryx fengingensis (IVPP v 15336; Zhang and Zhou, 2000), Sapeornis 104
chaoyangensis (IVPP V13276; Zhou and Zhang, 2002), Vescornis hebeiensis (NIGP 130722; 105
Zhang et al., 2004), Yanornis martini (IVPP V 13358; Zhou and Zhang, 2001), Yixianornis 106
grabaui (IVPP V13631; Clarke et al., 2006), Zhongornis haoae (DNHM D2455/6; Gao et al., 107
2008). Because Archaeopteryx and Confuciusornis are known from numerous specimens which 108
display a wide range of body size, a large and small representative of each species was chosen, 109
bringing the total number of Mesozoic birds examined in this study to nineteen. 110
Ecological Characterization 111
In order to characterize the ecology of the modern birds as completely as possible, 112
multiple ecological characters were used. First, birds were assigned to an ecological bin based on 113
foraging strategy. Since evolutionary pressures are also active in other aspects of ecology, 114
modern birds were also assigned to secondary categories based on reproductive strategies, 115
migration habits, and swimming strategies (where applicable). The primary bins used in this 116
study are the same as those used in Bell et al. (2010). Classification into all categories was made 117
on the basis of the ecologic description provided in del Hoyo et al. (2007). When necessary, the 118
primary literature was consulted for clarification on specific habits. Primary ecologic bins are as 119
follows: 120
380

1. Terrestrial foragers – birds which predominately forage for food on the ground. 121
2. Arboreal foragers – birds which predominately forage for food in trees. 122
3. On-the-wing hunters – birds which forage for food during flight, as well as birds which 123
utilize flight to capture prey. 124
4. Aquatic foragers – birds which forage for food associated with water, either fresh or 125
marine. These birds are subdivided into the following subcategories bins: 126
a. Foot-propelled divers – birds which carry out extended dives underwater using their 127
legs for propulsion. 128
b. Wing-propelled divers – birds which carry out extended dives underwater using their 129
wings for propulsion. 130
c. Plunge divers - birds which hunt by soaring over the water and then diving from the air 131
to capture prey within the upper few meters of the surface. 132
d. Surface swimmers – birds which primarily swim on the surface of the water while 133
foraging and only briefly (if at all) submerge. 134
e. Waders – birds which forage for food in shallow water by walking. 135
In addition, all birds were assigned to two secondary bins on the basis of their nesting and 136
migration strategies: 137
1. Nesting strategy - birds were classified as either: a) nesting on the ground, or b) nesting 138
in trees. 139
2. Migration strategy - birds were classified as either: a) non-migrants, or b) long-distance 140
migrants. Birds which may be considered regional, partial, or short-distance migrants were 141
grouped with the non-migratory birds, as these birds do not migrate in a fixed yearly 142
pattern. The definition of these terms coincides with their usage by the International Union 143
for the Conservation of Nature (IUCN). 144
381

Principle Component Analysis (PCA) 145
The data was used in a Principle Component Analysis (PCA) of the 10 log-adjusted variables, 146
carried out in PAST (Hammer & Harper, 2006). Following this analysis, principal components 1 147
and 2 were plotted against each other in order to compare the placement of different taxa within 148
the overall morphospace as defined by PC 1 and 2. 149
150
RESULTS 151
The results of the PCA can be found in Table 1 as well as Figure 1. The greatest variation was 152
expressed in PC 1 (87.48%), while over 90% of the variance was expressed by PC 1 and 2 153
combined (Tables 1 & 2). 154
Modern Avian Ecomorphology 155
When the taxa in the modern dataset are identified by their foraging strategy (primary 156
ecologic bin), some basic trends are observed in the resulting plot of PC 1 versus PC 2 (Fig. 1). 157
Arboreal birds cluster fairly closely together within the second quadrant, while terrestrial birds 158
group more loosely in the first and fourth quadrants. Aerial foragers cluster primarily in the third 159
quadrant but also extend into the second quadrant. Aquatic birds have no discernible segregation 160
to a particular region of morphospace and instead overlap with all other ecological bins. Figure 2 161
shows the aquatic birds subdivided into more specific ecologies (foot-propelled diving, wing- 162
propelled diving, surface swimming, and wading): wading birds are distributed only across the 163
morphospace occupied by terrestrial foragers, while the plunge and wing-propelled divers are 164
restricted to the area occupied by the aerial foragers. Given that both waders and traditional 165
terrestrial birds forage by walking (albeit with different substrates), these categories can be 166
effectively combined, thus moving wading birds from the aquatic foraging bin to the terrestrial 167
foraging bin. Also, both plunge divers and aerial foragers hunt by spotting prey from a soaring 168
382

flight, therefore these categories can be effectively combined as well. Since the wing-propelled 169
divers also plot in this area, the region of morphospace consisting of aerial hunters, plunge 170
divers, and wing-propelled divers could be thought of instead as the 'wing-powered foragers' 171
(Fig. 3). The fact that all these birds rely on their wings to capture food (whether soaring in the 172
air or in water) exposes them to similar selective pressures which might explain their proximity 173
in morphospace. After these adjustments to the original foraging bins, the foot-propelled divers 174
and surface swimmers are left as the only truly aquatic birds, thus drastically reducing the size of 175
this foraging bin to the central region of overlap between the other foraging bins (Fig. 2). The 176
exception to this trend are the penguins, which plot with other flightless terrestrial birds instead 177
of other wing-propelled divers (Fig. 2, inset). 178
In addition to aquatic specializations, this study also looked for evidence of migration or 179
nesting habits reflected in the distribution of modern taxa in morphospace. Figure 3 shows the 180
distribution of long-distance and non-migrants within the original morphospace. Figure 4 shows 181
the distribution of arboreal and terrestrial nesting birds. 182
Mesozoic Avian Ecomorphology 183
With few exceptions (see below), Mesozoic birds included in this analysis plotted within 184
the morphospace defined by modern birds (Fig. 6). The four Mesozoic outliers were 185
Cathayornis, Longipteryx, Protopteryx, and Zhongornis. The primitive non-pygostylian bird 186
Zhongornis fell slightly outside the modern morphospace near the boundary between terrestrial 187
and arboreal birds (Fig. 6b). The enantiornithines Cathayornis, Longipteryx, and Protopteryx 188
plotted just outside of the arboreal morphospace (Fig. 6c). As none of these birds lies far from 189
the morphospace of modern birds, it is possible that expansion of the modern taxa would result 190
in an expanded modern morphospace which would include these four taxa. 191
383

Long-tailed birds and basal pygostylians (Fig. 6b). In addition to Zhongornis, the 192
analysis presented here included two other non-pygostylians, taxa of the basal-most long-tailed 193
birds Archaeopteryx and Jeholornis and four basal pygostylians: Changchengornis, 194
Confuciusornis, Eoconfuciusornis, and Sapeornis. Of these, Jeholornis and the large 195
Confuciusornis (D2454) plot firmly in the central region of overlap. Eoconfuciusornis, 196
Changchengornis, and the small Confuciusornis (BM.Av.1168-1171) plot on the border of the 197
aquatic birds, overlapping with the terrestrial birds. Shenzhouraptor plots at the margin of the 198
aquatic birds, overlapping with both the aerial and arboreal birds. Two of these primitive birds 199
fall within unique foraging strategies: both Archaeopteryx specimens plot with the terrestrial 200
birds while Sapeornis plots within the aerial foraging category . 201
Enantiornithine birds (Fig. 6c). The analysis presented here includes six enantiornithine 202
birds: Cathayornis, Longipteryx, Neuquenornis, Pengornis, Protopteryx, and Vescornis. With the 203
exception of the three outliers (see above), Neuquenornis plotted in the central region of overlap 204
while Vescornis plotted with terrestrial birds and Pengornis fell in the region of overlap between 205
arboreal and aerial birds. 206
Ornithuromorph birds (Fig. 6d). Four ornithuromorphs were included in the present 207
analysis: Hongshanornis, Patagopteryx, Yanornis, and Yixianornis. Of these, Yanornis and 208
Yixianornis fell within or very close to the central region of overlap. Patagopteryx and 209
Hongshanornis both plotted toward the outer extent of terrestrial birds. 210
211
DISCUSSION 212
Trends in the Morphometric Data, as revealed by PCA 213
The principle component analysis carried out on the morphometric dataset of 10 skeletal 214
measurements found 87% of the variance was expressed by the first principle component axis 215
384

