Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The ecology of eight species of intertidal crabs of the family xanthidae in the Marshall Islands
(USC Thesis Other)
The ecology of eight species of intertidal crabs of the family xanthidae in the Marshall Islands
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
THE ECOLOGY OF EIGHT SPECIES OF
INTERTIDAL CRABS OF THE FAMILY XANTHIDAE
IN THE MARSHALL ISLANDS
by
Alan Douglas Havens
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Biology)
February 1974
UN IVER SITY O F S O U TH E R N C A LIFO R N IA
TH E G R A DU A TE SCHO O L
U N IV E R S IT Y PARK
LOS A N G ELES. C A L IF O R N IA 9 0 0 0 7
This dissertation, written by
........ A_LAN _DOU_GLAS HAVENS........
under the direction of h.i=§... Dissertation Com
mittee, and approved by a ll its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
DISSERTATION COMMITTEE
Chairman
CONTENTS
CHAPTER I. INTRODUCTION........... 1
The Diversity of Xanthidae on Coral Reefs (1)
— The Niche Concept (2)--The Species Diversity
Problem (6)— Factors Influencing Diversity of
Decapod Crustaceans (8)— Factors Contributing to
Diversity of Reef Organisms (12)— Background and
Focus of the Present Study (17)— Previous Literature
on Indo-Pacific Xanthidae (21)— Methods and Materials
(21)— Definition of Terms (24)— Statistical Techniques
(26)— Summary (33)
CHAPTER II. LOCATION OF STUDY AREAS AND DESCRIPTION
OF THE PHYSICAL AND BIOTIC ENVIRONMENT...................35
Introduction (35)— Descriptions of the Atolls (40)
— Descriptions of the Study Areas (50)— Comparisons
of the Reefs in the Northern and Southern Marshalls
(105)— Summary (107)
CHAPTER III. TAXONOMY, BIOMETRICAL, AND POPULATION
DENSITY DATA FOR THE STUDY SPECIES...................... 108
Introduction (35)— Descriptions of the Study
Species (109)— Descriptions of Other Intertidal
Species (149)— Species Incertae Sedis (157)—
Discussion and Conclusions (159)
CHAPTER IV. VERTICAL ZONATION .......................... 161
Introduction (161)--Methods and Terminology
Employed (167)— Infratidal Fringe Xanthids (170)
— Intertidal Xanthids on Reef and Conglomerate
Flats (173)— Intertidal Xanthids on Eroding
Conglomerate Edges and Similar Topographic
Features (178)— Intertidal Xanthids on Lagoon
Beachrock (184)— Comparison of Zonation on Different
Types of Consolidated Rock Habitats (187)— Inter
tidal Xanthids on Rubble Beaches and Flats (188)
— Comparisons Between Consolidated Rock and Loose
Rubble Habitats (199)— Temperature and Vertical
Zonation (200)— Conclusions (202)
ii
CHAPTER V. SUBSTRATE
206
Introduction (206)— Substrate-related Crab Behavior
in Reef Situations (206)— Defensive Behavior
Associated with Holes (210)— The Crab-Hole Hypothesis
and Methods of Analysis (211)— Crab-Hole Size and
Shape Relationships (217)— Ecological Separations
Due to Hole Size and Hole Shape (223)— Ecological
Separations Due to Differences in Sediment Cover
and Topography (231)— Substrate-related Crab Behavior
in Rubble Situations (254)— Functional Morphology
of Rubble Crabs (260)— Ecological Separations Due to
Rock Size (262)— Ecological Separations Due to Loose
ness of Rocks and Under-Rock Sediment in Rubble
Situations (270)— Comparisons of Crabs in Consoli
dated and Loose Substrates (273)— Conclusions (278)
CHAPTER VI. FEEDING HABITS AND BEHAVIOR .............. 282
Introduction (282)— Observations on Temporal and
Spatial Differences in Activity (282)— Ecological
Separations Due to Behavioral Differences (288)—
Observations on Feeding Habits (289)— Stomach
Content Analyses (294)— Feeding Habits of the
Individual Species, Based on Stomach Contents (303)
--Categories of Feeding Behavior (306)— Ecolo
gical Separations Due to Feeding Habits (317)—
Conclusions (319)
CHAPTER VII. CONCLUSIONS .............................. 324
Ecological Characteristics of the Eight Study
Species (324)— Comparisons with Previous Literature
on Coral Reef Xanthids (331)— Previous Records of
Hole-related Crab Behavior (343)— The Relative
Importance of Factors Responsible for Ecological
Separations Between Intertidal Xanthid Crabs (344)
— The Role of Intertidal Xanthid Crabs in the Reef
Ecosystem (360)— Predation on Xanthid Crabs (363)
— Predation and Diversity of Xanthid Crabs (369)—
Interpretation of the Xanthid Crab Findings in the
Light of Previous Work on Crab Ecology and on the
Diversity of Reef Organisms (371)
CHAPTER VIII. SUMMARY .................................. 376
WORKS CITED................................................ 382
APPENDIXES 401
LIST OF TABLES
Table Page
1. Species Collected in the Marshall Islands,
with Listing of Habitats......................... 17
2. Mean Widths of Crabs.............................112
3. Mean Size of Male and Female Crabs......... 114
4. Size Range and Sex Ratios of Male and Female
Crabs .•.••••».*• •••.. •••• 115
5. Population Density Estimates ........... 117
6. Maximum Population Densities of Intertidal
Crabs on Reef Flats and Higher Conglomerate
Platforms ........................................177
7. Ecological Separations with Respect to
Vertical Zonation on Reef Flats and Higher
Conglomerate Platforms ......................... 179
8. Representative Population Density Data for
Intertidal Crabs on Eroding Algal Ridge
Remnants and Margins of Conglomerate
Platforms at Different Vertical Levels .... 183
9. Ecological Separations with Respect to Vertical
Zonation on Eroding Algal Ridge Remnants and
Margins of Conglomerate Platforms ........... 185
10. Population Densities of Crabs on Rubble
Beaches, with Respect to Vertical Zonation . . 193
11. Ecological Separations with Respect to
Vertical Zonation Between Rubble Crabs at
Station 4 3 ........................................194
12. Ecological Separation with Respect to
Vertical Zonation Between Rubble Crabs at
Station 5......................................... 196
13. Summary of Vertical Zonation Data for Eight
Species of Crabs................................ 204
iv
Table Page
14. Size of Crab Holes for 5 Intertidal
Species...........................................218
15. Shape of Crab Holes for 5 Intertidal
Species...........................................219
16. Ecological Separations with Respect to
Hole Size.........................................232
17. Ecological Separations with Respect to
Hole Shape ...............................234
18. Ecological Separation Between Dacryopilumnus
rathbunae and Zozymodes biunguxs wxth Respect
to Topography and Sedxment Cover 24 0
19. Ecological Separation of Dacryopilumnus
rathbunae from Pseudozius caystrus, Xantho
leptodon, and Xantho sanguineus with Respect
to Topography......................................243
20. Ecological Separation of Eriphia scabricula
from Xantho leptodon and Xantho sanguxneus
with Respect to Topography andSediment
Cover............................................... 247
21. Ecological Separation Between Xantho leptodon
and Xantho sanguineus with Respect to
Topography and Sediment Cover .................. 250
22. Ecological Separation Between Lydia annulipes
and Pseudozius caystrus with Respect to
Topography ~ ......................................255
23. Size of Rocks Inhabited by 5 Intertidal
Species.............................................263
24. Numbers of Crabs Under Rocks in Rubble
Habitats...........................................264
25. Ecological Separations with Respect to Rock
S i z e ............................................... 268
26. Ecological Separation Between Xantho leptodon
and Xantho sanguineus with Respect to
Looseness of Rocks Under 23cm in Diameter . . . 272
27. Ecological Separation Between Pseudozius
caystrus and Xantho gracilis with Respect to
Type of Sediment Under R o c k s ....................274
v
Table Page
28. Ecological Separation of Xantho gracilis
from Dacryopilumnus rathbunae and Eriphia
scabricula with Respect to Substrate Type . .279
29. Summary of Substrate Data for the Eight
Study Species..................................28 0
30. Percentage of Crabs Containing Various Kinds
of Algae.......................................... 297
31. Percentage of Crabs Containing Various Kinds
of Animal Foods, Compared with Total Algae . . 300
32. Percentage of Crabs Containing Algae and
Animal Foods .................................... 308
33. Ecological Separations with Respect to
Differences in Utilization of Algae and
Animal Foods .................................... 311
34. Ecological Separations with Respect to
Differences in Utilization of Siphonaria
and Sipunculids................................. 316
35. Summary of Behavioral Data and Feeding Habits
for the Eight Study S p ecies ....................322
36. Summary of Habitats and Habits of the Eight
Study Species....................................329
37. Relative Importance of Different Ecological
Factors in Ecological Separation .............. 348
vi
LIST OF FIGURES
Figure Page
1. Diagram of Percentage of Ov e r l a p..............2 9
2. Diagram of Relative Abundance ................... 2 9
3. Diagram of Typical Windward Reef in the
Marshall Islands ................................ 38
4. Map of Eniwetok Atoll............................. 41
5. Map of Kwajalein A t o l l ...........................47
6. Map of Majuro Atoll..............................4 8
7. Diagram of the Seaward Reef at Eniwetok
I s l a n d .............................................53
8. Algal Ridge at Eniwetok Island Station 2 . . . 58
9. Decadent Algal Mound Behind Algal Ridge
at Eniwetok Island Station 2 58
10. Inner Reef Flat at Eniwetok Island
Station 2 .......................................... 61
11. Reef Flat at Eniwetok Island Station 4 .... 61
12. Inner Reef Flat at Eniwetok Island
Station 4 .......................................... 64
13. Innermost Reef Flat at Eniwetok Island
Station 4 .......................................... 64
14. Upper Surface of Rock Groin at Aniyaanii
Island Station 1 3 A ............................... 69
15. Lower Surface of Conglomerate Slab at
Runit Island Station 1 4 B ........................ 69
16. Pitted Zone on Innermost Reef at Eniwetok
Island Station 5 D ................................. 72
17. Margin of Conglomerate Platform at Rigili
Island Station 1 8 ................................. 72
vii
Figure Page
18. Eroding Conglomerate Remnants at Japtan
Island Station 1 2 A ...............................74
19. Boring Organisms from Conglomerate at Japtan
Island Station 1 2 A ...............................74
20. Beach Conglomerate at Rojoa Island
Station 1 6 A ........................................7 6
21. Loose Beachrock Slabs Along Shore at
Runit Island Station 1 4 B 7 6
22. Lagoon Beachrock at Eniwetok Island
Station 7 B ........................................78
23. Eroding Edge of Lagoon Beachrock at Runit
Island Station 1 4 D ...............................78
24. Reef Block on Outer Reef Flat at Igurin
Island Station 2 1 ................................. 81
25. Rubble Field at Japtan Island Station 12B . . . 81
26. Rubble Beach at Japtan Island Station 12C . . . 85
27. Rubble Field South of Rock Groin at
Aniyaanii Island Station 13A 85
28. Cut-Off Pool Behind Rubble of Seaward
Reef at Igurin Island Station 2 1 ............... 88
29. Algal Ridge at Arniel Island Station 43 .... 93
30. Innermost Reef Flat at Uliga Island
Station 4 7 ........................................ 93
31. Rubble-Filled Tide Pool at Kwajalein
Island Station 3 1 ................................. 96
32. Reef Flat with Overhanging Ledges at
Dalap Island Station 4 8 A ........................ 96
33. Distant View of Rock Groin at Uliga Island
Station 47, from Hotel Roof......................99
34. Dark Green Pitted Zone, Seaward Margin of
Rock Groin at Uliga Island Station 47 ..... 99
35. Porous Flanks of Rock Groin at Uliga
Island Station 4 7 ................................ 102
viii
Figure Page
36. Eroding Conglomerate Along Inner Reef
Channel at Rotoin Island Station 51 102
37. Boulder on Reef North of Kwajalein Island
Station 3 3 ............. 103
38. Rubble in Cut-Off Pool Behind Seaward Reef at
Dalap Island Station 4 8 A ........................ 103
39. Dacryopilumnus rathbunae ....................... 110
40. Eriphia scabricula .............................. 121
41. Lydia annulipes..................................123
42. Pseudozius caystrus ............................ 128
43. Xantho gracilis..................................131
44. Xantho leptodon..................................134
45. Xantho sanguineus .............................. 139
46. Zozymodes biunguis .............................. 143
47. Eriphia sebana.................................... 150
48. Pseudozius pacificus ............................ 152
49. Zozymodes pumilis .............................. 155
50. Reef Flat Xanthids from Station 4 7 ..............158
51. Diagram of Putty Plug and Crab H o l e ........... 214
52. Diagram of Crab and Hole Measurements .... 220
53. Scatter Diagram of Crab and Hole Size,
for Two Reef Crabs............................... 222
54. Hole Size Distribution, for Crevice-
Inhabiting Xanthids.............................224
55. Hole Shape Distribution, for Crevice-
Inhabiting Xanthids.............................228
56. Carapace Shapes of Two Species of
Reef Crabs........................................ 237
ix
Figure Page
57. Rock Size Distribution, for Rubble-
Inhabiting Xanthids.......................... 26 6
58. Substrate Preference of Pseudozius caystrus
and Xantho gracilis...........................275
x
LIST OF APPENDIXES
Appendix Page
I Station L i s t ..................................401
II Reef cind Beach Transects.................... 406
III Temperature Measurements in Various
Intertidal Habitats in 197 0 ................ 411
IV Mean Crab Size and Hole Size Compared
for 5 Species of Reef Crabs................. 413
V Formula for Correlation Coefficient
(Correlation of Paired XY Data) ........... 414
VI Correlation Between Crab Size and
Hole Size......................................415
VII Mean Crab Shape and Hole Shape Compared for
5 Species of Reef Crabs, and Correlation
Between Crab Shape and Hole Shape...........417
VIII Relationship Between Crab Shape and
Habitat T y p e ................................. 418
xi
ACKNOWLEDGMENTS
I wish to thank first of all my major professor,
Dr. John S. Garth, who was instrumental in enabling me to
work at Eniwetok, for his help with the taxonomic work;
without his initial investigations of the crab fauna of
Eniwetok, this study would not have been possible. I also
wish to thank the other members of my committee, Dr. Ber
nard C. Abbott, Dr. Gerald J. Bakus, Dr. William H. Easton,
and Dr. Basil G. Nafpaktitis, for reading my dissertation
and offering their helpful advice. Dr. Bakus and I have
had many discussions of tropical marine ecology; he has in
spired many graduate students, including myself, with his
enthusiasm for this subject. Special thanks are also due
to Dr. Easton for his help in interpreting the geology of
reef habitats included in this study.
I would also like to thank Dr. William Stephenson for
reading my dissertation and offering his critical advice;
Dr. Michael a . Chartock, with whom I spent three summers of
field work at Eniwetok, for his assistance in the field and
for many discussions of marine ecology; and Mr. Elliott A.
Norse for fruitful discussions of some of the theory in
volved. Thanks are also due to Dr. Michael Risk for aid in
some of the statistical work, and to Dr. Bernard C. Streh-
xii
ler for the use of a computer. Finally, I would like to
thank Mr. Douglass Birch for his help on some of the other
statistical analyses, and for his aid in some of the photo
graphic work, and my father, Mr. Kenneth B. Havens, for
his help in photographic reproduction of some of the il
lustrative material, and for proofreading parts of the
dissertation. Special thanks are also due to my chief
typist, Mrs. Diane Woods, for researching the correct form
for the final typing of the dissertation.
xiii
CHAPTER I
INTRODUCTION
The Diversity of Xanthidae
on Coral Reefs
Crabs of the family Xanthidae are among the most di
verse* and abundant members of tropical coral reef commun
ities, living in live and dead corals, in holes and
crevices in the reef rock, and under loose coral rubble.
Borradaile (1903), in discussing crabs from the reefs of
the Maldive and Laccadive Islands, noted that of all the
families, the Xanthidae are most numerous in terms of
genera and species, and the most varied in form. Garth
(1964) has stressed the preponderance of xanthid crabs at
Eniwetok Atoll in the Marshall Islands; he noted that of
14 families, 67 genera, and 112 species of crabs known
from Eniwetok at that time, the family Xanthidae alone com
prised 30 of the genera (45 percent) and 63 of the species
(56 percent). In addition to being numerous in terms of
species, they are also very abundant in terms of indivi-
*The term "species diversity" as it is used here
refers to total number of species per unit area (species
variety or species density), and not to a theoretical index
which takes into consideration the relative abundance of
species as well as the number of species.
1
2
duals; one species included in this study was found in
population densities of up to 242 crabs/m^ on the flat
reef surface.
Why are these crabs so diverse? Despite their great
abundance and apparent importance in tropical reef eco
systems, they have until recently received scant attention
from ecologists. Knudsen (1967) carried out an experi
mental study of the obligate coral commensals Trapezia
and Tetralia at Eniwetok, demonstrating the trophic rela
tionship between the crabs and their hosts. However,
little is known about the ecology of free-living forms,
and what factors are most influential in determining the
niches of the various species.
The Niche Concept
Before crab ecology is discussed, the niche concept
will first be considered. Evidently the first authors to
define the term niche were Grinnell (1924) and Elton (1927) .
According to Grinnell(1924), the ecologic or environmental
niche is an ultimate distributional unit, occupied by
just one species or subspecies. Grinnell (1917) had pre
viously discussed the environmental factors which delimit
species (e.cf. temperature, humidity, food supply, shelter)
and noted the critical nature of these in the evolution,
persistence, and extermination of species. He showed
that examination in detail of the ecology of a species
demonstrates the operation of one or several of many pos
sible factors in limiting distribution, noting that dif
ferent factors may constitute barriers in various sectors
of a species' range, and that the shifting of two such
critical factor-lines towards one another will eliminate
the species.
Elton (1927) used the term niche for the status of an
animal in its community; its place in the biotic environ
ment, and relations to food and enemies. Allee et al.
(1949) broadened the term niche to include plants as well
as animals, and distinguish a "habitat niche," which
defines the home or part of the physical environment
where an organism lives, from the "food niche," which
describes its feeding role in a community.
Closely related to the niche concept is the principle
of competitive exclusion. Darwin (1859) was clearly
aware of this and noted that the struggle for life is more
severe between the individual species of one genus, which
are similar in structure and habits, than between the
species of distinct genera. The hypothesis of competitive
control, that two species with the same niche requirements
cannot form stecidy-state populations in the same region
(Hutchinson and Deevey, 1949), was first shown theoret
ically by Volterra and confirmed experimentally by Gauss
(Hutchinson, 1944). This principle has found support in
observations on natural populations: for instance, Lack
(1944) found that closely related bird species often
occupy different geographical regions; if they inhabit the
same region, they occur in different habitats, and if they
are found in the same habitats, they have different
feeding habits and/or differ in size (species differing in
size eating mainly different foods).
Hutchinson (1957) has more recently speculated that
the "fundamental niche" of a species can be defined as an
N-dimensional hypervolume, defined by the limiting values
of a series of environmental variables. Hence, for
variable x^, the species lives between x'^ and x"-^; for
variable X2 , it occurs between x '2 and x"2 / and so on for
as many variables (xn) as affect the distribution of the
species. While this definition is not without problems,
as Hutchinson himself admits, it represents a means of
fitting the niche concept into a mathematical framework.
Some authors (e.cf. Simpson, 1964) apparently consider
the niche as a theoretical entity which exists whether or
not there is a species present to occupy it. This concept
originated with Grinnel (1924), who believed that when a
new niche arises, or when one is vacated, a new occupant
usually fills it. However, it must be noted that while
one may predict the "potential niche" of the species which
takes the place of a community member which becomes ex
tinct or reaches its geographical limits, the niche of
such a replacement species may be determined largely by
the functions of still-extant community members. However,
in cases when totally new environments appear or when
existing communities are drastically impoverished by mass
extinction, it would be difficult to predict what will
be the niches of future species when the community becomes
fully differentiated. It is important then not to con
fuse theoretical "potential niches" with known niches of
extant organisms.
Henceforth, the term "niche" will be restricted in
this study to the role of a species in a given community:
specifically, where it lives and what it does there. It
must be emphasized that a species' niche delineates its
role only in the context of its available food, competi
tors, predators, parasites, type of climate, and sub
stratum or topography present at a given time and in a
given part of its geographical range. Should any of these
variables change, the niche can also change. Support for
this concept is provided by Hobbs and Marchand (1943), who
have shown that the niches of certain fresh water crayfish
change radically with difference in latitude: thus
Procambarus blandingi acutus occurs in stagnant bogs,
ponds and slow streams in Indiana and Oklahoma, while in
Louisiana it inhabits swift-flowing pine wood streams
with sand bottoms; and Orconectes immunis in New York
lives on muddy bottoms in stagnant water, whereas in
Tennessee it is found on rocky bottoms in swift-flowing
6
streams.
There has been considerable discussion of the species
diversity question in recent ecological literature; Pianka
(1966) has discussed many of the theories (e.g. time,
spatial heterogeneity, stability, predation, productivity)
which have sought to explain regional differences in
diversity and in particular the question of why there are
so many species in the tropics. Some of these theories lie
beyond the scope of the present study, for the following
reasons: xanthid crabs are a predominantly tropical group;
only six species have been recorded as far north as
Chesapeake Bay in the North Atlantic, as compared with 66
species from Florida and the Bahamas, and 11 species have
been recorded from California as compared to 35 from
Panama on the Pacific coast of the Americas (Rathbun,
1930). It is possible that present-day instability com
bined with the severity of northern winters may slow the
evolution of this group in the temperate zone, or that
historical factors have reduced the number of species
there; but an alternative explanation may be that the
group originated in the tropics and due to certain factors
in the genetic make-up of its members, the latter have
been less successful than some other crustacean groups in
adapting to cold climates. As the object of this investi
gation is to look into the factors which allow the exis-
tence of many sympatric species of xanthid crabs in one
area of the Pacific, the question of why these crabs are
more diverse in the tropics requires little further dis
cussion here.
Several of the theories listed by Pianka (1966),
however, do have considerable bearing on this study.
Spatial heterogeneity has often been cited as an important
factor in allowing the coexistence of species in a given
geographical region; for example, MacArthur (1964) has
demonstrated that both horizontal and vertical components
of spatial heterogeneity (with respect to vegetation)
affect bird species diversity. The "predation theory,"
attributed by Pianka to Paine (1966) has considerable
bearing on the relation of spatial heterogeneity to di
versity. Paine theorized that local species diversity is
related to the efficiency with which predators prevent
the monopolization of major environmental requisites by
one species. This concept is based on the observation that
competition for space between sessile rocky-shore organisms
in the temperate zone is lessened by predation; Paine
showed that the removal of the starfish Pisaster resulted
in the crowding out by Mytilus and Mitella of other or
ganisms not eaten by the starfish, resulting in a reduction
in diversity. Similarly, Fryer (1959, 1960) believes
that in large African lakes, large offshore predators re
strict smaller fish species to inshore waters, facilitating
8
the isolation of population; in the absence of predation,
competition for food and living space would be increased,
exterminating some species and hence lowering diversity.
Factors Influencing Diversity
of Decapod Crustaceans
Most of the available information concerning the fac
tors which ecologically separate related species of crabs
and other benthic decapod crustaceans is in the form of
descriptive accounts of their habitats. Studies on fresh
water crayfishes indicate that habitat type is of great
importance in their ecology. Hobbs et al. (19 67) found
that loose rocks, debris, and banks suitable for burrowing
are important limiting factors; for instance, the abun
dance of Cambarus longulus longulus was found to be rough
ly proportionate to the number of rocks in the area where
they live. Rhoades (1962) found two Ohio stream species
to be separated by bottom type, one form living on lime
stone, the other on shale and sandstone stream bottoms.
Havens (19 66) found that neither rock cover nor presence
of fine sediment per se was responsible for separating two
Ohio stream crayfishes; Orconectes rusticus hides in the
crevices beneath rocks, even where they lie over soft mud,
while Cambarus bartoni may burrow under rocks in the same
localities, or in hard mud banks (but not where the mud
is too soft for burrowing). Hence, the mode of taking
cover may be as important as type of bottom in separating
9
species.
In the case of marine crabs, there is probably more
information available on intertidal grapsoid crabs than
any other group. MacNae and Kalk (19 62) and MacNae (19 63)
have provided some habitat data for species of Sesarma
and Uca in their discussion of the South African mangrove
faunas of Inhaca Island and Umngazana. For four species
of Sesarma, the patterns of ecological separation appear
complex; it seems that vertical zonation, tolerance of
fresh water, extent of vegetation cover, sediment type,
and the use of burrows or loose debris for refuge are
important factors. On the other hand, each of the five
species of fiddler crabs (Uca) appears to be specific for
one of the following habitats: lower sand flats, higher
dry sand flats, drier semi-shaded peripheries of mangroves,
semi-shaded muddy stream banks, and the dense shade within
mangrove thickets.
Crane (1941) has provided a number of ecological notes
on species of Uca on the west coast of the American tro
pics. Her data suggest that the 23 mainland species can
be separated by salinity gradients into at least two
groups, those from sheltered seashores and tidal flats,
and those from brackish and fresh water streams. For crabs
of the first group, the following apparently constitute
distinct habitat types (inhabited by one or more species):
sandy beaches with stones; muddy sand beaches or flats,
10
soft mud of open flats, mangrove mud partly shaded by new
shoots, deeply shaded mangrove mud, and clay-like mud flats
among mangroves. There is also suggestion that some mud-
flat species prefer locations near firmer sand-mud beaches,
upon which they display, while others do not have this re
quirement. For lower salinity situations, muddy banks and
white clay banks may be distinguished as habitat types.
While it is evident that spatial heterogeneity, in
volving sediment type and plant cover, is a major factor in
the ecology of these crabs, this may in part be due to the
water-retaining properties of different sediment grades or
the effects of different degrees of shading on evaporation
rates. Altevogt (in MacNae and Kalk, 1962) has suggested
that the organic content of the sediment as related to
efficiency in finding food is important in separating two
of the South African Uca species.
For xanthid crabs, there is relatively little infor
mation available. Ryan (1956) has published data on the
distribution of Chesapeake Bay mud crabs; Neopanope texana
sayi and Panopeus herbstii were found in higher salinity
water, and Eurypanopeus depressus and Rhithropanopeus har-
risii in waters of lower salinity, the last named ranging
into fresh water. While E. depressus was found to be more
abundant on oyster bars, R. harrisii was collected from a
greater variety of types of shelter. While Ryan (1956)
indicates that Panopeus feeds upon thin-shelled oysters,
11
the feeding habits of the other species are not given.
According to Tweed (personal communication, 1973), research
by the New Jersey Oyster Research Laboratory indicates that
mud crabs occur in population densities of 150 animals/m^
on oyster beds. The crabs appear to feed on a wide variety
of animal life, causing heavy mortality to oyster spat,
and feeding upon barnacles and sabellarid worms as well;
no obvious differences between species were discerned.
Observations I have made on Delaware Bay confirm the
importance of salinity in determining the distribution of
these species. Neopanope and Eurypanopeus are very abun
dant in oyster beds and in the red sponge Microciona; it is
probable that the shelter provided by these habitats is in
part responsible for the large populations of these crabs.
While the larger Panopeus is less common on oyster' beds, I
have found it in abundance in holes in peat-like Spartina
root-mud banks in salt marshes at Absecon Bay, New Jersey,
and under glacial cobbles at Davis Creek, Towd Point, Long
Island. Further study may reveal that substrate type as
well as salinity is important in separating these species.
Knudsen (I960), in his study of four California xan-
thids, found two distinct habitat types for species which
live on rocky ocean shores: the space between rocks and
the sediment below (inhabited by Cycloxanthops novemden-
tatus), and higher crevices between adjacent boulders
(where Paraxanthias taylori occurs). While Knudsen inves
12
tigated the feeding habits of the crabs and found that al
gae is their principal food, no differences between species
were indicated. Finally, some preliminary work I have
done on intertidal xanthid crabs at Puerto Penasco, Sonora,
Mexico, indicates that substrate and vertical zonation are
important factors in the ecological separation of species
in the northern Gulf of California.
Factors Contributing to Diversity
of Reef Organisms
Spatial heterogeneity can be expected to be of great
importance in the ecology of coral reefs, which have a
complex three-dimensional structure. Clausade (1970) has
commented on the proliferation of invertebrates (including
xanthid crabs) in crevices in the reef, and Garth (19 64)
believes the great variety of xanthids to be due in part
to the protection afforded its smaller members by branching
corals. Randall (1963) has shown the importance of the
reef structure as a shelter for numerous fishes; and Talbot
(1965) found the abundance of fish, in terms of both
species and individuals, increased with the percentage of
coral cover on the reef floor. The latter study indicates
that the type of cover provided is also important; reefs
with more coral species (or growth forms?) have more
species of fish.
Randall (1963) attributed the abundance of fishes on
his artificial reef to the shelter it provided them from
13
large, predatory fish, which prevented reef fishes from
lining on the adjacent, open sea-grass flats. He noted
that few small reef fishes occurred on this reef, probably
because the 5" square openings of the blocks used in its
construction were not small enough to exclude certain
predators (small groupers, moray eels). Bakus (1964, 1969)
believes that predation by fishes is the reason for the
widespread protective mechanisms and habits of coral reef
invertebrates; some of these defensive "strategies,"
notably burrowing nocturnal activity, cryptic coloration
and form, and rapid movements, are utilized by crabs.
Clausade (1970) considers the proliferation of inverte
brates on the reef to be due to the fact that they are
saved from the bulk of predatory fishes by their crevice-
dwelling habits.
The idea that predation is responsible for increasing
diversity on coral reefs has much merit, but may not be
applicable in all situations. The activities of grazing
and browsing fish, which eat many kinds of invertebrates
on exposed intertidal reef surfaces, either as larvae or
adults (Stephenson and Searles, 1960, Bakus, 1966), may
be responsible for decreasing the diversity of many groups
of sessile organisms. On the other hand, if the prolifer
ation of boring organisms on the reef (Otter, 1937) is
indeed an evolutionary response to the activity of fish
which consume exposed organisms, as Bakus (1969) suggests,
14
then the lowered diversity of exposed organisms may be
partially or wholly compensated by the increase in di
versity of boring forms. Insofar as other organisms may
live in unoccupied burrows, the increase in porosity re
sulting from the activity of boring forms may lead to an
increase in the diversity of other groups. In any case,
it is evident that the effect of predation on the diversity
of reef organisms must be related in part to spatial heter
ogeneity.
The fact that there are numerous hiding places for
many kinds of organisms in the reef environment does not
in itself mean that spatial heterogeneity, working in con
cert with predation, will necessarily be the primary
factor which allows the ecological separation of species
(i.e., that varying topography allows different species to
take refuge in habitats which differ in their dimensional
properties). There are alternative explanations for the
high diversity of various groups of reef organisms. Coral
reefs are among the most productive marine communities
(Bakus, 1969); high primary productivity, working in con
cert with the great amount of shelter available to reef
animals, may allow large numbers of individuals to exist
in a given area, and, given suitable conditions for speci-
ation, allow more species to be accommodated. The niche
parameters which actually separate species could relate
to feeding habits or other patterns of behavior, rather
15
than variations in the physical properties of refuges.
Spatial heterogeneity might for some groups be an indirect
cause of diversity in a different way: if dietary habits
are relatively restricted, the members of a group may live
in different habitats not because they have become adapted
to the physical properties of their respective habitats,
but because their specific food organisms have done so.
Thus far, little of the work done on tropical shallow-
water marine groups has been aimed at discovering how so
many sympatric species are able to coexist. Kohn (1959)
concluded that species of Conus differ significantly in at
least two of the following: type of food, nature of and
relation to the substratum, or distributional pattern, and
that these are the primary factors differentiating the
ecological niches of cone snails. More recently, Kohn
(1971) has found a higher diversity of cone snails on sub-
tidal reefs (9 to 24 co-occurring species), which have a
complex topography, a patchy substratum, and an equitable
climate, than on intertidal marine benches (5 to 9 co
occurring species), which provide a smooth, uniform sub
stratum and a more severe and variable climate. The di
versity is still lower (1 to 6 co-occurring species) in the
sandy substratum of bays, which provides space for burrow
ing but little topographic relief (Kohn, 1967). These
findings strongly suggest that habitat complexity is of
considerable importance (directly or indirectly) in influ-
16
encing the diversity of this group.
However, Kohn (1971) also reported more overlap in
microhabitat than in food type when he compared species
in several geographical localities. Although food and mi
crohabitat together were found to provide the most adequate
explanation of niche dimensionality, Kohn noted that co
occurring species specialize more for different foods than
for substrate, and food is likely to be somewhat more
important in allowing avoidance of interspecific compe
tition between adults. The diets of species which occur on
benches were determined to be more specialized than those
of reef species, which Kohn (1968) relates to accessibility
of food in the more uniform habitat; he later (1971) sug
gested that food specialization is a more efficient "feed
ing strategy" on benches, where the snails are more gener
alized in the use of substrate type.
Kohn (1971) has speculated that co-occurring predatory
invertebrates may tend to adopt a "strategy" of apportion
ing resources by specializing for more different prey
species than by subdividing habitats distinctly. Chartock
found that the four most common species of brittle stars
of the genus Ophiocoma at Eniwetok are non-selective
detritus feeders, occupying different microhabitats (in
Kohn, 1971); Kohn suggests that it may be more efficient
for such detritus-eating animals to specialize for micro
habitat patch types rather than for food type, whereas it
17
may be more efficient for predators to respond to appropri
ate food items.
Background and Focus of
tne Present Study
In 1968, I spent several months at the Eniwetok Marine
Biological Laboratory (EMBL) at Eniwetok Atoll, in the
Marshall Islands. In the course of an extensive program
of general collecting in a variety of intertidal and
shallow water habitats, I was struck not only by the great
diversity of xanthid crabs, but also by their secretive
habits and intimate relation to the substratum. Many of
the species were sufficiently abundant that they could be
associated with one or another reef habitat; these forms,
which will be mentioned later in this paper, are listed in
Table 1.
TABLE 1
SPECIES COLLECTED IN THE MARSHALL ISLANDS,
WITH LISTING OF HABITATS
Listing of Habitats
1. Algal Ridge.
2. Eroding Algal Ridge Remnants.
3. Reef Flats without Corals.
4. Eroding Margins of Conglomerate Flats.
5. Conglomerate Flats.
6. Seaward Beachrock.
7. Lagoon Beachrock.
8. Reef Blocks and Boulders.
9. Rubble Flats and Beaches„
10. Rubble Heaps.
11. Protected Coral Reefs.
12. Gravel Bars.
(M) Collected or Observed on Majuro Atoll
only.
18
TABLE 1 - Continued
Family and Species Habitat
Calappidae
Calappa hepatica (Linn.) 12
Majidae
Micippa sp. 3
Parthenopidae
Unidentified parthenopids 3
Atelecyclidae
Kraussia rugulosa (Krauss) 3
Portunidae
Portunus longispinosus (Dana) 3,11
Portunus sp. 3
Thalamita admete (Herbst) 3,11,12
Thalamita picta Stimpson 3,11
Thalamita sp. 1,3,7,10,11
Xanthidae
Actaea rufopunctata (A. Milne 1
Edwards)
Actaea speciosa (Dana) 1
Actaea tomentosa (H. Milne 3,10,11
Edwards)
Carpilius convexus (Forskitl) 11
Carpilius maculatus (Linn.) 1 (M)
Carpilodes bellus (Dana) 3,10,11
Carpilodes pallidus Borradaile 11
Chlorodiella cytherea (Dana) 3,10,11
Chlorodiella laevissima (Dana) 1,11
Chlorodiella nigra (Forsk^l) 11
"Chlorodius1 * miliaris A. Milne
Edwards 3 (M)
Cycloxanthops cavatus Rathbun 1
Dacryopilumnus emerita Nobili 1
Dacryopilumnus rathbunae Balss 2,4,7,8
Daira perlata (Herbst) 1 (M)
Domecia sp. 1/11
Eriphia' scabricula Dana 2,3,4,7,8
Eripbia~ sebana (Shaw and Nodder) 3,4,7,9,10
Etisus ‘ bifrontaiis (Edmondson) 3,9,11
Etisus demani Odhner 9,10,11
Etisus dentatus (Herbst) 10
Etisus electra- (Herbst) 11
Etisus frontai~is Dana 3 (M)
Etisus sp. 3 (M)
TABLE 1 - Continued
Family and Species Habitat
Globopilumnus globosus (Dana) 1
Lachnopodus s'ubacutus (Stimpson) 9,10,11
Lachnopodus tahitensis De Man 10
Liocarpilodei integerrimus (Dana) 1
Lydia annulip'es (H. Milne Edwards) 2, 4, 5, 6, 7
Medaeus elegans A. Milne Edwards 11
Medaeus" simplex A. Milne Edwards 11
Paraxanthias notatus (Dana) 1
Phymodius laysanl Rathbun 1
Phymodius nitidus (Dana) 10
Phymodius ungulatus (H. Milne Edwards) 10,11
Pilodius areolatus (H. Milne Edwards) 3,9,10,11
Pilodiui- flavus Rathbun 1
Pilodius pugil Dana 1,10,11
Pilumnus purpureus A. Milne Edwards 3 (M)
Pilumnus vespertTTio (Fabricius) 3 (M)
Pilumnus sp. 3 (M)
PseudozTus caystrus (Adams and White) 5,6,9,10
Pseudozius pacxfi c u s Balss 4
Tetralia sp. 1,11
Trapezia sp. 1,11
Xanthias lamarcki (H. Milne Edwards) 3,9,11
Xantho danae Odhner 9,10
Xantho gracilis (Dana) 6,9
Xantho leptodon (Forest and Guinot) 3,9
Xantho sanguineus (H. Milne Edwards) 3,6,9,10
Xantho waialuanus (Rathbun) 9
Zozymodes biunguis (Rathbun) 2,3,4,6,7
Zozymodes pumilis (Jacquinot) 2,4,8
Zozymus aeneus (Linn.) 3
Palicidae
Palicus sp. 3,12
Grapsidae
Pachygrapsus minutus A. Milne Edwards 1,2,3,4
Pachygrapsus planiffons De Man 9
Pachygrapsus plicatus (H. Milne Edwards) 1,2,3
Percnon abbreviation (Dana) 1
Percnon pilimanus CA. Milne Edwards) 1
Percnon planissTmum (herbst) 3,10
Plagusi'a speciosa Dana 1
Ocypodidae
Macropthalmus bosci (Audouin) 3
20
As a result of this preliminary investigation, eight
species of xanthid crabs which are abundant in the inter
tidal zone above the low spring-tide level were selected
for study: Zozymodes biunguis (Rathbun, 1907), Xantho
gracilis (Dana, 1852), Xantho leptodon (Forest & Guinot,
1961), Xantho sanguineus (H. Milne Edwards, 1834), Dacryo
pilumnus rathbunae Balss, 1938, Eriphia scabricula Dana,
1852, Lydia annulipes (H. Milne Edwards, 1834), and Pseudo-
zius caystrus (Adams & White, 1848). In many ways these
species are ideal for study, because of their relatively
small size and the accessibility of the habitats in which
they live. They are varied in form and habits, and range
from the low spring-tide level to the highest tidal
reaches; they provide an excellent opportunity to study the
ecology of certain common members of an important tropical
group.
It is the object of this study to investigate the
niches of eight species of tropical Indo-Pacific Xanthidae,
and ascertain which of certain easily measured factors in
their ecology (substrate, vertical zonation, food, and
temporal-spatial differences in behavior) is most important
in determining the ecological separations between species.
It is hypothesized that the spatial heterogeneity (sub
strate type) is the most important factor separating these
xanthid species.