(PC1), while 5% was expressed by PC2, about 3% by PC3, and almost 2% by PC4 (Table 1). 216
The remaining axes each expressed less than 1% of the variance. For the purposes of this study, 217
PC1 and 2 were considered together to constitute the significant portion of the variance in these 218
data. Examination of the variance-covariance loadings for PC1 and 2 allows interpretation of the 219
results of this PCA (Table 2). 220
The loadings of PC1 for each variable are all very similar, ranging from 0.243 to 0.362. 221
This implies that the variables are linearly related, as would be expected if PC1 were 222
representing size changes in the data. Anecdotal evidence for interpreting PC1 as indicating body 223
size changes comes from the inclusion of large and small examples of two Mesozoic taxa, 224
Archaeopteryx and Confuciusornis. In both cases the large specimens plotted separately from 225
the small specimens, being displaced primarily along the PC1 axis (Fig. 6b). As body size is the 226
main difference between the large and small individuals (however some slight differences in 227
element length proportions do exist), it seems reasonable that PC1 is expressing body-size 228
differences within the taxa in the dataset. 229
The loadings of PC2 separate into two different groups. Elements of the hindlimb range 230
from -0.122 to -0.519 while elements of the forelimb range from 0.253 to 0.419. This implies 231
that PC2 is expressing shape or proportion differences in the allometric relations between the 232
limbs. As the strongest separation in the data is seen as the segregation of leg-using foragers 233
(terrestrial birds and waders) above the PC1 axis and wing-dominated foragers (aerial, arboreal, 234
plunge-diving, etc.) below PC1, it appears that the reliance of a bird on one functional unit over 235
another (e.g., leg vs. wing) when foraging is quantitatively reflected in the morphology of that 236
bird (Fig. 2b). Similarly, the birds falling in the central region of overlap (along the PC1 axis) all 237
utilize both their wings and legs in their daily ecology, and so don't plot in either the terrestrial or 238
flying regions of morphospace, but rather between them. 239
385

Correlation between Morphology and Ecology 240
Primary foraging bins. The original primary foraging bins (terrestrial, arboreal, aquatic, 241
and aerial) were successful to varying degrees in describing the partitioning of morphospace for 242
modern birds (Fig. 1). The initial aquatic bin was the least delineated, as birds assigned to this 243
category were found across almost the entire morphospace, overlapping with birds from all other 244
foraging strategies. This was not unexpected, as the birds which were defined as 'aquatic' 245
included such diverse forms from wading birds such as spoonbills to flightless divers like 246
penguins. Clearly these birds have very different ecologies and should not be considered together 247
in a single ecological bin. 248
The subdivision of aquatic birds into waders, plunge divers, surface swimmers, and 249
wing- and foot-propelled divers was much more successful in revealing patterns in morphospace 250
which correlate to ecology (Fig. 2). From this partitioning it was evident that wading birds plot 251
with the terrestrial foraging birds, while plunge and wing-propelled divers plot with the aerial 252
birds, leaving the surface swimmers and foot-propelled divers in the aquatic foraging bin (Fig. 253
2). This implies that there is a fundamental difference between birds which rely on their legs as 254
their primary foraging strategy and birds which rely on their wings. In the most general sense 255
this is seen in the data by the restriction of all birds which forage using their legs to the area 256
above the axis representing PC 1 (x-axis) and the restriction of all wing-foragers to below the PC 257
1 axis (Fig. 2). 258
Even after the reorganization of the aquatic birds, there still remained a region of overlap 259
of most ecologic bins in the central region of the morphospace (Fig. 2). Within this region almost 260
all the surface swimmers and foot-propelled divers were found, as well as some birds from all of 261
the other foraging categories. This is not surprising, as many birds spend a great deal of their 262
time either using their legs in multiple ways (such as ducks, some of which swim on the surface, 263
386

make short dives underwater, and walk on land to forage) or using both their legs and wings to 264
forage (such as parrots, some of which spend equal amounts of time foraging in trees and on the 265
ground). This implies that the central region of overlap can be thought of as the 'generalist' 266
region of morphospace, while the outer regions segregate into 'specialist' regions which can be 267
correlated to a particular foraging strategy. The existence of generalized and specialized 268
morphospace has important ramifications for the identification of Mesozoic bird ecology in the 269
absence of a highly specialized morphology (e.g., the specialized hindlimbs of hesperornithiform 270
birds are characteristic of foot-propelled divers; Marsh, 1880): only fossil taxa which fall within 271
a specialized region, and not the central generalized region, should be considered as candidates 272
for that particular ecology. 273
Secondary foraging bins. In addition to the aquatic sub-categories discussed above, 274
birds were categorized into a number of secondary ecological bins. These bins were designed to 275
describe other aspects of avian lifestyle that might or might not be congruent with specialized 276
foraging morphologies. The first secondary category investigated was migration behavior, with 277
birds being considered as either non- or long-distance migrants. Most long-distance migrants 278
plotted in the region defined by the aerial foragers (Fig. 3), suggesting that the necessity of flying 279
extended distances every year imposes a similar selective pressure as using powered flight to 280
capture prey. Additionally, all of the birds which are not categorized as aerial foragers but which 281
plot in the aerial morphospace are long-distance migrants. However, migration does not appear 282
to be a strong enough factor to supersede certain foraging strategies, as terrestrial birds which 283
undergo long-distance migration still plot with other terrestrial birds, although they do fall in the 284
region of morphospace closer to the aerial birds. Not surprisingly, the birds which do not 285
undergo a long-distance migration do not show any evident pattern--all of these birds simply plot 286
with their foraging bin. 287
387

The second secondary ecological bin investigated was the nesting habitat of the birds, 288
categorized as either arboreal or terrestrial (Fig. 5). Most arboreal nesting birds fell in the 289
generalist area and the area defined by the arboreal birds, while terrestrial nesting birds plotted 290
throughout the remaining morphospace. This division may in part be due to the make-up of the 291
dataset, as almost all arboreal nesters were also characterized as arboreal foragers. The arboreal- 292
nesting, terrestrial-foraging birds plot with either the other terrestrial birds or in the generalist 293
region. 294
Therefore, with the exception of the consolidation of the long-distance migrants with the 295
aerial foragers, it appears that secondary ecologies such as nesting do not have a strong 296
morphological signature in these data. 297
Phylogenetic signal. One particular issue which has been raised regarding studies such 298
as this is the necessity of determining the extent to which the quantitative data supporting the 299
resulting clusters is due to shared evolutionary history (Garland et al., 1992; Garland et al., 300
1999). While a number of statistical methods have been proposed to investigate the effects of 301
phylogeny in morphometric datasets (Purvis and Rambaut, 1995; Lavin et al., 2008), the 302
application of such methods to this study was complicated due to the availability of a 303
comprehensive phylogeny of modern birds. While a number of phylogenies have been proposed 304
for modern birds (Sibley and Ahlquist, 1990; Clarke and Middleton, 2008; Hacket et al., 2008), 305
the taxa used in these studies does not align with the taxa used in the present study. In order to 306
look at possible causes of clustering (phylogeny versus ecomorphology) in the data presented 307
here, clades were analyzed which contained multiple taxa representing at least two foraging bins. 308
Clades were taken from the phylogeny of Hacket et al. (2008), as higher-order taxonomic 309
groupings of birds have been found to be paraphyletic in many cases (Hacket et al., 2008; Clarke 310
and Middleton, 2008). 311
388

Several clades of modern birds did not cluster together, with members of these clades 312
clustering instead within their respective foraging categories. Recent molecular studies have 313
suggested that the Phoenicopteridae and the Podicipedidae are sister clades, and are in turn most 314
closely related to the Phalacrocoracidae (Hacket et al., 2008). If phylogeny were the stronger 315
signal in the morphological data, then these groups would cluster together regardless of their 316
very different foraging strategies. However, Figure 7 shows this is not the case, with the 317
flamingo and the flightless phalacrocoracid Nannopterum plotting with other terrestrial birds 318
(specifically with the penguins, see previous discussion), while the grebes plot with foot- 319
propelled divers in the generalist morphospace. With the exception of Nannopterum, the 320
Phalacrocoracidae plot with the wing-propelled divers in the aerial morphospace. This is also 321
seen in the members of the Ciconidae, Gaviidae, Pelecanidae, Procellaridae, and Spheniscidae 322
(Fig. 8), all of which form a monophyletic clade (Hacket et al., 2008). Despite being closely 323
related to each other, members of these orders plot with their foraging bin as opposed to with the 324
other members of their order. The exception to this are the penguins, which plot with the 325
terrestrial birds, as previously discussed. 326
In contrast to the above examples, in a few cases in this dataset taxa cluster with their 327
close relatives despite having disparate ecologies. For example, within both the Psittacidae and 328
the Columbidae the terrestrial and arboreal taxa plot in a single cluster near the border of the 329
terrestrial/arboreal morphospace (Fig. 9a). Also, the terrestrial Pteroclididae plot with the 330
arboreal and terrestrial Columbidae (Fig. 9b), whom they are most closely related to (Hacket et 331
al., 2008). 332
333
334
389

Comparison to Previous Studies 335
Previous studies of modern birds. While the statistical approach used in this study has 336
not been used previously, similar quantitative methods have been used to investigate the 337
partitioning of modern avian morphospace and its correlation with ecology. In particular, the 338
study of Zeffer and colleagues concluded that the correlation between element lengths and 339
ecology varied in strength among different types of foragers (Zeffer et al., 2003a). For example, 340
the average length index of the tibiotarsus (after adjustment for body mass and remaining 341
element lengths - see Zeffer et al., 2003a for specific details) in some, but not all, ecological 342
groups was significantly (P>0.0001) different from the average length index of all birds. While 343
the specific findings of this study are not directly comparable to the study presented here, both 344
studies show that morphometrics may be useful as a means of quantifying the morphometric 345
differences between birds with distinct ecological strategies. 346
Several non-statistical studies have utilized ternary diagrams to investigate the 347
relationships of element proportions of the limbs to ecological strategies and the use of 348
morphospace for modern birds (Gatesy and Middleton, 1997; Middleton and Gatesy, 2000). The 349
findings of these previous studies are in accordance with those presented here in concluding that 350
among modern birds, fore- (Middleton and Gatesy, 2000) and hind- limb (Gatesy and Middleton, 351
1997) length patterns correlate with functional utility. 352
Previous studies of Mesozoic birds. To date, there have not been any studies which are 353
directly comparable to that presented here in terms of the database and methods used. However, 354
the recent study of Dyke and Nudds (2009) discussed above is somewhat similar in terms of 355
approach. The main difference in the methods of these studies is the use of ternary diagrams to 356
analyze the forelimb and hindlimb separately in the previous study (Dyke and Nudds, 2009), 357
while the present study uses statistics and a single combined dataset containing the same fore- 358
390

and hind-limb data but with the addition of the third pedal phalanx as well. Unfortunately, while 359
several Mesozoic taxa were used in both these studies, the specific goals of this study differed 360
from that of Dyke and Nudds (2009), making specific comparisons impossible. 361
Implications for Early Avian Evolution 362
Comparison to modern diversity. Of the 17 species of Mesozoic birds examined in this 363
study, four plotted outside of the morphospace defined by the modern avian database 364
(Cathayornis, Longipteryx, Protopteryx, and Zhongornis). This implies that most Mesozoic birds 365
were capable of utilizing foraging techniques that were essentially like those employed by 366
modern birds--the ecological niches and trophic guilds that characterize the latter clearly have a 367
deep evolutionary origin. Only six of the nineteen Mesozoic birds analyzed here fall into the 368
'generalist' area of the morphospace, the region in which three different foraging strategies 369
overlap. The Mesozoic birds in this region represent long-tailed birds and basal pygostylians as 370
well as ornithuromorphs and enantiornithines. Of the remaining Mesozoic birds, eight fell in a 371
region of morphospace with a unique foraging bin, while the remaining taxa fell in an area where 372
two foraging bins overlapped. These results demonstrate that early birds were well diversified 373
into specialized ecological niches as well as retaining some generalist forms. 374
By plotting the possible foraging strategies of Mesozoic birds on a cladogram (Fig. 10), it 375
can be seen that by the Early Cretaceous the modern ecological bins detailed in this study had 376
already originated. This implies that the evolutionary diversification of Cretaceous birds was 377
accompanied by a substantial ecological diversification. Furthermore, the present analysis shows 378
that a great deal of ecological specialization also characterized the early evolution of the two 379
major clades of Mesozoic birds, the enantiornithines and the ornithuromorphs (Fig. 10). 380
Examination of individual taxa. In addition to analyzing general trends between ancient 381
and modern birds, this study allows the identification of specific modern taxa which most closely 382
391

resemble a given Mesozoic taxon in terms of the morphological parameters analyzed in this 383
study (Table 3). This section will examine some of these comparisons, with the understanding 384
that other birds may exist which are more similar to a given Mesozoic specimen but which were 385
not included in this study. 386
Of the Mesozoic birds used in this study, six plotted within or very near the generalist 387
region (Fig. 6a): the basal pygostylian Confuciusornis (Dalian specimen), the long-tailed birds 388
Jeholornis and Shenzhouraptor, the enantiornithine Neuquenornis, and the ornithuromorphs 389
Hongshanornis and Yixianornis. Modern birds which fall in this generalist region are a mix of 390
aerial hunters, like the buzzard Buteo swainsoni, surface swimmers like members of the duck 391
genus Anas, foot-propelled divers of the auklet genus Fratercula, and terrestrial and arboreal 392
birds, like members of the pigeon genera Columba and Ducula. In light of the wide diversity of 393
morphology and ecology exhibited by the modern birds in this generalist region, it is not possible 394
to use these data to ascribe more specific ecologies to the Mesozoic birds which plot in this 395
region of morphospace. 396
Three Mesozoic specimens plotted in a region of morphospace in which two foraging 397
strategies overlapped. The Berlin specimen of Confuciusornis plotted at the uppermost margin of 398
the aquatic birds where they overlapped with the terrestrial birds, plotting near the modern 399
charadriforms Limnodromus griseus and Tringa incana as well as the pigeon Ptilinopus perlatus. 400
The confuciusornithid Changchengornis fell on the border of terrestrial and arboreal birds, 401
closest to the parrot Cyanoramphus novaezelandiae as well as two pigeons of the Ptilinopus 402
genus, P. melanospila and P. solomonensis. Finally, the enantiornithine Pengornis fell well 403
within the arboreal foragers, but overlapped with the outermost edge of the morphospace for the 404
aerial foragers. Pengornis plotted closely to the modern gull Chroicocephalus philadelphia as 405
well as the parrot Cacatua haemateropygia and the petrel Pterodroma cookii. While the overlap 406
392

of ecologies around these Mesozoic birds makes assigning specific ecologies for them difficult, 407
some generalizations can be made. For instance, Pengornis was unlikely to be a terrestrial bird 408
and probably utilized flight to a high degree in its daily activities. 409
Of the remaining Mesozoic specimens, seven plotted in a region of the morphospace with 410
minimal or no overlap between different foraging strategies. These taxa may therefore be 411
considered as having the most in common morphologically with modern birds of that particular 412
foraging category, and so may have inhabited a similar ecological niche. Additionally, all of 413
these birds overlapped very closely with one or more modern taxa, indicating almost exact limb 414
proportion similarities with these birds. 415
The basal pygostylian Sapeornis plotted with the aerial foraging birds, and so may be 416
considered to have had a similar ecology. More specifically, Sapeornis overlaps with both the 417
shearwater Calonectris diomedea and the buzzard Buteo lagopus. Both of these modern birds are 418
strong fliers which forage primarily by spotting prey from soaring flight, and so perhaps 419
Sapeornis had a similar capability for prolonged, soaring flight. This quantitative observation 420
supports the qualitative observations made by Zhou and Zhang (2002), who concluded from the 421
elongate wing and well-developed deltoid crest of the humerus that Sapeornis would have been 422
capable of soaring flight. 423
The remaining Mesozoic birds which did not fall in a region of foraging overlap all 424
plotted with the terrestrial modern birds. Both specimens of Archaeopteryx plotted near the 425
center of the terrestrial morphospace, however they did not plot in close proximity to each other. 426
The larger London specimen plotted closest to the charadriform Tringa semipalmata and the 427
pheasant Dendragopus obscurus while the smaller Eichstatt specimen plotted near the pheasant 428
Arborophila javanica and the pigeon Columba luzonica. Eoconfuciusornis also plotted within the 429
393