21
Previous Literature on Indo-
Pacific Xanthidae
There have been numerous papers on the taxonomy and
geographical distribution of tropical Indo-Pacific xanthid
crabs; those papers most pertinent to the present study
will be mentioned in Chapter 3. Articles which treat the
ecology of free-living xanthids are mainly descriptive in
nature; the most detailed of these is Ward's (1933) study
of the crabs of Heron Island, on the Great Barrier Reef of
Australia, in which a fairly detailed account of the habi
tats of a number of crab species is presented. Other
papers of importance are those of T. A. Stephenson et al.
(1931), Edmondson (1946), Gibson-Hill (1947), Tweedie
(1950), Morrison (1954), and Ward (1965). The findings of
these and other authors will be considered at some length
in Chapter 7, where they will be compared with the results
of this study; they need not be discussed in greater detail
now. There are no detailed accounts of the biology of
free-living xanthids of the Indo-West Pacific comparable to
Knudsen's (1960) study of the California Xanthidae, which
provides detailed descriptions of habitats as well as
behavior.
Methods and Materials
The collecting techniques, means of identification of
crabs and other reef organisms of importance to this study,
and statistical methods which are necessary for an under
22
standing of the study as a whole are discussed in this
section and those immediately following. Other techniques
which apply only to one or another part of the study will
be covered in those sections where their discussion is
most relevant.
Xanthid crabs were sampled mainly at low tide, either
by turning over rocks and capturing them by hand, or by
saturating small areas of the reef or beachrock with forma
lin to drive hidden crabs from their holes. Collecting
with the latter technique revealed that many areas which
are superficially barren of animal life support large crab
populations. In some places, where the reef rock is very
hard, it is difficult or very time-consuming to collect
these crustaceans by any other means.
When tide pools are poisoned with formalin, xanthid
crabs and alpheid shrimps are among the first creatures
to emerge from their hiding places, and are captured
relatively easily. Soon after application of the formalin,
there is considerable commotion in the pools as animals
escape from their holes and attempt to find refuge in
places that are free from the noxious chemical; eventually
a large number of dead crabs can be retrieved from the
bottoms of the pools. Shrimps other than alpheids are
difficult to collect until they are thoroughly inactivated
by the formalin, and small fish often jump completely out
of the water to escape. Generally, the last organisms to
23
appear are axiids, which often re-enter their holes, pre
sumably to die inside. Although this method is relatively
effective as compared to other techniques for obtaining
specimens from well-solidified rock, 20 or 30 minutes may
intervene between the initial application of poison and
the cessation of movement in the pools, even though full-
strength formalin is used.
This sampling technique has the disadvantage that some
crabs may die inside their holes because they either did
not come out at all, or were frightened by movements of the
collector. The latter source of error can be reduced by
allowing the animals to emerge fully, and then blocking
off their hole apertures with a pair of forceps (though
this may be difficult when many animals appear simultan
eously) . When formalin is used, the population density
measurements so obtained are probably always underesti
mates, for these reasons.
Quadrats of l/16m and l/4m were employed in collect
ing data on population densities, the quadrats being laid
out in a random manner; in some cases, measured areas
larger than these quadrats were sampled. When smaller
quadrats were used, at least eight samples were collected
at each locality. Waterproof field gear (e.g.. bottles,
preservative, hammer and chisel) were transported in an old
laundry bag; unlike a bucket, the bag can be dropped acci
dentally without the danger of overturning and spilling its
24
contents.
A large number of xanthid and other crab species were
identified with the aid of the reference collections in
the Allan Hancock Foundation, which are based on earlier
field collections by John S. Garth, Fred C. Ziesenhenne,
and Jens W. Knudsen. In addition, papers by Borradaile
(1903), Rathbun (1906), Odhner (1925) , Balss (1938), Sakai
(1939), Forest and Guinot (1961), Guinot (1964), and Takeda
& Miyake (1968) were of help in identifying crab material,
and Dr. Garth identified some of the rarer crabs personal
ly. Some of the algae and other organisms were identified
with the aid of the E.M.B.L. reference collections. Sev
eral of the more valuable references on the algae of the
Marshall Islands and the Indo-Pacific in general are
Taylor (1950), Dawson (1954, 1956, 1967), Hollenberg
(1968), and Saito (1969). Encrusting coralline algae were
identified on the basis of gross morphology and by exami
nation of cell structure following decalcification with
FAA, Perenyi solution, or EDTA, embedding in paraffin, sec
tioning, and staining with hematoxylin. Boring barnacles
were identified with the help of Cannon (1935).
Definition of Terms
A number of terms denoting topographic features and
sediment types are used throughout the text that follows;
these will be defined here to avoid possible misunderstand
ing over their meaning. Loose sediments are classified by
25
the Wentworth Scale according to particle size (Green,
1968). In this study, sand (0.0625 to 2mm) was the finest
sediment encountered in investigations of crab habitats.
Sediments made up of particles coarser than sand but finer
than cobbles are termed "gravel;" this includes granules
as well as pebbles (2 to 64mm). The term "rubble" is
used to indicate a mixture of larger pebbles, cobbles, and
boulders; "rocks" is used here as a catch-all term for par
ticles of these dimensions. When it was of importance to
crab ecology, the actual dimensions of the rocks were
recorded.
The term "boulders" as used here refers to well-sorted
sediment of boulder size (>256mm) but of somewhat rounded
configuration; flattened pieces of rock, of whatever
length or width, are referred to as "slabs." McKee and
Weir (1953) have published a classification of stratified
rocks according to their cross-bedding; practically all
the "slabs" referred to in the course of this study are
derived from beachrock and are, therefore, of sedimentary
origin, falling into the slabby (5 to 60cm) and flaggy
(1 to 5cm) categories. The term "reef block" is used to
indicate large masses of rock of boulder size, but of ir
regular rather than rounded shape; when flat, they are not
of sedimentary origin.
The term "conglomerate" will be used to indicate a
type of rock found above the reef-flat level, and composed
26
of rounded fragments of varying size cemented together.
The term "rock groin" refers to an elongated platform of
conglomerate which is oriented perpendicular to the reef
edge, and "rock bar" indicates conglomerate or beachrock
which lies parallel to it.
"Algal turf" means a low mat (at least 2mm in thick
ness) of macroscopic filamentous algae, usually growing
upon the rock surface. "Algal film" refers to microscopic
algae which form a slippery coating on the rock, not over
1mm thick.
Tidal heights are given in meters above datum, unless
otherwise indicated by a minus sign (see Chapter 4).
Statistical Techniques
Wherever practicable, statistics were used in analyz
ing data on the distribution of crabs on the reef. Several
methods can be used to demonstrate differences in the dis
tributions of individual species with respect to environ
mental variables. In the course of the investigation, both
the extent of these differences and their statistical sig
nificance are shown.
Different methods were found to be necessary for deal
ing with habitats as opposed to micro-habitats. For pur
poses of this study, a habitat is a larger area with dis
tinct physical properties such that a large number of
individuals of a given species are able to live within it
(e_.c[. , reef flat, rubble-strewn beach); a micro-habitat is
27
also set off by physical characteristics but is such a
small area that only one or a few individuals of a given
species are able to live within it (e.g.., holes on the reef
flat, rocks on the beach). Within a habitat, there is two-
dimensional continuity in the spacing of the individuals of
a species in a given area? on the other hand, the same area
might have several micro-habitats, the individual units of
which are randomly distributed, so that several species
could exist in the area, each with a discontinuous distri
bution. It should be noted that two species within the
same habitat can be expected to live in different micro
habitats; on the other hand, two species which have the
same micro-habitat requirements can be expected to live in
different habitats.
For the micro-habitats considered here (hole size,
rock size), one finds a continuum in size from the largest
to the smallest of these refuges. The most suitable method
of analysis of the ecological separations between species
which are separated by differences in hole or rock size is
to collect paired data on the dimensions of the crabs and
their individual hiding places and divide the total range
of hole or rock measurements into convenient size classes.
A frequency distribution can then be tabulated, followed by
a calculation of the percentage of the total sample of a
given species which occurs in each size class. One can
then compare the distributions of two species (let us call
them species A and species B) and determine the amount of
overlap between them by adding together the percentage of
species A in size classes where higher percentages of
species B occur and the percentage of species B in size
classes where higher percentages of species A occur (Figure
1). This constitutes the percentage of overlap between
the two species; a corresponding percentage of separation
is derived by simply subtracting this value from 100 per
cent. This percentage of separation is an estimate of the
extent to which species A and species B are free from one
another for the environmental requisite in question. A
variant of analysis of percentage overlap will be used in
dealing with differences in feeding habits; the details of
this will be given in Chapter 6, where its discussion will
be more germane.
In dealing with habitat differences, percentage of
overlap is not a useful measure, because there is often a
sharp physical discontinuity between one habitat and
another. It is difficult if not impossible to estimate the
percentage of the total population of a given species
which lives in one habitat or another, because the pre
valence of certain habitats varies considerably from place
to place on one atoll, and from one atoll to another. In
comparing two species which live predominately in different
habitats, then, it is necessary to determine the relative
abundance of these forms in two habitats where they occur.
29
Species A
Percentage
of Crabs
Continuous Environmental
Variable
Hatching indicates
extent of overlap.
Figure 1 - Diagram of Percentage of Overlap,
100 Percent Level
r
Species B Species A
Species A
Species B
J
Habitat 1 Habitat 2
Hatching indicates extent of
overlap in each habitat.
Figure 2 - Diagram of Relative Abundance.
30
This is done by taking quantitative samples of crabs in
each habitat, and for each, dividing the total number of
individuals of the less common species by that of the more
common species and multiplying by 100 (Figure 2).
The analyses described above define the extent of the
differences in habitat preference between species, but do
not indicate whether sufficient data have been collected
to render the results statistically significant. For this
2
reason Chi tests are also employed, following the con
straint that expected numbers of observations should not
be less than 10 (Fisz, 1963). In all Chi2 calculations,
a 2 X 2 contingency table was used (with 1 degree of free
dom) , to allow two species to be compared for two habitat
or food types. Only when there is a difference in distri
bution that is significant at the .001 level are the re
sults of analysis of percentage overlap or relative abun
dance considered valid. However, as Kohn (1959) pointed
out, Chi tests may indicate significant differences even
when the degree of overlap is large, when the actual
numbers involved are large. For this reason, it is neces
sary to determine the actual extent of the overlap.
In determining which of the four factors (substrate,
vertical zonation, feeding habits, and temporal-spatial
patterns of behavior) is of greatest importance in deter
mining the ecological separations between species, two
methods are employed. One is to examine each factor and
31
determine how many categories (e.g;. for substrate these
might include large rocks and small rocks) there are with
respect to that factor; each category must be characterized
by at least one species of crab. The number of categories
for each factor can be compared, and the factor with the
greatest number of categories can be considered the one
most important in separating species.
The establishment of such categories requires some
mathematical cut-off point to be employed in assessing
the results of analyses of percentage of separation or
relative abundance. A level of 50 percent has been chosen
for this study; while this is admittedly arbitrary, it
enables comparisons of various environmental factors to be
made at the same level of significance. In the case of
percentage separation, two species which show more than 50
percent overlap (less than 50 percent separation) are not
considered to be separated with respect to the variable in
question. In the case of analysis of relative abundance,
two species A and B are not considered to be significantly
separated unless species A is less than 50 percent as abun
dant as species B in one habitat and species B is less than
50 percent as abundant as species A in the other habitat.
The second method of analysis of the relative impor-
I.
tance of factors is the comparison of each Species with
every other species and the evaluation in each case of
which factor is of the greatest importance in determining
32
the ecological separation between a pair of species. By-
totalling the number of cases in which a given factor is
most responsible for maintaining ecological separations
between species, one can establish which of the factors de
termines the greatest number of separations. This factor
can then be considered the one most responsible for the di
versity of the group. This method allows a determination
to be made of the extent to which the categories within
each factor contribute to the ecological separations be
tween species.
It should be noted that this method of analysis re
quires no arbitrary cut-off points to be employed. Rather,
in each case, the amount of overlap with respect to each
factor could be compared, and the factor which is respon
sible for the greatest amount of separation can be con
sidered the one most responsible for separating the pair of
species, even if there is broad overlap with respect to
this factor. In fact, as can be seen, it is not always
necessary to employ statistical data on percentage of sepa
ration or relative abundance: in cases in which there is
complete isolation with respect to substrate, the tidal
ranges of the species under consideration may suffice for
vertical zonation comparisons. In cases when measurements
of certain ecological factors either were not or could not
be taken, the closest subjective approximation must be
employed.
33
In determining which factor is most important, it must
be noted that in cases in which a pair of species never or
very seldom occur in the same habitats, the question of
behavioral differences between the species need not be
considered unless it is shown that the two species feed
predominately upon organisms which are limited to their
respective habitats. Should this be the situation, then
the habitat factor and feeding habits can be considered
together as primary factors. Only if there is no dif
ference in habitat or micro-habitat can a behavioral fac
tor be considered alone as a primary factor. If a pair of
species occurs in different proportions in several habi
tats, and also differes significantly in behavior, then the
habitat and behavioral factor may again be of equal sig
nificance.
Summary
Xanthid crabs are one of the most diverse groups of
tropical reef organisms, despite which their ecology is
but little understood. This study will seek to determine
how large numbers of sympatric xanthid crabs are enabled
to coexist and which of several factors (substrate, verti
cal zonation, feeding habits, and temporal-spatial dif
ferences in activity) is most important in determining the
ecological separations between species.
The concept of niche is briefly discussed, and some
of the factors which are known to be influential in de-
34
termining ecological separations in other benthic decapod
crustaceans are mentioned. While there is relatively
little information on xanthid crabs, the fact that spatial
heterogeneity is of considerable importance to crayfish
and grapsoid crabs, together with the complex three-
dimensional structure of coral reefs, suggests that spatial
heterogeneity (substrate type) will prove to be the most
important factor separating species of xanthid crabs as
well.
Eight intertidal species common at Eniwetok Atoll,
in the Marshall Islands, were selected for study; methods
of collecting and identification are described, including
a technique for poisoning with formalin which proved ex
tremely successful. Finally, the statistical methods
(analysis of percentage separation and relative abundance)
used to determine the extent of separation between species
are discussed, along with two means of assessing which
factor is of the most importance in ecologically isolating
species.
CHAPTER II
LOCATION OF STUDY AREAS AND DESCRIPTION OF
THE PHYSICAL AND BIOTIC ENVIRONMENT
Introduction
Field work was carried out on three Atolls in the
Marshall islands: Eniwetok, Kwajalein, and Majuro. Eni-
wetok is located west of the main group of the Northern
Marshall Islands, at 11° 18-43' N and 162° 02-26' E.
Kwajalein, positioned at 8° 42' to 9° 24' N and 166° 49'
to 167° 47' E, is one of the Ralik (Sunset) chain of
atolls, while Majuro, at 7° 4-13' N and 171° 3-23' E be
longs to the Ratak (Sunrise) chain; both are in the South
ern Marshalls.
The weather in the Marshall Islands is dominated by
tradewinds, which blow from the east-northeast most of the
year, except for August and September when calms occur,
with light winds coming from a variety of directions. Ac
companying the trade winds is the North Equatorial Current,
flowing steadily from the east. The average air tempera
ture in the Northern Marshalls is 82°F (27.7°C); the sun
is intense and the sky usually clear with scattered cumulus
clouds. Rainfall increases from north to south (towards
the equator) in the Marshalls group; mean annual pre
35
36
cipitation is 50" (127cm) at Eniwetok, increasing to 96"
(244cm) at Kwajalein, and is heaviest in the fall. Rela
tive humidity is high throughout the year (Fosberg, 1956).
During the period when the trade winds blow, the
climate on the windward side of the atolls is pleasant,
owing to the strong breezes. At this time great clouds of
spray blow over the windward reef and island??. The dol
drums set in at Eniwetok in late July; the wind dies and
the climate becomes uncomfortable. On the leeward side,
the air normally feels hot and humid; in places where
trees cut off the slight breezes which characterize this
side of the atoll, it is especially unpleasant. During
the doldrum period there may be strong swell on the
south-facing reefs, caused by storms in the southern hemi
sphere (Emery et al., 1954); this was very noticeable on
Majuro, which is near the equator. It is not unusual for
the wind to shift to the south at this time, accompanied by
an overcast sky, storms, and heavy rains.
The tidal range in the Marshall Islands is greater
than that of most of Micronesia and Polynesia. Tides are
of the semi-diurnal type, there being two highs and two
lows per day. Mean sea level occurs at 2.6' (0.79m), 3.0'
(0.9m), and 3.2' (0.98m), respectively, for Eniwetok,
Kwajalein, and Majuro; these figures represent height above
datum. Mean spring tide ranges for these atolls are given
as 3.9' (1.28m), 5.0' (1.52m), and 5.3' (1.74m), in the
37
same order as above (U.S. Department of Commerce, 1970).
Henceforth, in this study, all tidal zonation will be given
in meters above or below datum; unless indicated by a
"minus" sign, all tidal heights are above datum.
Prior to describing the localities where collections
were made, it is necessary to provide a brief outline of
those morphological features of reefs that pertain to
atolls in the Marshall Islands. An atoll is essentially
a ringing reef (sometimes circular, but often of irregular
outline) reaching close to or above sea level, and en
closing a central lagoon whose bottom is markedly deeper
than that of the water on the reef flat. While it is
often difficult to distinguish between an atoll with a
shallow lagoon and a platform or table reef with a shallow
central depression, a useful criterion is the presence or
absence of a distinct hiatus or "drop-off" between the
enclosing reef and the depressed central area. When there
is such a hiatus, one can define the reef as an atoll;
when not, a platform reef. In actuality, this question of
lagoon depth does not present a problem in the Marshalls,
where maximum lagoon depths are usually over 66m.
The windward, or east-facing reef (Figure 3) is ex
posed to the pounding heavy surf most of the year, and
where there are no islands on the reef a strong current
sweeps across the reef flat to the lagoon at high tide.
The outer edge of the reef is the algal ridge, which
38
uhe
CROSS SECTION
Figure 3 - Diagram of Typical Windward Reef in
the Marshall Islands.
Vertical scale exaggerated.
39
varies in morphology from place to place: in some areas
it is a low-lying zone of vigorous coral and algal growth,
and in others, a series of high buttresses separated by
surge channels. Behind this is a shallow moat, called the
back-ridge trough, which lies just below the level of the
lowest tides; certain corals grow here in abundance. This
is followed by an intertidal reef flat, characterized by
tide pools; this zone is superficially barren-looking, and
small, filamentous algae are the dominant sessile organ
isms on the exposed rock surface. Corals cannot survive
on the intertidal reef flat, which often constitutes the
broadest zone on the windward reef.
The islands or islets, where they occur, are located
on the back part of the reef, and are formed from accumu
lations of coral rubble cast up by storm waves, or of
sand; in many places they adjoin platforms of coral con
glomerate which extend out across the reef from the island
shore. The lagoon sides of many windward islands are
sandy, and beachrock is often present; the latter is
formed by solidification of sand or coarser sediments
into a stratified rock and undoubtedly contributes to the
stabilization of the shoreline in many localities. Sand
or rubble horns or lobes often extend lagoonward from the
ends of the islands, and into the channels between the
islands. The sheltered lagoon water of these channels may
support a lush coral growth.
40
The leeward or west-facing reef is protected from the
heavy wave action which characterizes the windward side,
although there are waves of reduced size which have been
refracted around the atoll. A low algal ridge, with broad
pool-like areas rather than well defined surge channels is
often present on the leeward reef edge; corals predominate
over algae in many places. The reef flat is lower than
that of the windward side, and lush coral growth may be
present over much of this surface. Islands are less com
monly encountered than on the windward side. As this sec
tor of the reef is protected from the influence of the
North Equatorial Current, movement of water across the reef
is controlled for the most part by the tides.
Reefs having a southern exposure are more likely to
have islands than are those facing west or northwest.
They receive more wave action than other leeward reefs, and
in this sense resemble reefs of the windward side. They
lack, of course, the strong current which moves across the
east- and northeast-facing reefs at high tide.
Descriptions of the Atolls
Eniwetok
Eniwetok Atoll is 25 miles (40km) long by 20 miles
(32km) wide, and is roughly circular in outline but elon
gated in a north-northwest to south-southeast direction
(Figure 4). It has 39 islands and three navigable passes.
Eniwetok has been the scene of much recent human activity,
•ROMIT-
Kl&lU
^ m N
TWKY-
SRND
'BofrftMEGM
l&UK(N
ENlvJETOK
O
f c *
I Z. “ 3 4 S
■ * • ■ 1
IP
■KTOTUTE' rtlLeS
o ( 2.MS" 10
~ m d
KILOMETER
Figure 4 - Map of Eniwetok Atoll.
Collections were made at
islands indicated on the map.
42
both military and scientific; a major battle was fought
there during World War II, and later, nuclear testing was
conducted. For this reason, natural land vegetation re
mains only on the six western islands and on a few windward
ones (e.£. Aniyaanii). In other localities, a scrub vege
tation of Scaevola and Messerschmittia, with occasional
coconut palms, has grown where the original forest cover
was denuded.
There has in addition been some damage to the coral
reefs; Emery et al. (1954) noted the decadent nature of the
reefs from Japtan to Eniwetok Island, and off Engebi;
Tracey believed this was due at least in part to oil re
leased from damaged ships during the war (in Emery et al_.,
1954) . Taylor (1950), who visited the Northern Marshalls
in 1946, also noted the accumulation of oil, tar, and de
bris on the reefs adjacent to the southeastern islands of
this atoll and attributed the oil to wrecked tankers.
The shallow-water reefs of Eniwetok Atoll are divided
into three segments, broken by the Deep Channel, three-
quarters of a mile (1.2km) wide, between Parry and Japtan
Islands; the Wide Passage, six miles (9.6km) in width,
between Eniwetok and Igurin Islands, and the Southwest
Pass, just south of Rigili Island. The latter is partly
obstructed by patch reefs. Due to the large ocean swells
which come in through the Deep Channel, and also because
of the relatively great width of the lagoon, the latter has
43
a somewhat oceanic character: moderate-sized swells make
passage by small boats difficult at times, and influence
the biota as well. During the doldrum season, waves may
enter the lagoon through the Wide Passage when the weather
comes from the south.
The large southeastern islands of the atoll, Eniwetok
and Parry, are characterized by narrow reef flats, and the
islands themselves are elongate in a direction parallel to
that of the reef edge. The Eniwetok Marine Biological
Laboratory is located on Eniwetok Island. In 19 68 and
1969 it was near the middle point of the island, on the
lagoon side; in 1970 it was relocated to the seaward side,
adjacent to a new three-story dormitory farther north.
Although the reef has suffered destruction from the effects
of the war and subsequent construction work, this pertains
largely to reef-building organisms, not to the intertidal
crab fauna. A large number of crab habitats can be found
conveniently close to the marine laboratory, and for this
reason, a number of collections and most of the observa
tions on live crabs were made here. An especially good
locality for crab studies is the rock quarry at the north
end of the island, where artificial pools, beaches and
rubble piles facing the seaward reef simulate natural con
ditions elsewhere.
Parry Island is somewhat like Eniwetok, and like the
latter has a rock quarry on the reef, in this case at the
44
at the south end. There is a long stretch of reef between
Eniwetok and Parry, which can be waded at low tide; the la
goon edge of this reef has extensive Porites coral beds,
not seen adjacent to the islands themselves. A small tri
angular islet called Sand Island lies just north of Eniwe
tok; this provided an important study area. Farther north
along the same reef (midway between Sand and Parry Islands)
is an interesting complex of intertidal beachrock, conglom
erate, and rubble on the back part of the reef (station 10).
Most of the remaining windward islands of this atoll
are roughly triangular in shape, with the base of the tri
angle forming the lagoon shore and the apex pointing
towards the sea. These islands face east and northeast and
adjoin broad intertidal reef flats. A characteristic fea
ture of these reefs is the presence of numerous conglom
erate platforms jutting out from the island shores. Many
of these are elongate in a direction perpendicular to that
of the reef edge and may be termed rock groins. Where the
islands are relatively narrow, as at Aniyaanii Island, the
narrow channels between these platforms shunt the current
which flows across the reef at high tide in a diagonal di
rection, around the islands.
Aniyaanii Island lies roughly in line with the lagoon-
ward part of the reef; but the islands farther north, such
as Piiraai, Rojoa, and Biijiri, tend to lie behind the reef,
and are flanked by water at least 6m deep, with numerous
45
patch reefs. Runit Island is elongate, and formed in part
by sandy portions which connect several rock groins. The
seaward island shore faces a submerged or continuously
pooled inner reef flat, access to the exposed outer flats
being gained by walking out on the groins.
Japtan Island differs from the previously mentioned
windward island types in that it is rather round in shape,
and faces not only a seaward reef flat but also a narrow
reef next to the Deep Channel.
On the leeward side, there are several small tri
angular islands with a northwestern orientation; one of
these, Bogallua, was visited. Here, the seaward reef is
wider than elsewhere on the atoll, and a great part of it
is dominated by the blue coral, Heliopora. This reef does
not take the form of a solid flat, but is a maze of indi
vidual coral heads, pools, and meandering channels. On the
lagoon side, there is a fairly smooth, erosional reef flat
off the east end of the island, and farther west lies a
continuous, solid Porites reef which stretches far to the
west of the island.
The other leeward islands face southwest and are posi
tioned along a fairly narrow reef flat. Rigili is a small
island north of the Southwest Pass, and Grinem is similar
though south of the pass. The southernmost islands on
the leeward side, Boganegan and Igurin, are larger and
elongated. Igurin Island consists of three wider portions
46
connected by narrower necks of sand on the lagoon side; ad
jacent to the latter are vast seaward rubble fields which
retain water much above the low tide level in shallow pools.
There is no well-defined lagoon reef along this sector of
the atoll, though much loose rubble has accumulated in the
form of detrital flats along the northeast-facing island
shores.
Kwajalein
Kwajalein (Figure 5) is a very large atoll, 76 miles
(122km) long and roughly triangular in shape, with a north
west trend. Work on this atoll was restricted to the main
island, also named Kwajalein, which is C-shaped. In gen
eral, the reefs here are similar to those of Majuro; how
ever, a distinguishing feature of the reefs off Kwajalein
Island is the presence of a host of sea urchins (mainly
Echinometra mathaei) in numerous holes in the reef surface.
i
Majuro
Majuro Atoll (Figure 6) was much more extensively in
vestigated than Kwajalein; it is 25 miles (40km) long and
7 miles (11.2km) wide, and oriented in a west-northwest to
east-southeast direction. There are only two navigable
passes on the north side of the atoll; as the lagoon is
almost completely enclosed by reefs, and furthermore the
entire south side of the atoll is a continuous island, with
no connection between the sea and lagoon, the latter is ex
tremely calm— at times almost glassy. Majuro is presently
47
r
00
ihTisr'
> < * >
to
/V
/N
stbtutc ( wles
o j . 4 & y / o
I, i - —I— i— i— i
Kilometers
k wa jalei n
Figure 5 - Map of Kwajalein Atoll.
Southeastern End of
Atoll only Depicted.
m
STftTUTF ttll£ 5
/N
Figure 6 - Map of Majuro Atoll.
> c *
CO
49
the United Nations Trust Territory headquarters for the
Marshall Islands; there is a large native population, and
considerable commercial activity is centered there. Unlike
Eniwetok and Kwajalein, no battles were fought for posses
sion of this atoll during World War II, and there is .little
evidence of damage to the reefs.
Majuro is downwind of nearby Arno Atoll, and the wind
ward side faces to the north. This stretch of reef is
typified by wide reef flats and numerous islands (separated
by narrow interisland channels) on the back part of the
reef. The convex ocean shores of the islands often face
conglomerate platforms; lagoon shores tend to be concave,
with rubble and sand spits extending into the lagoon on
either side of the channels. The latter are floored by
intertidal reef rock to seaward, but deepen to subtidal
levels towards the lagoon.
The reef west of Darrit Island and extending past Ejit
and Arniel Islands is of the above type. Darrit and Uliga
are located at the eastern end of the windward reef; they
are presently joined by a road embankment at a point just
north of the Mieco Hotel on Uliga which was used as a base
of operations. For reasons of convenience, many collec
tions were made in this locality.
The southeastern part of the atoll has intertidal reef
flats facing both the ocean and the lagoon, and in places
there are islands of irregular shape extending across the
50
lagoonward flats. Between these islands are parallel, so
lidified rock bars which comprise the seaward and lagoon
ward shores of their respective reef flats; big pools of
water lie at a level considerably above low tide in the
area enclosed by these bars, at least at Dalap and Ajuro-
take Islands. There is presently a road running along a
high embankment between these bars; it extends the full
10-mile (16km) length of this part of the atoll. In all
probability, the rock bars represent the outlines of is
lands which were removed by a catastrophic storm; a similar
phenomenon occurred on nearby Arno Atoll, and has been
attributed to the typhoon of 1905 (Wells, 1951).
The southwestern part of the atoll is now one contin
uous island, 17 miles (22km) long, connected by the above-
mentioned road. At the western tip of the atoll, at Majuro
Island, the reef widens and some low-lying, coral-covered
reefs stretch northward. Otherwise the seaward reef flats
along this segment of the reef are intertidal in nature.
During both summer visits to Majuro (August 1969; Septem
ber 1970), waves of considerable size were observed on
the south-facing reefs.
Descriptions of the Study Areas
Detailed descriptions of the various reef habitats in
vestigated are deemed an essential part of this study. As
there are a great number of collecting areas, each local
ity has been assigned a station number, these being ar
51
ranged in order of their geographical position on each
atoll. A station list is provided in Appendix I, providing
information on the location, orientation, and type of habi
tat collected at each station. In addition, the width of
each zone mentioned in the text is given in Appendix II.
Below, descriptions of topography, biotic associa
tions, and vertical zonation (where known) are given for
each major type of reef feature. Xanthid crabs and other,
associated Brachyura are listed for each locality. Crab
habitats are arranged in a sequence which corresponds (at
least approximately) to their normal order of occurrence
from the algal ridge to the lagoon.
Most of the study areas described in this section are
inhabited by truly intertidal crabs, i.e. those which
usually live above the level of the low spring tides;
however, information is also provided for those localities
where the crab fauna is essentially a subtidal one. This
will give some idea of the differences in diversity be
tween different horizontal and vertical zones. As a
thorough account of the distribution of crabs at Eniwetok
is currently in preparation by John S. Garth and Jens W.
Knudsen, only the commoner crabs collected at low spring
tide levels will be mentioned here; rare or uncommonly
collected species that were collected as part of the pre
sent study from shallow water situations will be covered
in their paper.
52
Eniwetok Study Areas
Algal Ridge,- On Eniwetok, the development of a mass
ive, windward algal ridge is limited to certain islands.
A high algal ridge is well developed off the southern half
of Eniwetok Island; however, extensive collections were
made on the algal ridge at stations 1 and 3. This feature
begins at station 3 (Figure 7), and becomes progressively
better developed as one follows the reef edge in a south
western direction.
At station 3, porous algal rock buttresses jut out,
wedge-like, into the surf. The surge channels between
buttresses narrow rapidly and continue shoreward to termi
nate in high algal mounds with blowholes. A typical but
tress is 7m wide (width here means the distance from surge
channel to surge channel); surge channels are commonly lm
wide at the inner end, where they are at least lm deep.
The porous surge channel rims are about 1.5m wide; terminal
algal mounds are often about 4m in diameter. While this
algal ridge reaches up to the 0.9m level in places, the
back-ridge trough (which includes a dead coral zone) lies
at the low spring-tide level.
Farther south, at station 2, the algal ridge and back-
ridge trough widen somewhat (Figure 7); at station 1, the
reef is quite narrow, and the surge channels nearly reach
the island shore (or perhaps it is the case that the island
has encroached upon the outer reef) as the island reaches
K/f
b a t f r f c « S 0 6
dead Coral zonjz,
a l g a l rit^e a t
V T l S l
l *K8?
zone
| g
Station 2
X’ . , ^ )
a l g a l r id g f
coral
zmp
i;|
^ ■ ■ ■ - . 1 | <
owter /Znd orarue varl aUe 4«rf
vuL W r -V n o u g n subs+rtdum zone
inner reef - f t a t
Station 3
4 4 ^ *"*** &
Iwpr reef f l d f "
me+ers
Figure 7 - Diagram of the Seaward Reef at
Eniwetok Island
54
its greatest width to the southwest.
At present, this algal ridge is largely a dead struc
ture, with live coralline algae of the encrusting kind
(superficially, at least, resembling Porolithon onkodes)
limited mainly to the insides of channels; most of the
area is covered by a thick turf of brown, red, purple, and
green filamentous algae, many buttresses being darkened by
such growth. Dead corals, covered with a brownish algal
scum, are seen behind the buttresses and between the surge
channels (an area continuous with the backridge trough);
live corals (small Pocillopora and Acropora heads) cover at
most 5 percent of the area. An extensive, pink veneer of
coralline algae is seen only on low-lying shelf areas that
occlude the mouths of certain surge channels or on the tops
and sides of some of the more seaward algal mounds, where
the thin encrustation is sustained by jets of water from
the blowholes.
At low tide, the waves break over the buttresses, dis
appearing into the numerous holes in the rock. A portion
of each wave is however funneled into the surge channels,
to be ejected in geyser-like spurts of water, some 6m high,
from the terminal blowholes (in the case of live algal
mounds).
Conditions were seldom optimal for collecting crabs
on this part of the reef; heavy surf made work on the algal
ridge hazardous, and the constant spray was an added annoy-
55
ance. Except during very calm weather, the collector must
limit his activity to periods when the waves temporarily
abate, moving back to the centers of the buttresses (whose
porosity effectively absorbs most of the large swells) when
the waves are high (which may be most of the time). While
the porosity of the rock facilitates the removal of live
and dead corals as well as masses of brittle, cavitated
coral-algal rock, which can be broken apart later in search
of crabs, it is difficult to collect the latter by poison
ing the rock, as this same porosity, combined with heavy
wave action, makes it difficult to confine the formalin in
one place long enough for it to affect the hidden fauna.
Elsewhere on the windward side there is a low algal
ridge which is uncovered only on low spring tides; this is
a rough zone of composite coral and algal knobs, and but
few surge channels. Some samples were taken from this
type of algal ridge at Biijiri Island (station 16C), but
the algal ridge just north of Eniwetok Island (station 7A)
was much more thoroughly investigated. Here the reef edge
is a low, flat surface covered by dead, algal-overgrown
corals, only a few yellow encrusting corals remaining
alive. Crabs were collected in this locality both by poi
soning and by prying up thin, brittle slabs of encrusting
coral. As the surf striking a low algal ridge may be as
strong as that along a high reef margin, collecting there
may be even more hazardous, as there are no protecting but-
56
tresses, and the reef surface is very difficult to walk
upon.
On the leeward side, collections were made at Bogane-
gan Island (station 20), where the reef edge is a narrow,
convex algal rim covered by coralline algae and Pocillopora
corals and cut by narrow channels. A back-ridge trough is
present, in which Acropora heads thrive. Both kinds of
coral and bunches of the green calcareous alga Halimeda
were broken apart for crabs at this locale. It should be
noted that the seaward reef edge on these southwestern
reefs may be a dangerous place to go collecting during the
doldrum season, even though the surf on the windward side
may be fairly calm at this time.
Of the xanthid crabs taken from the algal ridge, the
commonest species collected was Paraxanthias notatus,
taken from both dead corals and algal rock. Liocarpilodes
integerrimus and the less common Cycloxanthops cavatus were
also found in both situations. In live corals, species of
Trapezia and Tetralia, and occasionally Domecia hispida
were collected; Chlorodiella laevissima was commonly taken
from dead corals. Less commonly obtained from coral heads
were Pilodius pugil, Actaea speciosa, and Actaea rufopunc-
tata.
In algal rock situations on the high windward-side
algal ridge, Globopilumnus globosus and Dacryopilumnus
emerita were commonly collected, though a few Phymodius
57
laysani and Pilodius flavus were also taken. Of the non-
xanthid crabs, a species of portunid, Thalamita sp., and
the grapsids Pachygrapsus plicatus, Pachygrapsus minutus,
Percnon pilimanus, Percnon abbreviatum, and Plagusia speci-
osa were found. The present data do not reveal an obvious
difference in species composition between leeward and wind
ward sides of the reef; it is on the other hand apparent
that certain species are associated with dead corals and
others with algal rock.
Algal Ridge Remnants.- Behind the high type of algal
ridge, there are often remnants of former reef growth.
While the crab fauna just described occupies the buttress
zone and the rims of active surge channels, the inner,
eroding parts of the algal ridge, located in the back-ridge
trough, are inhabited by a truly intertidal species assem
blage, otherwise found farther back on the reef.
At station 3 on Eniwetok, Eriphia scabricula was found
on the tops of algal mounds 19m back from the reef edge,
overlapping the algal ridge fauna proper. Farther south,
at station 2, many of the surge channels are decadent and
filled with rubble; the rims of these channels (Figure 8),
which are over 3m high and riddled with holes several cen
timeters in diameter, are inhabited by the xanthid crabs
Eriphia scabricula, Dacryopilumnus rathbunae, Zozymodes
biunguis, and Zozymodes pumilis. The rock surface here is
covered with an olive-colored algal scum of varying thick-
Figure 8
Figure 9
- Algal Ridge at Eniwetok Island Station 2.
Foreground: Decadent Surge Channel with
raised rims.
Background: Algal Ridge Buttresses.
- Decadent Algal Mound Behind Algal Ridge at
Eniwetok Island Station 2.
59
ness, though a rich purple algal turf occurs in the larger
holes and crevices.
These channels terminate in dead algal mounds, some of
which reach as high as the 1.22m level, 40 to 60m from
shore (Figure 9). For the most part the algal mounds seem
to be undergoing an erosional process, with small, sharp-
rimmed solution basins on top; a vermetid-algal association
occurs on the steeply sloping sides. Lydia annulipes is
the only xanthid which lives on the barren tops of these
mounds, though the surge-channel fauna mentioned in the
above paragraph occurs on the lower sides of these struc
tures. At station 2, L. annulipes was found on several
high rock mounds very close to the edge of the reef, in an
unexpectedly exposed location for this species.
While some of these surge channels and their associ
ated rock formations may have declined following wartime
damage, it is probable that there is a continual process of
headward growth of the buttresses, accompanied by the de
cline of the structures which have been left behind, where
the surge channels become filled by debris or pinched off
by seaward growth.
Reef Flats.- Behind the algal ridge lies the backridge
trough; at Eniwetok station 2, this forms a broad, barren-
looking zone of dead corals and smoother, pavement-like
rock which surrounds the algal mounds and the inner ends
of surge channels. Much of the surface is covered by a
60
thin, tan algal turf; as the reef merges gradually with the
intertidal shoreward flat, it is difficult to delineate the
back-ridge trough as a separate zone. Farther south, at
station 1, where the reef narrows towards the end of the
island, the back-ridge trough extends all the way to shore.
Sampling at this latter locality, where the reef flat
probably lies below the low neap-tide level, was carried
out by poisoning broad, shallow tide pools with a number of
narrow cracks and much sand and gravel in their bottoms.
The common xanthid crabs in this type of habitat were Zozy-
modes biunguis, Xanthias lamarcki, Etisus bifrontalis, and
Pilodius areolatus. Other crabs collected in this zone in
clude the atelecyclid, Kraussia rugulosa, the portunids
Thalamita picta, Thalamita admete, Portunus longispinosus,
and Portunus sp. (without long lateral spines), the grap-
sids Pachygrapsus plicatus, Pachygrapsus minutus, and Perc-
non planissimum, the ocypodid Macropthalmus bosci, the ma-
jid Micippa sp., and a small parthenopid. The same xan
thid fauna was found in some rather similar tide pools on
the outer (lower) reef flat at Japtan Island station 12C.
The inner reef flat at station 2 (Figures 7, 10) is
exposed at low tide and is a bleak, smooth rock surface
which stands in striking contrast to the rugose topography
of the algal ridge and its erosional remnants which lie to
the seaward. This inner reef is of a uniform orange color,
due to the thin, slick film of cyanophytes which covers
61
Figure 10 - Inner Reef Flat at Eniwetok Island Station 2.