terrestrial birds, overlapping with the modern charadriforms Limnodromus scolopaceus and 430
Gallinago delicata. 431
A single enantiornithine, Vescornis, was identified as terrestrial, overlapping with the 432
modern pigeons Columbina passerina and C. picui as well as the charadriform Calidris pusilla. 433
Two ornithuromorph birds, Patagopteryx and Hongshanornis, were also identified as terrestrial. 434
Patagopteryx plots well within the morphospace defined by modern terrestrial foragers and well 435
away from any other foraging category, overlapping with the pheasant Francolinus afer and 436
plotting closely to the rail Eulabeornis cajaneus and the partridge Alectoris melanocephala. 437
Patagopteryx has long been recognized as a flightless terrestrial bird due to the extreme 438
reduction of the forelimb (Chiappe, 1991), however it is interesting that Patagopteryx does not 439
plot most closely to other flightless or poor-flying terrestrial birds, such as the ratites or 440
tinamous, but rather plots with rails and pheasants. Hongshanornis plots near the outer limit of 441
modern terrestrial birds, overlapping with the charadriforms Actitis macularia and Calidris 442
mauri. 443
While all of the remaining Mesozoic taxa fell outside of the modern morphospace, three 444
of those taxa (the enantiornithines Cathayornis, Longipteryx, and Protopteryx) were only slightly 445
outside of the modern boundaries. Expansion of the modern database might well expand those 446
boundaries the small amount needed to include these birds. Longipteryx and Protopteryx plotted 447
very near each other, practically overlapping at the margin of the arboreal birds. The modern 448
taxa closest to these birds are the parrots Aratinga canicularis and A. weddellii. Cathayornis 449
also plotted just outside the arboreal birds, falling closest to the parrots Cyclopsitta diophthalma, 450
and Loricula galgulus. 451
452
453
394

CONCLUSIONS 454
The quantitative data presented here was successfully correlated with foraging behavior 455
for a large array of modern birds. Broad distinctions between walking and flying -dominate 456
foragers were observed in the clustering patterns of modern birds, as well as finer distinctions 457
between wing-dominated or aerial foragers and arboreal foragers. Secondary ecologies, such as 458
nesting or migratory behaviors, did not supercede primary foraging strategies. 459
460
The use of the modern database in conjunction with Mesozoic taxa enabled both general 461
comparisons between modern and Mesozoic birds as well as specific insights into possible 462
ecological roles for various Mesozoic birds. Archaeopteryx, Patagopteryx, Eoconfuciusornis, 463
Vescornis, and Hongshanornis were all identified as terrestrial foragers, while Sapeornis was 464
identified as a wing-forager. Of the remaining taxa, three fell outside of the modern morphospace 465
and eight fell within the generalist region of overlap between multiple foraging bins. These 466
results indicate that while morphometric studies offer tantalizing opportunities for the ecological 467
characterization of Mesozoic birds, these are largely limited to taxa that cluster within well- 468
defined specialized categories of modern birds and not in a generalist cluster. 469
470
While the results of this study are promising, the methods used did impose limits. The main 471
drawback to this sort of study is the requirement of a complete dataset for the principal 472
component analysis. When dealing with Mesozoic birds most are incomplete, thus limiting the 473
scope of the database. Additionally, the impact of phylogenetic signal and body size of known 474
fossils must be taken into consideration when interpreting the results of the PCA. 475
476
477
395

FIGURE CAPTIONS 478
Figure 1. Plot of PC1 vs. PC2 showing modern taxa divided into primary ecological bins based 479
on foraging strategy. 480
481
Figure 2. Subdivision of aquatic foraging taxa into more specialized subcategories. A) All 482
aquatic birds superimposed on the non-aquatic foraging bins from Figure 1 (shaded regions). B) 483
Simplified aquatic foraging bin after the re-organization of the waders to terrestrial birds and the 484
wing-propelled and plunge divers with the aerial birds in the new 'wing-powered' category. 485
486
Figure 3. Flow chart showing the original foraging bins used in the study and the re-arrangement 487
of aquatic subcategories to better reflect the distribution of specialized aquatic birds in 488
morphospace. 489
490
Figure 4. Long-distance migrants versus non- or regional- migrants. The shaded regions 491
represent the morphospace regions as defined by primary foraging strategy (terrestrial, arboreal, 492
wing-powered, and aquatic). 493
494
Figure 5. Terrestrial-nesting versus arboreal-nesting birds. The shaded regions represent the 495
morphospace regions as defined by primary foraging strategy (terrestrial, arboreal, wing- 496
powered, and aquatic). 497
498
Figure 6. Mesozoic birds plotted over modern morphospace. A. All 19 Mesozoic specimens 499
included in the analysis. B. Long-tailed birds and basal pygostylians. C. Enantiornithine birds. D. 500
Ornithuromorph birds. 501
396

Figure 7. Distribution of the sister groups Phoenicopteridae and Podicepidae and their closest 502
relatives, the Phalacrocoracidae. Nannopterum, the only flightless phalacrocoracid, is labeled. 503
504
Figure 8. Distribution of the monophyletic clade containing the Ciconidae, Gaviidae, 505
Pelecanidae, Procellaridae, and Spheniscidae, as identified by Hacket et al., 2008. 506
507
Figure 9. Clades with members that cluster together despite different foraging strategies. A. 508
Psittacidae. B. Pteroclididae and Columbidae. 509
510
Figure 10. Cladogram of Mesozoic birds included in this study (adapted from O'Connor et al., 511
2009), plotted with respect to possible foraging strategies identified in this study. 512
513
TABLE CAPTIONS 514
Table 1. Variance-Covariance matrix from the PCA analysis of the modern dataset presented 515
here. 516
517
Table 2. Variance-Covariance loadings for each variable included in the analysis. 518
519
Table 3. Mesozoic specimens included in this analysis and the modern taxa they are located near 520
in morphospace, as determined by the PCA analysis. 521
522
523
524
397

Figure 1. 525
526
Figure 2. 527
528
529
398

Figure 3. 530
531
Figure 4. 532
533
399

Figure 5. 534
535
400

Figure 6. 536
537
538
539
401

Figure 7. 540
541
542
Figure 8. 543
544
402

Figure 9. 545
546
Figure 10. 547
548
403

Table 1. 549
axis Eigenvalue % Variance
1 0.562401 87.47
2 0.0353564 5.499
3 0.0190645 2.9651
4 0.0128572 1.9997
5 0.00465194 0.72352
6 0.00299897 0.46643
7 0.00267639 0.41626
8 0.00155893 0.24246
9 0.00111971 0.17415
10 0.00027764 0.043181
550
551
404

Table 2. 552
variable
Axis
1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 Axis 7 Axis 8 Axis 9 Axis 10
Digit 3,
phalanx 1 0.323 -0.146 0.323 -0.410 0.143 -0.141 -0.482 -0.402 -0.340 -0.052
Digit 3,
phalanx 2 0.301 -0.170 0.473 -0.317 0.265 -0.104 0.627 0.244 0.145 0.051
Digit 3,
phalanx 3 0.258 -0.122 0.511 0.311 -0.675 0.300 -0.088 0.080 0.033 -0.001
Tarso-
metatarsus 0.342 -0.519 -0.547 -0.320 -0.352 -0.001 0.078 -0.121 0.230 0.120
Tibiotarsus 0.291 -0.319 -0.189 0.228 0.265 0.097 -0.222 0.656 -0.339 -0.230
femur 0.243 -0.290 -0.002 0.686 0.300 -0.188 0.164 -0.474 0.038 0.091
Carpo-
metacarpus 0.301 0.294 0.012 0.088 -0.110 -0.692 -0.296 0.277 0.307 0.256
ulna 0.355 0.391 -0.175 0.015 -0.183 -0.145 0.272 -0.148 -0.113 -0.726
radius 0.361 0.419 -0.202 0.017 -0.029 0.241 0.221 -0.013 -0.470 0.569
humerus 0.362 0.253 -0.018 -0.036 0.352 0.526 -0.273 -0.067 0.567 -0.053
553
554
405

Table 3. 555
Mesozoic taxa overlapping modern taxa
foraging strategy of
modern taxa
Primitive
Archaeopteryx
lithographica Arborophila javanica terrestrial
Birds small - Eichstatt Specimen Gallicolumba luzonica terrestrial

Archaeopteryx
lithographica Tringa semipalmata terrestrial (wader)
large - London Specimen Dendragopus obscurus terrestrial
Changchengornis
Cyanoramphus
novaezelandie arboreal
hengdaoziensis Ptilinopus melanospila arboreal
Ptilinopus solomonensis arboreal
Confuciusornis sanctus Ptilinopus occipitalis arboreal
large - Dalian Specimen Ptilinopus porphyrea arboreal
Confuciusornis sanctus Limnodromus griseus terrestrial (wader)
small - Berlin Specimen Tringa incana terrestrial (wader)
Ptilinopus perlatus arboreal
Eoconfuciusornis zhengi Limnodromus scolopaceus terrestrial (wader)
Gallinago delicata terrestrial
Jeholornis prima Puffinus griseus surface swimmer
Anas poecilorhynchus surface swimmer
Sapeornis chaoyangensis Calonectris diomedea aerial (plunge diver)
406

Buteo lagopus aerial
Shenzhouraptor sinensis Puffinus nativitatis aerial (plunge diver)
Larus ridibundus terrestrial
Sterna paradisaea aerial (plunge diver)
Zhongornis haoae none
Enantiornithine Cathayornis yandica Cyclopsitta diophthalma arboreal
Birds Loricula galgulus arboreal
Longipteryx chaoyangensis Aratinga canicularis arboreal
Aratinga weddellii arboreal
Neuquenornis volans Podiceps auritus foot-propelled diver
Aythya collaris foot-propelled diver
Anas smithii surface swimmer
Pengornis houi
Chroicocephalus
philadelphia aerial (plunge diver)
Pterodroma leucoptera aerial (plunge diver)
Cacatua haematuropygia arboreal
Protopteryx fengingensis Aratinga canicularis arboreal
Aratinga weddellii arboreal
Vescornis hebeiensis Columbina passerina terrestrial
Columbina picui terrestrial
Calidris pusilla terrestrial
Ornithurine Hongshanornis longicresta Actitis macularia terrestrial
Birds Calidris mauri terrestrial
407

Patagopteryx deferrisi Francolinus afer terrestrial
Alectoris melanocephala terrestrial
Eulabeornis cajaneus terrestrial
Yanornis martini Phalaropus fulicarius surface swimmer
Arenaria interpres terrestrial

Brachyramphus
brevirostris foot-propelled diver
Yixianornis grabaui Aethia cristatella foot-propelled diver
Rollandia rollandia foot-propelled diver
Cyclorrhynchus psittacula foot-propelled diver
556
557
408

ACKNOWLEDGEMENTS
I would like to thank my advisors, Dr. Luis Chiappe and Dr. Dave Bottjer, for their assistance
and guidance throughout the course of my dissertation.
I would also like to thank the following institutions and their respective collections managers for
access to specimens: American Museum of Natural History, British Museum of Natural History,
Field Museum of Natural History, Fort Hays Sternberg Museum, Harvard Museum of
Comparative Anatomy, Smithsonian Institute, South Dakota School of Mines Museum of
Geology, Southern Arkansas University, T-Rex Discovery Center of the Royal Saskatchewan
Museum of Paleontology, University of California Museum of Paleontology, University of
Kansas Museum of Paleontology, University of Nebraska State Museum, Yale Peabody
Museum.
A number of students and interns provided invaluable assistance in collecting measurements:
Chris McDonald, Vicky Torres, and Sean Vera Cruz. I would also like to thank Garreth Dyke,
Mike Everhart, Jesus Marugan-Lobos, and Larry Martin for their comments and advice on
various sections of this work.

409

REFERENCES CITED
Bell, A., Chiappe, L. M., Erickson, G., Suzuki, S., Watabe, M., Barsbold, R., and Tsogtbaatar,
K., 2010, Description and ecologic analysis of Hollanda luceria, a Late Cretaceous bird
from the Gobi Desert (Mongolia): Cretaceous Research, vol. 31, p. 16-26.
Bell, A., and Everhart, M., 2009, A new specimen of Parahesperornis (Aves:
Hesperornithiformes) from the Smoky Hill Chalk (Early Campanian) of western Kansas:
Transactions of the Kansas Acadey of Science, vol. 112, p. 7-14.
Bininda-Emonds, O. R. P., Bryant, H. N., Russell, A. P., 1998, Supraspecific taxa as terminals in
cladistic analysis: implicit assumptions of monophyly and a comparison of methoids:
Bilogical Jounral fo the Linnean Society, vol. 64, p. 101-133.
Brodkorb, P., 1963, Birds from the Upper Cretaceous of Wyoming in Sibley, C. G., Hickey., J.
J., and Hickey, M. B., eds., Proceedings of the XIII International Ornithologuical Congress:
American Ornithologists Union, Baton Rouge, p. 55-70.
Brodkorb, P., 1971, Origin and evolution of birds: in Farmer, D., King, J., eds, Avian biology,
Academic Press: New York, p. 19-55.
Chiappe, L. M., 1991, Cretaceous birds of Latin America: Cretaceous Research, vol. 12, p. 55-
63.
Chiappe, L. M., 1996, Late Cretaceous birds of southern South America: anatomy and
systematics of Enantiornithes and Patagopteryx deferrariisi, in G. Arratia (ed.),
Contributions of southern South America to vertebrate paleontology, Muinch, Geowiss,
vol.30, p. 203–244.
410

Chiappe, L. M., 2002, Basal bird phylogeny: problems and solutions, in Chiappe, L. M., and
Witmer, L. M., eds., Mesozoic birds: above the heads of dinosaurs: University of California
Press, Berkeley, California, p. 448-475.
Chiappe, L. M., 2007, Glorified dinosaurs: the origin and early evolution of birds, John Wiley &
Sons, New Jersey, 261 pp.
Chiappe, L. and Dyke, G., 2002, The Mesozoic radiation of birds: Annual Review of Ecology
and Systematics, vol. 33, p. 91-124.
Chiappe, L. M. and J. O. Calvo, 1994, Neuquenornis volans, a new Enantiornithes (Aves) from
the Upper Cretaceous of Patagonia (Argentina), Journal of Vertebrate Paleontology, vol. 14,
p. 230-246.
Chiappe, L.M., S. Suzuki, G.J. Dyke, M. Watabe, K. Tsogtbaatar, and R. Barsbold, 2007, A new
enantiornithine bird from the Late Cretaceous of the Gobi Desert, Journal of Systematic
Paleontology, vol. 5, p. 193-208.
Clarke, J., 2004, Morphology, phylogenegtic taxonomy, and systematics of Ichthyornis and
Apatornis (Avialae: Ornithurae): Bulletin of the American Museum of Natural History, vol.
286, 179 p.
Clarke, J.A. and Norell, M.A., 2004, New avialan remains and a review of the known avifauna
from the Late Cretaceous of Mongolia, American Museum Novitates, vol. 3447, 12 pp.
Clarke, J. A. and Middleton, K. M., 2008, Mosaicism, molecules, and the evolution of birds:
results from a Bayesian approach to the study of morphological evolution using discrete
character data, Systematic Biology, vol. 57, p. 185-201.
411