Left: Tide pools.
Right Foreground: Parallel spines of rock
modified by fish grazing.
Figure 11 - Reef Flat at Eniwetok Island Station 4.
Foreground: Outer reef flat with emergent
rims of tide pools.
Background: Secondary trough and emergent
inner reef flat.
62
most of the area. A distinctive feature is the presence
of numerous tooth markings made by the surgeon fish, Acan-
thurus guttatus, and to a lesser extent by large parrot
fish (Chartock, personal communication, 1970, Bakus 1967b)
which swim over the reef flat in big schools at high tide
and rasp the rock surface.
Sharp-spined, parallel ridges of rock, 5cm high, ori
ented normally to the beach, are another peculiar feature
of this zone. Tide pools are indistinct, being merely low
spots on the reef flat which retain water at low tide. At
one point the tidal level of this orange inner reef zone
was found to increase from the 0.5m level where it adjoins
the tan back-ridge trough to the 0.69m level along the
shore. In keeping with the featureless appearance of this
part of the reef, crabs are few in number and variety;
only the xanthid Zozymodes biunguis and the small grapsid
crab Pachygrapsus minutus live in the minute holes that
are sparsely distributed across the reef.
The reef off the northern half of Eniwetok Island
(station 4) is narrow, like that off the southern half
(see Appendix II), but the former occurs at a relatively
higher level in the intertidal. The algal ridge in this
location is low, and its knobby surface is exposed between
the waves only on the lowest tides. Waves come in and
break over the ramp-like outer reef flat at most times, al
though they rapidly diminish in size over this zone.
63
Behind the outer reef flat lies a secondary trough or
moat, up to 7cm deep, and uniformly covered by a low, tan
algal turf; this always holds water, as the only drainage
is across the high outer reef flat to the sea. Crabs are
scarce here, though occasionally Micippa sp. can be observ
ed wandering across the bottom.
The next zone has smooth-bottomed tide pools, up to
1m wide and 3 cm deep, covering up to half the area; to
wards the outside of this zone, pool bottoms are covered
by a tan algal turf, which gives way to a slick orange film
closer to shore. Between these larger pools, the reef sur
face is made up of small tide pools, up to 10cm wide and
2cm deep, with rough rims. These small pools and rims are
veneered by a knobby grow.h of the coralline alga Gonioli-
thon sp. , growi.ng in close association with Dictyosphaeria.
The coralline algae, which thrive in the outer part of this
zone (which will henceforth be called the Goniolithon zone)
is probably able to proliferate here because of its extreme
ly porous growth form (Figure 12) which allows water to be
retained in a multitude of minute pools and crevices as
the tide runs out. To shoreward, small black vermetids
become abundant and form rough rims around the smaller
pools. In this zone, which gradually increases in height
from the 0.7m level just behind the secondary trough,
Zozymodes biunguis is the dominant xanthid, although Eriphia
scabricula also occurs, as do the grapsids Pachygrapsus
Figure 12 - Inner Reef Flat at Eniwetok Island Station 4.
Goniolithon zone; tide pool filling
center of picture.
Figure 13 - Innermost Reef Flat at Eniwetok Island Station4
Zone of large tide pools.
6" (15cm) rule in background.
65
plicatus and Pachygrapsus minutus.
The inner reef flat reaches the 0.88m level along the
island shore; it is characterized by large tide pools, up
to 4m long and 10cm deep. These are irregular in shape and
have shelving edges under which sand, gravel, and small
rocks accumulate (Figure 13). A tan algal turf and sand-
encrusted worm tubes line the edges of the pools; the pool
bottoms have a thin layer of sand of the same color, loose
ly bound by miniscule green algal filaments. A network of
purple sponge, covered by detritus, also occurs here.
Loose algal material washed in from the outer reef drifts
about in these pools, which occupy about half the area.
The remaining part of the reef surface is smooth and cover
ed by the usual orange cyanophyte film, and dries at low
tide. Numerous blackened knobs of rock up to 20cm high
protrude above the general surface. The xanthid crabs
Zozymodes biunguis and Xantho leptodon are very common, and
Xantho sanguineus, Eriphia scabricula, and Eriphia sebana
also live here, along with the grapsids found in the pre
ceding zone.
At about the midpoint of the island (station 3), the
reef changes in character. The shore zone looses its por
osity, though larger, smooth-bottomed pools occur along the
edge of the beach. Where beach sand and tan algal turf
cover the inner reef, small Xantho leptodon are abundant,
and the portunids Portunus longispinosus and Portunus sp.,
66
the ocypodid Macropthalmus bosci, and the palicid, Palicus
sp. can be found hiding in the sediment.
Coralline algae become less abundant; the high outer
reef flat (here behind a massive algal ridge) is covered by
a slick, yellow-green cyanophyte film, with little, irides
cent green algal clumps and small, brown felt-like colonies
everywhere. Zozymodes biunguis and Pachygrapsus minutus
are the characteristic outer reef crabs here. South of
this point, the outer reef curves inward and away from the
reef edge, and appears to be continuous with the orange
zone which forms the inner reef at station 2. It should be
noted, therefore, that the terms "outer reef" and "inner
reef" denote location, and not homology or common origin.
Many of the other windward reef flats at Eniwetok
Atoll show a similarity in horizontal zonation to that of
station 4, though in general the crab fauna is less common
and varied on the intertidal flats. At Aniyaanii Island,
for instance, Zozymodes biunguis is the only common inter
tidal xanthid, being very abundant at station 13B off this
island, where the surface is highly perforated by small
(about 2cm wide) holes with jagged edges. However, much of
the neighboring reef area is smooth and veneered by a pink,
paste-like material composed of sand bound by microscopic
algae, and here even Z_. biunguis is uncommon.
North of Eniwetok and Sand Islands, a strong current
passes across the reef flat at high tide and during some
67
low tides. Most of this reef probably lies at or somewhat
below the 0.6m level. A richer algal flora occurs here
than on reefs adjacent to islands; in places it is light
and spongy, forming a mat several centimeters thick. Some
Eriphia scabricula were found in small tunnels in this algal
mat, and Etisus bifrontalis was collected from the rubble of
gravel-filled, pothole-like pools which are characteristic
of this current-swept habitat. The brightly colored, pois
onous Zozymus aeneus is also occasionally seen here, wan
dering about in broad daylight; other xanthids common on
this reef are Zozymodes biunguis and Xanthias lamarcki.
Higher Conglomerate Flats.- This term covers all flat
areas made up of solidified coral rock higher than the level
of the reef flat, and which shows either a lack of obvious
stratification or nearly horizontal strata, to distinguish
it from beachrock. In distinguishing higher conglomerate
deposits from reef flats, the most useful criterion is the
presence of standing water at low tide. On reef flats,
water is contained in tide pools, which generally cover at
least 5 0 percent of the area; on higher conglomerate plat
forms, tide pools are rare and standing water covers only
a very small percentage of the surface, at the most about
25 percent.
These conglomerate flats are extremely barren and for
bidding in appearance. Commonly, the rock surface is sculp
tured with sharp points and small solution basins. At
68
Japtan Island (station 12A), one such platform, lying at
the 0.98m level, is isolated from shore on the more pro
tected part of the reef; it is veneered with a light green
to blackish scum of algae that tends to crack and peel off
when dry. In such places the dominant xanthid is Lydia
annulipes , which lives in holes and crevices in the rock.
On the windward side, exfoliation of slabs can be seen
in several places. One is at Sand Island (station 9B),
where the brownish-black conglomerate surface lies behind
and above the reef flat, above the 0.9m level. Another
example is at Aniyaanii Island (station 13A), where loosely
cemented slabs occur on the flat upper surface of a long
rock groin, which lies at the 0.98m level (Figure 14). In
such places the slabs, some over lm long and 14cm thick,
can be pried off to reveal the conglomeratic nature of the
strata, with embedded, nodular pieces of coral pebbles and
branches, shells, and sandy material showing on their lower
surfaces (Figure 15). Under these slabs lies a thin layer
of gravel and a little sand; none of the loose pebbles is
over 25mm long. Some of the gravel is derived from the
thick, maroon spines of the large echinoid Heterocentrotus,
which lives on the algal ridge, suggesting that some of the
loose under-rock material may be washed in from the outside,
though much of it may be derived from the undersides of the
slabs themselves. The xanthid crabs Pseudozius caystrus
and Lydia annulipes live in the spaces beneath these slabs,
69
Figure 14 - Upper Surface of Rock Groin at Aniyaanii Island
Station 13A.
Showing layers of exfoliating conglomerate.
12" (30cm) rule in foreground.
Figure 15 - Lower Surface of Conglomerate Slab at Runit
Island Station 14B.
Showing Coral branches embedded in
underside of slab.
70
along with the rock oyster Isognomen legumen, large sipun-
culids, and hermit crabs. A yellow encrusting growth on
the normally hidden rock surface completes the picture.
Eroding Borders of Conglomerate Platforms.- An impor
tant feature of the higher conglomerate is a zone of tran
sition to the reef flat below, usually taking the form of a
narrow ramp or wall on the outer margin of the conglomerate
flat.
On the windward side, the seaward margin of the con
glomerate flats is an eroding ramp, covered with sharp
ridges and pinnacles of pitted rock. At Aniyaanii Island
(station 13B), the surface relief in this zone is at least
30cm. The xanthid crabs Zozymodes biunguis and Lydia
annulipes were collected here in 19 69; however, on the
outer borders of the southernmost groin at Runit Island,
where the topography is similar, the dominant crab is the
large xanthid species Eriphia sebana, which inhabits big
holes in the rock surface.
At Eniwetok Island, there is a much flatter (maximum
relief: 15cm) pitted zone shoreward of the reef flat at
the north end of the rock quarry (station 5D). This area
is so nearly level that it is difficult to distinguish it
from the reef flat proper. The entire surface is covered
by an orange algal film, like the reef; however, there are
no distinct tide pools. There are numerous 1 to 2cm wide
pits all over, some occupied by zoanthids, and others lead
71
ing to the smaller holes of boring sipunculids (Figure 16).
The xanthid crab Zozymodes biunguis occurs in large numbers
here, and is accompanied by the grapsid Pachygrapsus
minutus.
On the leeward side, the seaward margins of conglom
erate platforms tend to have a steep slope; at Rigili
Island (station 18), this zone takes the form of a nearly
vertical wall, extending from about the 0.3m to the 0.9m
level (Figure 17). Although large, basin-like areas sug
gest erosion, deposition of new rock is also in evidence:
the small vermetids which are abundant on this wall form a
layer as much as 10mm thick in many places. The vermetid-
blackened surface of this zone is perforated by many old
sipunculid borings and inhabited by the xanthid crabs
Dacryopilumnus rathbunae, Zozymodes pumilis, Zozymodes
biunguis, and Lydia annulipes.
A similar habitat occurs on the windward side in shel
tered locations. At Japtan Island (station 12B), the edge
of the island conglomerate is occupied by the same assem
blage of crabs, with the addition of Eriphia scabricula,
Eriphia sebana, and Pseudozius pacificus. There are also
old rims of rock in this locality, outlying the present
conglomerate platforms where the reef faces the Deep Chan
nel; these are probably the eroded remnants of similar
platforms which were reduced by enlargement of the large
basin-like areas seen along the edges of extant conglom-
Figure 16 - Pitted Zone on Innermost Reef at Eniwetok
Island Station 5D.
6" (15cm) rule in foreground.
Figure 17 - Margin of Conglomerate Platform at Rigili
Island Station 18.
Foreground: Inner reef flat.
Center: Vermetid-encrusted conglomerate
wall.
Background: Conglomerate flat.
73
erate flats. An impressive field of these rock rims, some
14m long, is seen at station 12A, where it is closer to the
reef edge than to the shore (Figure 18). The vertical rims
are about 50cm high, extending from the reef flat up to the
0.64m level and enclosing interconnecting basins 1 to 2m in
diameter. The lower sides of these wall-like structures
are inhabited by Zozymodes biunguis; the convex upper sides
are lighter, and much bored by sipunculids and barnacles
iLibhotrya) (Figure 19). Abandoned sipunculid holes are
occupied by Dacryopilumnus rathbunae, and barnacle holes by
Eriphia scabricula.
Seaward Beachrock.- Beachrock is formed by the solidi
fication of calcareous beach deposits in the intertidal
zone, often at some depth below the surface; subsequent re
moval of any overlying sediment exposes a stratified rock,
with the same dip as the layers of sand or gravel from
which it was formed. On atolls, beachrock is often found
along the lagoon shores of windward islands, where sandy
beaches are common, and less often along seaward shores.
When coarser sediment makes up the bulk of beachrock, it
is considered to be beach conglomerate; when it is composed
mainly of sand, it is beach sandstone. As we shall see,
beachrock occurs in a variety of forms.
Along the ocean shores of some of the northern islands
of Eniwetok Atoll one finds a beach conglomerate which is
undergoing a process of exfoliation similar to that des-
74
Figure 18 - Eroding Conglomerate Remnants at Japtan Island
Station 12A.
Center: Conglomerate rims.
Distant background: Algal ridge and deep
channel.
Figure 19 - Boring Organisms from Conglomerate at Japtan
Island Station 12A.
Upper left: Sipunculid.
Below left and right: The barnacle Lithotrya.
75
cribed above for horizontally-bedded conglomerate. At Ro-
joa Island (station 16A)., the grey-to-black strata slope
upward towards shore (Figure 20). Between the layers there
is much gravel and some sand, and the xanthid crabs Pseudo-
zius caystrus and Lydia annulipes can be found with rock
oysters, sipunculids, and other organisms.
At Runit Island (station 14B), a curving beachrock bar
extends out from shore into a shallow pooled area on the
reef off the southern part of the island; it probably rep
resents the outlines of a former sand bank. The 4.6m wide,
slanting outer surface is cracked in a mosaic pattern; the
inner face is a 46cm-high wall showing many thin strata in
cross section. This rock is soft and friable. Zozymodes
biunguis proved to be the most common xanthid inhabiting
the low crevices between the layers, associated with a few
Pseudozius caystrus.
Along the south shore of Biijiri Island (station 16B),
a massive bed of beachrock faces the channel between is
lands; the sloping upper surface is extremely rugose and
completely covered by 6cm-wide solution basins with sharp
rims. The outer face forms a vertical wall about .9m high,
on which Lydia annulipes has been observed wandering at
night.
In sheltered locations, beach sandstone and beach con
glomerate slabs often lie loose over sand, gravel, and
beachrock (Figure 21). At Sand Island (station 9A), many
76
Figure 20 - Beach Conglomerate at Rojoa Island Station 16A.
Figure 21 - Loose Beachrock Slabs Along Shore at Runit
Island Station 14B.
A pink, paste-like algal-sand material
covers many of the slabs.
6" (15cm) rule in foreground.
77
such slabs, some well over lm long and 20cm thick, lie be
tween the 0.6m and 1.22m levels and provide cover for the
xanthid crabs Xantho sanguineus, Xantho gracilis, and Pseu-
dozius caystrus.
Lagoon Beachrock.- On the lagoon shores of islands,
beachrock provides a different kind of habitat than seaward
beachrock; it is actually more exposed than is the latter,
for the Eniwetok lagoon is subject to a heavy swell, while
seaward beachrock is in a sheltered position behind the
reef flat.
A study area of major importance was provided by the
lagoon beachrock at station 7B on Eniwetok Island (Figure
22). Numerous observations on live crabs were made here
along a stretch of rock 11m long. In this locality the
beachrock extends from the 0.44m to the 1.35m level, and
slopes upward towards shore, as it follows a former beach
profile. Diagonal cracks run inward from the lagoon edge,
and there are indistinct pools in places. While most of
the smooth beachrock surface is covered with a slick, dark
green-to-blackish algal film which flakes off when dry,
there is a rich flora of filamentous red and green algae in
cracks and under ledges. This beachrock is exposed to con
siderable wave surge; a zone 8m wide may be alternately
covered and exposed by successive waves. This is the
beachrock habitat of the xanthid crab Eriphia scabricuia,
which lives in holes in and around the cracks and under
78
5 * *n . •" !
Figure 22 - Lagoon Beachrock at Eniwetok Island Station 7B.
Showing overhanging ledges.
6" (15cm) rule in foreground.
Figure 23 - Eroding Edge of Lagoon Beachrock at Runit
Island Station 14D.
79
shelving edges.
Other beachrock on the lagoon shore of this island is
generally similar; where the collections were made, very
little sand is present, and there is a considerable drop
into the lagoon just off shore. At station 7A, Eriphia
scabricula extends down to the 0.3m level in holes along
the deep vertical cracks which extend into the subtidal
zone. At stations 7C and 7E, there are many horizontal
cracks between beachrock strata which provide a habitat for
Lydia annulipes and young Eriphia sebana. In the higher
cracks where the former species is common, a rich algal
turf and occasional rock oysters occur; neither is found on
exposed rock surfaces.
In some places, the lagoonward margin of the beachrock
constitutes a separate zone from the rest of the surface.
On the lagoon side of Runit Island (station 14D), the edge
of the beachrock is grooved and undercut, with basin-like
depressions up to lm long and .5m deep behind a narrow,
convex rim which reaches the 0.6m level (Figure 23). Even
on a low spring tide the lagoon swell may wash up over the
rim. The rock surface is covered by a yellow-green algal
film, and its characteristic xanthid inhabitant is Dacry-
opilumnus rathbunae.
At station 10C, a line of beachrock impedes the flow
of water across the reef flat and into the lagoon. The
beachrock strata slope upward towards the lagoon, and the
80
lagoonward edge takes the form of a vertical or undercut
wall, cut by diagonal clefts. The latter are often en
larged and basin-like, functioning as spillways which allow
water to drop from the 0.6m level of the seaward flat to a
lagoon level as low as 0.15m on a low spring tide. A rich
algal flora occurs on this beachrock "dam" except on the
convex upper rim, where the vermetid-covered rock has a
bare, bleached appearance. The commonest xanthid crabs
here proved to be Dacryopilumnus rathbunae and Zozymodes
biunguis, and a few Eriphia scabricula and Lydia annulipes
were also collected when this locality was studied in 1970.
Reef Blocks.- Reef blocks are large masses of rock de
posited on the surface of the by storms, but too large to
be moved by ordinary waves. They occur sporadically on the
leeward side of Eniwetok Atoll. On the seaward reef flat
at Igurin Island (station 21), one example lies to the sea
ward of the rubble field and measures 2.7m long by 2.1m
wide and 1.2m high (Figure 24). It is very porous and
shows corals in growth position (now tilted at an angle).
Dacryopilumnus rathbunae and Zozymodes pumilis were col
lected from holes on the side of this block.
Rubble Heaps.- Along the edges of the rock quarry at
the north end of Eniwetok Island (station 5D), there are
masses of rock, ranging from boulders down to pebbles and
heaped one upon another, several layers deep. These rocks
are piled up along the sides of the quarry and in cracks
Figure 24
Figure 25
- Reef Block on Outer Reef Flat at Igurin Island
Station 21.
Foreground: Coral zone surrounding block.
Distant Background: Rubble Field.
- Rubble Field at Japtan Island Station 12B.
Showing boulders and heaps of smaller rocks.
6" C15cm) rule in foreground.
82
along its margin; they are firmly wedged together, and
slightly cemented by encrusting organisms. This habitat
lies at or above the level of the reef flat, and is dupli
cated along the shore of the Parry Island quarry (station
11A) .
At night, crabs were often observed in the crevices be
tween the rocks of the uppermost layers at the Eniwetok
quarry; when disturbed, they moved rapidly down towards
the bottoms of these rubble heaps, and had to be pursued by
removing rocks, layer by layer, through progressively
finer grades of material until gravel and sand were reached
at the base. The animals were not restricted to the low
pool level of the quarry, as the porosity of their habitat
allowed them to move up with the tide at night. Many spe
cies were taken from this type of situation, but the com
monest xanthids are: Pilodius areolatus, Pilodius pugil,
Chlorodiella cytherea, Phymodius nitidus, Phymodius ungula-
tus, Etisus dentatus, Lachnopodus subacutus, Lachnopodus
tahitensis, and Carpilodes bellus. This appears to be a
fauna that normally lives at or below the lowest intertidal
levels, here elevated in vertical zonation in the main
quarry pool, which is several meters deep.
In a sheltered situation along the south side of Japtan
Island (station 12B), there is a field of rubble lying on
the inner reef flat, at approximately the 0.6m level. On
this flat, the rocks are wedged together in shallow, semi-
83
consolidated heaps (Figure 25). Their exposed upper sur
faces are covered by a black-colored algal scum; however, a
thick green and purple algal turf grows on the sides of the
rocks along with masses of sand-encrusted worm tubes. Xan
tho sanguineus is the dominant xanthid, and readily digs
into the sand under the rocks when they are pried apart.
Some Xantho danae live under loose rocks on the outer mar
gin of the flat, and Pseudozius caystrus and Eriphia sebana
occur along shore at the foot of the rubble beach.
Rubble Flats and Beaches.- Loose rubble, ranging up to
boulder size, occurs on the surface of the sand and gravel
of beaches and flats, on the seaward or on the lagoon side.
This habitat is very barren in appearance; the only life
visible on the surface of the rocks is a film of grey-to-
black algae on exposed surfaces. Under rocks, however,
live several species of crabs as well as polychaetes.
In places where rubble lies in large pools on the sea
ward reef, a colorful wash of pink and purple coralline al
gae and other encrusting organisms on the rocks and pebbles
reveals a richer biota than would otherwise exist where
loose, coarse sediments occur on the reef. This is the
case at stations 10A and 1QB, on the seaward fringes of the
nearly level rubhle fields on the back part of the reef.
The lagoonward flow of water, impeded by a continuous rock
bar behind the rubble, forms an elevated pooled zone at the
0.6m level. Here, and along the ocean side of the main
84
quarry pool (station 5D), the xanthid fauna includes Xan-
thias lamarcki, Pilodius areolatus, Etisus bifrontalis,
Etisus demani, Xantho waialuanus, and Lachnopodus subacutus/
which would ordinarily not occur above the low spring-tide
level in intertidal rubble. At station 10B, Pseudozius cay
strus occurred in piles of loose rocks and gravel which dry
at low tide, just above the pooled zone, though at station
10A, the transition to mid-tidal levels is less abrupt and
Xantho gracilis is the dominant xanthid on the sandy rubble
flat behind the pooled zone.
On the windward side of Eniwetok Atoll, drab-appearing,
rubble-covered beaches overlie the high inner reef, which
lies at approximately mean sea level. At Japtan Island
(station 12C), rocks up to 60cm in diameter cover the sand
and gravel of the lower beach; this zone is covered by
water at high tide and the rocks are accordingly blackened
by a film of algae (Figure 26). The upper beach in this
locality is made up of white coral gravel which forms a
steep rampart of loose material that is encroaching upon
the island vegetation. The characteristic xanthid crab
fauna of such beaches is made up of Pseudozius caystrus,
Xantho leptodon, Xantho gracilis, and Xantho sanguineus; in
addition, the small grapsid crab Pachygrapsus planifrons is
conspicuous on the intertidal beach at low tide, and the
grapsids Pseudograpsus sp. and Cyclograpsus sp. can be
found under rocks on sand in the suprat.idal fringe, at the
85
Figure 26 - Rubble Beach at Japtan Island Station 12C.
Left: Reef flat.
Center: Intertidal rocks.
Right: Rubble rampart of island shore.
6" (15cm) rule in foreground.
Figure 27 - Rubble Field South of Rock Groin at Aniyaanii
Island Station 13A.
12" (30cm) rule in foreground.
86
interface between the intertidal zone and dry land.
In certain localities, there are variations of the gen
eral picture. At Parry Island (station 11B), it is only
where the beach overlying the reef flat is concave to the
sea that there is any exposed rubble, at the reef/beach in
terface; here only Xantho leptodon and Xantho sanguineus
occur. Rubble can also be found on sandy beaches at the
ends of windward islands, as at Eniwetok Island station 6
in 1969 (in 1970 it was a sandy beach with no loose mater
ial) . Here, and on the rubble-covered lagoon beaches of
leeward islands, such as that at Grinem Island (station
19B), the xanthid assemblage is the same as that given
above for Japtan Island.
There are several places on the windward side of Eniwe
tok Atoll where beds of rubble on the back part of the reef
have such a gentle slope that they can almost be considered
rubble flats. Just south of the groin at Aniyaanii Island
(station 13A), the rubble lies at approximately the 0.82m
level, with numerous jagged coral cobbles (about 13cm in
diameter) and gravel scattered about over the sand (Figure
27). At station 13C, and on the drying part of the rubble
flat at station 10A, the rubble slopes gradually downward
towards the sea; Xantho gracilis is very abundant in such
places, and is accompanied by Pseudozius caystrus on the
inner, higher parts of the rubble beds. At station 13C, a
number of Eriphia sebaha were also collected.
87
In some places, true rubble flats occur, cutting off
pools of water above the reef flat level in pools that
never drain completely at low tide. This is the case at
Igurin Island (station 21), where broad, shallow pools
floored by sand and rubble lie separated from the seaward
reef by an extensive rubble field (Figure 28). At Eniwetok
Island, a small (9m by 9m) pool is cut off from the reef
flat by conglomerate, at the south end of the quarry (sta
tion 5A). This afforded an opportunity to observe crabs in
a similar, albeit artificial habitat. The floor of this
pool is one of mixed sand, gravel, and small, flattened
rocks under 15cm in diameter. The low water line is at
about the 0.82m level, leaving part of the rubble-filled ba
sin filled with water at all times, and part exposed at low
tide. During neap-tide periods, the pool is not refreshed
by ocean water during daytime high tides, although spring
tides may cover the area completely up to the 1.22m level.
Xantho leptodon is the dominant xanthid in pools of this
kind, being replaced by Xantho gracilis and Pseudozius cay
strus in the drying area.
Gravel Bars.- At Bogallua Island (station 17B), there is
a low-lying seaward gravel bar, which was sampled by poison
ing small tide pools in the gravel. While the commonest
species collected was the portunid Thalamita admete, a few
specimens of the calappid crab Calappa hepatica and the
palicid Palicus sp. were also taken here. No xanthid spe-
88
Fiqure 28 - Cut-Off Pool Behind Rubble of Seaward Reef at
Igurin Island Station 21.
Left: Rubble field.
Center: Rubble-filled pool.
Right: Sand beach connecting central
and southern sectors of island.
89
cies were obtained from this gravel.
Protected Coral Areas.- A number of collections were
made of crabs living in association with live and dead
corals in sheltered coral reefs, to provide further infor
mation on the crab fauna inhabiting the lowest part of the
reef which is exposed by the tides.
Branching corals (Acropora and Pocillopora) were col
lected from the Eniwetok Island quarry (station 5D), and
at Biijiri Island (station 16B), where isolated corals
occur on the sloping shore of the channel (dead corals
here being partly buried by fine sediment). The commonest
xanthid crabs collected from living corals were species of
Trapezia, Tetralia, and Domecia. In dead corals, the xan
thid crabs Pilodius areolatus, Xanthias lamarcki, Chloro-
diella nigra, and Chlorodiella cytherea were found to be
common, along with the portunid Thalamita admete. Pilo
dius pugil was collected in some numbers from coral micro
atolls on the inner seaward reef flat at Bogallua Island
(station 17B).
Beds of the nodular coral Porites lutea provide ex
cellent collecting grounds for crabs and other Crustacea.
In shallow water, this coral tends to grow up to the low-
tide level and then expand in an outward direction, as it
is prevented by desiccation from extending any higher.
When this occurs, the resulting Porites heads resemble
lily pads. Sometimes thin plates of this coral grow over
90
sand and gravel, providing cover for numerous animals in
the space beneath. Along the lagoon shore of Bogallua
Island (station 17C) there is a Porites reef of this type,
with sand and gravel"floored tide pools several centimeters
deep between the growing coral colonies. Off the ocean
shores of Piiraii Island (station 15) and Rigili Island
(station 18) there are smooth, low-lying reef flats at
about the 0.3m level which may have originated through the
coalescence of adjacent Porites heads (at any rate, some
live coral still grows around the edges of pools in these
places). A hollow crevice extends some distance under the
reef flat from the edges of these pools; the thin shelf of
rock can be broken away to reveal sand and rubble under
neath. The numerous xanthid crabs that live in this space
can be obtained by poisoning the pools and breaking away
the margins. The commonest of these are Xanthias lamarcki,
Pilodius areolatus, Etisus bifrontalis, Etisus electra, and
Carpilodes bellus. In addition, Zozymodes biunguis is
abundant at stations 15 and 18, where the pools are inci
dental on a solid reef flat. Several portunid crabs are
also common in these pools: Thalamita admete, Thalamita
picta, and Portunus longispinosus.
Another type of Porites habitat can be seen just
north of Eniwetok Island, on the inter-island reef that
extends over to Sand Island (station 8C). Here the coral
is most abundant on the lagoon side; the substratum is a
91
complex of live and dead, loose and attached Porites slabs
(which often tilted up on one end) with rocks, gravel, and
sand between and under the slabs. These Porites beds were
extensively collected during low spring tides in 1968;
they lie low in the intertidal, between the 0.15 and 0.29m
levels. This area provided the richest collecting, in
terms of variety of species, done at Eniwetok. The com
monest xanthids taken here were Pilodius areolatus, Pilo-
dius pilumnoides, Xanthias lamarcki, Carpilodes bellus,
Carpilodes pallidus, Chlorodiella cytherea, Chlorodiella
laevissima, Medaeus simplex., Medaeus elegans, Etisus bi-
frontalis, Etisus demani, Lachnopodus subacutus, and small
Carpilius convexus. Other frequently collected crabs from
this reef were the portunid Thalamita picta and a small
red portunid.
Subtidal Lagoon Areas.- Information on strictly sub-
tidal habitats is limited, as it was far easier to collect
on reefs which dry at low tide. Some collecting was done
by snorkeling in shallow water, and in addition divers oc
casionally brought up specimens from water as deep as 80
feet (24.4m). In all, about 30 specimens belonging to 20
species were recorded for the family Xanthidag; these ani
mals were taken under rocks, from within coral heads, or
from Tridacna shells. With the exception of one or two
specimens of Xantho sanguineus, all. of these crabs were
forms which live in the lowest part of the intertidal, or
92
which had not been seen in the intertidal at all.
Study Areas on Majuro and Kwajalein
Algal Ridge.- On Majuro, the algal ridge is well de
veloped from Uliga to Arniel Island, and is made up of
narrow buttresses and surge channels, although its height
varies. In most places, globular and encrusting coralline
algae thrive. At night, numerous crabs come out of their
holes in the porous rock of the buttresses; these include
the large xanthids Carpilius maculatus and Daira perlata
which are difficult to capture as they retreat rapidly
through large holes into the interior of the buttresses.
As at Eniwetok, the common small xanthid is Paraxanthias
notatus.
Algal Ridge Remnants.- On Majuro, remnants of buttres
ses, up to at least 46cm high, form jagged rows, oriented
perpendicular to the reef edge where there is a high algal
ridge. These are eroded into porous, brown-colored masses
of rock, with finger-like knobs and numerous holes. Be
hind the partially dead algal ridge at Arniel Island (sta
tion 42), such relict buttresses are often covered by a
striking steel-grey soft coral (Figure 29). The xanthids
Dacryopilumnus rathbunae and Eriphia .scabricula are com
mon on such erosional remnants at Uliga Island (station
47) .
Reef Flats.- On Majuro, the windward reef flat is
generally similar to that of Eniwetok. The outer reef
93
Figure 29 - Algal Ridge at Arniel Island Station 43.
Showing eroding remnants of buttresses.
Background: Intertidal reef flat.
Figure 30 - Innermost Reef Flat at Uliga Island Station 47.
4" (10cm) rule in foreground.
94
slopes gradually upward from the back-ridge trough, and is
generally covered with a rich mat of green and brown-
colored algae, over 1cm thick. At Darrit Island (station
46), tide pools in this zone are often lm wide, shallow
and indistinct; and shallow grooves, perhaps former surge
channels which have become filled in, cross the entire zone
in places. The common xanthids on this outer reef flat are
Xanthias lamarcki, Etisus sp.*, and Pilumnus purpureus;
other xanthids collected here include Etisus bifrontalis,
Etisus demani, Pilodius pilumnoides, and Zozymus aeneus.
At Uliga Island (station 47), Actaea tomentosa and Pilodius
areolatus are abundant in crevices and among the gravel of
small tide pools on the outer reef.
The secondary trough is not as well marked a feature
on these reefs as it is at Eniwetok. The inner reef flat
which lies behind this is narrower where there are islands
behind the reef than where there are inter-island channels,
as the islands occupy an area which would otherwise be part
of the reef flat.
At Uliga Island (station 47), the major study area was
located east of an outlying rock groin. The greater part
of the inner reef flat appears dark green in color, due to
tiny filamentous algae in minute pits in the exposed rock
* This small species is superficially similar to
Etisus electra and E. frontalis, but differs in having nar
row, forward-pointing teeth and a deep median interorbital
notch.
95
surfaces between tide pools; the tide pools are 30cm long
and often have a thin, tan algal turf covering their bot
toms. This inner reef flat reaches the 0.9m level, and the
xanthid crabs Zozymodes biunguis, Eriphia scabricula, and
Xantho sanguineus live on this part of the reef. The in
nermost reef zone is slightly lower and is characterized by
a thick sand and algal turf of light color (Figure 30).
Tide pools are larger and rather indistinct here. Xanthid
crabs, principally Xantho leptodon, Etisus frontalis, and
"Chlorodius" miliaris live in this zone; the portunids
Thalamita admete and Portunus longispinosus, and the ocy-
podid Macropthalmus sp. also occur.
Many of the windward reefs in the southern Marshalls
show considerable evidence of scour in the tide pools. On
Kwajalein Atoll, the reef flat off the east end of Kwaja-
lein Island (station 31) displays this feature close to
shore near the bend in the island; tide pools here are up
to 2 0cm wide and 13cm deep, and small, rounded rocks about
13cm in diameter rest on the bottoms of many of them (Fig
ure 31). These tide pools tend to coalesce, and are ar
ranged in rows up to 2m long, running normal to the beach.
A variety of algae, including Padina, grows on the narrow
rims between pools. On Majuro, the reef west of Darrit
Island (facing the interisland channel) is rather similar;
here the scour is evidently due to the current crossing the
reef at high tide. A characteristic feature of this reef
96
Figure 31 - Rubble-Filled Tide Pool at Kwajalein Island
Station 31.
Showing rocks in pool bottom and Padina
around rim.
Figure 32 - Reef Flat with Overhanging Ledges at Dalap
Island Station 48A.
6" (15cm) rule in foreground.
97
flat is an abundance of 5cm wide, maroon-colored Jania
clumps. Both of these reef flats probably reach mid-tidal
levels, and are inhabited by the xanthid crabs Zozymodes
biunguis, Eriphia scabricula, and Xantho sanguineus, along
with the portunid Thalamita admete.
On the south side of Majuro Atoll, there is a fairly
high intertidal reef flat. At Dalap Island (station 48A),
the seaward reef is similar to the dark green reef zone at
Uliga Island, except that the tide pools are larger, up to
lm long and with numerous low, overhanging ledges along
their margins (Figure 32). The xanthid crabs Zozymodes
biunguis, Xantho sanguineus, Eriphia scabricula, and Xan-
thias lamarcki are characteristic of this area.
Off the west end of Majuro Island (station 52), a
lower-lying reef flat occurs. This reef is probably not
much above the 0.24m level, and is light pink in color, due
to a covering of sand. Numerous boulders are strewn across
the inner reef, where there is abundant leaf litter derived
from the island vegetation. A rich crab fauna occurs on
this reef; the xanthids Pilumnus sp., Etisus sp. (the same
as that found on the outer reef at station 46), Etisus bi-
frontalis, Pilodius pilumnoides, Carpilodes bellus, Chloro
diella cytherea, and Actaea tomentosa were taken from sand
and gravel-filled pools, and Eriphia scabricula was col
lected from the intervening algal-covered reef rock.
Finally, there is a very sheltered lagoon reef which
98
runs along the southeastern end of the atoll. At Dalap
Island (station 48B), this reef appears to lie for the most
part below the 0.44m level, sloping downward into the la
goon. This flat is very porous, with numerous holes all
over its surface; it is coated by a brownish detrital ma
terial, which also covers its characteristic xanthid inhab
itant, Pilumnus vespertilio.
Higher Conglomerate Flats.- Many of the windward is
lands on Majuro Atoll are fronted by a very coarse, grey,
parched conglomerate rock along their ocean shores. Nu
merous jagged coral skeletons can be seen in an irregular
orientation on the surface of this rock. This habitat was
sampled near a small islet west of Arniel Island (station
41); the coral rocks can be pried off to reveal small yel
low stalked barnacles, rock oysters, and the xanthid crabs
Pseudozius caystrus and Lydia annulipes underneath.
At Uliga Island (station 47), the conglomerate has a
different appearance. The outlying rock groin near the
hotel was studied in some detail; it is roughly C-shaped,
with a sandy area occupying the concave side next to shore
(Figure 33). The outer, higher parts of this groin (lying
between the 1.68 and 1.83m levels) form a smooth platform,
covered by a thick turf of blackish cyanophytes; on the
somewhat lower inner margins of the groin there is a crum
bly, greenish-brown crust of algal-bound sediment. Both
zones are inhabited by Lydia annulipes.
99
Figure 33 - Distant View of Rock Groin at Uliga Island
Station 47, from Hotel Roof.
Center: Higher conglomerate platform.
Left and Right: Intertidal reef flat,
viewed at high tide.
*S23wmw>!
Figure 34 - Dark Green Pitted Zone, Seaward Margin of Rock
Groin at Uliga Island Station 47.
Left and center: Eroding conglomerate.
Right: Tide pools at edge of reef flat.
Distant background: Emergent reef flat and
channel-like pools.
100
Eroding Borders of Conglomerate Platforms.- The mar
gins of windward conglomerate flats in the Southern Mar
shalls generally take the form of a rough, pitted, dark
green ramp (Figure 34). This zone is very irregular in
topography, with miniature ridges, valleys, and knobs, and
is covered by numerous pits under 1cm in diameter; much of
the rock surface is encrusted by a tiny, dark green alga.
The inclination of the rock surface is as much as 45° in
places, but generally it is much less; typically there is a
rise of 6 0cm in 3 meters, the usual width of the zone. In
various places this conglomerate margin extends between the
0.9m and 1.52m levels. This habitat was sampled in a num
ber of places, such as the windward islands of Darrit (sta
tion 46) and Kwajalein (station 32), and the eastern end
of the south-facing reef at Dalap Island (station 48A).
The crab fauna includes the xanthids Pseudozius pacificus,
Dacryopilumnus rathbunae, Zozymodes biunguis, and Lydia
annulipes, and the grapsid Pachygrapsus minutus. At Uliga
Island (station 47), some Xantho sanguineus were taken
from orange-bottomed tide pools in this zone, on the sea
ward face of the groin.
Along the more protected sides of conglomerate plat
forms which face inter-island channels, the surface is also
rough, though very different in nature. Here there are
numerous, interconnected holes up to 5cm in diameter, and
jagged coral skeletons show in relief; the exposed rock is
101
light pink in color, due to a mineral deposit over much of
the surface (Figure 35). At Uliga Island (station 47), the
flanks of the groin are of this nature, and have a steep
slope, dropping 60cm in 1 meter. Lydia annulipes is very
abundant in this situation, and Pseudozius caystrus also
occurs. A similar conglomerate border is seen along the
outer face of the lagoonward bar at Dalap Island, where
mussels are very abundant in the rock crevices.
Much farther west along the south side of the atoll
there is a massive bed of conglomerate which is believed to
be located at Rotoin Island (station 51); this was visited
in 1969 only. This formation extends halfway across the
narrow seaward reef, and the exposed outer parts are ve
neered by pink coralline algae. On an incoming tide, waves
surge through narrow channels in the conglomerate, which
undercut the rock in places. Rounded boulders are seen in
relief along the walls which face these channels (Figure
36). The bleached, vermetid-covered rock surfaces of these
walls are inhabited by xanthid crabs such as Dacryopilumnus
rathbunae, Lydia annulipes, and Zozymodes pumilis.