Clarke, J. A., Zhou, Z., and Zhang, F., 2006, Insight into the evolution of avian flight from a new
clade of Early Cretaceous ornithurines from China and the morphology of Yixianornis
grabaui, Journal of Anatomy, vol. 208, p. 287-308.
Cracraft, J., 1982, Phylogenetic relationships and monophyly of loons, grebes, and
hesperornithiform birds, with comments on the early history of birds: Systematic Zoology,
vol. 31, p. 35-56.
Cumbaa, S. L., Schroder-Adams, C., Day, R. G., and Philips, A. J., 2006, Cenomanian bonebed
faunas from the northeastern margin, Western Interior Seaway, Canada in Lucas, S. G. and
Sullivan, R. M., eds., Late Certaceous vertebrates from the Western Interior: New Mexico
Museum of Natural History and Science Bulletin, vol. 35, p. 139-156.
del Hoyo, J., Elliot, A., and Sargatal, J. (eds.), 2008, Handbook of the Birds of the World, Vols 1
– 13, Lynx Edicions, Barcelona.
Dyke, G., Malakhov, D., and Chiappe, L. M., 2006. A re-analysis of the marine bird
Asiahesperornis from northern Kazakhstan: Cretaceous Research, vol. 27, p. 947-953.
Dyke, G. J. and Nudds, R. L., 2009, The fossil record and limb disparity of enantiornithines, the
dominant flying birds of the Cretaceous, Lethaia, vol. 42, p. 248-255.
Eisenmann, V. and Baylac, M., 2002. Extant and fossil Equus (Mammalia, Perissodactyla)
skulls: a morphometric definition of the subgenus Equus: Zoologica Scripta, vol. 29, p. 89-
100.
Elzanowski, A. and Galton, P. M., 1991, Braincase of Enaliornis, and Early Cretaceous bird
from England: Journal of Vertebrate Paleontology, vol. 11, p. 90-107.
412

Everhart, M., and Bell, A., 2009, A hesperornithiform limb bone from the basal Greenhorn
Formation (Late Cretaceus; Middle Cenomanian) of north central Kansas: Journal of
Vertebrate Paleontology, vol. 28, p. 952-956.
Fox, R., 1974, A middle Campanian, nonmarine occurrence of the Cretaceous toothed bird
Hesperornis Marsh: Canadian Journal of Earth Sciences, vol. 11, p. 1335-1338.
Farlow, J., Hurlburt, G., Elsey, R., Britton, A., and Langston, W., 2004, Femoral dimensions and
body size of Alligator mississippiensis: estimating the size of extinct mesoeucrocodylians,
Journal of Vertebrate Paleontology, vol. 25, p. 354-369.
Fielder, W., 2005, Ecomorphology of the external flight apparatus of blackcaps (Sylvia
atricapilla) with different migration behaviors, Annals of the New York Academy of
Science, vol. 1046, p. 253-263.
Galton, P. and Martin, L., 2002, Enaliornis, an early hesperornithiform bird from England, with
comments on other Hesperornithiformes in Chiappe, L. M., and Witmer, L. M., eds.,
Mesozoic birds: above the heads of dinosaurs: University of California Press, Berkeley,
California, p. 317-338.
Gao, C., Chiappe, L. M., Meng, L., O’Connor, J., Wang, X., and Liu, J., 2008, A new basal
lineage of early Cretaceous birds from China and its implications on the evolution of the
avian tail: Palaeontology, vol. 2008, p. 1-17.
Garland, T., Jr., P. H. Harvey, and A. R. Ives, 1992. Procedures for the analysis of comparative
data using phylogenetically independent contrasts. Journal of Systematic Biology, vol. 41,
p. 18-32.
413

Garland, T., Jr., P. E. Milford, and A. R. Ives, 1999. An introduction to phylogenetically based
statistical methods, with a new method for confidence intervals on ancestral values.
American Zoologist, vol. 39, p. 374-388.
Gatesy, S. M. and K. M. Midddleton, 1997. Bipedalism, flight, and the evolution of theropod
locomotor diversity, Journal of Vertebrate Paleontology, vol. 17, p. 308-329.
Glen, C. and M. Bennett, 2007. Foraging modes of Mesozoic birds and non-avian theropods.
Current Biology, vol. 17, p.r911-r912.
Goloboff, P., Farris, J., and Nixon, K., 2008, TNT: A free program for phylogenetic analysis:
Cladistics, vol. 24, p. 774-786.
Habib, M. and Ruff, C., 2008, The effects of locomotion on the structural characteristics of avian
limb bones, Zoological Journal of the Linnean Society, vol. 153, p. 601-624.
Hackett, S., Kimball, R., Reddy, S., Bowie, R., Braun, E., Braun, M., Chojnowski, J., Cox, W.,
Han, K.-L., Harshman, J., Huddleston, C., Marks, B., Miglia, K., Moore, W., Sheldon, F.,
Steadman, D., Witt, C., and Yuri, T., 2008, A phylogenetic study of birds reveals their
evolutionary history: Science, vol. 320, p. 1763-1767.
Hammer, Ø. and Harper, D. A. T., 2005, Paleontological data analysis, Blackwell Publishing,
Oxford, UK, 368 pp.
Heilman, G., 1926, Origin of Birds: London, Witherby, 209 p.
Hertel, F. and Campbell, Jr., K.E., 2007, The antitrochanter of birds: form and function in
balance, The Auk, vol. 124, p. 789-805.
Hopson, J., 2001, Ecomorphology of avian and nonavian theropod phalangeal proportions:
implications of the arboreal versus terrestrial origin of bird flight. in Gauthier, J. and L.F.
Gall (eds.), New Perspectives on the Origin and Early Evolution of Birds: Proceedings of
414

the International Conference in Honor of John H. Ostrom, New Haven, Peabody Museum of
Natural History, p. 210 - 235.
Hou, L., 1999, A new hesperornithid (Aves) from the Canadian arctic: Vertebrata PalAsiatica,
vol. 7, p. 228-233.
Hou, L., Zhou, Z., Martin, L. D., and Feduccia, A., 1995, A beaked bird from the Jurassic of
China, Nature, vol. 377, p. 616-618.
Houde, P., 1987, Histological evidence for the systematic position of Hesperornis
(Odontornithes: Hesperornithiformes): Auk, vol.104. p. 125-129.
Kearney, M., and Clarke, J. M., 2003, Problems due to missing data in phylogenetic analyses
including fossils: a critical review: Journal of Vertebrate paleontology, vol. 23, p. 263-274.
Hurlburt, A.H., 2004, Species-energy relationships and habitat complexity in bird communities.
Ecology Letters, vol. 7, p. 714-720.
Keast, A. and Saunders, S.,2008, Ecomorphology of the North American Ruby-crowned
(Regulus calendula) and Golden-crowned (R. satrapa) Kinglets, The Auk, vol. 108, p. 880-
888.
Kurochkin, E., 1996, A new enantiornithid of the Mongolian Late Cretaceous, and a general
appraisal of the infraclass Enantiornithines (Aves), Palaeontological Institute, Special Issue;
Moscow.
Lavin, S. R., Karasov, W. H., Ives, A. R., Middleton, K. M., and Garland, T.,Jr., 2008,
Morphometrics of the avian small intestine compared with that of nonflying mammals: a
phylogenetic approach, Physiological and Biomechanical Zoology, vol. 81, p. 526-550.
Livezey, B. C., 2009, Morphometric patterns of variation in recent and fossil penguins (Aves,
Sphenisciformes): Journal of Zoology, vol. 219, p. 269-307.
415

Losos, J.B., 1990, Ecomorphology, performance capability, and scaling of West Indian Anolis
lizards: an evolutionary analysis, Ecological Monographs, vol. 60, p. 369-388.
Lucas, F. A., 1903, Notes on the osteology and relationship of the fossil birds of the genera
Hesperornis, Hargeria, Baptornis, and Diatrima: Proceedings of the United States National
Museum, vol. 26, p. 545-556.
Marsh, O.C., 1870, Notice of some fossil birds, from the Cretaceous and Tertiary Formations of
the United States: American Journal of Arts and Science, vol. 49, p. 205.
Marsh, O.C., 1872, Description of Hesperornis regalis, with notices of four other new species of
Cretaceous birds: The American Journal of Science and Arts, vol. 3, p. 1-7.
Marsh, O.C., 1876, Notice of new Odontornithes: The American Journal of Science and Arts,
vol. 11, p. 509-511.
Marsh, O.C., 1877, Characters of the Odontornithes with notice of a new allied genus: The
American Journal of Science and Arts, vol. 14, p. 85-87.
Marsh, O.C., 1880, Odontornithes, a monograph on the extinct toothed birds of North America:
Government Printing Office, Washington, D.C.
Marsh, O.C., 1893, A new Cretaceous bird allied to Hesperornis: American Journal of Science,
vol. XLV, p. 81-82.
Martin, L., 1984, A new hesperornithid and the relationships of Mesozoic birds: Transactions of
the Kanas Academy of Science, vol. 87, p. 141-150.
Martin, J., and Cordes-Person, A., 2007, A new species of the diving bird Baptornis from the
lower Pierre Shale Group of southwestern South Dakota: The Geological Society of
America, Special Paper 427.
416

Martin, L. and Lim, J., 2002. New information on the hesperornithiform radiation: Proceedings
of the 5th Symposium of the Society of Avian Palaeontology and Evolution, p. 165-174.
Martin, L. and Tate, J., 1967, A Hesperornis from the Pierre Shale: Nebraska Academy of
Science Proceedings, vol. 77, p. 40.
Martin, L. and Tate, J., 1976, The skeleton of Baptornis advenus (Aves: Hesperornithiformes):
Smithsonian Contributions to Paleobiology, vol. 27, p. 35-66.
Martin, L., Kurochkin, E. N., and Tokaryk, T. T., 2012, A new evolutionary lineage of diving
birds from the Late Cretaceous of North America and Asia: Palaeoworld, vol. 21, p. 59-63.
Mayr, G., Pohl, B., and Peters, D. S., 2005, A well-preserved Archaeopteryx specimen with
theropod features: Science, vol. 310, P. 1483-1488.
Middleton, K. M. and S. M. Gatesy, 2000, Theropod forelimb design and evolution, Zoological
Journal of the Linnean Society, vol. 128, p. 149-187.
Miles, D.B. and Ricklefs, R.E., 1984, The Correlation between ecology and morphology in
deciduous forest passerine birds, Ecology, vol. 65, p. 1629-1640.
Nessov, L. and Prizemlin, B., 1991, A large advanced flightless marine bird of the order
Hesperornithiformes of the Late Senomanian of Turgai Strait – the first finding of the group
in the USSR: USSR Academy of Sciences, Proceedings of the Zoological Instititute, vol.
239, p. 85-107.
Nessov, L. and Yarkov, A., 1993, Hesperornithiforms in Russia: Russian Journal of
Ornithology, vol. 2, p. 37-54.
Nudds, R. L., Dyke, G. J., and Rayner, J. M. V., 2004, Forelimb proportions and the
evolutionary radiation of Neornithes, Proceedings of the Royal Society of London B
(Supplement), vol. 271, p. S324-S327.
417

O’Connor, J., Gao, K.-Q., and Chiappe, L. M., 2010, A new ornithuromorphs (Aves:
Ornithoraces) bird from the Jehol Group indicativwe of higher-level diversity: Journal of
Vertebrate Paleonotlogy, vol. 30, p. 311-321.
O’Connor, J., Chiappe, L. M., and Bell, A., 2011, Pre-modern birds: avian divergences in the
Mesozoic: in Dyke, G. and Kaiser, G., eds., Living dinosaurs: the evolutionary history of
modern birds: Wiley-Blackwell, West Sussex.
O'Connor, J. K., X. Wang, L. M. Chiappe, C. Gao, Q. Meng, X. Cheng, and J. Liu, 2009.
Phylogenetic support for a specialized clade of Cretaceous enantiornithine birds with
information from a new species, Journal of Vertebrate Paleontology, vol. 29, p. 1-17.
Ostrom, J., 1976, Archaeopteryx and the origin of birds: Biological Journal of the Linnean
Society, vol. 8, p. 91-182.
Peczkis, J., 1994, Implications of body-mass estimates for dinosaurs, Journal of Vertebrate
Paleontology, vol. 14, p. 520-533.
Peters, D.S and E. Gorgner, 1992. A comparative study on the claws of Archaeopteryx in K.
Campbell (ed.), Proceedings of the Second International Symposium of Avian Paleontology,
L.A. Museum of Natural History Press, Los Angeles, p. 29-37.
Prendini, L., 2001, Species or supraspecific taxa as terminals in cladistic analysis? Groundplans
versus exemplars revisited: Systematic Biology, vol. 50, p. 290-300.
Purvis, A. and A. Rambaut, 1995. Comparative analysis by independent contrast (CAIC): an
Apple Macintosh application for analyzing comparative data, Cabios, vol. 11, p. 247-251.
Qiang, J., L. M. Chiappe, and J. S. An, 1999. A new late Mesozoic Confuciusornithid bird from
China. Journal of Vertebrate Paleontology, vol. 19, p. 1-7.
418

Rees, J. and Lindgren, J., 2005, Aquatic birds from the upper Cretaceous (Lower Campanian) of
Sweden and the biology and distribution of hesperornithiforms: Palaeontology, vol. 48, p.
1321-1329.
Russell, D., 1967, Cretaceous vertebrates from the Anderson River NWT: Canadian Journal of
Earth Science, vol. 4, p. 21-38.
Sanchez, J., 010, Late Cretaceous (Cenomanian) Hesperronithiformes from the Pasquia Hills,
Saskatchewan, Canada [M.S. thesis]: Ottawa, Carleton University, 257 p.
Seeley, H.G., 1876, On the British fossil Cretaceous birds: Quarterly Journal of the Geological
Society of London, vol. 32, p. 496-512.
Shufeldt, R., 1915, The fossil remains of a species of Hesperornis found in Montana: The Auk,
vol. 32, p. 290-294.
Sibley, C. G., and Ahlquist, J. E., 1990, Phylogeny and classification of birds: a study in
molecular evolution, Yale University Press, New Haven, Connecticut, 976 pp.
Simpson, G. G., 1980, Fossil birds and evolution in Campbell, K.E., ed., Papers in avian
paleonotology honoring Hildegarde Howard: Contributions in Science: Natural history
Museum of Los Angeles Co., vol. 330, p. 3-8.
Stolpe, M., 1935, Colymbus, Hesperornis, Podiceps: ein Vergleich ihrer hinteren extremitat:
Journal fur Ornithologie, vol. 83, p. 115-128.
Storer, R. W., 1958, Evolution in the diving birds: Proceedings of the 12
th
Intyernational
Ornithological Congress, vol. 2, P. 694-707.
Slowinski, J., 1993, ‘Unordered’ versus ‘ordered’ characters: Systematic Biology, vol. 42, p.
155-165.
419

Thiele, K., 1993, The holy grail of the perfect character: the cladistic treatment of morphometric
data: Cladistics, vol. 9, p. 275-304.
Tickle, P.G., A.R. Ennos, L.E. Lennox, S.F. Perry, and J.C. Codd, 2007, Functional significance
of the uncinate processes in birds. The Journal of Experimental Biology, vol. 210, p. 3955-
3961.
Tokaryk, T., Cumbaa, S., and Storer, J., 1997, Early Late Cretaceous birds from Saskatchewan,
Canada: the oldest diverse avifauna known from North America: Journal of Vertebrate
Paleontology, vol. 17, p. 172-176.
Wellnhofer, P., 2008, Archaeopteryx, Der Urvogel von Solnhofen (in German), Verlag Friedrich
Pfeil, Munich.
Wiens, J. J., 2003, Missing data, incomplete taxa, and phylogenetic accuracy: Systematic
biology, vol. 52, p. 528-538.
Wiens, J. J., 2005, Missing data and the design of phylogenetic analyses: Journal of Biomedical
Informatics, vol. 39, p. 34-42.
Wilcox, H. H., 1952, The pelvic musculature of the loon, Gavia immer: The American Midland
Naturalist, vol. 48, p. 513-557.
Witmer, L., 1990, The craniofacial airsac system of Mesozoic birds (Aves): Zoological journal
of the Linnean Society, vol. 100, p. 327-378.
Whetstone, K. N., 1983, Braincase of Mesozoic birds: I. new preparation of the ‘London’
Archaeopteryx: Journal of Vertebrate paleontology, vol. 2, p. 439-452.
Young, J.Z., 1950, The Life of Vertebrates, Oxford University Press, London.
420