Reef Boulders.- On Kwajalein Atoll, the inter-island
reef flat north of Kwajalein Island (station 33), which is
easily traversed at low tide, is covered with numerous
rounded boulders. A typical example is lm long and 6 6cm
high (Figure 37). These boulders harbor a most diversified
assemblage of organisms, including large acorn barnacles
102
Figure 35 - Porous Flanks of Rock Groin at Uliga Island
Station 47.
4" (10cm) rule to left.
Figure 36 - Eroding Conglomerate Along Inner Reef Channel
at Rotoin Island Station 51.
6" (15cm) rule on rock in foreground.
Figure 37 - Boulder on Reef North of Kwajalein Island
Station 33.
Showing acorn barnacles and fleshy
encrusting algae.
Figure 38 - Rubble in Cut-Off Pool Behind Seaward Reef
at Dalap Island Station 48A.
Showing rubble and Jania clumps;
reduced zone appears in hole in sand
just above 6" (15cm) rule.
104
(several centimeters wide), a tough, maroon encrusting al
gae, boring barnacles and clams, and large vermetids.
Dacryopilumnus rathbunae and Eriphia scabricula constitute
the xanthid fauna. Farther north on the reef, beyond the
first small islet, these boulders are extremely numerous;
evidently this is an important habitat type on Kwajalein.
Rubble Heaps.- On Majuro, a considerable amount of
rubble is strewn across the back parts of the windward reef
flats, between islands. Much of this material is formed
into rubble trains, oriented parallel to the direction of
current flow across the reef at high tide. One example
west of Arniel Island (station 42) was 3m wide and about
60m long, and composed of rocks up to 60cm in diameter
piled together with gravel. Such masses of rock must be
torn apart piecemeal to obtain the crabs that live within.
The xanthid species Xantho danae, Etisus demani, Pilodius
areolatus, Actaea tomentosa, and Chlorodiella cytherea
were collected here.
Rubble Flats and Beaches.- On the windward side of
Majuro, sand and rubble spits trail off into the lagoon at
the ends of the islands, and the beaches facing the inter
island channels in such places are rather steep. One such
beach on the east side of Arniel Island (station 43) rises
1.8m in 10m. The lower beach is covered with rocks up to
30cm long, lying on sand and gravel, whereas above the
1.22m level, the rocks are smaller and more rounded and lie
105
upon gravel with little sand. Extensive collections were
made here at various levels; Xantho danae was found at the
foot of the beach, and Xantho sanguineus, Xantho gracilis,
and Pseudozius caystrus higher up.
On the south side of Majuro Atoll, a rock bar cuts off
a pooled zone at a level above that of the seaward reef
flat, at Dalap Island (station 48A). Small rocks, up to
13cm long, are embedded in white sand composed of broken
segments of the coralline alga, Jania, which is abundant on
the reef (Figure 38). Removal of these rocks reveals a
black zone of reduced organic matter under the surface; the
dominant xanthid crab, Xantho leptodon, burrows readily
into this layer.
Protected Coral Areas.- On Majuro, a few of the chan
nels on the lagoon side of the windward reef were found to
have a prolific growth of coral at the level of the low
spring tides. One coral, the finely branched, grey Monti-
pora racemosa, forms large reef platforms which are compact
enough to support a man's weight. One such platform, at
Ejit Island (station 45), was broken apart for crabs, which
live between the branches. The xanthids Pilodius areola-
tus, Pilodius pilumnoides, Actaea tomentosa, Chlorodiella
cytherea, and Phymodius ungulatus were collected.
Comparison of the Reefs in the Northern
and Southern Marshalls'
There is in general a very strong similarity between
the reefs at Eniwetok and those of Majuro and Kwajalein.
106
All of these atolls have a very similar horizontal and ver
tical zonation on the windward reef; and the erosional al
gal ridge remnants, conglomerate flats and their eroding
margins, and other features of the reef are much the same
and have a similar fauna. Both Eniwetok and Majuro have
elongate islands and high, cut-off pools on the southward-
facing reefs.
There are a few minor differences in the crab fauna,
however. One major habitat type, the sheltered lagoon reef
of the southeastern sector of Majuro Atoll, was not found
at Eniwetok; and the characteristic crab inhabitant of this
reef, Pilumnus vespertilio, was likewise not found there.
Several other crabs, notably Actaea tomentosa, "Chloro-
dius" miliaris, and Pilumnus purpureus, which are impor
tant on Majuro, were not found on Eniwetok. Certain inter
tidal forms (Eriphia scabricula, Lydia annulipes, and Pseu-
dozius pacificus) seem to be much more abundant on Majuro,
probably due to a more heterogeneous topography (with more
crevices for cover) in certain reef habitats in the south
ern Marshalls. Otherwise, the crab populations of the two
atolls are largely the same, and the eight study species
constitute the major part of the intertidal xanthid fauna
in both places. In general, the survey made of crabs on
Majuro and Kwajalein strongly supports the findings of the
major study at Eniwetok.
Summary
In this section descriptions have been given of the
geography, climate, and character of the reefs of three
Marshallese atolls, Eniwetok, Kwajalein, and Majuro, which
were visited in the course of this study. Some of the more
important differences between islands and reefs with a dif
ferent exposure are indicated; and details are given for
each important study area. For the most part the interti
dal xanthid fauna, and the habitats occupied by this fauna,
are the same for different atolls in the Marshall Islands.
CHAPTER III
TAXONOMY, BIOMETRICAL, AND POPULATION DENSITY
DATA FOR THE STUDY SPECIES
Introduction
As some of the species under consideration are
poorly known or in an uncertain taxonomic position, a
historical review of the nomenclature is provided, along
with the justification for the generic names used in fc&is
paper. The geographical distribution reported in the
literature is also given, along with any new range
extensions.
A description is given of the morphology and color
ation of each species with special regard to characters of
value in distinguishing closely related or easily con
fused species. Data on the mean size of the crabs are
reported; these data are tabulated separately for major
habitat types in cases where there appear to be habitat-
related differences in size. The mean size and size
range of males and females are also given, along with the
extent of sexual dimorphism in size, the relative numbers
of males and females, and the abundance of ovigerous
females. Finally, population density data are presented
for a number of habitat types and localities.
108
109
In addition to the eight study species, there are
four other species which are likely to be encountered in
the intertidal zone above the infralittoral fringe in the
Marshall Islands; as they may be confused with the study
species, partial descriptions for these species are also
provided.
Descriptions of the Study Species
Dacryopilumnus rathbunae Balss, 1938
{Figure 39)
Dacryopilumnus emerita, Rathbun, 1911, p. 228, Pi. 16,
fig. 6, 7.
Dacryopilumnus sp. (emerita adult?), Balss, 1932,
p. 514-515.
Dacryopilumnus sp. (? rathbunae), Balss, 1938, p. 67.
Nullicrines amplifrons Edmondson, 1935, p. 32, fig.
lOa-e, PI. Ila.
Dacryopilumnus rathbunae, Sakai, 1939, p. 525-526,
Pi. XCIX, fig. 2.
This species was first tentatively recognized by
Balss (1932) on the basis of a specimen illustrated by
Rathbun (1911); Balss regarded it as either a possible
adult of the smaller Dacryopilumnus emerita (Nobili, 1907)
or as a new species. Should the latter be the case, Balss
suggested in his text that it should be named D. rathbunae.
Sakai (1939) was the first to regard this species as
unquestionably distinct from D. emerita and provided a
diagnosis that is useful in distinguishing the two species.
110
METRIC 1
Figure 39 - Dacryopilumnus rathbunae.
Range. Peros, Coin (Rathbun, 1911), Christmas Island,
Indian Ocean (Gibson-Hill, 1947), Cocos Keeling Atoll
(Tweedie, 1950), Amboina (Balss, 1932), N. Daitozima
(Sakai, 1939), Christmas Island, Central Pacific Ocean
(Edmondson, 1935). This species is abundant on Eniwetok,
Kwajalein, and Majuro Atolls, in the Marshall Islands
(this study).
Description. Carapace high relative to length, convex as
viewed from the side. No interorbital or anterolateral
teeth. Eyes large, placed at sides of carapace. Carapace
and outsides of claws covered with low tubercles, the
latter being surrounded by very short setae (especially
on claws). Claws small in size, fitting closely against
carapace. Fingers short, curved inwards towards tips;
cutting edges provided with thin, blunt, interdigitating
teeth. Meri of claws not provided with setae; setae of
legs scattered, bristle-like. Few setae on carapace
around incurrent openings to gills.
Coloration. Carapace greenish anteriorly, mottled with
white, yellow, pink, purple, and brown posteriorly; often
with two brownish blotches on either side between orbits,
two other lateral blotches, and a medial posterior blotch.
Eyes purple, eyestalks green. Carpus of claws yellowish
with brown blotch; propodus greenish (may be brownish on
large claw) with white tubercles proximally. Fingers
black, with white distal teeth. Legs white, yellow, or
112
TABLE 2
MEAN WIDTHS OF CRABS
Species Habitat Mean
Width (mm)
No.
Crabs
Dacryopilumnus
rathbunae
Algal Mounds, Eroding
Conglomerate, Beachrock
6.4 138
Eriphia
scabricula
Reef Flats, Algal Mounds,
Eroding Conglomerate,
Beachrock
13.1 106
Eriphia sebana Rock Groin, Station 14C 44.6 15
Lydia annulipes Algal Mounds, Eroding
Conglomerate
9.1 124
Beachrock, Cemented
Conglomerate Slabs
13.1 40
Pseudozius
caystrus
Loose Beachrock,
Rubble
13.9 147
Cemented Conglomerate
Slabs
11.3 95
Pseudozius
pacificus
Dark Green Pitted
Eroding Conglomerate
7.4 27
Xantho gracilis Beachrock, Station 8A 11.3 69
Other Rubble 12.5 200
Xantho leptodon Rubble Flats, Beaches 14.3 98
Reef Flats, Stations 4,
47
10.0 107
Edge of Beach, Station 3 6.4 66
Xantho Beachrock Slabs 23.0 45
sanguineus Rubble Beaches 21.5 74
Rubble Heaps, Station 12B 21.2 46
Reef Flats, Stations 4,
11B, 47, 48A
14.2 46
Scoured Reef Flat
Pools, Station 31
11..0 34
1
Width of Crabs Measured Across Widest Point on Carapace.
113
TABLE 2- Continued
Species Habitat Mean
Width (mm)
No.
Crabs
Zozyxnodes Inner Reef, Station 2 4.6 42
biunguis Inner Reef, Station 4 5.2 102
Eroding Conglomerate,
Stations 12A, 12B
4.5 50
Eroding Edge of Rock
Groin and Reef Flat,
Station 13A
5.2 42
Zozymodes
pumilis
Algal Mounds, Eroding
Conglomerate
6.2 26
"Chlorodius"
miliaris
Inner Reef, Station
47
7.0 21
114
TABLE 3
MEAN SIZE OF MALE AND FEMALE CRABS
Species Habitat Mean
Males
Width (mm)
Femalesl
Sexual Di
morphism2
No.
Crabs
Dacryopilumnus
rathbunae
all 5.9 7.0
(7.4)
1.20 138
Eriphia
scabricula
all 13.2 14.0
(16.7)
1.06 97
Lydia
annulipes
all 12.3 13.4
(18.1)
1.10 144
Pseudozius
caystrus
all 13.4 14.3
(14.8)
1.06 201
Xantho
gracilis
all 14.0 11.7
(11.4)
1.20 122
(25)
Xantho
leptodon
Rubble 15.2 12.1
(12.0)
1.25 78
Reef
Flats
11.1 10.8
(12.2)
1.03 88
Xantho
sanguineus
Beach
rock
25.6 22.3
(22.6)
1.15 40
Other
Rubble
22.8 19.4
(19.6)
1.17 118
Reef
Flats2
18.8 16.7
(19.2)
1.09 32
Zozymodes
biunguis
Reef
Flats^
5.9 5.4
(6.0)
1.09 186
Reef
Flats2
4.9 4.7
(5.0)
1.04 137
•^•Ovigerous Females in Parentheses.
9
Mean Width of Larger Sex in mm/Mean Width of Smaller Sex.
in mm.
^Except Station 31.
^Mean Width of Crabs Over 5mm.
^Mean Width of Crabs Under 5mm.
115
TABLE 4
SIZE RANGE AND SEX RATIOS OF
MALE AND FEMALE CRABS
Species Size Range
Males
in mm 1
Females
Percent of
Females
No.
Crabs
Dacryopilumnus
rathbunae
4.0-7.6 4.2-11.1
(5.2)
62.5
(27.6)
168
Eriphia
scabricula
6.1-24.3 5.1-23.9
(10.2)
49.5
(32.7)
97
Eriphia
sebana
up to
60.3
up to
58.7
Lydia
annulipes
5.6-20.8 5.4-23.6
(14.4)
61.0
(12.6)
144
Pseudozius 6.3-22.2 7.1-25.2 61.7 201
caystrus (9.8) (22.5)
Pseudozius 4.0-14.7 5.4-17.6
pacificus (6.9)
Xantho 7 .4-20.6 6.9-14.7 51.0 236
gracilis (8.5) (13.0)
Xantho 5.3-24.9 5.1-19.6 55.0 118
leptodon (8.5) (60.0)
Xantho 7.8-39.6 9.1-31.0 61.5 190
sanguineus (15.2) (47.8)
Zozymodes
biunguis
2.8-10.6 2.9-9.7
(3.4)3
54.9
(37.3)
303
Zozymodes
pumilis
3.6-7.3 5.6-8.0
(6.3)
"Chlorodius" 4.8-11.6 6.4-9.3
miliaris
1Smallest Ovigerous Female in Parentheses.
TABLE 4- Continued
^Based on Total Number of Crabs Whose Sex Could Be
Determined. Percentage of Ovigerous Females Given
in Parentheses (Based on Total Number of Females).
^One Unusually Small Ovigerous Female Was 2.9mm Wide.
117
TABLE 5
POPULATION DENSITY ESTIMATES
Species Station Habitat Population Density
(No. Crabs/m^)
Dacryopilumnus
rathbunae
Eriphia
scabricula
Lydia
annulipes
Pseudozius
caystrus
Pseudozius
pacificus
Xantho
gracilis
Xantho
leptodon
2
12A, B
18
14D
7C
4
4
4
46
47
12A
7C
32, 46,
48A
18
9B
7C
9B
5A
5A
6
32, 46,
48A
5C
6
13A
4
47
48A
Algal Ridge Remnants 14.4
Conglomerate Rims 19.7
Edge of Conglomerate 16.0
Flat
Rimmed Beachrock 23.3
Sloping Beachrock 5.2
Outer Reef Flat 8.6
Goniolithon Zone 2.7
Inner Reef Flat 1.0
Inner Reef Flat 4.0
Dark Green Inner Reef 1.4
Conglomerate Rims 3.1
Sloping Beachrock 15.0
Algal Ridge Remnants 9.7
Dark Green Pitted Zone 17.0
of Conglomerate
Wall-like Margin of 16.0
Conglomerate Flat
Exfoliating Slabs of 9.8
Conglomerate
Sloping Beachrock 1.7
Exfoliating Slabs of 61.5
Conglomerate
Under Rocks on Gravel 27.9
Sandy Rubble Beach 21.0
Sandy Beach with 1.8
Boulders in High Tide
Zone
Dark Green Pitted Zone 18.8
of Conglomerate
Rubble Beach 15.0
Rubble Beach 2.0
Rubble Flat 13.5
Inner Reef Flat 29.2
Inner Reef Flat 25.6
Rubble in Pond behind 19.2
Reef Flat ______________
118
TABLE 5- Continued
Species Station Habitat Population Density
(No. Crabs/m^)
5C Rubble Beach 15.6
Xantho 4, 11 Inner Reef Flat 3.7
sanguineus 48A Inner Reef Flat 16.9
46 Inner Reef with Rubble 4.0
12B Rubble Heaps 11.8
5A Under Rocks on Gravel 5.4
5C Rubble Beach 1.7
Zozymodes 2 Inner Reef Flat 33.4
biunguis 4 Goniolithon Zone 242.0
4 Inner Reef Flat 104.0
48A Inner Reef Flat 127.8
13A Reef Flat with Pink
Pasty Material
32. 0
31 Scoured Inner Reef 58.0
46 Scoured Inner Reef 22.9
5D Pitted Zone of Inner
Reef
215.0
12A, B Conglomerate Rims 146.0
Zozymodes 2 Algal Ridge Remnants 1.2
pumilis 12A Conglomerate Rims 4.6
119
green, with brown bands and bluish articulations.
Variations: some individuals tend towards yellow, orange,
red, or purple.
Biometrical Data. The mean width of this species is 6.4mm
(table 2); males range up to 7.6mm, females to 11.1mm
(table 4). Females are decidedly larger than males
(table 3).
Population Densities. This species attains a population
density of 23.3 crabs/m^ (table 5); it is most abundant in
the lower half of the intertidal on eroding algal rock,
conglomerate, and rimmed beachrock margins. It is much
less common on sloping beachrock; the figure given for
Eniwetok Island station 7C is probably an over-estimate,
as this species was seldom seen on the beachrock north or
south of this point.
Adaptive Morphology and Coloration. The widely spaced
eyes of these crabs enable them to spy from the apertures
of their holes in the daytime, when they are very wary.
Some of the color phases appear to camouflage the crabs
well; for instance, specimens from the eroding beachrock
at Runit Island (station 14D) tended towards a yellow
color similar to the algal film on the rocks, and crabs
from Rigili Island (station 18) tended towards a purple
color similar to the background of vermetid-covered
conglomerate. Individual crabs with brighter coloration
were easily seen in some localities, however.
Eriphia scabricula Dana, 1852
(Figure 40)
Eriphia scabricula Dana, 1852, p. 247, 1955, PI. 14,
figs. 5a, b.
, Sakai, 1939, p. 523, Pi. 99,
fig. 3.
Range. From the Red Sea and the East African Coast east
to Southern Japan, Wake, Hawaii, Fanning and Palmyra
Islands, and the Tuamotu Archipelago (Edmondson, 1925,
Balss, 1938, Me Neill, 1968).
Description. Carapace relatively narrow; eyes widely
separated. Six pairs of sharply pointed antero-lateral
teeth; anterior and lateral carapace and outer surfaces
of claws covered with low tubercles and scattered, bristle
like setae. Claws massive; fingers with pointed tips and
long, thin cutting teeth. Heterochelous; one claw
slightly larger, with heavier, blunter teeth towards base
of fingers. Legs relatively elongate, with long, bristle
like setae. Little setation around incurrent openings
to gills.
Coloration. Carapace and claws mottled pink, green, white
and black, with numerous black dots. Outside of hand
with some yellow coloration; fingers light. Legs
yellowish-white with black, pink and green mottling; setae
on carapace and limbs pink, orange, or yellow. Variations
122
some specimens tend towards pink or green; sometimes a
bright pinkish-red.
Biometrical Data. The mean width of this species is 13.1
mm (table 2); males range up to 24.3mm, females to 23.9mm
(table 4). There is little sexual dimorphism (table 3).
Population Densities. This species was found in maximum
population densities of 8.6 crabs/m^ on the outer reef
flat at Eniwetok station 4, and 15 crabs/m^ on lagoon
beachrock (table 5). It is not abundant on many reef
flats on Eniwetok Atoll, probably in part due to the
absence of holes large enough to accommodate it. Whereas
it is somewhat more characteristic of algal rock, eroding
conglomerate, and beachrock on Eniwetok, it is fairly
common on a number of reef flats on Majuro and Kwajalein
Atolls.
Adaptive Morphology and Coloration. The widely spaced
eyes of these crabs enable them to spy from the apertures
of their holes before emerging; this species is very wary
during daylight hours, and unlike other xanthid species
can run very rapidly, proceeding with agile, catlike move
ments. The variegated coloration of this species appears
to camouflage it well on some reef flats and beachrock.
Lydia annulipes (H. Milne Edwards, 1834)
(Figure 41)
Figure 41 - Lydia annulipes.
124
Ruppellia annulipes H. Milne Edwards, 1834, p. 422.
, Dana, 1852, p. 246, 1855, Pi.
14, fig. 4a-c.
Ozius (Euruppellia) annulipes, Alcock, 1898, p. 188.
, Odhner, 1925, p. 85.
Lydia annulipes, Edmondson, 1925, p. 42.
, Balss, 1938, p. 66.
, Sakai, 1939, p. 521, Pi. 64, fig. 3.
Range. From the coast of East Africa to Southern Japan,
Wake, Hawaii, Fanning and Palmyra Islands, and the
Tuamotu Archipelago (Edmondson, 1925, Mac Neill, 1968).
Description. Carapace with six pairs of bluntly-pointed
antero-lateral teeth; anterior and lateral regions of
carapace deeply sculptured. Eyes large. Carapace and
outer surfaces of claws finely granulated; outer surface
of propodus of claw wrinkled in appearance. Markedly
heterochelous; small claw slender, with downward-bent
fingers set with low, sharp, recurved teeth and pointed
tips. Large claw massive; fingers not downward-bent,
with heavy, blunt teeth (especially the large tooth at
base of dactyl) and bluntly-pointed tips. Few setae on
carapace or limbs, except for a thick, very short
covering on dactyls of walking legs.
Coloration. Carapace dull yellow, with purple spots on
medial and lateral regions, on anterolateral teeth, and
around orbits. Gastric regions with a pair of larger
125
blotches, and posterior carapace with a median, unpaired
blotch. Claws yellowish, with purple spots along upper
part of carpus and propodus and around distal margins of
carpus. Fingers of large claw white, those of small
claw reddish. Legs with purple bands on proximal portion
of merus and on articulations distal to this. Very small
individuals mottled white and light olive, with fingers
of small individuals mottled white and light olive, with
fingers of small claw pink.
Biometrical Data. The mean width of this species ranges
from 9.1mm in pitted algal rock and conglomerate to 13.1mm
in beachrock and under slabs of conglomerate (table 2);
males range up to 20.8mm, females to 23.6mm (table 4).
This species exhibits moderate sexual dimorphism, females
being larger (table 3). In many localities where Lydia
annulipes is abundant, only a small percentage of the
population is of adult size, probably because of the small
size of the holes in some of the eroding conglomerate
margins. An estimation of the number of sexually mature
individuals can be made by counting all those which are
equal to or greater in size than the smallest ovigerous
female. In localities where the mean width of the crabs
is over 12mm, 62 percent of the crabs are estimated to be
mature; in places where the mean width is 10.2mm or less,
only 12.6 percent are apparently mature. The fact that
juveniles predominate in many of the populations may help
explain why such a small percentage of females was
recorded as ovigerous (table 4).
Population Densities. This species occurs in a maximum
population density of 17 crabs/m2 in eroding conglomerate
(table 5); it is very abundant in this type of habitat,
and in the algal rock at station 2. It is less abundant
in beachrock. The value given in table 5 for exfoliating
conglomerate slabs at Sand Island station 9B is based on
the number of crabs per unit area of the slabs removed;
if allowance is given for the actual area of the flat
covered by these slabs, a much smaller corrected figure of
0.5 crabs/m is obtained. However, on Majuro, certain
conglomerate flats have numerous eroding coral heads
which probably allow this form to occur in much greater
population densities.
Adaptive Morphology and Coloration. This species is
particularly slow and deliberate in its movements, and when
disturbed will hold onto the rock surface with tenacity.
The tufts of setae on the dactyls of the walking legs may
be an adaptation which aids it in gripping the rocks. Its
coloration blends in fairly well with the dull orange
color of the rock surfaces in the high inter-tidal in
many localities, although it is relatively inactive in
the daytime. It should be noted that the adults of both
127
Lydia annulipes and Eriphia sebana, which appear to be
camouflaged for the same kind of background, have a more
solid color pattern than do small juveniles, which are
often mottled.
Pseudozius caystrus (Adams and White, 1848)
(Figure 42)
Panopeus caystrus Adams and White, 1848, p. 42, PI.
9, fig. 2.
Pseudozius planus Dana, 1852, p. 233, 1855, PI. 13,
fig. 6.
Pseudozius caystrus, Alcock, 1898, p. 181.
" " , Balss, 1938, p. 64.
, Sakai, 1939, p. 514.
Range. From the Red Sea to the Bonin Islands, Wake,
Hawaii, Fanning and Palmyra Islands, and the Tuamotu
Archipelago (Edmondson, 1925, Balss, 1938, Sakai, 1939,
Forest and Guinot, 1961)
Description. Carapace smooth and flat, with hardly any
sculpturing; three pairs of greatly reduced anterolateral
teeth. Eyes very small. Claws smooth, heterochelous.
Smaller claw with sharp, recurved teeth on fixed finger
and reduced teeth on dactyl; fingers tips pointed. Large
claw with a large, blunt tooth proximally on dactyl and
a blunt tooth slightly distal to it on fixed finger;
finger tips bluntly pointed. Little setation on
carapace or limbs except for some distal setae on legs.
Lef t:
Right:
Figure 42 - Pseudozius caystrus«
Light color phase typical of females.
Dark color phase typical of males.
Coloration. Females generally darker than males.
Females often purple or dark blackish-brown; posterior
carapace and legs often lighter. Some dark-colored
crabs with anterior carapace and legs purple, posterior
carapace orange; others occasionally with white claws.
Some females with coloration similar to usual male
pattern. Males often white with light gray or brown
mottling on anterior carapace and claws; legs purple,
purple-white, or white with purple spots. Some males
solid white in coloration. Immature specimens white,
dull orange, or light brown, with purple legs. Claws
with black fingers.
Biometrical Data. The mean size of this species ranges
from 11.3mm to 13.9mm (table 2); males range up to 22.2mm,
females to 25.2mm (table 4). There is little sexual
dimorphism (table 3).
Population Density Data. In loose rubble, this species
has been found in population densities of up to 27.9
crabs/m^ (table 5). It is very common where there is
adequate rock cover, as at Eniwetok Island station 5;
where there are few covering rocks, as at station 6, it
is relatively rare. The comments given above in
discussion of the calculation of population densities of
Lydia annulipes on exfoliating conglomerate flats also
apply here; in this case, a corrected figure of 1.8
130
2
crabs/m is obtained. Like L. annulipes, this form is
probably found in much greater population densities on
certain conglomerate flats on Majuro Atoll.
Adaptive Morphology and Coloration. The relatively
small eyes of this species apparently relate to its
secretive habits. The white color phase may afford
protective resemblance against a sand background, while
crabs having the dark coloration may resemble dark
pebbles. The dark phase may also be advantageous to
crabs living in poorly illuminated crevices or in rubble
heaps. It is uncertain why females are consistently
darker than males.
Xantho gracilis (Dana, 1852)
(Figure 43)
Chlorodius gracilis Dana, 1852, p. 210, 1855, PI.
II, fig. 13
Leptodius gracilis, De Man, 1888, p. 287, PI. XI,
fig. 2.
" " , Rathbun, 1906, p. 848, PI.
IX, fig. 2.
, Balss, 1938, p. 42.
" " , Forest and Guinot, 1961, p. 64,
fig. 57, 58a, PI. II, fig. 4.
Xantho (Leptodius) gracilis, Sakai, 1939, p. 465,
pi. 91, fig. 2.
Xantho gracilis, Tweedie, 1950, p. 115.
" " , Holthuis, 1953, p. 27.
131
METRIC 1 .
I»#PS8
Figure 43 - Xantho gracilis.
Range. From the Red Sea and the coast of East Africa to
Southern Japan, Wake, Hawaii, Fanning and Palmyra Islands,
and the Tuamotu Archipelago (Edmondson, 1923; Balss, 1938;
Forest and Guinot, 1961).
Description. Carapace with five pairs of bluntly-pointed
anterolateral teeth; third pair decidedly rounded at
anterior angle. Transverse crests of first and second
medial regions diagonal in orientation; crests of first
in parallel with crests of second on same side. Crests of
second medial regions on opposite sides not in parallel
with each other. Carapace smooth and polished in
appearance, with very fine granualtion. Claws finely
granualted; one claw slightly larger than the other.
Claws with blunt teeth along edges of fingers and with
spatulate finger tips. Carpus and propodus of last pair
of walking legs relatively short and broad. Bushy setae
abundant on meri of claws and legs, and below anterolateral
and posterolateral margins of carapace.
Coloration. Light or dark gray, green-, yellow-, white-,
or pink-gray, or olive; mottled patterns common. Often
with large spots on carapace; variations include: gray
with black median spot on posterior carapace; gray with
brown patch between eyes; and mottled green-gray with
white median spot on anterior carapace and brown posterior
spot. Claws similar in color to carapace, often with
133
white spot on carpus. Fingers black.
Biometrical Data. The mean size of the species is 11.3
to 12.5mm (table 2); males range up to 20.6mm, females
to 14.7mm (table 4). This species is decidedly sexually
dimorphic, males being larger (table 3).
Population Density Data. This species was found in a
maximum population density of 15.0 crabs/m (table 5).
In places where there are numerous rocks which provide
these crabs with cover, as at Eniwetok Atoll stations
5 and 13A, they are very abundant; where the rock cover
is sparse, as at station 6, this species is uncommon.
Adaptive Morphology and Coloration. The relatively
broad segments of the last pair of walking legs may relate
to the digging habits of this form. These crabs are well
camouflaged against a background of loose rubble; when
exposed by removal of their covering rocks, they closely
resemble pebbles. The spotted color patterns may be a
form of disruptive coloration. These remarks apply also
to Xantho leptodon and X. sanguineus.
Xantho leptodon (Forest and Guinot), new combination
(Figure 44, 49)
Leptodius exaratus, Nobili, 1907, p. 389.
Leptodius leptodon Forest and Guinot, 1961, p. 65-68,
fig. 55, 56, 59a, b, PI. II, fig.
3, 196-2, p. 64-65.
M fTKIC i t
mjjijifl
Xantho exaratus (H. Milne Edwards, 1834), a species
closely related to the present form, was first placed
in Leptodius, a subgenus of Chlorodius, by A. Milne
Edwards (1863). De Man (1888) considered Leptodius to be
a genus, in which he placed Xantho sanguineus, another
related form. Alcock (1898) regarded Leptodius as a
subgenus of Xantho; he distinguished Leptodius from Xantho
proper on the basis of the hollowed finger tips, a
character originally given by A. Milne Edwards (1863)
for Chlorodius, in which Leptodius was formerly included.
Holthuis (1954), who was instrumental in having the name
Xantho (Leach) added to the "Official List of Generic
Names in Zoology," considers Leptodius to be synonymous
with Xantho, and has referred Xantho gracilis and X.
sanguineus to this genus (Holthuis, 1953). Forest and
Guinot (1961) recently named the present species
Leptodius leptodon, which they regard as close to Xantho
exaratus and X. gracilis; it is appropriate therefore to
refer their new species to the genus Xantho.
Range. As this species was only recently described, it is
undertain how extensive is its range because previous
material was probably identified as Xantho exaratus.
Forest and Guinot (1961) examined material from Vahitahi,
Hikueru, and the Caroline Islands; in a later paper
(Forest and Guinot, 1962), the range of Xantho
leptodon was given as Melanesia, Polynesia, and
western Micronesia. However, they also indicate that
136
X . exaratus occurs in these areas, as well as the
Marshall Islands and other parts of the Pacific; this
is apparently based on previous literature which dates
to a period before X. leptodon was described.
This broad distribution for Xantho exaratus in the
Pacific is highly suspect. All of the material collected
from Eniwetok, Kwajalein, and Majuro Atolls in the
Marshall Islands in the course of this study is refer
able to Xantho leptodon; if X. exaratus occurs in the
Marshall Islands at all, it must be very rare. Some
material in the Allan Hancock Foundation collections
collected by F.C. Ziesenhenne on Samoa was identified
as X. leptodon, as was a specimen collected by R. Kantor
and R. Emerson on Rarotonga. Some Xantho specimens from
the Ryukyus on the other hand correspond very well with
descriptions of X. exaratus.
According to Forest and Guinot (1961), another
species with which Xantho leptodon may easily be confused
is Xantho australis (Ward); a specimen in the Allan
Hancock Foundation collections from the Great Barrier
Reef (received from A. McLean) corresponds to Ward's
(1936) description and illustration (though these are not
clear) and can be distinguished from X. leptodon by the
shape of the anterolateral teeth. Xantho australis
should probably be redescribed and contrasted with both
■H* leptodon and X. exaratus.
Description. Similar to X. gracilis, but carapace
narrower and flatter, with more relief and with sharper
anterolateral teeth. Transverse crests of first medial
regions decidedly anterior in position to those of second
medial; crests of first medial diagonal in orientation.
Crests of second medial regions on opposite sides in
parallel with one another and transverse in orientation;
not in parallel with those of first medial on same side.
Coloration. Yellow-tan, green, green-gray, or gray;
often mottled with various combinations of white, green-
grown and/or black spots. Darkest recorded individual
mottled steel-blue and black. Often with large spots
on carapace; variations include: light with two large
dark spots along midline (one specimen white with black
spocs), white sides with dark brown center region, green
with white center strips, and green with white band
across front (between orbits). Small individuals
sometimes solid in color, white or pink.
Biometrical Data. The mean size of this species is
generally from 10.0mm on reef flats to 14.3mm for crabs
on rubble beaches; in places where the mean size is much
smaller than these figures (as at station 3), most of
the animals are immature (table 2). Males range up to
24.9mm, females to 19.6mm (table 4). This species is
138
sexually dimorphic, males being larger than females; it
should be noted that sexual dimorphism is less on reef
flats than on rubble beaches (table 3), where the crabs
attain a larger mean size.
Population Density Data. This species attains a maximum
population density of 19.2 crabs/m2 on rubble beaches and
flats (table 5). On reef flats it may be found in
densities as high as 29.2 crabs/m , though xt should be
noted that it is abundant on relatively few reef flats.
Xantho sanguineus (H. Milne Edwards, 1834)
(Figure 45)
Chlorodius sanguineus H. Milne Edwards, 1834,
p. 402.
, Dana, 1852, p. 207, 1855, PI.
II, fig. 11a, b.
Leptodius sanguineus, De Man, 1888, p. 285.
, Rathbun, 1907, p. 39.
" " , Forest and Guinot, 1961, p. 63,
fig. 50a, b.
Xantho (Leptodius) sanguineus, Alcock, 1898, p. 119.
, Sakai, 1939, p. 464,
PI. 90, fig. 3.
Xantho sanguineus, Tweedie, 1950, p. 117.
" " , Holthuis, 1953, p. 27.
Range. From the Red Sea and the coast of East Africa to
Southern Japan, Marcus Island, Hawaii, Fanning and Palmyra
Islands, and the Tuamotu Archipelago (Edmondson, 1925,
Figure 45 - Xantho sanguineus.
Forest and Guinot, 1961, Sakai, 1965).
Description. Similar to Xantho gracilis and X. leptodon,
but with six pairs of anterolateral teeth; sixth tooth
behind maximum width of carapace. Carapace well
sculptured; with distinct crests on frontal lobes,
immediately before and paralleling those of first medial
lobe. Medial regions otherwise as in X. Leptodon.
Coloration. White-, yellow-, green-, or pink-gray, white
brown, or green and black; coloration solid or finely or
coarsely mottled. Often with large spots on carapace;
variations include: green-gray with an anterior and a
posterior median maroon spot, yellow-gray with an
anterior and a posterior black spot, white or gray with
a median posterior brown spot, yellowish with white
stripe along center of carapace, or red-brown with light
yellow patch between eyes. Claws same as carapace or
purplish with pale orange spots on outside of propodus;
legs same as carapace or dark red or brown. Setae around
margins of carapace and on bases of legs often orange.
Small individuals often white (one specimen pink) with
red or purple on legs distal to merus (one specimen with
first three pairs of legs only red).
Biometrical Data. The mean size of this species ranges
from 11mm on scoured reef flats and 14.2mm on other reef
flats to 21.2-23.Omm in loose rubble situations (table 2)
141
Males range up to 39.6mm in width, females to 31.0mm
(table 4). This species exhibits sexual dimorphism,
males being larger; however there is less dimorphism in
reef-flat crabs than in animals from rubble habitats,
where the mean size is larger (table 3).
Population Density Data. In loose rubble, this species
was found in greatest abundance (11.8 crabs/m ) in semi
consolidated rubble heaps at Japtan Island (station 12B),
and was also abundant at station 5A, where large rocks
provided continuous cover over gravel (table 5).
Generally, it is much less abundant than in these
localities; the figure given for station 5C, where the
rubble was loose and more thinly spread on the beach, may
be more typical. On reef flats, it was found in greatest
abundance (16.0 crabs/m ) on the Dalap Island reef
(station 48A), where there was an abundance of holes and
crevices for the crabs to hide in. This species was also
fairly common on the vermetid-Jania zone at Darrit Island
(station 46), where many of the crabs were taken from
under rocks. There are many reef flats, however, where
X. sanguineus is rare or absent.
Zosymodes biunguis (Rathbun, 1906)
(Figure 46)
Xanthodius biunguis Rathbun, 1906, p. 849, fig. 12,
PI. 8, fig. 10.
" " , Edmondson, 1925, p. 50.
142
Zozymodes biunguis, Odhner, 1925, p. 83.
" " , Forest and Guinot, 1961, p. 52,
fig. 38.
Zoozymodes biunguis, Balss, 1938, p. 38.
Liocarpilodes biunguis, Guinot, 1964, p. 23, 65.
This species was originally placed in the genus
Xanthodius by Rathbun (1906), who considered it closest
to Leptodius (Xanthodius) cristatus Borradaile, trans
ferred by her to Xanthodius (Rathbun, 1906). Odhner
(1925) placed the latter species, which he recognized
as identical with Zozymodes pumilis Jacquinot, with X.
biunguis in the genus Zozymodes. The latter genus was
first erected by Heller (1861) for the species Z_.
carinipes, now called Z_. xanthoides (Krauss) , which is
almost identical with Z_. pumilis (Forest and Guinot, 1961) .
Balss (1938) also included these three species in the
same genus, spelled by him Zoozymodes, although he stated
Z^. biunguis is very aberrant for this genus, as the upper
surfaces of the pereiopods are only feebly keeled. Guinot
(1964) considers this species to be a member of the genus
Liocarpilodes Klunzinger, on the basis of the form of
the first pleopod of the male.
If a comparison is made between the male first
pleopods of Zozymodes pumilis, 1Z. xanthoides, and Z_.
biunguis figured in Forest and Guinot, (1961, p. 53, figs.
36a, b, 37a, b, and 38) with that of Liocarpilodes
METRIC 1
Figure 46 - Zozymodes biunguis.
144
integerrimus, illustrated by Guinot (1964, fig. 36a, b),
it is not clear that Guinot's reassignment of Z_. biunguis
to Liocarpilodes is justified. While Z_. biunguis lacks
the recurved tip of the first pleopod which characterizes
the other Zozymodes species, the distal spination of this
appendage in Z_. biunguis is similar to the subterminal
spination of these forms, and unlike the distal spination
seen in the corresponding position on the pleopod of
Liocarpilodes integerrimus, which Guinot had used in her
comparison.
As the evidence of the male first pleopods is
equivocal, a comparison of other morphological characters
should be made before a decision can be reached as to
the generic assignment of Zozymodes biunguis. One of
the diagnostic features of the genus Zozymodes given
by Heller (1861) , but lacking in Z_. biunguis is the strong
keeling of the walking legs; while this is a striking
feature of Z. pumilis and other species considered to
belong to this genus by Guinot (1964), the genus
Zozymodes was described on the basis of a single species,
and has never been redescribed taking into consideration
characters of other species subsequently assigned to it.
Indeed, in many respects Zozymodes biunguis seems
intermediate between Z_. pumilis and Liocarpilodes
integerrimus. It has a much less strongly areolated
145
carapace than Z_. pumilis, but the anterolateral borders are
generally similar; whereas L. integerrimus has a smooth,
almost featureless carapace, and only faint traces of
anterolateral teeth. The tips of the fixed and movable
fingers of the chelipeds are smoothly bowed in Z_. biunguis
and Z^. pumilis, while in !L._ integerrimus they are set with
blunt teeth. Another species of Lipcarpilodes, 1. armiger
(Nobili) , is extremely dissimilar to Z^. biunguis and to
other Zozymodes species. For these reasons, then, it is
doubtful that Rathbun's species should be assigned to Lio
carpilodes; it is therefore retained here in Zozymodes.
Range. Hawaii (Rathbun, 1906); Lisianski, Laysan,
Johnston, and Wake (Edmondson, 1925). This species was
found to be extremely abundant at Eniwetok, Kwajalein,
and Majuro Atolls in the Marshall Islands (this study),
and extends west to the Marianas, from which the Allan
Hancock Foundation has several specimens received from
E.R. Tinkham (unpublished).
Description. Carapace granulated, sparsely covered with
setae; five pairs of anterolateral teeth. Claws
granulated, relatively large and massive. Carpus of
claw held at a distance from anterolateral border of
carapace in large specimens (11mm wide). Fingers of claws
bowed, with a few blunt teeth near base; finger tips
spatulate. Heterochelous; one claw longer and more
massive than the other. Dactyls of walking legs with
two points at tip. Posterior surface of propodus of
walking legs with curved, flange-like distal extension,
over which hooks a proximal, finger-like extension from
dactyl. Thick, bushy setae on meri of walking legs and
claws, either side of buccal region, and below antero
lateral borders of carapace.
Coloration. Sides of carapace white, yellow, or greenish
center of carapace with broad brown or black area (solid
or mottled), beginning behind orbits and narrowing
towards posterior margin. Some specimens with three dark
stripes. Claw background color white or yellow, with
closely-spaced brown or black tubercles on middle of
carpus and proximally on propodus. Base of movable
fingers of claw with brown or black spot, surrounded by a
white area; fingers black, legs greenish, mottled or
banded with brown.
Biometrical Data. This is a small species, the mean
width varying from 4.6mm to 5.2mm in various localities
(table 2); males range up to 10.6mm, females to 9.7
(table 4). This species exhibits moderate sexual
dimorphism, especially on reef flats where the mean size
of the crabs is greater (table 3).
Population Density Data. This species may be found in
very high population densities, up to 242 crabs/m^
(table 5); it is unquestionably the most abundant
intertidal xanthid, being found on most reef flats.
Population density estimates for Z_. biunguis vary
considerably from one place to another. Where there is
evidence of erosion of the reef surface, it is relatively
uncommon; this is the case off the south half of Eniwetok
Island (station 2) where the rock surface is very smooth
(presumably due to the activity of grazing fishes), and
in places where there is considerable scour by loose
rubble on the reef surface, as on the reef west of Darrit
Island (station 46) and that off the south end of
Kwajalein Island (station 31). It is also rather uncommon
in places where a pink pasty material composed of algae
and sand covers much of the reef surface, as at Aniyaanii
Island (station 12A). In all of these places, relatively
few holes or crevices in which crabs might hide were
visible.
By contrast, these animals occur in much greater
abundance where there are numerous overhanging crevices
around the borders of tide pools, as on the inner reef
at Eniwetok Island station 4, or at Dalap Island station
48A. Other situations where they occur in high densities
in level areas of the reef are the coralline algal-
covered zone at station 4, where the rock surface is
148
very porous in appearance, and the pitted inner reef at
station 5D. They may also be very common in the eroding
rims of surge channels or in eroding conglomerate, such
as that found along the south side of Japtan Island
(station 12); the figure given for this latter locality
is a maximum density estimate taken from the lower sides
of (and other concave places along) the eroding conglo
merate rims standing out on the reef flat.
Adaptive Morphology and Coloration. The bi-partite tips
of the walking legs may aid in gripping the rock surface
of the reef; this species will adhere to the rocks with
great tenacity when disturbed. The peculiar finger-
and-flange arrangement on the distal ends of the walking
legs functions as follows: when the dactyl is extended
or retracted, its proximal extension follows the flange
on the propodus. Using preserved specimens, one can lock
the dactyl into a retracted position, by pushing it back
ward in relation to the propodus; the finger-like
extension from the dactyl fits into a notch in this
position. While it is not certain how this structure is
actually used in life, it is possible that it aids a
struggling crab in maintaining its grip on the rock
surface. The coloration of Z^. biunguis may afford it
some protective resemblance, but it is uncertain how
beneficial this may be, because this form is seldom seen
149
in the open during the day. It may serve to camouflage
the crabs (which resemble small pebbles), which may forage
during the day in small deposits of gravel in the tide
pools in some localities.
Descriptions of Other Intertidal Species
Eriphia sebana (Shaw and Nodder, 1809)
(Figure 47)
Cancer sebana Shaw and Nodder, 1809, PI. 591.
Eriphia laevimana Dana, 1852, p. 240, 1855, PI. 591.
" " , Alcock, 1898, p. 214.
, Sakai, 1939, p. 522, Pi. 99,
fig. 1.
" " , Tweedie, 1950, p. 124.
Eriphia sebana, Rathbun, 1907, p. 57.
" " , Holthuis, 1953, p. 20.
Range. From the Red Sea and South Africa to Southern
Japan, Hawaii, Fanning and Palmyra Islands, and the
Tuamotu Archipelago (Edmondson, 1923, Forest and Guinot,
1961, McNeill, 1968).
Description. Similar to Eriphia scabricula but lacking
in setation on carapace and claws, and with more antero
lateral teeth. Lateral parts of carapace with tubercles
similar to and continuous with anterolateral teeth. Claws
smooth, lacking tubercles. Heterochely pronounced;
smaller claw with slender fingers and indistinct teeth,
150
151
and larger claw with heavy, blunt teeth.
Coloration. Adults brown or purple-gray, with bright red
eyes (hence the popular name, "red-eye crab"). Very small
individuals variable; one example mottled pink and olive
with black dots, claws purple with black dots, small claw
with red fingers, large claw with pink fingers, legs pur
ple. Another small individual with mottled purple carapace
and yellow claws with purple speckling; others white. Small
specimens lack the bright red eyes of the adults.
Biometrical Data. This is the largest intertidal xanthid
in the Marshalls; the mean width of specimens from Runit
Island, where adults are abundant, is 44.6mm (table 2).
Males range up to 60.3mm in size, females to 58.7mm (table
4). In certain places, like the beachrock at Eniwetok
station 7, it is doubtful whether many individuals reach
adult size.
Adaptive Morphology and Coloration. The adult coloration
may provide some protective resemblance on the dull,
orange-colored rock surface where it often lives.
However, the bright red eyes easily reveal its presence
to the human observer.
Pseudozius pacificus Balss, 1938
(Figure 48)
? Pseudozius inornatus Dana, 1852, p.236, 1855, Pi.
13, fig. 8.
152
\ < ! • . ' ;
V ^ . '
v »
Figure 48 - Pseudozius pacificus.
Showing solid and striped color phases.
153
Pseudozius pacificus Balss, 1938, p. 64-65, PI. II,
fig. 5.
Range. According to Balss (1938), this form is known
from Amboina (south of Ceram) and Jaluit Atoll, Marshall
Islands. It is also found on Eniwetok, Kwajalein, and
Majuro Atolls in the Marshall Islands (this study).
Description. Similar to Pseudozius caystrus in lacking
carapace sculpturing and in having reduced anterolateral
teeth. Anterior and lateral parts of carapace heavily
granulated. Fingers of claws short, with pointed tips;
smaller claw or both claws with an incipient tooth on
inside of claw tip, indicating an incipient spatulate
condition.
Coloration. General color dark brown or purple. Often
with three white or yellow longitudinal stripes, two
lateral and one median; median stripe sometimes reduced
to an anterior or posterior spot. Some individuals
solid dark in color or with a broad orange area in center
of carapace in lieu of stripes. Very small crabs with
four dark stripes on a light background.
Biometrical Data. The mean size of this species from the
southern Marshall Islands is 7„4mm (table 2); somewhat
larger animals were collected from Japtan Island,
Eniwetok (station 12B). Males range up to 14.7mm, females
to 17.6mm (table 4).
154
Population Density Data. A population density of 18.8
crabs/m^ was obtained for the dark green pitted borders
of conglomerate flats on Kwajalein and Majuro Atolls
(table 5). However, this species is very limited in
distribution on Eniwetok Atoll, being collected only
in eroding conglomerate at Japtan Island and under
cemented slabs high in the intertidal at the north end of
the rock quarry at Eniwetok Island (station 5).
Adaptive Morphology and Coloration. The similarity in
coloration between this species and Zoozymodes pumilis
is striking. While both species are found (sometimes in
association) in eroding conglomerate and appear to be
secretive in habits, it is uncertain what adaptive value
this coloration may confer.
Zozymodes pumilis (Jacquinot, 1852)
(Figure 49)
Zozymodes pumilis Jacquinot, 1852, PI. 4, fig. 1.
" " , Jacquinot and Lucas, 1853, p. 20,
PI. 3, fig. 1.
Leptodius (Xanthodius) cristatus Borradaile, 1902,
p. 252, fig. 51.
, Rathbun, 1911,
p. 216, PI. 17,
fig. 9.
Xanthodius cristatus, Rathbun, 1907, p. 41.
Zozymodes carinipes, Nobili, 1907, p. 388.
155
METRIC 1,
Figure 49 - Zozymodes pumilis.
156
Zozymodes pumilis, Miyake, 1939, p. 177, PI. 13,
fig. 2, text fig. 5, p. 208.
" " , Forest and Guinot, 1961, p. 52,
fig. 36a, b.
" " , Guinot, 1964, p. 22, 23.
Zoozymodes pumilis, Balss, 1938, p. 38.
" " Tweedie, 1950, p. 115.
Range. From Madagascar east to the Tuamotu Archipelago
and north to the Caroline Islands (Balss, 1938, Guinot,
1964).
Description. Somewhat similar to Zozymodes biunguis, but
with greater carapace relief. Regions of carapace well
delineated; five pairs of blunt anterolateral teeth.
Carapace and claws coarsely granulated. Fingers short;
claw tips spatulate. Carpal and propodal segments of
walking legs with crests along upper edges; distal ends
of legs with interlocking flange-and-finger arrangement
as in Z_. buinguis.
Coloration. General color dark brown or purple; often
with two green, yellow, or white stripes on either side of
mid-line and continuous anteriorly with anterolateral
borders of same color. Posterior ends of stripes some
times cut off as spots. Some individuals with solid dark
color except for a few orange spots on anterior carapace.
Claws sometimes tend towards bluish-white.
Biometrical Data. The mean size of this species is 6.2mm
157
(table 2); males range up to 7.3mm, females to 8.0mm
(table 4).
Population Densities. This form occurs in a maximum
population density of 4.6 crabs/m (table 5). It is
common nowhere, but fairly widespread in occurrence,
regularly being found in association with Dacryopilumnus
rathbunae.
Species Incertae Sedis
Another small species is A. Milne Edwards' Chlorodius
miliaris (figure 50). Guinot (1964) considers this form
not to be a member of the genus Zozymodes, in which
context Balss (1938) had discussed it, and has provided
an excellent photograph of the type specimen from New
Caledonia. As the generic affinities of this species
are unknown, it will be referred to here under its
original name. This species was collected only on Majuro
Atoll, in the course of this study.
In contrast with Zozymodes biunguis, Z_. pumilis, and
small Xantho leptodon, the intertidal species with which
it is most likely to be confused, there are numerous
cross-striations on the carapace, and the anterolateral
teeth are more pronounced, the third and fourth pairs
being rather sharply pointed. The claws have rows of
tubercles on the propodus, rather than an even, granulated
covering. The clawtips are spatulate, and there is an
Figure 50 - Reef Flat Xanthids from Station 47.
"Chlorodius" miliaris.
(upper left)
Etisus frontalis.
(upper right)
Xantho leptodon.
(He low)
159
abundance of setae on the legs, the meri of the claws,
and below the anterolateral margins of the carapace. In
Z. biunguis and X. leptodon, the carapace is flatter, with
shallower grooves between the regions; and pumils has
keeled walking legs, unlike "Chlorodius" miliaris. The
latter has the flange-and finger arrangement seen on the
distal ends of the legs of the Zozymodes species, although
it is not as robust a structure as it is in biunguis;
this is another feature which will aid in distinguishing
"Chlorodius" miliaris from X. leptodon, which lacks the
structure.
Preserved specimens of "Chlorodius" miliaris show
dark mottling on the medial region and on the posterior
and lateral regions of the carapace. The mean size of
this species is 7.0mm (table 2); males range up to 11.6mm,
females to 9.3mm (table 4). It is rather abundant on
the inner reef at Uliga Island (station 47), where there
is a very thick sandy algal turf, being associated with
Xantho leptodon.
Discussion and Conclusions
The commonest intertidal species in the Marshall
Islands, Zozymodes biunguis, is in an uncertain taxonomic
position, as is another species which is locally
abundant, "Chlorodius" miliaris; a third common species,
Xantho leptodon, was only recently described. The
160
geographical boundaries of these species, and probably
also of Dacryopilumnus rathbunae, Pseudozius pacificus,
and Zozymodes pumilis, are not known with any degree of
certainty. These facts demonstrate how poorly known are
the crab faunas of the tropical Indo-Pacific, although
many collections have been made in this area since the
middle of the 19th century.
Observations on the coloration of intertidal xanthid
crabs indicate that most of the species are camouflaged
in at least some of the habitats where they occur, even
in the case of species which are largely nocturnal.
Biometrical data show that certain species vary in size
from one habitat to another; this is probably due to the
small size of the holes in certain places on the reef, and
because of the spatial restriction provided by holes as
compared to the spaces beneath loose rocks or large slabs.
Population density estimates show that there is consider
able variation in the abundance of certain species from
one locality to another; this is believed to be largely
due to the number of holes or rocks available for cover.
CHAPTER IV
VERTICAL ZONATION
Introduction
Vertical zonation is a factor of major importance in
the ecology of organisms on the reef. As it is the major
criterion employed in distinguishing the intertidal zone
from the subtidal, and hence in deciding which species of
crabs would be emphasized in this study, it is appropriate
to consider vertical zonation before other ecological
factors and to discuss the tides in some detail.
As vertical zonation data were collected on both
Eniwetok and Majuro Atolls, which have different tidal
ranges, a comparison of absolute levels in the intertidal
is deemed less meaningful than comparisons based on a
relative scale. This is especially appropriate for the
present study, which is concerned mainly with crab ecology
in reef habitats which are for the most part not exposed
to the heavy surf which strikes the reef edge.
Theoretically, on a tideless shore, vertical zonation
would be determined by the height to which waves reach,
and an absolute scale might be possible (assuming shores
with the same degree of exposure were compared). In
actuality, the waves are a complicating factor which must
161
162
be considered in any discussion of tidal zonation on
exposed cliffs or wave-swept parts of reefs (e,.gi. the
algal ridge).
A relative scale was calculated for each absolute
value which could be determined with some certainty.
A scale ranging from 0 to 1, which correspond to the
mean low and high spring-tide values, respectively, is
used; all values which fall in between represent
fractions of the mean spring-tide range. These relative
values are given below in parentheses, the abbreviation
"r.l." signifying "relative level." To avoid confusion
between absolute and relative tidal levels, absolute
values below lm are given as "O.XXm," and relative levels
below 1 as "r.l. .XX."
The mean of the low spring tides has been calculated
to be 0.68' (0.204m) for Eniwetok, and 0.55' (0.168m) for
Majuro; this corresponds to r.l. zero. The mean of the
high spring tides has be calculated to be 4.52' (1.384m)
for Eniwetok, and 5.85' (1.79m) for Majuro; this
corresponds to r.l. 1. It will be noted that the mean
of the low spring tides is higher than datum. Datum is
mean lower low water for Hawaii and the Philippines, but
in other Central and Western Pacific Ocean areas it is
generally mean low water during springs, Indian spring
low water, or lowest possible low water (U.S. Dept.
163
Commerce, 1970). Similarly, a diurnal tidal range is
given in the Coast and Geodetic Survey tide tables for
the Philippines, whereas a spring-tide range is published
for Kwajalein (upon which daily predictions for Eniwetok
and Majuro are based, applying a correction factor). As
datum in the Marshall Islands is considerably below mean
low water during springs, it probably corresponds to
lowest possible low water, or a similar level.
During the study period at Eniwetok in the summer
of 1970, the greatest rise or fall during two consecutive
tides within the two-week cycle having the greatest
spring-tide range was 1.46m the corresponding figure for
the period with the smallest spring-tide range was 1.13m.
For neap tide periods, the greatest rise or fall between
two consecutive tides at the peak of the neaps was 0.49m
for the two-week cycle with the greatest neap-tide range,
and 0.40m the period with the smallest neap-tide range.
At the latter time, the smallest drop between a consecu
tive lower high tide and higher low tide was 0.15m, so
that during the day (when this occurred), the tides
appeared to be nearly stationary (although there was a
noticeable drop from the higher low tide which came at
night).
During springs, higher high tides reached levels of
1.43 to 1.59m (r.l. 1.03 to 1.16), and lower lows, 0.3 to
0.12m (r.l. .082 to -.072). During neap-tides, higher
164
high tides reached 1.07m (r.l. .723), and lower lows, 0.52
to 0.64m (r.l. .262 to .364). The minimum tidal rise
(during a neap) was from a higher low tide of 0.6 7m
(r.l. .390) to a lower high tide of 0.82m (r.l. .518).
Majuro Atoll was visited in 1970 during a spring
tide period with a maximum range of 1.65m, the highest
high tide being 1.83, (r.l. 1.03) and the lowest low
0.18m (r.l. .010); at other times of the year however,
greater spring-tide ranges are encountered.
The lower low tides occurred from 1136 to 1148 hours,
at the peaks of the spring-tide periods; considering the
intense sunlight, which can heat shallow-water pools
much above normal temperatures, this must impose severe
conditions on exposed intertidal organisms living very
low in the intertidal zone. At higher reef-flat levels,
most low tides leave the flat exposed about two hours,
so that forms which live above low neap-tide levels will
not only be exposed more often but also for longer periods
during a given low-tide period.
High intertidal organisms may face severe conditions
imposed by the neap-tides rather than by the spring
tides , as there may be periods of 3 days when the tide
does not reach above the 1.07m level (r.l. .723) on
Eniwetok Atoll. At the same time, daytime high tides
will be under 0.91m (r.l. .595), leaving the area above
165
this level completely exposed.
Vertical zonation on windward reefs is complicated
by wave pressure. When the trade winds are blowing,
a head of water is built up on the reef flat; the water
level increases, departing from predicted values of low
tide. For this reason, a person walking a constant
distance behind the reef edge will notice an apparent
increase in the depth of the water as he approaches
an island. Especially where there are obstacles to the
lagoonward flow of water across the reef flat, the effect
of this head of water will be to elevate vertical
zonation on lower parts of the reef. A striking example
of this is seen at station 10, where a narrow "dam" of
beachrock (which has the same effect on current flow as
an island) on the back part of the reef allows the water
level on the seaward reef to be 0.46m higher than that
on the lagoon side, during a 0.15m low spring tide.
Differences in reef topography where the reef flat
is backed by the same island can also influence the
height of water on a reef. This is the case at Eniwetok
Island, where the reefs at stations 2 and 4 were compared
during the same high and low tides. At station 2, there
is a high outer reef flat, and the reef level is generally
higher. At high tide, the waves sweeping across the flat
are much higher at station 2 than at station 4, making it
166
hazardous to wade; evidently, the high algal rim does not
impede waves on the reef at high tide nearly as much as
does the presence of a high outer reef flat. At low
tide, conditions are somewhat reversed, and low waves
come much closer to shore at station 4 than at station 2,
although the reef is wider and the level higher at
station 4. Apparently the numerous surge channels at
station 2 allow good drainage of water from the reef at
low tide, when the algal ridge is an effective wave
barrier; on the other hand, the ungrooved reef at station
4 allows water to be swept up over the ramp-like outer
flat and retained at a higher level by sustained wave
pressure.
There is yet another in which the relationship
between reef topography and currents may affect reef
zonation. Where there is a long enough barrier to the
lagoonward flow of water (e.g;. Eniwetok Island) , tide-
pool water which has been heated by sunlight at low tide
will not immediately be dissipated by the incoming tide,
but will be pushed shoreward and remain for some time as
a tepid zone at high tide. Where the barrier is fairly
short, as at station 10, there is a rapid "downhill"
flow of water around the flanks of the obstacle,
allowing cool ocean water to flow rapidly over the back
part of the reef. This may explain in part why the crab
167
fauna of the outer rubble flat in this latter locality
contain so many species which would otherwise be expected
at a lower level.
Methods and Terminology Employed
Due to the complication imposed by the elevation
of tidal zonation by wave pressure on the windward side,
it is not at most times possible to determine vertical
zonation by observing the water line at low tide.
Estimations of tidal zonation made in this way only
approach accuracy on completely placid days (which occur
at times during the doldrum season) or on sheltered
shores such as those of the leeward side or the lagoon.
Measurements of reef-flat levels in the main study
areas at Eniwetok Island were carried out by determining
the depth of the water over certain reef zones at high
tide and, using predicted high-tide levels, subtracting
this distance to obtain the level of the reef below. Due
to wave-induced fluctuations in the high-water level, a
mean value was in each instance determined by averaging
a series of measurements taken at 15- second intervals.
On some beaches and reef flats, a large series of
measurements was made with a simple inclinometer (using
lm intervals) to determine height profiles to be related
with known tidal levels. In other places, reef and
beach profiles were made (in cooperation with Michael
168
Chartock) using a transit and pole; these measurements
were related to local bench marks.
As noted in Section 1, three vertical zones are
recognized in this study: the infratidal fringe, the
low intertidal, and the high intertidal. The term
infratidal fringe is a synonym for the infralittoral
fringe of Stephenson and Stephenson (19 49); they defined
this as the lowest part of the rocky shores, covered by
low neap-tides, but uncovered by springs, and only
exposed to the air during calm weather in places where
the wave action is strong. These authors considered the
infralittoral fringe to extend up to the upper edge
of the coral growth on coral reefs. This makes it
possible to equate the algal ridge with lower-lying
coral areas in sheltered localities.
At least some, and probably most, of the crabs
which inhabit this reef zone in the Marshall Islands are
subtidal species which extend upward into the lowest
part of the intertidal; until adequate subtidal collections
are made, it is very unwise to assert that any of these
crabs is a true intertidal form. Without making any
assumptions about their zonation, however, one can
consider the species which live at low spring-tide levels
in sheltered parts of the reef to constitute an infratidal
fringe fauna. In exposed places, this assemblage of
169
crabs may occur at a higher level; on the algal ridge it
may range up to about mean sea level (the height of
the highest outer buttresses). Some eight species of
xanthids which live on the algal ridge have also been
collected at low spring-tide levels in protected places;
these include Globopilumnus globosus and Paraxanthias
notatus, both common algal ridge species which appear
in quiet water reefs as stragglers, and Actaea speciosa,
Chlorodiella laevissima, and Pilodius pugil, which are
seemingly abundant both on the algal ridge and at lower
levels elsewhere. These forms can be considered
indicator species of relative vertical zonation.
The remainder of the intertidal zone is the part
exposed during the low neap tides. From the viewpoint
of this study, it can best be divided into a low inter
tidal zone, which extends from the lowest neap tide levels
to mean sea level, and a high intertidal zone, from mean
sea level to the level of the high spring tides. Low
intertidal xanthids are those whose maximum abundance is
above the level of the low spring tides, generally from
the low neap-tide level to about mean sea level. High
intertidal forms are those whose maximum abundance occurs
from about mean sea level to the highest part of the
intertidal zone.
170
Infratidal Fringe Xanthids
At Eniwetok, there are only 11 xanthid species
which live predominately in the low and high intertidal
zones (taken together), and it is unusual to find more
than five of these at any given level in the same
habitat. By comparison, there is a great variety of
crabs in the infratidal fringe. Thirty-five species*
of xanthids were taken from Porites beds north of Eniwetok
Island (station 8C), at levels ranging from 0.15m to
0.29m (r.l. -.05 to .07). Of these, 16 were common or
repeatedly taken on different occasions. In other
localities where samples were taken at a comparable
level in the lowest part of the intertidal, still other
xanthid species, not seen at station 8C, were collected.
A detailed species list of infratidal fringe forms
will not be given here; it will suffice to say that the
species listed in Chapter 2 for the algal ridge, live
coral areas in general, the rubble piles in the rock
quarries at Eniwetok Atoll, and the rubble trains on the
reef at Majuro Atoll belong to the infratidal fringe.
In the backridge trough, on windward outer reef flats
and lower-lying leeward and lagoon reefs where live corals
are absent, and on low-level rubble flats and beaches the
situation is not as clear, as there may be a mixture
of infratidal fringe and low intertidal species.
*This list includes only one known coral symbiont
171
On Majuro Atoll, the outer reef reaches approximately
the 0.7m level (r.l. .331), and has a fauna distinctly
different from the higher, inner reef on the windward
side. On Eniwetok, one infratidal fringe species,
Xanthias lamarcki, ranges up to the 0.7m level (r.l.
.416), on the outer reef at station 4, although low
intertidal species are dominant on this part of the
reef. On lagoon and leeward reef flats, an infratidal
fringe fauna was collected at levels approximating 0.37m
(r.l. .138) on Eniwetok and 0.4m (r.l. .140) on Majuro.
Some of the crabs which live in this lowest tidal
zone on the reef are probably ecological equivalents
(at least in part) of certain low intertidal forms upon
which this study is focussed. It is worth making a
few remarks here about the commoner infratidal-fringe
xanthids which occur on reef flats or in loose rubble
and noting which of the true intertidal crabs they most
resemble in morphology and habits.
Of the algal ridge species, Dacryopilumnus emerita
is closely related to the low intertidal form D. rathbunae,
which it sometimes overlaps on the back part of the ridge.
Dacryopilumnus emerita has always been collected from
holes in the surface of coral-algal rock, like its
larger cousin. Just as Zozymodes biunguis is the
ubiquitous small xanthid on low intertidal reef flats,
172
Paraxanthias notatus is similarly abundant on the algal
ridge, being taken in almost every collection from dead
corals or coral-algal rock: these two crabs may well be
ecological equivalents in their respective zones.
Xanthias lamarcki lives in holes on the surface of
consolidated reef flats, in crevices under well-cemented
Porites slabs, and in loose rubble (especially around
Porites beds). In size, color, and general appearance
it is somewhat similar to Eriphia scabricula with which
it may co-occur in places; the two differ somewhat in
behavior, however, as only the latter species is active
out of water at low tide.
Etisus bifrontalis and Pilodius areolatus are common
where gravel occurs in pools on the outer and lower
parts of the reef flat; and these species, along with
Etisus demani, are among the most abundant crabs among
rubble in flat areas (especially Porites beds) on the
lagoon side of the reef. These crabs closely resemble
the low intertidal Xantho leptodon and X. sanguineus
which live among the rubble flats and beaches at a
generally higher level. On Majuro Atoll, Pilodius
areolatus and the larger Actaea tomentosa are often seen
among rubble in pothole-like tide pools, and dart into
holes in the rock when disturbed. In this they
resemble the above-mentioned Xantho species, whose
173
behavior on the inner reef is very similar. Finally,
Xantho danae is very abundant at the foot of sheltered
beaches and on low-lying flats (but not in the vicinity
of Porites), where it lives under rocks on gravel. This
species is dominant at the 0.37m level (r.l. .104) on
the lagoonward beaches of windward islands on Majuro
Atoll, being replaced by the previously named X. leptodon
and X . sanguineus at higher levels. It should be noted
that the apparent replacement of two low intertidal
species by five (and probably more) similar species in
the infratidal fringe is in line with the generally
higher diversity of crabs in the latter zone.
Intertidal Xanthids on Reef and Conglomerate Flats
On Eniwetok Atoll, the highest measured reef-flat
level was that of the inner reef at station 4, where the
flat lies at the 0.88m level (r.l. .570). On Majuro
Atoll, the inner reef at Uliga Island (station 47) lies
at the 0.9m level (r.l. .462); this is probably typical
for the windward side of this atoll. Most reef flats
appear to lie somewhat below mean sea level. Higher
conglomerate flats were found to range from the 0.9 8m
level (r.l. .647) at Aniyaanii Island, Eniwetok Atoll,
to the 1.83m level (r.l. 1.13) at Uliga Island, Majuro;
all appear to be above mean sea level.
Six species of xanthid crabs are common in these
situations. Zozymodes biunguis has been collected in
abundance from the highest reef flats on the windward
side to a level as low as 0.3m (r.l. .082 on Eniwetok,
.085 on Majuro) in sheltered places. It has not been
taken from the lower, other reef flat on the windward
side of Majuro, however; this is probably due in part
to the general elevation of the level of the infratidal
fringe (to which this outer reef belongs) in wave-exposed
locations. Eriphia scabricula also has a rather broad
vertical range, extending from the highest levels of
windward flats to the low spring-tide level on the leeward
reef at Majuro Island. It has been collected in greatest
abundance on the outer reef flat at station 4, at the
0.7m level (r.l. .416). Xantho leptodon has been found
only at the highest levels to which the reef flat reaches
on both Eniwetok and Majuro (r.l. .570 and .462,
respectively), perhaps due to lack of suitable substratum
at lower levels. Xantho sanguineus occurs at the same
level as and at lower levels than its smaller cousin.
Lydia annulipes and Pseudozius caystrus both inhabit
the flat upper surfaces of conglomerate platforms, and
apparently their vertical range in this habitat coincides
completely with that of the conglomerate, at least where '
the latter is found between the tides. Lydia annulipes
175
can be observed wandering over a wide vertical range at
night; it is even seen (rarely) on the inner reef at this
time, but has never been extracted from holes there. The
reef-flat species Zozymodes biunguis and Xantho sanguineus
also occasionally live in tide pools on the upper surface
of conglomerate flats, on algal mounds, or in depressions
in the eroding margins of rock groins, but these are
exceptional cases; there are few pools above reef-flat
levels and where they do occur, the holes are often
occupied by Lydia annulipes instead.
Generally speaking, then, the low intertidal xanthid
fauna which occurs on reef flats is composed of Zozymodes
biunguis, Eriphia scabricula, Xantho leptodon, and Xantho
sanguineus; and the high intertidal xanthid fauna which
lives on the higher conglomerate flats is made up of
Lydia annulipes and Pseudozius caystrus. Another species,
Eriphia sebana, is also seen on reef flats in a few
places. It was taken from the high inner reef at
Eniwetok Island station 4, but it is doubtful whether it
is restricted to this reef level, as there appear to be
few holes large enough to accomodate it farther out on
the flat. Otherwise, it has not been collected on reef
flats, except for individuals which wandered out from
under boulders or piles of rubble onto the adjacent reef.
176
In order to evaluate the extent of ecological
separation between these forms, population-density
measurements were used. In this case, the task is
simplified by the fact that for the areas studied,
none of the randomly-chosen quadrats used revealed any
overlap between low and high intertidal species. Table
6 depicts the maximum population densities for the various
study species in the higher and lower zones.
Quantitative analyses were also used to determine
2
the significance of these findings; Chi data were
chosen from as many localities as possible to implement
this. In comparing larger species with one another, any
data could be used. However, in evaluating the ecological
separation between the small Zozymodes biunguis and the
remaining forms, data for Z_. biunguis were taken only
from localities where larger crabs of one or another
species were collected in some abundance. This was done
because larger crabs do not occur on many of the reefs
where Zozymodes biunguis lives, evidently because of an
absence of holes of adequate size for them to hide in;
high intertidal forms would not show up in these places
even if vertical zonation allowed them to live on the
reef flat. For each locality, those quadrats which had
the greatest number of individuals of: the larger species,
relative to the number of Z. biunguis, were chosen, to
177
TABLE 6
MAXIMUM POPULATION DENSITIES OF INTERTIDAL
CRABS ON REEF FLATS AND HIGHER
CONGLOMERATE PLATFORMS
Species
No. Crabs/m^
Reef Flats Conglomerate
R.L. .047-.570 R.L. .647-1.13
Eriphia
scabricula
8.6 0
Lydia
annulipes
0 9.8
Pseudozius
caystrus
0 61.5
Xantho
leptodon
29.2 0
Xantho
sanguineus
16.9 0
Zozymodes
biunguis
242.0 0
R.L.=relative level.
178
maximize the probability that the (larger) high intertidal
crabs would have been collected, should they occur at
reef flat levels.
Reef-flat and conglomerate-flat species prove to
be significantly different (at the ,001 level) in their
vertical distribution (table 7). Therefore, there is no
question but that there is a sharp separation between
the low intertidal and high intertidal xanthid faunas,
the former living at or below a relative level of .570
and the latter at or above a relative level of .647.
It is probable that this boundary is determined in part
by the physical, chemical, and historical factors which
have limited the height to which the reef flat reaches.
Intertidal Xanthids on Eroding Conglomerate
Edges and Similar Topographic Features
The margins of conglomerate rock platforms, decadent
algal mounds .and surger.channel rims, the eroding edges
of beachrock beds, and the sides of reef blocks and
boulders are all similar in terms of crab fauna. On
these reef features, there is very little retention of
water at low tide, as most of the surface is at a steep
angle to the horizontal.
Of the two larger species which occur in these places,
Eriphia scabricula has been found up to approximately the
0.76m level (r.l. .467) at Eniwetok Island station 2,
179
TABLE 7
ECOLOGICAL SEPARATIONS WITH RESPECT TO
VERTICAL ZONATION ON REEF FLATS AND
HIGHER CONGLOMERATE PLATFORMS
Species Compared Co-occurrence Chi2 Sig.
Eriphia scabricula-
Lydia annulipes
None 49 .001
Eriphia scabricula-
Pseudozius caystrus
None 51.8 .001
Eriphia scabricula-
Xantho leptodon
Complete — ”
Eriphia scabricula-
Xantho sanguineus
Complete — —
Eriphia scabricula-
Zozymodes biunguis
Complete —
Lydia annulipes-
Pseudozius caystrus
Complete — —
Lydia annulipes-
Xantho leptodon
None 61.9 .001
Lydia sangulipes-
Xantho sanguineus
None 50.2 .001
Lydia annulipes-
Zozymodes biunguis
None 55.3 .001
Pseudozius caystrus-
Xantho leptodon
None 65.3 .001
Pseudozius caYstrus-
Xantho sanguineus
None 52.8 .001
Pseudozius caystrus-
Zozymodes biunguis
None 59 .001
Xantho leptodon-
Xantho sanguineus
Complete
- -
Xantho leptodon-
Zozymodes biunguis
Complete - -
180
TABLE 7- Continued
Species Compared Co-occurrence Chi2 Sig.
Xantho sanguineus- Complete
—
Zozymodes biunguis
Chi2 DATA
Species
Number of Crabs
Low
Intertidal
High
Intertidal
Number
of
Stations
Eriphia
scabricula
23 0 10
Lydia
annulipes
0 26 6
Pseudozius
caystrus
0 29 2
Xantho
leptodon
36 0 5
Xantho
sanguineus
24 0 4
Zozymodes
biunguis
48 0 7
181
and at Japtan station 12B; in the former locality it
seems to extend down to the low spring-tide level in
porous surge-channel rims. It proved to be very abundant
at the 0.64m level (r.l. .365) on eroding conglomerate
outliers at Japtan station 12A. Lydia annulipes
occupies the zone above this species, from about the 0.61m
to the 1.22m level (r.l. .339 to .851) on Eniwetok Atoll.
On Majuro Atoll, L. annulipes occurs as high as the 1.52m
level (r.l. .840) on the margins of conglomerate platforms.
Of the smaller species, Dacryopilumnus rathbunae
ranges from the 0.34m to the 0.95m level (r.l. .108 to
.613) at Rigili Island, Eniwetok; on Majuro Atoll, it
has been taken as high as the 1.22m level (r.l. .651)
or higher. Zozymodes biunguis has a similar vertical
range, extending up to the .98m level (r.l. .647) on
Eniwetok; however, it occurs as low as the 0.2m level
(r.l. zero) at Japtan Island, where conglomerate merges
with a low-lying reef flat.
Zozymodes pumilis lives at the same levels as
Dacryopilumnus rathbunae, with which it is usually
associated, though it is much rarer. Another small species
Pseudozius pacificus, is fairly common on Majuro Atoll
from the 0.9m to the 1.52m level (r.l. .462 to .840),
occurring mainly with Lydia annulipes. The large
Eriphia sebana is occasionally seen where large holes are
182
present in the eroding borders of conglomerate platforms;
it does not extend much above the level where the
conglomerate overlies the reef, however. It does not
co-occur with Lydia annulipes in places where the latter
is very abundant at higher levels. Xantho sanguineus
and Pseudozius caystrus are rare on this kind of
topography, although where the rock is very porous, the
latter may occur in the same zone as L. annulipes.
In determining the ecological separations between
the four study species which occur in these habitats,
two tidal range categories were employed: a higher zone
(above r.l. .462) characterized by miniature ridges and
valleys and a considerable amount of pitting, and a
lower zone (below this level; samples taken from below
r.l. .365) with a smoother rock surface and large basin
like concavities in the rock. Population-density data
for crabs living in these topographically differing
vertical zones are given in table 8.
Although it is evident that the larger Eriphia
scabricula and Lydia annulipes are characteristic of the
lower and upper zones, respectively, the situation is not
as clear for Dacryopilumnus rathbunae and Zozymodes
biunguis, as these forms occur at both levels, but are
much more abundant in the lower zone. When typical
upper and lower zone data are compared, Zozymodes
183
TABLE 8
REPRESENTATIVE POPULATION DENSITY DATA FOR
INTERTIDAL CRABS ON ERODING ALGAL RIDGE
REMNANTS AND MARGINS OF CONGLOMERATE
PLATFORMS AT DIFFERENT VERTICAL LEVELS
Species
---5----------- " -------
No. Crabs/m'1
Algal Rock (1) Conglomerate (2)
R.L. 0 - .365 R.L. .462-.840
Dacryopilumnus
rathbunae
15.6 6.3
Eriphia
scabricula
2.4 0
Lydia
annulipes
0 17.0
Pseudozius
pacificus
0 18.8
Zozymodes
biunguis
20.4 3.6
Zozymodes
pumilis
1.2 0
R.L.=relative level.
Habitats: 1. Sides of Surge Channel Rims and Algal
Mounds, Eniwetok Island Station 2.
2. Dark Green Pitted Zone, Margins of
Conglomerate Flats, Flats; Kwajalein
Island Station 32, Majuro Atoll Stations
46, 48A.
184
biunguis proves to be 17.7 percent as abundant in the
upper as in the lower zone, and Dacryopilumnus rathbunae
40.0 percent as abundant in the upper as in the lower
zone; the latter species evidently plays a more important
role in high intertidal communities.
Ecological separations due to vertical zonation were
determined for three of the species by employing
population density data to compute relative abundance
(table 9); in comparing the two smaller species with
Lydia annulipes, data were taken only from localities
where larger crabs, of any species, occurred in the
2
quadrats. Chi tests confirmed that Zozymodes biunguis
and Dacryopilumnus rathbunae differ significantly in
their vertical distribution from L. annulipes, but not
from one another (table 9). Although there were
insufficient data to extend the comparison to Eriphia
scabricula, there is little doubt but that it is
separated from Lydia annulipes, but not from the other
two study species, which completely overlap its vertical
range.
Intertidal Xanthids on Lagoon Beachrock
A number of observations were made on beach rock on
the lagoon side of Eniwetok Island, where Eriphia
scabricula and Lydia annulipes are the characteristic
xanthids. The former species was collected from 0.3m
185
TABLE 9
ECOLOGICAL SEPARATIONS WITH RESPECT TO
VERTICAL ZONATION ON ERODING ALGAL
RIDGE REMNANTS AND MARGINS
OF CONGLOMERATE PLATFORMS
Habitat
and Level
Relative
Abundance Chi2 Sig.
(1)
R.L. .462-.840
Dacryopiluiunus rathbunae
37.1% as abundant as
Lydia annulipes
39.2 .001
(2)
R.L. 0-.365
Lydia annulipes absent
(1)
R.L. .462-.840
Zozymodes biunguis
57.2% as abundant as
Dacryopilumnus rathbunae
1.2 not sig.
at . 05
(2)
R.L. 0-.365
Dacryopilumnus rathbunae
76.5% as abundant as
Zozymodes biunguis
(1)
R.L. .462-.840
Zozymodes biunguis
20.2% as abundant as
Lydia annulipes
36.4 .001
(2)
R.L. 0-.365
Lydia annulipes absent
R.L.=relative level.
Habitats as in Table 8.
186
TABLE 9- Continued
Chi2 DATA
Species Number
Hab. 1
of Crabs
Hab. 2
Number of
Stations Compared with:
Dacryopilumnus 31 7 3 Lydia
rathbunae
II
44 7 6
annulipes
Zozymodes
Lydia 0 24 4
biunguis
Dacryopilumnus
annulipes
Zozymodes 19 71 4
rathbunae,
Zozymodes
biunguis
Dacryopilumnus
biunguis
I I
4 22 3
rathbunae
Lydia
annulipes
187
to the 0.9m level (r.l. .082 to .594), while the latter
was taken from about the 0.76m to the 1.31m level (r.l.
.467 to .928). Although there is a certain amount of
overlap (there being a topographic continuum from the
lower to the upper border of the beachrock), it is
evident that these forms are separated by vertical
zonation.
Dacryopilumnus rathbunae was collected on the beach-
rock at station 7C from about the 0.61m to the 0.78m
levels (r.l. .339 to .540), but was not common; and
small individuals of Eriphia sebana were taken from
about the 0.78m level in this locality, apparently
overlapping both E. scabricula and L. annulipes.
Comparison of Zonation on Different Types of
Consolidated Rock Habitats
Eriphia scabricula and Lydia annulipes form a useful
pair for comparison of vertical zonation on different
kinds of topography, as they occur in holes of roughly
the same size, and are often observed in the same locality,
but at different levels. Eriphia scabricula apparently
does not extend as high in the intertidal on conglomerate
edges (up to r.l. .467), where the water drains away at
low tide, as it does on reef flats (up to r.l. .570),
which retain water in tide pools, or on lagoon beachrock
(up to r.l. .595)m which is continuously swept by waves.
188
Lydia annulipes on the other hand is found as low as a
relative level of .339 on conglomerate edges while it
is not found on flats at or below a relative level of
.570. In many places, its lower limit is set by the
level at which a reef flat intersects the conglomerate.
Clearly vertical zonation is relative, being highly
modified by the effects of topography.
As the above two species occur on a wide range of
topographic features, they can be used as indicators
of tidal zonation; in this way, the lower and upper zones
on conglomerate edges and beachrock can be equated with
the low and high intertidal of flat areas. Similarly,
from the fact the Dacryopilumnus rathbunae corresponds
most closely to Eriphia scabricula and Zozymodes biunguis,
one can classify it as a low intertidal crab, a designa
tion which is supported by the fact that it has never
been found on higher conglomerate flats, while it has
been observed (rarely) on the smooth outer reef at
station 3.
Intertidal Xanthids on Rubble Beaches and Flats
Where loose rubble has accumulated on the surface
of a beach, there is no marked topographic discontinuity
with increase in elevation, except that the upper beach
may have a somewhat steeper slope, and in some cases the
lower beach merges very gradually into a sand flat.
189
From the viewpoint of crab ecology, sand flats differ
from beaches in that water is often retained in big pools
at low tide on flats, whereas beaches are well drained.
Xantho leptodon was found on the beach at Eniwetok
Island station 5C between the 0.6m and 0.9m levels
(r.l. .339 to .595), although it was less abundant than
its relative, X. gracilis, above the 0.7m level (r.l.
.416). X. leptodon also occurs in pools at a higher level,
and proved to be the dominant xanthid in a pool behind
a bar of conglomerate at station 5A, where the low water
line was at the 0.82m level (r.l. .519). It was still
common in this locality on the beach above low pool level
at the 0.9m level (r.l. .583), being largely replaced by
X. gracilis at the 0.98m level (r.l. .647). X. leptodon
was also the commonest crab in the large pooled area on
the back part of the rubble flat at Igurin Island (where
X. gracilis lived on the drying part of the flat) and in
the seaward rubble pool at Dalap Island, where the sand
flat lay above the level of the reef.
Yet another situation favoring X. leptodon occurs
where a beach overlies a high reef flat; at Eniwetok
Island station 5B, it was taken from the foot of the
beach, from the 0.8m to the 0.89m level (r.l. .500 to
.577). At Perry Island (station 11B), where it occurred
in the same situation, rill markings in the fine sand
190
at the foot of the beach indicated that at low tide, water
seeps out of the beach and onto the reef; it is probable
that beach runoff keeps this zone moist, allowing this
species to occur at an elevated level.
Xantho gracilis was the dominant xanthid in the
middle range of the intertidal at Arniel Island, where
it was common at the 0.9m and 1.22m levels (r.l. .462
to .651), occupying the zone below Pseudozius caystrus.
At Eniwetok station 5C, it was the commonest xanthid at
the 0.76m and 0.9m levels (r.l. .467 to .595), and at
Aniyaanii Island it was abundant on a rubble flat
lying at the 0.82m level (r.l. .517).
Xantho sanguineus has a very broad range in the
intertidal. On the beach at Arniel Island, it was
collected in decreasing numbers from the 0.6m level (r.l.
.274) to the 1.52m level (r.l. .840). At Eniwetok Island,
it was taken from the 0.6m to the 1.11m level (r.l.
.339 to .757) at stations 5B and 5C (considered together).
At station 12B on the south side of Japtan Island, X.
sanguineus was common at approximately the 0.6m level
(r.l. .339) on the rubble field on the inner reef, and
ranged considerably higher on the cobble beach above.
The vertical range of Pseudozius caystrus seems to
vary according to the type of substratum present. At
Arniel Island, it occurred from the 0.9m to the 1.83m
191
level (r.l. .462 to 1.13), being commonest at the 1.52m
level (r.l. .840). In this locality, however, the
sediment shifted towards gravel in the highest part of
the intertidal, which might account for its abundance
there. At Eniwetok Island station 5A, P. caystrus was
common at the 0.98m level (r.l. .647) under rocks on
gravel, while X. gracilis lived at the same level where
sand was predominant under rocks. However, higher up,
from the 1.13m to 1.43m level (r.l. .775 to 1) P.
caystrus was much more abundant than X. gracilis despite
the sandy substratum. Similarly, P. caystrus was the
only xanthid living under rocks on sand just below the
high-tide level on a 1.22m high tide on the upper beach
at station 5B. In some places on Eniwetok atoll, P.
caystrus was taken as low as the 0.6m level (r.l. .339),
but only where it was found in association with rubble
heaps or loose gravel.
Several other species have also been collected in
loose rubble, but are not especially common in this
situation. Zozymodes biunguis has occasionally been
taken from the lower part of beaches, where it is
associated with Xantho leptodon. Lydia annulipes some
times lives on the upper beach, where it is associated
with Pseudozius caystrus under rocks on gravel. Eriphia
sebana also occurs on rubble flats and beaches, where
192
it seems to correspond most closely to Xantho sanguineus
in its vertical range.
Before making an analysis of ecological separation
due to zonation for the four commoner species, it is
informative to examine some population-density data for
these crabs at different levels (table 10). Data from
station 5C demonstrate the replacement of Xantho
leptodon by X. gracilis with increasing height of the
beach; and data from station 5A show that Pseudozius
caystrus is much more abundant than is X. gracilis on
the upper beach. There is no obvious trend in the overall
abundance of crabs with change in elevation; a high
density of P. caystrus was found in the highest part of
the intertidal zone.
In comparing the relative abundance of the various
rubble-crab. species at different levels, quantitative
data from various beaches were used, as no one locality
presented optimal conditions for all species. Data from
Arniel Island indicate that Xantho sanguineus occupies
the lower part of the beach, X. gracilis the middle, and
Pseudozius caystrus the upper beach (table 11). Due to
the possible influence of change in substratum on vertical
zonation, another comparison was made using data from
Eniwetok Island (table 12). Here P. caystrus still
characterizes the uppermost beach zone, but there is
193
TABLE 10
POPULATION DENSITIES OF CRABS ON RUBBLE
BEACHES, WITH RESPECT
TO VERTICAL ZONATION
Species Station 5C
R.L. .339-.416
No. Crabs/m2
Station 5C
R.L. .46-.595
Station 5A
R.L. .775-1.0
Xantho
leptodon
15.6 4.8 0
Xantho
gracilis
3.9 15.0 5.4
Pseudozius
caystrus
21.0
R.L.=relative level.
Note: Substrate in all cases was sand mixed with some
gravel, under rocks. Figures are not given for
Pseudozius caystrus for Station 5C, where the
rocks were much smaller than the optimal size
for this species.
194
TABLE 11
ECOLOGICAL SEPARATIONS WITH RESPECT TO
VERTICAL ZONATION BETWEEN
RUBBLE CRABS AT
STATION 43
Level Relative
Abundance
Chi2 Sig.
1.83m
(R.L. .1.13)
Xantho gracilis
absent
84.7 .001
1.52m
(R „ L. .840
Xantho gracilis
3% as abundant as
Pseudozius caystrus
0.9-1.
(R.L.
22m
.462-.651)
Pseudozius caystrus
8.8% as abundant as
Xantho gracilis
1.83m
(R.L. 1.13)
Xantho sanguineus
absent
44.7 .001
1.52m
(R.L. . 840)
Xantho sanguineus
9.1% as abundant as
Pseudozius caystrus
0.9-1.
(R.L.
22m
.462-.651)
Pseudozius caystrus
46.5% as abundant as
Xantho sanguineus
0. 6m
(R.L. .274)
Pseudozius caystrus
absent
1.22-1
(R.L.
.52m
.651-840)
Xantho sanguineus
18% as abundant as
Xantho gracilis
11.3 .001
0 .9m
(R.L. .462)
Xantho sanguineus
36.8% as abundant as
Xantho gracilis
0. 6m
(R.L. .274)
Xantho gracilis
8.3% as abundant as
Xantho sanguineus
R.L.=relative level.
TABLE 11- Continued
195
Chi2 DATA
Species Vertical
Range
No. Crabs Vertical
Range
No.
Crabs
Pseudozius
caystrus
(R.L. .462-
.651)
57 (R.L. .840-
1.13)
1
Xantho
gracilis
5 44
Pseudozius
caystrus
(R.L. .274-
.651)
5 (R.L. .840-
1.13)
44
Xantho
sanguineus
23 3
Xantho
gracilis
(R.L. .274-
.462)
20 (R.L. .651-
.840)
44
Xantho
sanguineus
19 3
196
TABLE 12
ECOLOGICAL SEPARATION WITH RESPECT TO
VERTICAL ZONATION BETWEEN
RUBBLE CRABS AT STATION 5
Station and
Level
Relative
Abundance
--- -----
Chi Sig.
Stations 5A, B, C
1.13-1.43m
(R.L. .775-1.0)
Xantho gracilis
15.6% as abundant as
Pseudozius caystrus
52.7 .001
0.76-1.11m
(R.L. .467-.757)
Pseudozius caystrus
48.9% as abundant as
Xantho gracilis
Station 5C
0 .9m
(R.L. .583)
Xantho leptodon
13.2% as abundant as
Xantho gracilis
70.4 .001
0 .76m
(R.L. .467)
Xantho leptodon
18.5% as abundant as
Xantho gracilis
0. 6m
(R.L. .339)
Xantho gracilis
13.6% as abundant as
Xantho leptodon
Station 5A
0 .98m
(R.L. .647)
Xantho leptodon
9.5% as abundant as
Xantho gracilis
27.2 .001
0. 9m
(R.L. .583)
Xantho gracilis
27.3% as abundant as
Xantho leptodon
0.82m
(R.L. .519)
Xantho gracilis
absent
R.L.=relative level.
197
TABLE 12- Continued
Chi2 DATA
Station and
Species
Vertical
Range
No.
Crabs
Vertical
Range
No.
Crabs
Station 5A, B, C
Pseudozius
caystrus
(R.L. .467-
.757)
24 (R.L. .775-
1.0)
32
Xantho
gracilis
49 5
Station 5C
Xantho
gracilis
(R.L. .339) 13 (R.L. .467-
.595)
149
Xantho
leptodon
37 24
Station 5A
Xantho
gracilis
(R.L. .519-
.583)
3 (R.L. .647) 21
Xantho
leptodon
19 2
198
considerable overlap with X. gracilis lower down (at r.l.
.467-.757). Xantho leptodon was compared with X. gracilis
at several localities on Eniwetok Island (table 12)
and found to lie clearly below the later species in both
instances.
Chi^ tests run on the above data indicate that the
difference between Xantho sanguineus, X* gracilis, and
Pseudozius caystrus, and between X. le'-ptodon and X-
gracilis, are significant at the .001 level (table 11, 12).
By extrapolation, one can conclude that X. leptodon and X-
sanguineus lie in the same zone (both lie below the level
where the infratidal fringe species X- danae occurs), and
that X. leptodon is clearly below P. caystrus.
On beaches where there is an cibundance of sand under
the rocks, then, there are three vertical tiers of crabs
above the low spring-tide level. The lower tier is
characterized by Xantho leptodon and X. sanguineus, which
are both abundant around the low neap-tide level. The
middle tier is occupied by Xantho gracilis, which ranges
from about mean sea level to the reaches of the higher
high tides during neaps; and the upper tier is character
ized by Pseudozius caystrus, which lives above the level
of higher high tides during neaps and extends up to the
highest spring-tide levels. However, where gravel
predominates, Pseudozius caystrus completely overlaps the
199
vertical range of Xantho gracilis. For this reason,
both these species can be considered the high intertidal
xanthid fauna on rubble beaches, while X. sanguineus and
X. leptodon can be referred to as the low intertidal
fauna.
Comparisons Between Consolidated Rock and
Loose Rubble Habitats
The two species of rubble crabs which live in the
low intertidal on beaches and flats also occur in the
same zone on reef flats. While Xantho leptodon does not
extend as low in the intertidal on reefs as it does on
beaches, it may live in rubble-filled pools as high as
or higher than the level of the reef flat. Xantho
sanguineus has a broader vertical range both on beaches
and in consolidated rock situations.
Pseudozius caystrus is characteristic of the high
intertidal both on conglomerate rock platforms and in
loose rubble, though it may occasionally extend to low
intertidal levels on rubble beaches. Xantho gracilis,
which is considered here to be a high intertidal species,
may actually live at the same absolute level on beaches
as that at which its low intertidal cousins occur
on reef flats; this is not surprising, as beaches dry
at low tide, while the reef flat does not. It is more
valid to compare these species in the same habitat.
200
Temperature and Vertical Zonation
One factor expected to correlate closely with
vertical zonation is temperature. Daytime, low-tide
temperature measurements taken at Eniwetok and in the
Southern Marshall Islands in the summer of 1970 are
given in appendix 3. Data collected on submerged parts
of the reef flat, which are least affected by increases
in temperature at low tide, can be used in comparison
with those from exposed parts of the reef. In all
habitats, temperatures taken inside the mouths of holes
or under rocks and slabs were found to be never above
32.5° C, which is not much warmer than the highest
temperature on a submerged reef (31.8° C). On the other
hand, low-tide temperatures taken in pools on the reef
flat range up to 37.5°, comparable to temperatures of
36.5° C and 38° C for pools on the top of a conglomerate
and of a rubble flat, respectively. Exposed surface
rock temperatures of up to 36.5° C were recorded for the
reef flat, compared to 38.5° C for beach rock, 39.8° C for
conglomerate, and 39° C for the exposed surfaces of rocks
on rubble beaches.
Generally, then, there is no great difference
between exposed rock surfaces and pools on the same reef,
whereas holes and crevices exposed to the air are much
cooler. While somewhat higher surface temperatures were
201
recorded from higher conglomerate than from the reef
flat, an even more significant relationship is the
general increase in temperatures recorded from tide
pools and dry rock surfaces as the season progressed.
This is most likely due to the onset of the calms later
in the summer? earlier in the season, the strong trade
winds keep the intertidal zone (on the windward side
at least) from reaching the high temperatures recorded
during the doldrums. This seasonal change in reef
temperature suggests that the intertidal part of the reef
does not have as stable a climate as is usually assumed
for a coral reef.
The temperature data collected in the course of
this study do not indicate that high intertidal crabs
are exposed to higher temperatures than are those which
live at reef-flat levels, for as will be shown in Chapter
6, most of the species are relatively inactive during
daytime low tides, remaining in their holes or under rocks
where the temperature is lower. Of the three species
which occur in the high intertidal zone, only Lydia
annulipes was observed out in the open during the day,
and at that rather infrequently. On the other hand, of
the five low intertidal species, both Eriphia scabricula
and Dacryopilumnus rathbunae are active at low tide
(the former in tide pools as well as out on the open
202
rock), while the two Xantho species live on the reef
flat were also occasionally seen in tide pools during the
day. Therefore, of the eight study species, the low
intertidal forms are more likely to be exposed to high
daytime temperatures than are crabs which live higher
up. It is probable that for these crabs, vertical
zonation is dertermined by desiccation rather than by
temperature. On the other hand, temperature may well
be important in separating true intertidal species from
the infratidal fringe fauna.
Conclusions
Before analyzing the vertical zonation of intertidal
crabs, a relative zonation scale was set up. Three xones
were recognized: an infratidal fringe, where numerous
species of crabs live, a low intertidal zone, where
there are five species of true intertidal xanthids
(Zozymodes biunguis, Xantho leptodon, X. sanguineus,
Eriphia scabricula, and Dacryopilumnus rathbunae), and
a high intertidal zone, where three species (Xantho
gracilis, Lydia annulipes, and Pseudozius caystrus) occur.
In consolidated rock situations such as reef and
conglomerate flats, only two vertical tiers of crabs
(above the infratidal fringe) may be recognized, but
on beaches with an abundance of sand under the rocks,
three tiers can be distinguised. Therefore, the maximum
number of intertidal habitat types (each characterized
by one or more species of crab) determined by vertical
zonation is three.
Both in consolidated rock and in loose rubble,
vertical zonation is not absolute, but is considerably
influenced by wave exposure and/or topography. It is
probable that desiccation rather than temperature is
responsible for zonation in true intertidal crabs.
Vertical zonation data are summarized in table 13.
204
TABLE 13
SUMMARY OF VERTICAL ZONATION DATA FOR
EIGHT SPECIES OF CRABS
Species Vertical Zonation, by Habitat^
A . Predominantly High Intertidal Forms
Lydia
annulipes
Xantho
gracilis
Pseudozius
caystrus
1 .
2 ,
3,
4,
0.98m (E)-1.83m (M) (R.L. .647-1.13)
0.6-1.22m (E) (R.L. .339-.851)
0.9-1.52m (M) (R.L. .462-.840)
Mainly above R.L. .462
0.76-1.31m (E) (R.L. .467-.928)
0.6-1.13+m (E) (R.L. .339-.775+);
common from 0.76-to 0.9m (R.L. .467-
.595)
0.6-1.52m (M) (R.L. .274-.840);
common from 0.9-to 1.22m (R.L.
.462-.651)
1.
4.
0.98m (E)-1.83m (M) (R.L.
0.9m-1.43m (E) (R.L. .462
commonest above 1.13m
under rocks on sand,
common at 0.98m (R.L.
rocks on gravel
0.9-1.83m (M) (R.L. .462-1.13);
commonest at 1.52m (R.L. .840)
.647-1.13)
1.0) ;
.775) (R.L.
.647) under
B. Predominantly Low Intertidal Forms
Dacryopilumnus
rathbunae
2. 0.34-0.95m (E) (R.L. .108-.613)
Up to 1.22+m (M) (R.L. .651)
Commoner below R.L. .46 2
■^•Habitats: 1. Reef and Conglomerate Flats.
2. Eroding Algal Rock, Margins of
Conglomerate Flats, Beachrock Rims,
Reef Blocks and Boulders.
3. Sloping Lagoon Beachrock.
4. Rubble Beaches and Flats which Dry at
Low Tide.
5. Tide Pools in Rubble Situations.
6. Beaches above Intertidal Reef Flats.
TABLE 13- Continued
205
Species Vertical Zonation, By Habitat-1 -
Eriphia 1. Up to 0.88m (E) (R.L. .570)
scabricula Up to 0.9m (M) (R.L. .462)
2. Up to 0.76m (E) (R.L. .467)
3. 0.3-0.9m (E) (R.L. .082-.594)
Xantho 1. 0.88m (E) (R.L. .570)
leptodon 0.9m (M) (R.L. .462)
4. 0.6-0.9m (E) (R.L. .339-.595);
common up to 0.7m (R.L. .416)
5. 0.82-0.98m (E) (R.L. .519-.647);
common up to 0.82m (R.L. .519)
6. 0.8-0.89m (E) (R.L. .500-.577)
Xantho 1. Up to 0.88m (E) (R.L. .570)
sanguineus Up to 0.9m (E) (R.L. .462)
4. 0.6-1.11m (E) (R.L. .339-.757);
common at 0.6m (R.L. .339)
0.6-1.52m (M) (R.L. .274-.840);
commonest at 0.6m (R.L. .274)
Zozymodes 1. 0.3-0.88m (E) (R.L. .082-.570)
biunguis 0.3-0.9m (M) (R.L. .085-.462)
2. 0.2-0.98m (E) (R.L. Zero-.647)
Commoner below R.L. .462
(E)=Eniwetok Atoll. (M)=Majuro Atoll.
CHAPTER V
SUBSTRATE
Introduction
This section deals with the influence of substrate on
the distribution of the eight study species. Substratum is
used in a broad sense: the material, in or upon which the
animals live, and upon which they rely for refuge. Many
kinds of substratum differences are responsible for separ
ating xanthid crabs ecologically: variations in topography,
differences in the size and shape of holes in the rock (in
which the crabs hide), the extent and nature of the algal
and sediment cover on the reef, and variations in the size
and looseness of rocks.
Substrate-related Crab Behavior in Reef Situations
An aspect of crab ecology which has hitherto received
scant attention is the relationship between crabs which
inhabit naturally occurring holes and crevices in the reef
and the physical parameters of their refuge holes. In
1968, a considerable amount of general collecting was done
on the reef, and the present writer was greatly impressed
by the apparent dependence of many xanthid species upon
such holes. Subsequently, a number of observations were
made on crab behavior; these observations have lent sup
port to the idea that refuge is of considerable importance
206
207
to these animals, and can be generalized as follows:
Zozymodes biunguis, the smallest and most abundant
intertidal reef crab, is especially prone to spend much
of its time, during low tide at night when it is most
active, either in or near small holes in the reef rock,
darting rapidly into the holes when disturbed. Of 185
crabs counted in l/16m^ guadrats on the inner reef flat at
Eniwetok stations 3 and 4, 85 percent were seen inside or
in close proximity to the mouths of their holes, the re
mainder wandering across the bottoms of the tide pools.
During the day or at high tide in general, this species is
seldom seen in the open at all, and apparently remains
hidden well inside the holes.
Xantho leptodon and X. sanguineus behave in much the
same way as the above species, being most active at night
at low tide in the shallow reef pools. They often sit in
or nearby the apertures of their holes, and being very
wary are prone to retreat upon the slightest disturbance.
In some cases, where small rocks and gravel litter the
pool bottoms, the crabs will retreat into this material
and thereby gain access to hidden crevices. Where there
is an abundance of rubble in the pools, as at Kwajalein
Island (station 31) a considerable number of X. sanguineus
may be found hidden in it. As it is usually sandy where
X. leptodon lives, individuals of this species may be seen
cleaning the sediment from their holes, using the legs
208
extended on one side of the body as a plow.
Dacryopilmnnus rathbunae is active on exposed rock at
low tide during daylight hours; it does not appear to
wander far from cover, however. A close examination of
the reef surface where these crabs live will reveal that
many animals remain just inside the apertures of their
holes, keeping watch with the eye facing the opening. If
the observer remains at a distance and keeps fairly still,
they will come out to feed in close proximity to these
holes, seldom wandering off across the rock. On a rising
tide, they withdraw before incoming waves.
Eriphia scabricula often wanders widely over the reef
surface where it occurs. Numerous observations were made
on this species from a distance of 8m or so, with binocu
lars; it is otherwise difficult to obtain a clear picture
of its natural behavior during the day, as E. scabricula
is very wary. On sheltered, inner reefs, these animals
can often be seen emerging from their abodes in the rock
to meander some distance away. Under such circumstances
they may sometimes be trapped away from their holes, in
which case they will feign death, menace the collector
with their claws or dig rapidly backwards into a sandy
reef turf, if there is any. In the latter case, only the
anterior of the carapace is left exposed (the setae and
mottled coloration causing it to closely resemble the reef
surface).
209
However, along the rims of surge channels (as at
station 2), on lagoon beachrock (at station 7), and on ex
posed outer reef flats (as at station 4), they are highly
sensitive to the surf, and do not venture far; in such
places they advance and repeat between successive waves.
Only seldom is a crab caught in the open by an incoming
swell; should this happen, it will flatten itself against
its background. Hence, this species uses holes for refuge
not only from visible intruders, but also from the waves.
It will often extend one side of its body from the aperture
of its hole, keeping one eye out to watch for danger,
before completely emerging.
Lydia annulipes is most active at night, when it is
not infrequently encountered some distance from shelter,
though it will attempt to wedge itself into any handy
crack when disturbed. Should no crevices be available it
will fiercely grip the rock surface with its legs, or
failing all else draw up its claws and legs and mimic a
pebble. In the daytime these animals remain hidden deep
within their holes in the rock, though occasionally one is
seen just inside the aperture.
Pseudozius caystrus is abundant under slabs of con
glomerate but has never been observed away from its rock
shelter. This is the most cryptic in habits of all the
study species, and for this reason little is known of its
behavior in nature. However, it can at times be observed
210
at night, moving about under low ledges where it finds
continuous cover.
Defensive Behavior Associated with Holes
In addition to spending a considerable amount of time
nearby or just within the mouths of their holes or hidden
deep within, these crabs also exhibit a characteristic
defensive behavior associated with the holes, common to
all species. When pursued into the aperture of a hole or
crevice, the animals will wedge their bodies in the open
ing, bracing themselves with their claws and legs. So
great is their tenacity, that it is often difficult to
extract a crab from its hole without mutilating the speci
men. This behavior was observed in the field as well as
in an artificial aquarium habitat with holes in a block of
concrete.
Eriphia scabricula and Pseudozius caystrus will, when
trapped in a shallow hole, defend themselves vigorously
with their claws or block the aperture with the outside of
one claw. Lydia annulipes often adopts a characteristic
defensive posture when lying in the mouth of its hole in
the daytime, holding the slender, red fingers of the small
claw in the opening and pinching any intruding object.
When placed in the above-mentioned artificial crab
habitat, most of the crabs would dive into the nearest hole
without any concern whether it was occupied. However,
Eriphia scabricula displayed a form of exploratory behavior
211
in the laboratory: extending the legs and open claw of
one side into the opening, and holding onto the edge with
its remaining legs, it would move into the hole in a series
of jerking motions. In the field, Lydia annulipes was
observed to advance into holes while probing ahead with
its small claw. Actual eviction of small crabs by larger
ones was observed for Pseudozius caystrus in the aquarium;
the larger animal would run into a crevice, and after a
brief scuffle inside, the smaller would emerge uninjured.
There is evidence that at least some of these species
exhibit homing behavior, in the sense that a given indi
vidual will return to a particular hole repeatedly;
Eriphia scabricula and Dacryopilumnus rathbunae, which
were observed from some distance away during the daytime,
were seen darting in and out of the same holes a number of
times. It is not certain however how long a given crab
will occupy the same hole, as tagging experiments were not
carried out.
The Crab-Hole Hypothesis and Methods of Analysis
The various xanthid species considered here differ
from one another in both size and morphology; as they all
appear to be highly dependent upon holes for refuge, it is
logical to ask whether the various species inhabit holes
which closely correspond to the dimensions of the crabs.
Differences in the size and shape of the holes in which
212
these animals live could theoretically be important in
determining ecological separations between species.
In order to determine the validity of this idea, one
must measure both the crabs and their holes. While it is
most desirable to have complete data on dimensions of the
crabs' refuges, the extreme hardness of much of the in
tertidal reef rock makes it difficult to penetrate very
far below the surface. For this reason, it is only fea
sible to confine the investigation to the dimensions of
the hole apertures. However, as the crabs spend a con
siderable amount of time in the mouths of their holes,
and because this is the part of the hole which must af
ford them rapid and easy entry when threatened, it is con
cluded that a study of the hole apertures alone will con
tribute much useful information about their ecology.
Even measurement of the apertures of holes presents
certain problems. Fine forceps or calipers cannot be
used, since many of the holes lie under ledges, and many
others are at odd angles to the reef surface. A sampling
method is required which will allow a cast to be made of
the outer and most accessible part of the hole. Such a
method was devised in 1969, and after subsequent analysis
of that summer's data a decision was made to use the same
technique extensively in 1970.
The procedure is as follows: individual crabs are
213
poisoned from their hiding places with full strength
formalin, using a syringe (such as a pastry baster or
hypodermic needle); a putty impresion is then taken of the
hole aperture, (fig. 51), using a water-proof putty
(Scotch Seal Brand No. 1167 Synthetic Putty was used
during this study). Both crab and putty impression are
then stored in a vial of dilute formalin. Although the
putty appears to hold its shape indefinitely, it is wise
to measure all hole impressions as soon as possible, to
safeguard against possible accidental deformation.
The crabs and their putty hole impressions are then
measured with a dial-caliper (a Helios, which can be read
to 0.01mm was used in this case), most measurements being
made under a binocular scope. Crab height, length, and
width are all measured; each crab is prepared by removing
the legs and claws, and mounting the body on a putty base.
In the case of ovigerous females, the abdomen is held some
distance away from the body; in order to obtain a concrete
measurement, the eggs must be removed and the abdomen
pressed tightly against the body, to determine the minimum
space required by the crab within its hole.
Putty impressions of the holes are obtained by mount
ing each impression on its outer end (the part used as a
finger-hold in taking the impression in the field) and
positioning the intact cast of the hole upward to allow a
measurement to be taken at the point where the cast ceases
Figure 51 - Diagram of Putty Plug
and Crab Hole.
Showing usual position
of crab in hole aperture.
215
to taper. Maximum width is measured, and minimum height
taken at a right angle to this. Only impressions with
parallel sides can be used; tapering ones are abandoned in
the field, as they do not allow the minimum dimensions of
the hole aperture to be measured. Other holes are un
measurable due to an irregular shape, which prevents a
good impression from being taken.
As there is a great deal of variation in the appear
ance of the reef from one locality to the next, it was
decided that for this study, data should be collected from
a number of different localities (including different kinds
of reef topography when possible); this reduces the pos
sibility of bias in the data which could result from con
centration on a few localities, which might be atypical.
Of the study species, however, Xantho sanguineus and X.
leptodon were not found in abundance on the reef in many
localities, and it was only possible to collect adequate
data for the latter of the two. Pseudozius caystrus could
not be studied in this manner at all, as the continuous
crevices under the slabs where it lives are nearly impos
sible to measure.
At each locality where crabs were collected, a
certain area was systematically poisoned, moving from one
hole to the next until a quota of 10 crabs per locality
was met for each species. It was originally intended to
216
collect at five localities for each species; this often
had to be modified in practice, however.
In many of the localities, the holes were spaced so
far apart that it is probable that most ended blindly.
When the holes occupied by the crabs proved to be old
barnacle borings, like the holes utilized by Eriphia
scabricula at station 12A, or old sipunculid holes, as is
the case for Dacryopilumnus rathbunae in general*, it was
possible to chip away the soft rock in localities where
these crabs and boring animals lived and locate the ends
of the holes. On the other hand, in some localities, the
reef was so porous that many of the holes were probably
interconnected. The porous edge of the rock groin at
station 47, where Lydia annulipes was abundant, and the
coralline algal zone at station 4, inhabited by Zozymodes
biunguis, can be cited as examples of this. While data
from these places do not reflect accurately the relation
ship between individual crabs and their holes (as possible
accessory exits were not measured), these data can be used
to determine the average size and shape of holes occupied
by a given species, which is more important from the view
point of niche studies.
* At Ajurotake Island, Majuro Atoll, a very large
Dacryopilumnus rathbunae was collected from what was
probably a boring clam hole.
217
Crab-Hole Size and Shape Relationships
Before analysing the extent to which differences in
hole size and shape relate to the ecological separation of
species, it is instructive to examine the relationship
between crab and hole size and crab and hole shape. The
size and shape of the holes occupied by five of the study
species is depicted on tables 14 and 15. Crab height was
compared with hole height, and crab length to hole width
(fig. 52). Crab width is apparently unrelated to hole
width, as the animals always move in and out of their
holes sideways, with the legs of one side extended out of
the hole. In connection with this, it should be noted
that measurements of the depth of reef-flat holes made
with a wire probe indicate that most holes are at least
six times as deep as.they are wide, allowing the crabs to
move some distance away from the apertures of their holes.
Mean crab height was found to be 53.7 to 65.3 per
cent of mean hole height, and mean crab length 53.1 to
64.8 percent of mean hole width, depending upon the
species; this is not suprising, since the animals must
require a certain amount of spare room to maneuver as they
dart into their holes (see Appendix 4). As the crab-hole
data were collected in pairs (i.e. for each measured hole
there is a set of data on the dimensions of the crab which
had been hiding within), it was possible to correlate
218
TABLE 14
SIZE OF CRAB HOLES FOR 5 INTERTIDAL SPECIES
Mean Hole Size (Range in Parentheses).
Species and
Habitat
Hole Size
Width
(mm)
Height
No.
Crabs
No.
Stations
Dacryopilumnus
rathbunae
2 6.7
(4.1-12.5)
5.3
(4.1- 7.9)
52 7
Eriphia
scabricula
1 16.1
(5.7-32.4)
9.8
(3.9-18.1)
29 3
Eriphia
scabricula
2 15.0
(6.0-23.4)
8.5
(4.0-15.3)
30 6
Lydia
annulipes
2 15. 2
(4.7-35.5)
9.3
(3.5-18.4)
41 4
Xantho
leptodon
1 15. 3
(7.3-23.3)
9.1
(4.5-14.5)
21 2
Zozymodes
biunguis
1 7.4
(3.4-14.3)
4.6
(2.7- 9.8)
59 7
Zozymodes
biunguis
2 6.9
(3.5-18.2)
4.4
(3.0- 7.3)
19 5
Habitat Types: 1. Reef Flats, with Tide Pools.
2. Eroding Algal Ridge Remnants,
Margins of Conglomerate Flats,
and Beachrock; Tide Pools
Infrequent.
219
TABLE 15
SHAPE OF CRAB HOLES FOR 5 INTERTIDAL SPECIES
Mean Hole Shape (Range in Parentheses)
Species and
Habitat
Hole Shape
(Height/Width)
No.
Crabs
No.
Stations
Dacryopilumnus
rathbunae
2 .805 (.563-.962) 52 7
Eriphia
scabricula
1 .636 (.386-.943) 29 3
I I
2 .576 (.283-.897) 30 6
Lydia
annulipes
2 .637 (.226-.976) 41 4
Xantho
leptodon
1 .608 (.376-.911) 21 2
Zozymodes
biunguis
1 .663 (.296-.986) 59 7
I I
2 .698 (.327-.961) 19 5
Habitat types as in Table 14.
220
Figure 52 - Diagram of Crab and Hole
Measurements.
Showing relation of crab
carapace height (H), length
(L), and width (W) to hole
aperture height (H) and
width (W).
221
crab size and hole size, using data for individual crabs
of each species (appendix 6; correlation formula given in
appendix 5). In each case, crab and hole size was found
to be significantly correlated (p= .05).
Now, it is possible that a correlation of this kind
could come about simply because large crabs cannot live in
small holes, while small crabs could live in large or
small holes. However, when scatter diagrams of crab size
and hole size are constructed (e.g. fig. 53), it is found
that the smaller crabs do not tend to live in the larger
holes, suggesting some kind of competition between larger
and smaller animals for the larger holes.
Correlations were also run on crab and hole size
using the means of the five species (appendix 6), and a
significant relationship (also at p= .05) was found. As
expected, the larger species prove to live in the larger
holes. Mean crab and hole shape were also calculated,
using the crab height/length ratio as the index of crab
flatness most closely related to hole shape (height/width
of the aperture; these data are depicted for each species
in appendix 7). When crab shape and hole shape were cor
related, a significant relationship (p= .05), was once
more found indicating that the flatter crab species are
the ones which inhabit lower crevices.*
*Regression equations are given for all of these cor
relations, in the appropriate appendices.
Crab Size (Height + Length) in
O-JP
I 10 go 30
Hole Size (Height + Width) in mm
+ = Eriphia scabricula Reef Flat Data Only,
o = Zozymodes biunguis
Figure 53 - Scatter Diagram of Crab and Hole Size, for Two Reef Crabs.
222
223
Ecological Separations Due to Hole Size and Hole Shape
While correlation coefficients are useful in demon
strating some of the dimensional relationships between the
crabs and their holes, they do not provide very meaningful
information on ecological separations. Instead, it is
necessary to employ techniques such as analysis of per
centage separation which allow the differences between
individual species to be quantified.
Before percentage of overlap can be studied, it is
necessary to combine hole height and width data in some
way that will roughly approximate the average size of the
holes. This could be done by averaging height and width,
or by multiplying them to give an estimate of the cross-
sectional area of the hole apertures. As aperture cross-
sections were not measured using the actual impressions,
it is felt that there is no sense in multiplying linear
dimensions to calculate an estimated area; therefore the
averaging method is used. There is no need to take a true
average (height + width)/2); as these data are only used
for cross-comparisons between species, the measure of hole
height used is simply height + width.
In analysing hole-size overlap, the crab-hole data
(initially plotted on a graph with 1-mm hole size incre
ments) are divided into five convenient groups (depicted
in Fig. 54). The rationale for the hole-size categories
224
70- t
w
.a
( d
a
u
Dacryopilumnus rathbunae
o
Habitat 2
Number of Crabs: 52
•P
C
d )
o
p
0 )
1 0 -
10 < to — W - h —
H < lb < 25"<
Hole Size (Height + Width) in mm
l o O - I
So-
m
o
•p
a
< u
o
u
a )
/o < 10-
/v<
Eriphia scabricula
Solid Lines, Habitat 1
Number of Crabs: 29
Dashed Lines, Habitat 2
Number of Crabs: 30
Hole Size (Height + Width) in mm
Habitat types as in Table 5.1.
Figure 54 - Hole Size Distribution, for Crevice-
Inhabiting Xanthids.
Percent of Crabs Percent of Crabs
225
( , 0 - t
s o -
30-
n o -
/V- n - Z5H-
Lydia annulipes
Habitat 2
Number of Crabs: 41
Hole Size (Height + Weight) in mm
60-i
50-
Xantho leptodon
Habitat 1
Number of Crabs: 21
Hole Size (Height + Weight) in mm
Figure 54 - Continued
Percent of Crabs
226
40-
s o -
m-
30-
Z0 -
t o
Zozymodes biunguis
Solid Lines, Habitat 1
Number of Crabs: 59
Dashed Lines, Habitat 2
Number of Crabs: 19
/o< T o - W - 7 7“
/ ■
1 0 -
0
Dacryopilumnus rathbunae
Habitat 2
Number of Crabs: 52
.7c .74-
Hole Shape Index (Height/Width)
70-
60
S O
n o
n o
20
1 0
0
Eriphia scabricula
Solid Lines, Habitat 1
Number of Crabs: 29
Dashed Lines, Habitat 2
Number of Crabs: 30
•7 .74-
Hole Shape Index (Height/Width)
Habitat types as in Table 51.
Figure 55 - Hole Shape Distribution, for
Crevice-Inhabiting Xanthids.
229
20
10
2*0
O ‘ t o
4J 30
S 2 0
s»
0
•7- .7+
Lydia annulipes
Habitat 2
Number of Crabs: 41
Hole Shape Index (Height + Width)
80
0 1
n TO
r t f
H
6 0 -
U
50-
M - l
0
%■
+» 30
C
0 ) ZD-
O
10■
0 )
CM
Xantho leptodon
Habitat 1
Number of Crabs: 21
.7* . 7 4 -
Hole Shape Index (Height + Width)
o i 80
*§ ™1
u w
s o -
° < f o -
a
o
CM
1 0
Zozymodes biunguis
Solid Lines, Habitat 1
Number of Crabs: 59
Dashed Lines, Habitat 2
Number of Crabs: 19
•7< . 7 4 -
Hole Shape Index (Height + Width
Figure 55 - Continued
230
situations.
Although it may appear as if the second grouping of
habitats is an artificial one, this is not strictly true.
Dacryopilumnus rathbunae holes in beachrock are of the
same kind as those in conglomerate and algal rock; for
this species, all the hole data came from the same general
kind of topography (and the holes are probably of the
same origin, i.e. derived from the borings of sipuncu-
lids). It would probably be reasonable to compare hole
data for Dacryopilumnus rathbunae with any data for
Eriphia scabricula or Lydia annulipes, but it is believed
that a more precise comparison can be made using data for
these latter species only from habitats where Dacryop
ilumnus rathbunae actually occurs.
In the case of Eriphia scabricula and Lydia an
nulipes , the non-reef-flat habitats from which hole data
were collected are somewhat more heterogeneous in appear
ance, but approximately the same mix of algal rock, con
glomerate, and beachrock data are included; therefore the
comparison is a legitimate one.
The results of the analysis of percentage separation
are as follows: two hole-size categories can be estab
lished, one of smaller holes (under 17mm in combined
height and width) inhabited by Zozymodes biunguis and
Dacryopilumnus rathbunae, and one of the larger holes
231
(over 17 mm in combined height and width) inhabited by
Eriphia scabricula, Lydia annulipes, and Xantho leptodon
(table 16). Two categories of hole shape can be estab
lished, one of the rounder holes inhabited by Dacryopi
lumnus rathbunae, and one of flatter holes inhabited by
Eriphia scabricula and Xantho leptodon (with the other
two species overlapping both categories) (table 17, fig.
56) .
Hence, of ten possible species-pair comparisons, six
pairs can be significantly separated (p= .001) on the
basis of hole size, and only two on the basis of hole
shape. In both cases where there is a significant dif
ference in the shape of the holes inhabited by pairs of
species, there is also a significant hole-size difference;
hole shape alone does not account for any of the eco
logical separations between these species. Therefore,
there is little doubt but that hole size is the more
important of the two factors in separating these crabs
ecologically.
Ecological Separations Due to Differences in Sediment
Cover and Topography in Consolidated Rock Substrates.
Although the study of crab refuge holes discussed
in the preceding section provided much useful information
about the niches of certain xanthid crabs, it must be
noted that this by no means explains all of the substrate-
related differences in the ecology of xanthid crabs.
232
TABLE 16
ECOLOGICAL SEPARATIONS WITH RESPECT TO HOLE SIZE
Species and Habitats
Compared
Percent of
Separation Chi2 Sig.
Dacryopilumnus
rathbunae (2) ,
Eriphia scabricula (2)
76.2 42.1 .001
Dacryopilumnus
rathbunae (2)
Lydia annulipes (2)
71.8 55.3 .001
Dacryopilumnus
rathbunae (2),
Xantho leptodon (1)
86.7 29.3 .001
Dacryopilumnus
rathbunae (2),
Zozymodes biunguis (1)
27 .9 7.1 .01
Dacryopilumnus
rathbunae (2) ,
Zozymodes biunguis (2)
41.1
Eriphia scabricula
Lydia annulipes (2)
(2) , 1.2 0.8 Not Sig.
at . 05
Eriphia scabricula
Xantho leptodon (1)
(1) ,
0.7 0.0 Not Sig.
at .05
Eriphia scabricula
Zozymodes biunguis
(1) ,
(1)
72. 6 37. 8 .001
Eriphia scabricula
Zozymodes biunguis
(2) ,
(2)
70.8 “
Lydia annulipes (2)
Xantho leptodon (1)
t
3.6 0.1 Not Sig.
at . 05
Lydia annulipes (2)
Zozymodes biunguis
(1)
65.4 47.2 .001
Lydia annulipes (2)
Zozymodes biunguis
(2)
67.1
Xantho leptodon (1)
Zozymodes biunguis
(1)
80.3 20.8 .001
Habitat types as in table 51.
233
TABLE 16“ Continued
Chi^ Data
Species and Habitat
Small
No.
..."T
Holes
Crabs
Large
No.
Holes^
Crabs
Dacryopilumnus (>16) 46 (16+) 6
rathbunae (2),
Eriphia scabricula (2) 5 25
Dacryopilumnus (>18) 51 (18+) 1
rathbunae (2) ,
Lydia annulipes (2) 31
Dacryopilumnus (=-13) 38 (13+) 14
rathbunae (2),
Xantho leptodon (1) 1 20
Dacryopilumnus (>10) 9 (10+) 49
rathbunae (2),
Zozymodes biunguis (1) 22 37
Eriphia scabricula (2), (>24) 15 (24+) 15
Lydia annulipes (2) 16 25
Eriphia scabricula (1), (>24) 15 (24+) 14
Xantho leptodon (l) 11 10
Eriphia scabricula (1), (>16) 5 (16+) 24
Zozmodes biunguis (1) 50 9
Lydia annulipes (2), (>26) 22 (26+) 19
Xantho leptodon (1) 11 10
Lydia annulipes (2), (>17) 10 (17+) 31
Zozymodes biunguis (1) 54 5
Xantho leptodon (1), (>13) 1 (13+) 20
Zozymodes biunguis (1) 37 22
^Figures in parentheses represent size ranges
of holes (hole height + hole width) in mm.
234
TABLE 17
ECOLOGICAL SEPARATIONS WITH RESPECT TO HOLE SHAPE
Species and Habitats
Compared
Percent of
Separation Chi2 Sig.
Dacryopilumnus
rathbunae (2),
Eriphia scabricula (2)
60.8 29.1 .001
Dacryopilumnus
rathbunae (2)
Lydia annulipes (2)
46 „ 7 20.9 .001
Dacryopilumnus
rathbunae (2),
Xantho leptodon (1)
61.8 22.2 .001
Dacryopilumnus
rathbunae (2),
Zozymodes biunguis (1)
42.5 21. 3 .001
Dacryopilumnus
rathbunae (2) ,
Zozymodes biunguis (2)
28.2
Eriphia scabricula (2),
Lydia annulipes (2)
14.1 0.5 Not Sig.
at .05
Eriphia scabricula (1),
Xantho leptodon (1)
15.5 0.1 Not Sig.
at . 05
Eriphia scabricula (1),
Zozymodes biunguis (1)
2.8 0.3 Not Sig.
at .05
Eriphia scabricula (2),
Zozymodes biunguis (2)
32.6
— —
Lydia annulipes (2),
Xantho leptodon (1)
15.1 0.4 Not Sig.
at .05
Lydia annulipes (2)
Zozymodes biunguis (1)
3.2 0.1 Not Sig.
at . 05
Lydia annulipes (2),
Zozymodes biunguis (2)
17.5
— “
235
TABLE 17- Continued
Species and Habitats Percent of
9
Compared Separation Chi Sig.
Xantho leptodon (1)
Zozymodes biunguis (1)
18.3 2.1 Not Sig.
at . 05
Habitat types as in Table 51.
236
TABLE 17- Continued
Chi2 Data
Species and Habitat Flatter
No.
Holesl
Crabs
Rounder
No.
Holes^
Crabs
Dacryopilumnus (>0.700) 10 (0.700+) 42
rathbunae (2)
Eriphia scabricula (2) 24 6
Dacryopilumnus (>0.700) 10 (0.700+) 42
rathbunae (2)
Lydia annulipes (2) 27 14
Dacryopilumnus (>0.780) 18 (0.780+) 34
rathbunae (2)
Xantho leptodon (1) 20 1
Dacryopilumnus (>0.700) 10 (0.700+) 42
rathbunae (2),
Zozymodes biunguis (1) 37 22
Eriphia scabricula (2), (>0.600 18 (0.600+) 22
Lydia annulipes (2) 21 20
Eriphia scabricula (1), (>0.613) 15 (0.613+) 14
Xantho leptodon (1) 10 11
Eriphia scabricula (1), (>0.600) 17 (0.600+) 12
Zozymodes biunguis (1) 31 28
Lydia annulipes (2) , (>0.600) 21 (0.600+) 20
Xantho leptodon (1) 9 12
Lydia annulipes (2), (>0.700) 27 (0.700+) 14
Zozymodes biunguis (1) 37 22
Xantho leptodon (1), (>0.620) 13 (0.620+) 8
Zozymodes biunguis (1) 26 33
^ Figures in parentheses represent hole shape
ranges (hole height/hole width).
237
INCHI' S
Figure 56 - Carapace Shapes of Two Species
of Reef Crabs.
View from the side of crabs'
carapaces sans appendages
and mounted on putty base.
Left; Dacryopilumnus
rathbunae.
Right; Xantho leptodon.
238
This section deals with a number of other factors, such
as the nature of the sediment and/or algal cover, topo
graphy, or a combination of all of these, which may be
equally important in determining ecological separations
between crabs in reef situations.
In the case of Dacryopilumnus rathbunae and
Zozymodes biunguis, it is clear that topography is at
least indirectly involved. Although the former is almost
never found on reef flats, where the bulk of the
Zozymodes biunguis population lives, the two species occur
together in places where the water runs off of irregular
surfaces at low tide.
Upon close examination, there prove to be two
distinct types of rock surface on the eroding conglomerate
and algal ridge remnants where these two species overlap.
On the upper parts of steeply sloping to nearly vertical
walls and the convex upper edges of these walls there is
no visible algal turf or accumulated sediment (vermetids
being visible on the rock surface), the characteristic
species is Dacryopilumnus rathbunae, which is abundant in
old sipunculid holes. On the lower parts of these walls
and concave surfaces in general where there is a sand-
binding algal turf, the commoner xanthid is Zozymodes
biunguis; this form is also found where there are con
cavities in the rock higher up, producing similar
239
conditions.
When samples of both types of rock surface were ex
amined under a dissecting scope, it was found that the
former type, which is light green in color when wet, is
covered by a thin film of algae not over 1mm, and usually
no more than 200 or 300^ thick. The latter type on the
other hand is 2 to 5mm thick, and the algae (often Jania)
incorporate sand and detrital particles (up to 500yu* in
diameter).
An analysis of relative abundance, supported by
2
Chi tests (significant at the .001 level) indicates that
these two xanthid species are ecologically separated by
topography and algal-sediment cover (table 18). It is un
certain however whether sedimentation or desiccation is
ultimately responsible for this difference. The fact that
Dacryopilumnus rathbunae is not abundant on sediment-
covered portions of these exposed rock surfaces is in
keeping with its absence on the reef flat, where sand
often accumulates in crab holes; this may relate to the
lack of setation around the incurrent openings to the
gills of Dacryopilumnus rathbunae as compared to
Zozymodes biunguis, which is well provided with such
setae. On the other hand, it should be noted that the
thicker algal covering on the rock surfaces where
Zozymodes biunguis is the more abundant species may re-
240
TABLE 18
ECOLOGICAL SEPARATION BETWEEN DACRYOPILUMNUS
RATHBUNAE AND ZOZYMODES BIUNGUIS WITH RESPECT
TO TOPOGRAPHY AND SEDIMENT COVER.
Habitat relative abundance Chi2 Sig.
1 Zozymodes biunguis 51.2 .001
12.8% as abundant as
Dacryopilumnus rathbunae
2 Dacryopilumnus rathbunae
17.1% as abundant as
Zozymodes biunguis
Habitat Types: 1. Algal Film 1mm or less,
Without Bound Sediment;
Rock Surface Often Convex.
2. Algal Turf 2 to 5mm thick,
Generally Incorporating
Much Sediment; Rock Surface
Often Concave.
Both (1.) and (2.) occur on eroding
algal ridge remnants, conglomerate,
and beachrock rims.
Chi2 Data
Species
Number of Crabs
Habitat 1 Habitat 2
Dacryopilumnus
rathbunae
39 7
Zozymodes
biunguis
5 41
Number of Stations 7 4
241
duce desiccation in these places, whereas the surfaces
inhabited by Dacryopilumnus rathbunae clearly dry at low
tide. Perhaps the reduced claw size and globular carapace
(permitted by life in holes of cylindrical cross-section)
of Dacryopilumnus rathbunae are adaptations for reducing
the surface/volume ratio of this crab. Further work
would be required to answer these questions.
In comparing Zozymodes biunguis with other low inter
tidal species living in consolidated rock substrates, it
will be noted that this species has been shown to live in
significantly smaller holes than those of Eriphia
scabricula and Xantho leptodon, with which it often occurs.
No doubt it is also separated from Xantho sanguineus on
the same basis, as this latter species is even larger than
Xantho leptodon. Hole size is evidently the only sub
strate-related factor involved in the separation between
Zozymodes biunguis and these reef-flat species.
Zozymodes biunguis has also been shown to live in
smaller holes than those of Lydia annulipes, but the
latter is a high intertidal species and the vertical
zonation difference is evidently more important in separ
ating these species. Zozymodes biunguis also seems to
inhabit a different type of substrate (i.e. small holes in
the rock) than does Pseudozius caystrus, which occurs
under loosely cemented slabs with gravel underneath.
242
However, Pseudozius caystrus is a high intertidal species,
and its preferred habitat in conglomerate does not occur
at a lower level where Zozymodes biunguis reaches its
maximum abundance. These two species have been taken to
gether in low crevices between slabs of beach rock in a
few places (as at station 14B at Runit Island), which
suggests that there is some overlap in the substrate
preferences of these species (though it is difficult to
estimate how much).
When Dacryopilumnus rathbunae is compared with other
low intertidal forms, it is found that there is a clear
topographic separation between this species, which occurs
on algal rock, eroding conglomerate, and beachrock, and
Xantho leptodon and Xantho sanguineus, which live on reef-
flats (table 19). This difference can be supported
quantitatively by analysis of relative abundance and
Chi tests (significant at .001). On the other hand,
Dacryopilumnus rathbunae often occurs together with
Eriphia scabricula on the same topography, but as has
been previously seen, these two species are separated by
hole size.
In comparing Dacryopilumnus rathbunae with the high
intertidal Pseudozius caystrus (whose vertical range it
overlaps to some extent), a clear separation in substrate
type can be shown; Dacryopilumnus rathbunae has never
been found under slabs, and P s e u d o z iu s o a y s t r u s d n s s not
243
TABLE 19
ECOLOGICAL SEPARATION OF DACRYOPILUMNUS
RATHBUNAE FROM PSEUDOZIUS CAYSTRUS,
XANTHO LEPTODON, AND XANTHO SANGUINEUS
WITH TOPOGRAPHY.
Species and Habitats
Compared
Co-occurrence Chi2 Sig.
Dacryopilumnus
rathbunae (1),
Pseudozius caystrus (2)
None 81.0 .001
Dacryopilumnus
rathbunae (1),
Xantho leptodon (3)
None 101.8 .001
Dacryopilumnus
rathbunae (1),
Xantho sanguineus (2)
None 80.2 .001
Habitat Types: 1. Eroding Algal Ridge
Remnants, Margins of
Conglomerate Flats,
and Beachrock Rims.
2. Conglomerate Flats with
Cemented Slabs.
3. Reef Flats, with Tide Pools.
5------
Chi Data
Species
Number
Hab. 1
of Crabs
Hab. 2 Hab. 3
Dacryopilumnus
rathbunae
44 0 0
Pseudozius cavstrus 0 37 0
Xantho leptodon 0 0 67
Xantho sanguineus 0 0 36
Number of Stations 8 4 5
244
live in small round holes. In this case it was possible
to collect sufficient data on Dacryopilumnus rathbunae in
the high intertidal for use in demonstrating the lack in
substrate overlap between these species where they do
occur at the same levels (table 19). This finding is
supported by a Chi test (significant at .001). On the
other hand where Dacryopilumnus rathbunae overlaps
Lydia annulipes in the high intertidal, it is on the same
type of topography, though there is a certain degree of
hole-size separation between these forms.
In the case of the three larger species which occur
on reef-flats, it will be recalled that no difference in
hole size was found between Eriphia scabricula and Xantho
leptodon, and that no hole-size data are available for
Xantho sanguineus. However, each species can be shown to
be specific to certain areas of the reef, which differ
with respect to topography or extent of sediment cover.
Of these species, Eriphia scabricula is the one most
characteristic of non-reef-flat features, such as lagoon
beachrock, eroding conglomerate rims, and algal ridge
remnants. It is also very common on certain reef-flats,
such as the outer reef zone at station 4. Here an algal
covering at least 15mm thick covers most of the area. Of
the algae making up this turf, only samples of Jania were
found to incorporate any sand (and at that a negligible
245
amount); the other types of algae (e.g. Dictyosphaerea,
Microdictyon) proved to be clean of sediment. On the in
ner reef at station 4, on the other hand, Xantho leptodon
is the most abundant of the larger xanthids. Whereas
about half of the reef rock in this zone is covered by a
thin scum of algae (under 1mm thick), devoid of sediment,
most of the crabs live in the larger pools where there is
an abundance of sand and coarser sediment, which partly
fills many of the crabs' holes. The flat pool bottoms are
characterized by a flocculent coating of sand and algae
of variable thickness, covering 50 to 75 percent of the
surface area. Sand comprises at least 90 percent of the
volume of this material, which is loosely held together by
fine, threadlike green algal filaments which give the
entire mass a green tint, but no definite shape.
Another locality where Xantho leptodon proves to be
very abundant (and Eriphia scabricula almost absent) is
the innermost reef-flat at Uliga Island (station 47). A
sandy algal turf, 20 to 20mm thick in many places, covers
almost the entire area, and about half of its volume is
made up of sand. Unlike the inner reef at station 4,
tide pools are indistinct, and crevices as well as the
algal-sand turf are rather evenly distributed across the
reef surface. Apparently, the increase in the amount of
sediment to landward is a major factor controlling the
246
distribution of these two species. On the narrow reef
flat at station 4, there is also a gradient in wave
exposure which might offer an alternative explanation,
but this is not the case at station 47 where Eriphia
scabricula occurs on the dark green zone of the inner
reef, on a relatively sediment-clean part of the reef that
is as sheltered as the shoreward zone where Xantho
leptodon lives. This ecological separation (due to
sediment cover) can be documented statistically by an
analysis of relative abundance (table 20), supported by a
Chi test (significant at .001).
Xantho sanguineus was the dominant large xanthid on
the seaward reef at Dalap Island (station 48A). In this
locality, the reef surface is generally similar to that of
the dark green zone at station 47, only numerous crevices,
20 to 30cm long, undercut the sides of many of the tide
pools. Xantho sanguineus is abundant under the ledges
so formed, and is much commoner than Eriphia scabricula
in this locality. On the inner reef at station 4, Xantho
sanguineus is likewise more abundant than Eriphia
scabricula; this reef is similarly characterized by num
erous 15 to 30cm crevices around the edges of pools.
These species can be statistically separated on the basis
of topography, using an analysis of relative abundance
(table 20) and a Chi^ test (significant at .001); Xantho
247
TABLE 20
ECOLOGICAL SEPARATION OF ERIPHIA SCABRICULA
FROM XANTHO LEPTODON AND XANTHO SANGUINEUS
WITH RESPECT TO TOPOGRAPHY AND SEDIMENT
COVER
Habitat Relative Abundance
----- -----
Chi Sig.
1 Xantho leptodon absent 89.9 .001
2 Eriphia scabricula absent
1 Xantho leptodon absent 78.0 .001
3 Eriphia scabricula
3.2% as abundant as
Xantho leptodon
1 Xantho sanguineus absent 72.9 .001
4 Eriphia scabricula
11.8% as abundant as
Xantho sanguineus
Habitat Types: 1. Eroding Algal Ridge and
Conglomerate Remnants, and
Beachrock; and Reef Flats
Without Algal-bound Sand
and Lacking Undercut Ledges
(Station 4, Outer Reef).
2. Reef Flat Almost Completely
Covered by Sandy Algal Turf
(Station 47, Innermost Reef).
3. Reef Flat with 50-75% of
Tide Pools Covered by Sand
Very Loosely Bound by Fine
Algal Filaments (Station 4,
Innermost Reef).
4. Reef Flats with 15-30cm long,
Undercut Ledges (Station 4,
Innermost Reef, Station 48A.
248
TABLE 20- Continued
Chi^ Data
Species Number of Crabs
Hab. 1 Hab. 2 Hab. 3 Hab. 4
Eriphia scabricula 50 0 1 4
Xantho leptodon 0 40 31
-
Xantho sanguineus 0 - - 34
Number of Stations 6 1 1 2
249
sanguineus lives where these crevices occur, and Eriphia
scabricula where they are absent. It is uncertain whether
Xantho sanguineus is favored by this habitat because the
undercut ledges provide better cover while the crabs feed
in the open, or because there are larger holes underneath
(or both).
Yet another difference between Eriphia scabricula and
the two Xantho species is the fact that the former occurs
only in holes in the reef rock, whereas the latter may
also be taken from under loose rocks and among gravel on
the reef-flat.
Xantho leptodon and Xantho sanguineus appear to be
separated at least in part by the extent of sediment cover
on the reef-flat. At Dalap Island, where Xantho
sanguineus is dominant, there is little loose sediment in
the tide pools, although some sand is bound in a thin
algal turf. The innermost reef at Uliga Island, where
Xantho leptodon is the characteristic species, is almost
completely covered with sand, much of it bound in a thick
turf. The inner reef at Eniwetok Island station 4 re
presents an intermediate condition, though it is certain
that there is much loose sand lying in the tide pools;
both species occur here, but Xantho leptodon is much
commoner. Extent of sand cover can be shown by analysis
of relative abundance (table 21) and a Chi test (sig
nificant at .001) to be a major factor separating these
250
TABLE 21
ECOLOGICAL SEPARATION BETWEEN XANTHO LEPTODON
AND XANTHO SANGUINEUS WITH RESPECT TO
TOPOGRAPHY AND COVER
Habitat Relative Abundance Chi ^ Sig.
1 Xantho sanguineus absent 67.1 .001
3 Xantho leptodon absent
2 Xantho sanguineus
14.3% as abundant as
Xantho leptodon
46.4 .001
3 Xantho leptodon absent
1 Xantho sanguineus absent 63.2 .001
4 Xantho leptodon
20% as abundant as
Xantho sanguineus
Habitat Types: 1. Reef Flat Almost Completely
Covered by Sandy Algal Turf
(Station 47, Innermost Reef).
2. Reef Flat with 50-75% of
Tide Pools Covered by Sand
Very Loosely Bound by Fine
Algal Filaments; Much Loose
Sand under Ledges (Station
4, Innermost Reef).
3. Reef Flat with Low, Sand-
binding Algal Turf over about
50% of Area; Little Loose
Sediment, and Little Sand
under Ledges (Station 48A).
4. Reef Flats with Little Sand:
Bottoms of Pools with Gravel
and Rocks up to 21.5cm long
(Station 31, 46, Inner Reefs).
251
TABLE 21 - Continued
.... ■ T5----------
Chi Data
Species
Number of Crabs
Hab. 1 Hab. 2 Hab. 3 Hab. 4
Xantho leptodon
Xantho sanguineus
39 28 0 9
0 4 0 45
Number of Stations 1 1 1 2
252
two species on reef flats.
On reef flats where rocks and gravel occur in small,
scoured tide pools (such as the inner reefs at Kwajalein
and Darrit Islands), Xantho sanguineus again becomes
abundant. In these pools, the crabs live under rocks of 5
to 21.5cm in diameter, lying on the bare reef rock or upon
. o
some gravel. An analysis of relative abundance and Chi
test (significant at .001) was used to demonstrate the
ecological separation between Xantho sanguineus and Xantho
leptodon due to the kind of sediment on the reef; the
former is found where there are rocks and gravel but little
sand, the latter where there is much sand, few rocks, and
no evidence of scour (table 21).
It is difficult to compare reef-flat crabs with high
intertidal crabs with respect to substrate; there are types
of substrate not common to the high and low intertidal,
and the crabs in question are abundant on some of these.
Eriphia scabricula appears to approximate most closely the
high intertidal Lydia annulipes in terms of substrate re
quirements; the two live in holes of approximately the
same size and shape, and in some places (e.g_. the algal
rock at station 2, and lagoon beachrock), the general ap
pearance of the rock surface is also similar. In other
localities, however, the difference may be pronounced; for
instance the thick green algal turf of inter-island reefs
where Eriphia scabricula lives is very different from the
cyanophyte scum on the rock surfaces where Lydia annulipes
occurs. These two crabs differ in color, which probably
camouflages the animals for movement over their normal
background rocks; but this difference is probably due
primarily to vertical zonation. Eriphia scabricula and
Pseudozius caystrus appear to differ more in terms of sub
strate (P. caystrus seldom lives in isolated holes as does
Eriphia scabricula)but there may be some overlap as both
species are found in beachrock crevices. Xantho leptodon
and Xantho sanguineus occasionally live in pooled places
under the conglomerate slabs normally inhabited by
Pseudozius caystrus and Lydia annulipes (in addition to
which Lydia annulipes and Xantho leptodon occupy holes of
similar size); this suggests that all of these species may
overlap somewhat in their substrate requirements.
In the high intertidal, there are two principal kinds
of consolidated rock substrates inhabited by xanthids:
jagged, irregular rock surfaces with numerous holes and
pits (usually the eroding margins of conglomerate plat
forms) , and the gravel-filled spaces under slabs on con
glomerate flats. Lydia annulipes is characteristic of the
former type; data on the holes inhabited by this species
are given in table 14. Pseudozius caystrus is the most
abundant crab in the latter habitat, though Lydia
254
annulipes also regularly occurs there in smaller numbers.
Table 22 demonstrates through the use of analysis of re-
2
lative abundance and a Chi test (significant at .001) that
the nature of the topography is of considerable importance
in determining the distribution of these crabs. This dif
ference is in all probability due to the very secretive
habits of Pseudozius caystrus, which does not come out from
under cover in nature; the under-slab habitat should give
it adequate protection, while Lydia annulipes, which
wanders widely about the reef surface at night, can live
where there are only small holes available.
Substrate-related Crab Behavior in Rubble Situations
Four species of xanthids are very abundant in loose
rubble: Xantho leptodon, Xantho sanguineus, Xantho
gracilis, and Pseudozius caystrus. Before discussing the
ecological separations between these forms, it is worth
noting the way in which the substratum of loose rocks,
lying over sand and gravel, is used for cover.
On the exposed windward-side beaches of Eniwetok
Island, to which field observations of live animals were
restricted, these xanthids were never seen out in the open,
night or day, either at low tide, when the water had drain
ed from the beach, or at high tide under conditions of
moderate surf (at which time observations were made with a
water-glass). In places where water is retained in pools
255
TABLE 22
ECOLOGICAL SEPARATION BETWEEN LYDIA ANNULIPES
AND PSEUDOZIUS CAYSTRUS WITH RESPECT TO
TOPOGRAPHY
Habitat Relative Abundance
2
Chi Sig.
1 Pseudozius caystrus
3.4% as abundant as
Lydia annulipes
119.6 .001
2 Lydia annulipes
25.3% as abundant as
Pseudozius caystrus
Habitat Types: 1. Eroding Algal Mounds, Margins
of Conglomerate Platforms,
and Lagoon Beachrock; Crabs
in Individual Holes.
2. Flat Surface of Conglomerate
Platforms, and Seaward Beach
Conglomerate; Crabs Co-occur
ring under Continuous Cemented
Slabs.
Chi2 Data
Species Number of
Habitat 1
Crabs
Habitat 2
Lydia
annulipes
88 23
Pseudozius
caystrus
3 91
Number of Stations 9 5
256
at low tide, some Xantho leptodon were seen feeding in the
open at night. In aquaria kept in the laboratory, all of
these species were active to some extent (the Xantho spec
ies would even sit on top of their rocks in the quiet water
of the tank); when the water was drained from the tanks,
they would burrow or move under cover. Judging from these
observations, it appears that they tend to be more active
under water, coming out from their cover only under quiet
conditions.
A number of field observations were made on the be
havior of these crabs when their sheltering rocks were
removed to determine the nature of their escape behavior
and digging habits in nature. The most detailed observa
tions were made on Xantho leptodon in the pool at station
5A, under water at low tide. When a rock was lifted, the
crabs underneath would often scuttle away, running and
stopping in spurts. They would generally move along any
available cracks or crevices (as along the edge of a rock),
with either the ventral or dorsal side against the rock
(in the latter case, holding the last pair of legs against
the rock). Upon encountering an opening under a rock, a
crab would readily dart underneath; otherwise it would
often push vigorously into the sediment under a rock.
Some crabs would, in lieu of running away, dig rapidly
into the gravel, pushing up the pebbles with their claws.
257
If resting on sand they would push very gradually back
wards into the sediment, executing this movement so slowly
that their motion could be perceived only with difficulty.
This "melting away" into the sand no doubt reveals an un
detected crab's presence much less than would a more rapid
burial; it is apparently possible only in fine sediment,
however.
Crabs which were exposed on the adjoining beach at
low tide would act very differently, usually remaining
still for from 30 seconds to 3 minutes. Upon finally
moving, they would generally hide under rocks, but even
then their escape movement was a very sluggish one in
comparison to their behavior under water.
For Xantho gracilis, the escape pattern was essential
ly the same as that of Xantho leptodon, whether under water
or exposed to the air. Observations made on this species
at station 5B (a beach which dries at low tide) revealed
that the crabs behave in the same way on a calm lower high
tide with little wave action or on a higher high tide with
considerable turbulence in the water, generally running or
burrowing under a rock, or hiding in gravel. When rocks
are lifted from a drying beach, all three species of Xantho
may be found hiding in individual pockets in the sand
underneath; while it is evident that they are efficient
burrowers, in only one instance was a small mound of sand
258
found at the side of a rock at low tide, betraying in this
case the presence of a Xantho gracilis.
Some observations made on Pseudozius caystrus at
station 5B during a higher high tide (when wave action was
heavy) revealed that this species is unable to dig very
easily in the sediment under rocks, unlike Xantho gracilis
in the same locality. Individuals of Pseudozius caystrus
would generally run into openings beneath the rocks; some
simply wedged themselves between the rock and sand, or sat
at the bases of rocks, and a few were washed away by the
surge before they took any action. When exposed to the air
at low tide, this species would often "play dead" like the
Xantho species, but unlike the latter is much more prone to
run away rapidly (holding its claws up in a menacing posi
tion) as soon as the rocks are disturbed.
A number of observations on the actual mode of burrow
ing were made on captive animals in the laboratory. Xantho
gracilis and Xantho leptodon would dig under rocks by
moving sediment forward beneath their bodies with their
legs and pushing it away with their claws, in the process
blocking the entrance to the burrow. When digging down
ward into gravel, the crab often held its back against a
larger pebble. Sometimes the crabs would dig in sideways;
one Xantho leptodon was observed to use the claw on one
side (held against a rock) to force the opposite side of
259
its body under a pebble, meanwhile executing the normal
digging movements with its remaining limbs.
Pseudozius caystrus would dig very readily in the
quiet water of the aquarium. Nearly all of their burrowing
activity was carried out, however, while they were under
neath rocks, as this species would almost never venture
away from cover, unlike Xantho gracilis in the same tanks.
Pseudozius caystrus has never been observed to dig into
sand or gravel without having a large rock overhead. When
excavating in gravel under a rock, the pebbles were either
pushed in front of the crab with the legs of the leading
side, or dragged along by the legs of the trailing side;
also, the crabs were observed to pick up individual pebbles
with their pincers and drop them outside the burrow, or to
pass them across the top of the carapace, using the legs
and one claw.
When constructing a burrow in sand beneath a rock,
both claws would be used to push sand from the entrance.
In transporting sand under the rock, the claw on the lead
ing side was used as a plow to move sand ahead of the body,
while the legs on the trailing side were used to drag it
along behind. In concert with these motions, the posterior
legs on the leading side might be hooked against the rock
overhead to aid in pulling the crab's body forward until
the sand was moved to the edge of the rock. At this point
260
the sediment was pushed away from the excavation by the
leading legs. In walking along these tunnels the last pair
of legs was usually held up against the rock overhead.
When the burrows of two Pseudozius caystrus became acciden
tally joined due to the digging activity of one of the
crabs, both avoided contact, and the smaller one immedi
ately pushed up sand to plug the opening.
Functional Morphology of Rubble Crabs
The relationship between crab carapace shape and hole
shape was demonstrated in a previous section. One would
similarly expect a relationship between the shape of
xanthid crabs and the extent to which they burrow in loose
rubble; species which occupy the low crevices under rocks
should be flatter than those which live in holes. It is
difficult to quantify this however because of the problem
of trying to estimate what percentage of the total pop
ulation of certain species live on reefs and what percent
age inhabit beaches (this probably varies from one atoll
to the next, in accordance with the degree of porosity of
the intertidal reefs).
However, the eight study species can be ranked after
making a subjective estimate of their tendency to live in
loose rubble as opposed to consolidated rock substrates.
Appendix 8 demonstrates that rubble-inhabiting species are
indeed flatter than those which seldom live in rubble,
261
whichever index of flatness (crab height/length or crab
height/width) is used. It is probable that flattened
crabs can move about more efficiently in the spaces between
rocks and their underlying sediment.
The rugosity of the exoskeleton also shows some re
lationship to the type of habitat. Crab species which
usually inhabit holes in the reef tend to have granulated,
tuberculated, or sculptured carapaces, and often exhibit
well developed antero-lateral teeth; the claws are similar
ly rugose. This probably relates to their habit of wedg
ing themselves in their crevices when attacked.
When closely related intertidal xanthid crabs are
compared, the species which live in holes prove to be more
rugose than those which occur in loose rubble. Thus,
Eriphia scabricula, which lives only in holes, has tuber
culated claws whereas Eriphia sebana, which often lives in
rubble, has smooth claws; Xantho sanguineus and Xantho
leptodon, which live in holes in the reef as well as under
rocks, are more rugose than Xantho gracilis, which occurs
only under rocks, and Pseudozius pacificus, which does not
occur in rubble, is more rugose than Pseudozius caystrus,
which does. It is uncertain whether the relative smooth
ness exhibited by rubble crabs affords them any protection,
although it is apparent that half-buried Xantho gracilis
and Pseudozius caystrus resemble smooth pebbles.
262
Ecological Separations due to Rock Size
Although there is little reason to believe that the
slight variations in carapace shape exhibited by the var
ious species of rubble crabs bear any relationship to
heterogeneity in their habitat, these crabs do differ con
siderably from one another with respect to size, and it is
not illogical to expect crab size to correspond in some
way to the size of the rocks under which the animals hide.
This aspect of crab ecology was investigated as fol
lows: for each species, at least fifty crabs were collected
in five or more localities, and the dimensions of the
crabs and of their covering rocks were measured. These
collections were made on fields of loose slabs derived
from beachrock as well as in places where the rubble was
of the normal rounded kind (often derived from individual
heads of dead coral). It is evident from table 23 that
the crabs are much smaller than the rocks beneath which
they live. In fact, more than one crab on the average is
found under each rock (table 24); this is especially the
case with beachrock.
Tabulations were made of the lengths of rocks under
which individual crabs were collected, in 1cm size in
crements (each rock being recorded as many times as the
number of crabs found underneath); these data were then
grouped into several convenient size categories, depicted
263
TABLE 23
SIZE OF ROCKS INHABITED
BY 5 INTERTIDAL SPECIES
Mean Rock Size (Range in Parentheses)
Species and
Habitat
Rock
Length (cm)^
Mean Crab
Width (cm)
No.
Crabs
No.
Stations
Pseudozius
caystrus
1 34.9 (10- 79) 14. 8 43 4
1 !
2 56.1 (13-107) 13.0 30 1
Xantho
gracilis
1 32.1 ( 9- 63) 13.8 102 6
f l
2 40.5 (10-101) 11.0 66 1
Xantho
leptodon
1 19. 3 ( 8- 51) 10.8 70 7
Xantho
sanguineus
1 32.1 (10- 97) 18.4 59 6
I I
2 53.9 (43- 69) 23. 5 20 1
Habitats: 1. Rounded Rock Rubble.
2. Flattened Beachrock Slabs.
- L In tabulating data on mean rock length,
measurements of rocks which had more than
one crab underneath were repeated as many
times as the number of crabs they sheltered.
The number of rock measurements averaged
is the same as the number of crabs (above);
the actual number of rocks is given in
Table 24.
264
TABLE 24
NUMBERS OF CRABS UNDER ROCKS
IN RUBBLE HABITATS
Species and
Habitat
Mean Number of..
Crabs per Rock
Number
of rocks
Pseudozius
caystrus
1 1.4 22
t l
2 1.8 17
Xantho
gracilis
1 1.6 45
1 1
2 3.4 19
Xantho
leptodon
1 1.3 46
Xantho
sanguineus
1 1.6 36
I f
2 2.2 9
Habitats as in Table 23,
1 Number of crabs given in Table 23,
265
here in a histogram (fig. 57). The category of smaller
rocks (8 to 22cm) corresponds to the optimal rock-size
range for Xantho leptodon; larger rocks were divided be
tween two further categories (23 to 44cm, and 45 to 107cm)
to allow any less obvious differences between species to
be detected.
An analysis of percentage separation was then run on
the size of the rocks inhabited by the four species (table
25). Valid ecological separations (more than 50 percent
2
separation), supported by Chi tests significant at the
.001 level, were found only between Xantho leptodon and
Xantho sanguineus, and between Xantho leptodon and
Pseudozius caystrus.
As Xantho sanguineus and Xantho leptodon occur at the
same vertical level, for the most part, it is probable that
some form of competition for rocks of different sizes ac
counts for the separation between these species. On the
other hand, Xantho gracilis, which appears to be the
ecological equivalent of Xantho leptodon at a higher level
(resembling it in size, appearance, and behavior) overlaps
both Xantho leptodon and Xantho sanguineus in terms of rock
size. It is possible that lessened competition from the
larger species (Xantho sanguineus, which is much less
common higher up) allows Xantho gracilis to live under
larger rocks.
266
w
X i
< 0
p
u
m
o
+>
< u
u
p
0 )
d)
60
SO
W-
30-
# >-
1 0 -
Pseudozius caystrus
Solid Lines, Habitat 1
Number of Crabs: 43
Dashed Lines, Habitat 2
Number of Crabs: 30
S -2Z Z5~W
Rock Length in cm
w
X)
to
p
u
u-t
o
■ p
c
d )
o
P
a )
cw
so
4o-
3o-
90-
1 0 -
i Xantho gracilis
Solid Lines, Habitat 1
Number of Crabs, 102
Dashed Lines, Habitat 2
Number of Crabs, 66
2-Z L 25-W
Rock Length in cm
Figure 57- Rock Size Distribution, for
Rubble-Inhabiting Xanthids.
267
w
,Q
n J
L
U
O
-P
c
< D
O
u
0 )
P - 4
SO
w
yt r
s o
V o -
30-
20-
/o-
Xantho leptodon
Habitat 1
Number of Crabs, 70
1
0 - 2 Z Z 5 -W t r - lQ l
Rock Length in cm
f e o i
So-
w ^
n
n i
8 30-
in
0 20-
+>
a
n ( 0 -
o —a 23—vy vr— m
Rock Length in
Xantho sanguineus
Habitat 1
Number of Crabs, 59
cm
Figure 57- Continued
268
TABLE 25
ECOLOGICAL SEPARATIONS WITH
RESPECT TO ROCK SIZE
Species and Habitats
Compared
Percent of
Separation
Chi2 Sig.
Pseudozius caystrus (1) ,
Xantho gracilis (1)
18.8 22.6 .001
Pseudozius caystrus (2),
Xantho gracilis (2)
18.5 1.8 Not Sig
at .05
Pseudozius caystrus (1),
Xantho leptodon (1)
52.4 82.4 .001
Pseudozius caystrus (1) ,
Xantho sanguineus (1)
3.0 6.5 .02
Xantho gracilis (1),
Xantho leptodon (1)
34.5 7.5 .01
Xantho gracilis (1),
Xantho sanguineus (1)
21. 8 39.4 .001
Xantho leptodon (1),
Xantho sanguineus (1)
53.7 51.1 .001
Habitat types as in Table 24.
269
TABLE 25- Continued
Chi^ Data
Species and Habitat Small
No.
Rocks■ * ■
Crabs
Large
No.
Rocks'*"
Crabs
Pseudozius caystrus (1),
Xantho gracilis (1)
(
22) 10
18
(23+) 76
121
Pseudozius caystrus (2),
Xantho gracilis (2)
( 39) 13
35
(40+) 27
31
Pseudozius caystrus (1),
Xantho leptodon (l)
(
22) 10
77
(23+) 76
21
Pseudozius caystrus (1),
Xantho sanguineus (1)
(
29) 18
25
(30+) 68
37
Xantho gracilis (1),
Xantho leptodon (1)
(
22) 81
77
(23+) 121
21
Xantho gracilis (1),
Xantho sanguineus (1)
(
22) 81
13
(23+) 121
49
Xantho leptodon (1),
Xantho sanguineus (1)
(
22) 77
13
(23+) 21
49
" * " Figures in parentheses represent size
ranges of rocks (rock length) in cm.
270
Although it is true that Xantho leptodon and
Pseudozius caystrus occur mainly at different vertical
levels, at Eniwetok station 5C, where these species over
lapped at the 0.76 to 0.9m level, Xantho leptodon was found
mostly under rocks less than 30cm in length, and Pseudozius
caystrus under rocks larger than this . (although Xantho
gracilis was the most abundant form in this locality, under
rocks of all sizes).
It is not immediately obvious why Xantho leptodon and
Xantho sanguineus should be separated by rock size. Xantho
sanguineus will prey upon the other species when they are
confined together in captivity; this may explain why Xantho
leptodon does not thrive under larger rocks where its
larger cousin Xantho sanguineus occurs. It is less certain
why Xantho sanguineus does not live under small rocks which
favor Xantho leptodon; perhaps they are not stable enough
to afford cover to so large a crab.
Ecological Separations due to Looseness of Rocks
and Under-Rock Sediment in Rubble Situations
At station 12B on Japtan Island, Xantho sanguineus was
collected in abundance in fairly compact piles of rubble,
composed of rocks under 23cm in diameter with sand filling
the space underneath. The rocks making up these heaps were
firmly wedged together and provided a rather stable habitat
for the crabs. Rocks of the same size, but scattered
across the surface of a beach, or in loose piles, proved
271
to be inhabited by Xantho leptodon. An analysis of re
lative abundance, supported by Chi2 test (significant at
.001) demonstrates that these two species are separated on
the basis of the stability of their rock cover (table 26).
The type of sediment under the rocks is another factor
of importance in the distribution of crabs. Well-sorted
sediments were classified as either sand (.0625 to 2mm) or
gravel (2 to 64cm). In addition, in many cases sand and
gravel were found in a mixed situation, with sand filling
in most of the space between the pebbles. The two low
intertidal rubble crabs, Xantho leptodon and Xantho
sanguineus, appear to be equally common where the sediment
is sand or sand mixed with gravel. Likewise, they both
occur under rocks which overlie gravel or which rest
directly upon the reef rock at the foot of a beach; in all
of these situations, rock size rather than the type of
under-rock sediment controls their distribution.
In the case of Xantho gracilis and Pseudozius caystrus,
it will be remembered that the latter extends up above the
vertical range of the former; comparisons between these two
species are best made using data from vertical levels where
they both occur. An analysis of relative abundance, sup-
o
ported by a Chi test (significant at .001) demonstrates
that Xantho gracilis occurs mainly where there is sand (or
sand mixed with some gravel) under the rocks, and that
272
TABLE 26
ECOLOGICAL SEPARATION BETWEEN XANTHO
LEPTODON AND XANTHO SANGUINEUS WITH
RESPECT TO LOOSENESS OF ROCKS UNDER
23cm IN DIAMETER
Habitat Relative Abundance
----- ----
Chi Sig.
1 Xantho sanguineus
21.9% as abundant as
Xantho leptodon
47.0 .001
3 Xantho leptodon absent
2 Xantho sanguineus
21.2% as abundant as
Xantho leptodon
48 .1 .001
3 Xantho leptodon absent
Habitat Types: 1. Loose, Scattered Rocks
(Station 5C).
2. Piles of Loose Rocks
(Station 5C).
3. Piles of Tightly-wedged
Rocks (Stations 12C, 19B).
2
Chi Data
Species Hab. 1 Hab. 2 Hab. 3
Xantho leptodon 32 33 0
Xantho sanguineus 7 7 31
Number of Stations 1 1 2
273
Pseudozius caystrus is associated with gravel (table 27).
The preference of Xantho gracilis for sand, and Pseudozius
caystrus for gravel (fig. 58), may relate to the fact that
the latter has little setation around the incurrent open
ings to the branchial chambers, while the former has heavy
setation there. Alternatively, it is possible that there
is more desiccation under rocks on gravel, and that this
favors Pseudozius caystrus, which extends higher in the
intertidal than does Xantho gracilis
Comparisons of Crabs in Consolidated
and Loose Substrates
When comparisons are made between species which live
in both loose and consolidated substrates and those which
occur only in one or the other, the following patterns
emerge:
(1). In some cases, the species in question are ecologi
cally separated in reef situations, but one member of each
pair under consideration also lives in loose rubble; this
occurs when Zozymodes biunguis, Dacryopilumnus rathbunae,
and Eriphia scabricula are compared with Xantho leptodon
and Xantho sanguineus, and when Dacryopilumnus rathbunae
an(^ Lydia annulipes are compared with Pseudozius caystrus.
(2). In the case of Xantho gracilis and Pseudozius
caystrus, there is a substrate separation on beaches, but
the latter species also has a consolidated rock habitat.
In both (1) and (2) the information that one member
274
TABLE 27
ECOLOGICAL SEPARATION BETWEEN PSEUDOZIUS
CAYSTRUS AND XANTHO GRACILIS WITH RESPECT
TO TYPE OF SEDIMENT UNDER ROCKS.
Habitat Relative Abundance Chi2 Sig.
1 Xantho gracilis
27.6% as abundant as
Pseudozius caystrus
33.9 .001
2 Pseudozius caystrus
41.3% as abundant as
Xantho gracilis
Habitat Types: 1. Gravel Under Rocks.
2. Sand or Sand Mixed with
Gravel Under Rocks.
72
Chi Data
Species
Number of
Habitat 1
Crabs
Habitat 2
Pseudozius caystrus 58 19
Xantho gracilis 16 46
Number of Stations 3 4
275
Fig. 58 Substrate Preference of Pseudozius caystrus
and Xantho gracilis.
Two overturned beachrock slabs showing
under-rock sediment typical of the two
crab species.
A. Patch of gravel found under slab
with nodular fragments on under side,
favoring Pseudozius caystrus.
B. Patch of sand found under slab with
sandy consistency on under side,
favoring Xantho gracilis.
Commonly the sand and gravel habitat types are
found in different localities; in this
instance at Sand Island station 9A, it was the
sediment under slabs was sorted well enough
to allow the habitats of both species to be
photographed in close proximity.
&
276
of each pair has an additional habitat does not add much to
our knowledge of the ecological separation between the
species, as it is already known that there is a substrate
difference.
t3). There is broad overlap between Xantho gracilis and
its relatives Xantho sanguineus and Xantho leptodon on
beaches; however, the latter two species each have an add
itional habitat on reefs that affords complete separation
from Xantho gracilis (which does not live on reefs). In
this instance we have a partial ecological separation be
tween species which is probably not of great importance,
as the bulk of the populations of Xantho sanguineus and
Xantho leptodon probably occurs on beaches, where Xantho
gracilis is found.
(4). It is difficult to estimate how much substrate separ
ation there may be between the low intertidal Eriphia
scabricula and Zozymodes biunguis and the high intertidal
Pseudozius caystrus when these are compared in consolidated
rock situations; however Pseudozius caystrus also lives in
great abundance on beaches, where the former two species
are absent or uncommon. In these cases there is a partial
ecological separation due to substrate type.
(5). One further class of comparisons can be made: be
tween species restricted or nearly restricted to reef
situations, and the one species limited to beaches, Xantho
277
gracilis. Zozymodes biunguis is relatively uncommon on
beaches, being only 18.2 percent as abundant as the larger
Xantho leptodon at station 5C, where an entire strip of
beach was examined for crabs. It is difficult to compare
Zozymodes biunguis with Xantho gracilis, due to the verti
cal zonation difference; however it may be noted that the
former was only 8.9 percent as abundant as Xantho gracilis
at station 5C. The relative rarity of Zozymodes biunguis
on beaches indicates that it is separated from Xantho
gracilis by substrate as well as by vertical zonation.
Lydia annulipes is very seldom found in the same
substrate as Xantho gracilis; the latter almost never
occurs in holes in high intertidal conglomerate, where
Lydia annulipes is the dominant crab. Under slabs on con
glomerate flats, Xantho gracilis is only 13 percent as
abundant as Lydia annulipes (and only 3.2 percent as abun
dant as Pseudozius caystrus, the dominant species in that
habitat). On the other hand Lydia annulipes is rare on
beaches, where it is associated with Pseudozius caystrus
(being about 4 percent as common as that species) rather
than Xantho gracilis. Hence, Lydia annulipes and Xantho
gracilis can be considered to be separated by substrate
type.
Eriphia scabricula and Dacryopilumnus rathbunae never
occur in loose rubble, while Xantho gracilis almost never
278
occurs on reefs; these species are clearly separated in
terras of substrate type. This can be documented statistic-
ally using a Chi test (significant at .001) (table 28).
Conclusions
The behavior and morphology of intertidal xanthid
crabs has been found to be closely related to the type of
substrate in which the animals live. In general these are
cryptic animals, and much of the escape behavior is con
cerned with their ability to hide in holes or under rocks.
Carapace size and shape are particularly influenced by the
type of hiding place preferred.
In consolidated rock situations, there are five
distinct kinds of crab hatitats with respect to substrate
in the low intertidal: those of Zozymodes biunguis, Xantho
leptodon, Xantho sanguineus, Dacryopilumnus rathbunae, and
Eriphia scabricula. In the high intertidal there are two
demonstrable habitat types, those of Lydia annulipes and
Pseudozius caystrus. In loose rubble, there are two kinds
of habitats with respect to substrate in the low intertidal
those of Xantho leptodon and Xantho sanguineus; and two
kinds in the high intertidal, those of Xantho gracilis and
Pseudozius caystrus. Taking loose and consolidated sub
strates together, five species prefer distinctly different
habitats in the low intertidal, and three in the high in
tertidal. Substrate data are summarized in table 29.
279
TABLE 28
ECOLOGICAL SEPARATION OF XANTHO GRACILIS
FROM DACRYOPILUMNUS RATHBUNAE AND ERIPHIA
SCABRICULA WITH RESPECT TO SUBSTRATE TYPE
Species and Habitats
Compared Co-occurrence
. 2
Chi Sig.
Dacryopilumnus
rathbunae (1),
Xantho gracilis (2)
None 168 .001
Eriphia
scabricula (1),
Xantho gracilis (2)
None 231 .001
Habitat Types: 1. Reef Flats, Eroding
Algal Ridge Remnants,
Conglomerate, Beachrock,
and Reef Boulders.
2. Rubble Beaches and Flats.
. 2
Chi Data
Species Number of Crabs
Habitat 1 Habitat 2
Number of
Stations
Dacryopilumnus
rathbunae
92 0 11
Eriphia scabricula 50 0 6
Xantho gracilis 0 139 5
280
TABLE 29
SUMMARY OF SUBSTRATE DATA FOR THE EIGHT STUDY SPECIES
Species Substrate
Dacryopilumnus
rathbunae
Eroding Algal Rock, Conglomerate,
Beachrock Rims, Reef Blocks and
Boulders; Convex Rock Surfaces with
an Algal Film 1mm or Less in Thickness.
In Sipunculid Borings Whose
Aperture Size (Height + Width) is
Mainly Under 17mm, and Whose Aperture
Height is Mainly Over 70% of the
Aperture Width.
Eriphia
scabricula
Reef Flats without a Sand-binding
Algal Turf and without Ledges, in
Holes Whose Aperture Size (Height +
Width) is Mainly Over 17mm.
Eroding Algal Rock, Conglomerate
and Beachrock, in Holes Whose Aperture
Size (Heightt Width) is Mainly Over
17mm, and Whose Aperture Height is
Mainly Under 70% of the Aperture Width.
Lydia
annulipes
Eroding Algal Rock, Conglomerate, and
Beachrock, in Holes whose Aperture
Size (Height + Width) is Mainly Over
17mm.
Less Abundant Under Cemented Slabs of
Conglomerate and Beach Conglomerate,
in Crevices Filled with Gravel.
Pseudozius
caystrus
Rubble Beaches and Flats, Under Rocks
Whose Length is Mainly Over 23cm, or
Under Loose Beachrock Slabs.
Especially Under Rocks Over Gravel;
Less Common Where There is Sand or
Gravelly Sand Under Rocks.
Under Cemented Slabs or Conglomerate
and Beach Conglomerate, in Crevices
Filled with Gravel.
281
TABLE 29- Continued
Species Substrate
Xantho
gracilis
Rubble Beaches and Flats, Under Rocks
Whose Length is Over as Well as Under
23cm, or Under Loose Beachrock Slabs,
Mainly Where There is Sand or Gravelly
Sand Under Rocks.
Xantho
leptodon
Rubble Beaches and Flats, Under Rocks
Whose Length is Mainly Under 23cm.
Reef Flats with a Thick (up to 40cm)
Sand-binding Algal Turf or a Floccu-
lent Coating of Sand (Loosely Bound
by Fine Filamentous Algae); in Holes
Whose Aperture Size (Height + Width)
is Mainly Over 17mm.
Xantho
sanguineus
Rubble Beaches and Flats, Under Rocks
Whose Length is Mainly Over 23cm,
or Under Loose Beachrock Slabs; Also
Under 23cm in Length, with Encrusting
Filamentous Algae and Worm Tubes, and
Sand.
Reef Flats with 15-30cm Long Over
hanging Ledges (in Large Holes);
and Under Rocks on Reefs with Gravel
and Rocks up to 21.5cm long.
Zozymodes
biunguis
Reef Flats, in Holes Whose Aperture
Size (Height + Width) is Mainly Under
17cm.
Eroding Algal Rock, Conglomerate, and
Beachrock; Concave Rock Surfaces with
a Sandy Algal Turf (2mm or More in
Thickness), in Holes Whose Aperture
Size (Height + Width) is Mainly Under
14cm.
Occasionally Found Under Slabs of
Cemented Beachrock, in Low Crevices
without Gravel.
CHAPTER VI
FEEDING HABITS AND BEHAVIOR
Introduction
Field observations revealed a number of differences in
the habits of the various species, and further information
was gained through studies of crab stomach contents. The
kind of food eaten by these animals, and when and where it
is eaten, are doubtless very important aspects of their
ecology. However, it is much more difficult to assess
quantitatively the role that behavioral differences may
play in determining ecological separations between species
than it is to establish where each form occurs on the reef,
for reasons which will be explained later. It is neverthe
less important to include data on behavioral differences in
a study of this kind, especially when comparisons are made
between species which live in close proximity on the reef
(e.cf. those which live in the same area but occupy holes of
different sizes) and which are likely to compete for food.
Observations on Temporal and Spatial
Differences in Activity
A number of observations were made at high and at low
tide, night and day in order to determine when the various
282
283
study species are active, and where they carry out this ac
tivity in relation to their hiding places. Observational
methods varied considerably, depending upon the size of the
crabs, the locality, time of day, and stage of the tide.
For the smaller and less active xanthids living on reefs
and beaches, 3-minute observations on l/4m^ quadrats were
deemed sufficient. However, for day-active crabs which
come out of the water, it was necessary to watch with bi
noculars at a distance of 6 or 7m for periods of up to 25
minutes; such observations were made on beachrock, algal
mounds, and reef flats. At high tide, an underwater view
ing glass was used in the shallow waters covering reef
flats and beaches. A flashlight was employed at night,
when the larger and more wary crabs could be readily ap
proached; nocturnal counts were made of the number of crabs
seen during a traverse across the reef flat or on a given
stretch of beach rock or an algal mound.
Unfortunately, a uniform observational program could
not be employed, as field conditions at night and day and
at high and low tide were so different that it was imprac
tical to attempt to standardize the methodology. At high
tide, and especially at night, it was often difficult to
see anything, because of wave surge or turbulent water.
On the reef flat at Eniwetok station 4, Zozymodes bi
unguis was nocturnally active; 243 crabs were counted dur
ing seventy 3-minute observations at low tide at night,
284
compared with only 3 individuals in sixty-eight observa
tions at low tide during the day, and 4 in sixty-five ob
servations at high tide during the day. This species ap
peared to be much less active during high tide than during
low tide at night, though fewer observations were made at
this time. The behavior of this xanthid is in striking
contrast to that of the small grapsid crabs (Pachygrapsus
minutus and P. plicatus) seen in the same quadrats, which
readily came out at low tide, day or night. Zozymodes bi
unguis is relatively sedentary; it occasionally wanders
across the tide pools during low tide at night, but at
other times will seldom come more than part of the way
out of its hole.
Fewer data are available on the two Xantho species
that live on the inner reef at station 4. Although their
behavior appeared to be similar to that of Zozymodes biun
guis , they were more often seen during daytime low tides
than was £. biunguis, although the latter was known to be
more abundant in the same localities.
Eriphia scabricula is very active in the daytime at
low tide. On the outer reef at station 4, numerous indivi
duals of this species could be observed as they came out
to feed between successive waves; they were also active on
outer parts of the reef (associated with Eriphia sebana in
the near-shore zone). At night they were active on the in
ner reef and in the Goniolithon zone, but not on the outer
285
reef, over which the waves washed even on a very low tide.
At high tide, they were never seen in the open on the outer
reef flat, night or day; though conditions were far from
optimal for viewing these crabs at high tide, other forms
(the swimming crab Thalamita sp. and the majid Micippa sp.)
were seen at this time. On quiet parts of the reef, Eri-
phia scabricula could be observed wandering many meters
from their holes; they are decidedly amphibious, going in
to tide pools or walking across the drying flat. When in
pools, the crabs would often keep both eyes above the sur
face like periscopes, as if to watch for danger from above.
On the outer reef, they were seldom observed to go under
water or to wander far from their holes, owing to the wave
action.
On algal mounds at Eniwetok station 2, Dacryopilumnus
rathbunae and Eriphia scabricula both proved to be very
active at low tide in the daytime, and much less so at
night, probably because of the swells which pass over the
lower zone where they live. Lydia annulipes, on the other
hand, was very active during low tide at night, but infre
quently observed out in the open in the daytime.
On lagoon beachrock at Eniwetok station 7, Eriphia
scabricula was observed to be active during the day, and
Lydia annulipes active at night, with few exceptions. A
number of E. scabricula were seen inside their crevices at
low tide at night in this locality; they were probably so
286
restricted by the waves which sweep over the lower zone.
In the daytime, wave surge did not deter them, as they were
able to come out to feed between the waves. Observations
during high tide were difficult to make on the slick, slop
ing surface of this beachrock, and rather dangerous at
night. However, the few observations which were made at
this time did not reveal any crabs out in the open.
On higher conglomerate, Lydia annulipes was often seen
wandering about on the dry rock surface during low tide at
night in a number of localities (especially the rock groin
at Uliga Island, station 47). Pseudozius caystrus in the
same localities would never come out from beneath the
ledges or slabs they inhabited, though they could be obser
ved out of water near the edges of their crevices at night.
A number of attempts were made to observe the behavior
of crabs living on rubble beaches at the north end of Eni
wetok Island (stations 5 and 6); but crabs were not seen
out in the open, day or night, either at low tide when the
beaches are dry or at high tide when there is a certain
amount of wave surge. However, all three species of Xantho
as well as Pseudozius caystrus occurred under rocks in the
same quadrats where the observations were made. The small
grapsid Pachygrapsus planifrons was by contrast very active
on the surface of the rocks at low tide, day or night.
In the quiet water of cut-off pools in the rubble,
Xantho leptodon and X. sanguineus were sometimes seen feed
287
ing out in the open at night, during low tide. It is not
certain when the various Xantho species are most active on
beaches, as their activities are well hidden by their cryp
tic habits. In the aquarium, these crabs will come out
from under cover and sit on top of the substratum; they re
spond by digging when the water is drained off. This sug
gests that they are better adapted for activity at high
tide, when their escape behavior under field conditions is
much more vigorous. Pseudozius caystrus in the aquarium,
on the other hand, always remain under rocks or around
their bases, whether there is water in the tank or not; in
this regard its behavior seems to be somewhat different
from that of the other rubble crabs.
Some additional information on crab activity can be
inferred from stomach-content analyses, by examining crabs
collected at different times of day or at different stages
of the tide. The most interesting data obtained in this
way relate to Lydia annulipes and Pseudozius caystrus,
which occur together under conglomerate slabs in the high
intertidal. Of 59 Pseudozius caystrus collected at low
tide in the daytime from this habitat, 51 (86.5 percent)
had food in their stomachs; by contrast, only 5 of 22 Lydia
annulipes (22.7 percent) taken at the same times and places
had food in their stomachs. On the other hand, both spe
cies apparently feed at night; 60 of 80 L. annulipes (69
percent) collected in a variety of habitats at night had
288
food in their stomachs, while all of 35 P. caystrus (100
percent) collected from rubble beaches at night contained
food. This information suggests that Lydia annulipes feeds
infrequently during the daytime when it remains under cov
er, in contrast to Pseudozius caystrus which feeds under
the same slabs. Unless Lydia annulipes feeds under slabs
as well as out in the open at night, this indicates that
these species feed in different places.
Ecological Separations due to
Behavioral Differences
Crabs which feed both during the day and at night must
have some advantage over potential competitors which feed
only at night. Hence, Eriphia scabricula may have an ad
vantage over Zozymodes biunguis, and Pseudozius caystrus an
advantage over Lydia annulipes where these forms occur to
gether provided they feed upon the same foods. It should
be noted, however, that this behavioral difference does not
allow, in the first case cited above, co-occurrence in the
same habitat unless both species have holes of adequate
size for shelter.
In all instances, the nocturnal species are completely
overlapped behaviorally by species which are also active
during the day. The only exceptions occur where waves in
hibit the nightly activity of day-active species in the low
intertidal, leaving nocturnal, high intertidal crabs unaf
fected. This behavioral difference is not observed in
289
sheltered locations.
In general, there is little evidence that nocturnal-
diurnal activity patterns are the primary factors involved
in the ecological separation of any of the study species.
Similarly, there is little reason to believe that high
tide-low tide differences are of great importance. There
is, however, reason to believe that there is a spatial dif
ference in the feeding habits of Lydia annulipes and Pseu
dozius caystrus which may be important where they occur in
the same substratum.
Observations on Feeding Habits
Four of the species under consideration have claws
with moderately long fingers and spatulate tips, an adapta
tion for scraping algae from the rocks. Of these species,
Zozymodes biunguis was often observed feeding on the reef
flat at night. The spoon-like fingertips are actually used
in two ways, depending upon the type of algal material
growing near the crabs' holes. Where the rock is covered
by a thin, yellowish film of algae (mainly blue-greens or
Cyanophyta) the claws are used to scrape the surface by
holding the tips against the rock and bringing them togeth
er. On the other hand, where there is a thicker, macro
scopic sandy turf of algae (e.£. Jania, Polysiphon!a) the
claw tips are used to grip bunches of algal filaments and
pull them away from the rock.
While this feeding behavior is sometimes carried out
in the open, many of the crabs tend to remain partly inside
their holes, reaching out with one claw or both claws to
pick at the adjacent rock surface. They may assume any
posture while feeding, including an upside-down position
when feeding on algae overhead. The fact that so many of
these small xanthids feed in the area immediately around
the apertures of their holes probably explains why the rock
surface there often has a bare appearance (as opposed to
the effect of fish grazing, which affects very large areas
of the reef). On the narrow intertidal reef flat at the
foot of the conglomerate wall at Rigili Island (station 18)
holes inhabited by Zozymodes biunguis were often surrounded
by encrusting coralline algae; it is probable that the con
tinual scraping activity of these crabs allows the coral
lines to become established.
Xantho leptodon and X. sanguineus on the reef flat
have been observed to feed in much the same way as Zozy
modes biunguis. In rubble areas, the three Xantho species
are so secretive in their habits that it proved necessary
to make observations on captive animals in laboratory aqua
ria. Under these conditions, Xantho leptodon and X. graci
lis could be observed scraping the surface of the rocks
with their claws to remove the film of algae growing there.
They were also seen eating the algal coating on small peb
bles, by holding the latter to their mouths and rotating
them with their legs (or occasionally picking at them with
291
their claws). The pebbles were often carried along by the
crabs as they retreated beneath rocks, which may help ex
plain how these herbivorous crabs can feed while remaining
under cover in nature. They were also observed to pick up
minute bits of food from the sand, and to feed upon mater
ial obtained by cleaning growths from their claws.
It should be noted that much predation, or cannibalism
as the case may be, was observed in aquaria where Xantho
spp. were confined together; and Xantho sanguineus was ob
served preying upon the diminutive Zozymodes biunguis on
the reef flat at night. In the laboratory, Xantho leptodon
and X. gracilis vigorously competed with one another for
dead, immature manini (Acanthurus triostegas triostegas L.;
a surgeonfish, whose young are common in tide pools) which
were dropped in the tank; the crabs would apprehend this
food with their claws and draw it to their mouthparts, or
scrape the skin of the fish with their clawtips and pass
whatever was obtained to their mouths. Hence, despite the
special claw modification which adapts these animals so
well to an algal diet, they are opportunistic and by no
means restricted to plant matter.
Dacryopilumnus rathbunae has short fingers, which
while not actually spatulate are rather blunt and curved,
are used in feeding on algae in much the same way as those
of Zozymodes. DacryOpilumnus rathbunae were often obser
ved sitting halfway or completely out of their holes,
292
scraping the rock surface in the vicinity.
Eriphia scabricula was commonly observed to feed on
algae; as its claws have pointed tips and lack any special
modification, it is capable of taking only a thicker turf
of algae, pulling the material away from the rock. On the
outer reef at Eniwetok station 4, this species emerged be
tween successive waves to feed voraciously on the algae
which was picked with rapid movements of one or both claws.
Sometimes the algae would be gripped with both claws while
the crab would rock back and forth, probably a means of
prying the filaments from the rock. This energetic feeding
behavior may be due to the fact that there is relatively
little time available to the crabs for feeding on this
wave-swept reef, which is exposed only by low spring tides.
By contrast, the same species on sheltered reef flats at
low tide could be observed wandering freely across the reef,
feeding frequently.
Eriphia scabricula were commonly observed probing the
algal turf with their claws, or with the dactyls of the
legs on one side which were moved rapidly in succession
(much like a person tapping his fingers on a table top).
It is not known whether these movements were attempts to
locate small animals in the algae, but the crabs were often
seen grabbing at nearly invisible objects in the turf.
They would also lunge forward (both claws extended) at
other animals in tide pools (in one case a swimming crab,
293
Thalamita sp. was identified as the object of the attack);
and were observed eating smaller crabs (of the same or
other species) in the field and in the laboratory.
Lydia annulipes exhibits a peculiar claw dimorphism:
it has one massive claw with blunt teeth and one slender
claw with sharp, recurved teeth. This species has been ob
served using its small claw to feed on a thick, green
algal turf, or to probe around in the algae (perhaps hunt
ing for small animals). One individual was seen pulling a
sipunculid from a piece of coral rock left outside the
marine laboratory, using its small claw to drag the worm
from its hole. Another crab was observed in the process of
prying a Siphonaria (limpet-like gastropod) from the sur
face of an algal mound with its large claw; and on beach-
rock, others were seen grappling with small black mussels
(which live in cracks in the rock) or attempting to eat the
snail Nerita plicata. Like many of the other species,
Lydia annulipes is cannibalistic. In a grotesque display
of this behavior, one crab was observed eating alive an
other of the same species under a rock; the victim had lost
most of its legs and the fingers of one claw, which the
other crab was in the process of eating when discovered.
No observations on the feeding behavior of Pseudozius
caystrus are available; however, judging from the claw
morphology, it should be capable of feeding upon the same
items as Eriphia scabricula.
294
Stomach-Content Analyses
In order to better understand the feeding habits of
individual species, analyses were made of the stomach con
tents of formalin-preserved crabs. As these animals break
up their food with their mouthparts prior to ingestion into
the foregut, identification of some of the food items, es
pecially larger animals, presented problems. It was very
difficult to judge the relative quantities of some compo
nents of the crabs' diet, especially when the material was
broken down to a barely recognizable condition. For this
reason the only non-arbitrary measure that could be em
ployed to indicate the relative importance of different
food items was the number of crabs of a given species with
a certain item in their stomachs, expressed as a percentage
of the total number with any recognizable food in their
guts.
A great variety of algae is eaten by xanthid crabs.
While it did not prove practical to carry out detailed
identifications of algal material, the most commonly-
occurring algae can be placed into several categories.
Blue-green algae (such as Lyngbya) commonly appear in the
foreguts of crabs as greenish tubules from lj* to 10^>/ in
diameter, although a thicker yellow form 22 yvin diameter
are not uncommon in the stomachs of high intertidal crabs.
The fine, branching coralline alga Jania is also of con
siderable importance, appearing in the form of disarticu-
295
lated segments. Polysiphonia and similar algae occur as
intact, brown filaments; chunks of a branching filamentous
alga similar to Laurencia also occur, as do segments of
Ceratocentrum (or a close relative).
In many cases these algae are intact, but algal mater
ial is also present in a semi-digested condition. Blue-
greens and Polysiphonia are clear in color after break-down
of the cells, and masses of round or oblong objects, 2 to
in diameter are often found clustered around the larger
algal cells. Jania in particular appears to break down
into masses of these objects. Often the stomachs of the
more herbivorous crabs Zozymodes biunguis, Xantho spp.,
Dacryopilumnus rathbunae are filled with a light green
mass of this material, with occasional cells still recog
nizable .
Foraminifers of various kinds are often found intact,
although foraminifer fragments are also found. Small crus
taceans, insects, and mites are commonly intact or semi
intact; larger crustaceans, however, are evidently thor
oughly macerated by the crabs, leaving only the more resis
tant parts: setae, eyes, the dactyls of the walking legs
and the tips of the claws (when crabs have been eaten).
Polychaetes seldom occur in a whole condition; usually only
jaws and setae remain. Sipunculids can be recognized as
tubular objects with rows of hook-like brown teeth (part of
the introvert), though occasional intact sipunculids are
296
found. Bodies of the small, limpet-like gastropod Sipho-
naria appear as fleshy, greenish lumps, and occasionally
brown- and white-banded shell fragments are found. Unfor
tunately, some of the objects found in the crab stomachs,
especially certain shell fragments, cannot be positively
identified.
Despite the problems of identification and quantifica
tion of materials found in the digestive tracts of the
crabs, it is possible to obtain a fairly good idea of what
the various species feed upon. Table 30 gives some indi
cation of the relative importance of the algae most often
found in the stomachs of the eight species; data for crabs
collected in different localities are given separately to
show how the type of algae eaten varies from one place to
another. While for some localities the number of crabs
with food in their stomachs was small, it is felt that it
is better to present this information separately than to
lump data from different areas without respect to habitat.
It is apparent that the five major kinds of algae listed
in Table 30 are eaten by some crabs in all major habitat
types. Table 31 depicts the various kinds of animal foods
occurring in the stomachs of these xanthids. Where
practical, this information is broken down according to
habitat type. There is a certain relation between the type
of food eaten and the habitat, regardless of crab species.
TABLE 30
PERCENTAGE OF CRABS CONTAINING VARIOUS KINDS OF ALGAE
Species
Habitat and
Station
Blue-
greens Jania
Polysi
phonia1
Lauren-
cia1
Cerato-
centrum
Number of
Crabs2
Dacryopilumnus
rathbunae
Algal Rock,
Conglomerate,
Beachrock
65.3 1.4 6.9 22.2 72
Eriphia
scabricula
Reef Flats,
Sta. 3,4
68.2 40.9 50.0 22
11
Reef Flats,
Kwajalein and
Majuro
35 13.6 30.0 26.7 20.0 20
I I
Algal Ridge,
Sta. 2
57.2 14.3 28.6 21.4 14
I I
Beachrock,
Sta. 7
44. 8 25.0 75.0 41.7 16.7 24
Lydia
annulipes
Algal Ridge,
Sta. 2, Beach
rock, Sta. 7
16.1 3.2 31
1 1
Conglomerate 23.1 2.6 39
land related forms.
2with food in their stomachs.
NJ
VO
TABLE 30 - Continued
Species
Habitat and
Station
Blue-
greens Jania
Polysi
phonia
Lauren-
cia
Cerato-
centrum
Number of
Crabs
Pseudozius
caystrus
Conglomerate 39.4 18.2 48.5 9.1 33
It
Rubble Beaches 40.7 25.0 50.0 15.6 7.8 64
Xantho
gracilis
Rubble Beach,
Sta. 5A
92.3 46.2 19.2 26
II
Rubble Flat,
Sta. 21
100. 0 17
Xantho
leptodon
Reef Flat,
Sta. 4
20.0 70.0 18
It
Rubble Beach,
Pool, Sta. 5A
94.1 41.2 17
Xantho
sanguineus
Reef Flat,
Sta. 48A
30.8 23.1 15.4 15.4 13
11
Under Rocks
on Reef, Sta.
31, 46
27.8 11.1 42.1 18
II
Beachrock
Slabs, Sta.
14B
36.4 36.4 18.2 11
298
TABLE 30 - Continued
Species Station
Blue-
greens Jania
Polysi
phonia
Lauren-
cia
Cerato-
centrum
Number of
Crabs
Xantho
sanguineus
Rubble Flat,
Beach, Sta. 12B 84.5 69.3 38.5
46.2 13
Zozymodes
biunguis
Smooth, Or
ange Reef
Flat, Sta. 2,
3
67.7 6.5 12.9 6.5 31
II
Tan, Turf-
covered Reef
Flat, Sta. 3
27.8 16.7 63.9 25.0 36
II
Orange Pitted
Zone, Sta. 5
83.3 29.4 23.5 29.4 5.9 17
299
TABLE 31
PERCENTAGE OF CRABS CONTAINING VARIOUS KINDS OF ANIMAL
FOODS, COMPARED WITH TOTAL ALGAE
Species Habitat
Total
Algae
Foramin-
ifers
Poly-
chaetes
Sipun
culids
Siphon-
ana
Dacryopilumnus Algal Rock, Conglom 92.6 1.3 1.3
rathbunae erate, Beachrock
Eriphia
scabricula
Reef Flats 86.0 9.3 18.6 18.6
( 1
Algal Rock, Beachrock 71.5 2.0 10.2 16.2 2.0
Lydia
annulipes
Algal Rock,
Exposed Beachrock
33.4 3.2 48.4
I I
Conglomerate, Shel
tered Beachrock
30.4 7.7 10.3 58.0 2.6
Pseudozius
caystrus
Conglomerate 82.2 33.4 6.7 4.4
I I
Rubble Beaches, Flats 74.4 32.9 20.7 3.7
Xantho
gracilis
Rubble Beaches, Flats 98.8 3.7
Xantho
leptodon
Reef Flats 97.1 17.7
300
TABLE 31 - Continued
Species Habitat
Total
Algae
Foramin- Poly- Sipun- Siphon-
ifers chaetes culids aria
Xantho Rubble Beaches, Flats 98.4 1.7
leptodon
Xantho Rubble Beaches, Flats 96.7 6.5 2.2
sanguineus (mostly); Reef Flats
Zozymodes Reef Flats 100.0 8.0 0.9
biunguis
Small
Crus
tacea
Large
Crus
tacea
Insect
Larvae
Insect
Adults Mites
No.
Crabs
Dacryopilumnus
rathbunae
Algal Rock, Conglom
erate, Beachrock 11.8 1.3 1.3 7.9 76
Eriphia
scabricula
Reef Flats 25.6 9.3 7.0 4.6 43
I I
Algal Rock, Beachrock 38.8 2.0 16.3 8.2 2.0 49
Lydia
annulipes
Algal Rock,
Exposed Beachrock
16.1 6.4 6.4 31
I I
Conglomerate, Shel
tered Beachrock
2.6 5.1 5.1 12.8 5.1 39
301
TABLE 31 - Continued
Species Habitat
Small
Crus
tacea
Large
Crus
tacea
Insect
Larvae
Insect
Adults Mites
No.
Crabs1
Pseudozius
caystrus
Conglomerate 2.2 17.8 45
I I
Rubble Beaches, Flats 20.8 1.2 82
Xantho
gracilis
Rubble Beaches, Flats 6.1 1.2 82
Xantho
leptodon
Reef Flats 5.8 2.9 34
1 1
Rubble Beaches, Flats 1.7 3.3 60
Xantho
sanguineus
Rubble Beaches,
(mostly): Reef
Flats,
Flats 17.4 92
Zozymodes
biunguis
Reef Flats 2.7 112
■^with food in their stomachs.
302
303
Feeding Habits of the Individual Species,
Based on Stomach Contents
Stomach content analyses allow the following conclu
sions to be drawn concerning the type of food eaten by the
eight species:
Zozymodes biunguis is mainly herbivorous, although oc
casionally it eats foraminifers (which may be fed upon
accidentally, as they never constitute a significant pro
portion of the food found in the crabs). The data presen
ted on Table 30 indicate that crabs living on slick, orange-
colored reef flats feed largely upon blue-green algae which
form a coating on the rock, while those living in turf-
covered areas eat larger filamentous algae, such as Polysi
phonia . The blue-greens eaten vary considerably in size,
from 2 to 14y in diameter; the stomachs of crabs from a
given locality often contain a variety of these algae.
Xantho leptodon feeds upon a variety of algae, depen
ding upon its habitat. Crabs from the inner reef at sta
tion 4, like Zozymodes biunguis, had eaten largely Polysi
phonia and some foraminifera. At Dalap Island (station
48A), X. leptodon from the pool behind the inner, seaward
reef had eaten mainly Jania, whose disarticulated segments
comprised most of the sand in the pool; this algae appar
ently washed in from the seaward reef flat in this locality.
Crabs from certain beaches and flats (e.£. Eniwetok Island
station 5) had consumed mainly tiny blue-green algae whose
304
filaments varied from 2 to 8yv in diameter, and occasionally
other algae.
Xantho gracilis is most similar in feeding habits to
X. leptodon which live on beaches, eating the same kinds of
blue-green algae. Filamentous algae of other groups are
also eaten in some localities, however.
Xantho sanguineus appears to eat mainly larger kinds
of filamentous algae, such as Jania, Polysiphonia, and
and Ceratocentrum, in many localities. Blue-green algae,
8yw in diameter, are also eaten in some places. Crab re
mains are more often found in this species than in its
smaller relatives, X. leptodon and X. gracilis.
Dacryopilumnus rathbunae eats mainly filamentous blue-
green algae, from 2 to 14^ in diameter; some of these are
in the form of minute colonies, with the filaments radiat
ing from