Zeffer, A., L.C. Johansson, and A. Marmebro, 2003, Functional correlation between habitat use
and leg morphology in birds (Aves), Biological Journal of the Linnaean Society, vol. 79, p.
461-484.
Zeffer, A. and U. Lindhe Norberg, 2003, Leg morphology and locomotion in birds: requirements
for force and speed during ankle flexion, The Journal of Experimental Biology, vol. 206, p.
1085-1097.
Zhang, F. and Z. Zhou, 2000, A primitive enantiornithine bird and the origin of feathers, Science,
vol. 290, p. 1955-1959.
Zhang, F., P. Ericson, and Z. Zhou, 2004, Description of a new enantiornithine bird from the
Early Cretaceous of Hebei, northern China. Journal of Earth Sciences, vol. 41, p. 1097-
1107.
Zhang, F., Z. Zhou, and M. J. Benton, 2008, A primitive confuciusornithid bird from China and
its implications for early avian flight, Science in China Series D: Earth Sciences, vol. 51, p.
625-639.
Zhou, Z., 2004, The origin and early evolution of birds: discoveries, disputes, and perspectives
from fossil evidence: Naturwissenschaften, vol. 91, p. 455-471.
Zhou, Z., J. Clarke, and F. Zhang, 2008. Insight into diversity, body size, and morphological
evolution from the largest Early Cretaceous enantiornithine bird. Journal of Anatomy 212:
565-577.
Zhou, Z., F. Jin, and J. Zhang, 1992, Preliminary report on a Mesozoic bird from Liaoning,
China. Chinese Science Bulletin, vol. 37, p. 1365-1368.
Zhou, Z. and F. Zhang, 2001, Two new ornithurine birds from the Early Cretaceous of western
Liaoning, China, Chinese Science Bulletin, vol. 46, p. 1261-1269.
421

Zhou, Z. and F. Zhang, 2002, Largest bird from the Early Cretaceous and its implications for the
earliest avian ecological diversification, Naturwissenschaften, vol. 89, p. 34-38.
Zhou, Z. and F. Zhang, 2002, A long-tailed seed-eating bird from the Early Cretaceous of China,
Nature, vol. 418, p. 405-410.
Zhou, Z. and F. Zhang, 2005. Discovery of an ornithurine bird and its implications for Early
Cretaceous avian radiation, Chinese Academy of Sciences, vol. 102, p. 18998-19002.
Zhang, F., Z. Zhou, L. Hou, and G. Gu, 2001, Early diversification of birds: evidence form a new
opposite bird, Chinese Science Bulletin, vol. 46, p. 945-949.

Abstract (if available)

Abstract The discovery of abundant Mesozoic avian fossils, beginning with Archaeopteryx in the 1800’s and increasing dramatically since the 1990’s with the discovery of numerous Chinese fossils, has provided researchers with sufficient quantities of specimens to study the evolution and ecology of ancient birds using phylogenetic and morphometric methods. This study approaches the evolution and ecology of Mesozoic birds from two perspectives – a comprehensive analysis of the Hesperornithiformes, a highly specialized group of diving birds, and a series of morphometric analyses of modern and Mesozoic birds designed to find correlations between ecologic niche partitioning and morphometric trends. ❧ Despite being one of the most taxonomic, geographic, and stratigraphically diverse groups of Mesozoic birds, the Hesperornithiformes have received virtually no comprehensive study since the initial discovery of Hesperornis, Baptornis, and Enaliornis in the late 1800s. This lack of study has resulted in a confusing array of taxa organized into a taxonomic framework beset by errors in description, contradictions, and redundancy. Furthermore, little work has focused on evolutionary relationships among hesperornithiforms, leading to virtually no understanding of their phylogenetic interrelationships. While hesperornithiforms have an extensive fossil record that would be appropriate for morphometric analyses, none have been performed to date, despite the use of numerous morphological features that could be described quantitatively, but instead are treated qualitatively as diagnostic features in the current taxonomic framework. Therefore, the objectives of the first portion of this dissertation are to evaluate and update the current taxonomic framework of hesperornithiform birds (Chapter 1), conduct the first cladistic analysis of the Hesperornithiformes (Chapter 2), identify morphometric trends that may have diagnostic utility (Chapter 3), and integrate these studies to develop a new taxonomic framework informed by the phylogenetic relationships and morphometric patterns identified (Chapter 4). ❧ Among modern birds, morphometric data have been used in a variety of ways in an attempt to correlate ecology with morphology. This study seeks to build on previous work first through the analysis of a new Late Cretaceous ornithuromorph, Hollanda luceria (Chapter 5), and then through a broader analysis of a wide variety of Mesozoic birds (Chapter 6). The goals of these studies are to first test the correlation of ecologic niches with fore- and hind- limb measurements using multivariate statistics, and then to analyze Mesozoic birds in relation to the modern avian morphospace.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Benthic and pelagic marine ecology following the Triassic/Jurassic mass extinction

PDF

A systematic review of Enantiornithes (Aves: Ornithothoraces)

PDF

The early Triassic recovery period: exploring ecology and evolution in marine benthic communities following the Permian-Triassic mass extinction

PDF

Assessing the quality of the fossil record using a phylogenetic approach

PDF

Benthic paleoecology and macroevolution during the Norian Stage of the Late Triassic

PDF

Evolution of the dinosaur flight feather: insights from 3-dimensional fossil feathers

PDF

The evolution and functional significance of the cetacean pelvic bones

PDF

Dental evolution and tooth cycling in Mesozoic birds

PDF

A phylogenetic analysis of oological characters: A case study of saurischian dinosaur relationships and avian evolution

PDF

Integrated approaches to understanding diversification through time using sea urchins as a model system

PDF

Paleoenvironments and the Precambrian-Cambrian transition in the southern Great Basin: Implications for microbial mat development and the Cambrian radiation

PDF

Ecological recovery dynamics of the benthic and pelagic fauna in response to extreme temperature events and low oxygen environments developed during the early Triassic

PDF

Community paleoecology and global diversity patterns during the end-Guadalupian extinction (middle-late Permian) and the transition from the Paleozoic to modern evolutionary faunas

PDF

The geobiology of fluvial, lacustrine, and marginal marine carbonate microbialites (Pleistocene, Miocene, and Late Triassic) and their environmental significance

PDF

Bioturbation in Cambrian siliciclastic shelf strata: paleoecological, paleoenvironmental, and temporal patterns

PDF

Paleoenvironmental and paleoecological trends leading up to the end-Triassic mass extinction event

PDF

Stromatolites in the ancient and modern: new methods for solving old problems

PDF

Evolution of body size and sexually dimorphic traits in North American Gopherus tortoises

PDF

Paleoecology of Upper Triassic reef ecosystems and their demise at the Triassic-Jurassic extinction, a potential ocean acidification event

PDF

The geobiological role of bioturbating ecosystem engineers during key evolutionary intervals in Earth history

Asset Metadata

Creator Bell, Alyssa K. A. (author)

Core Title Evolution & ecology of Mesozoic birds: a case study of the derived Hesperornithiformes and the use of morphometric data in quantifying avian paleoecology

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Publication Date 03/27/2014

Defense Date 09/09/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Bird,Evolution,Hesperornithiformes,morphometrics,OAI-PMH Harvest,paleontology,phylogenetics

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bottjer, David J. (committee chair), Chiappe, Luis M. (committee chair), Corsetti, Frank A. (committee member), McNitt-Gray, Jill L. (committee member)

Creator Email alyssa_wjc@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-331584

Unique identifier UC11296993

Identifier etd-BellAlyssa-2055.pdf (filename),usctheses-c3-331584 (legacy record id)

Legacy Identifier etd-BellAlyssa-2055.pdf

Dmrecord 331584

Document Type Dissertation

Format application/pdf (imt)

Rights Bell, Alyssa K. A.

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 a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA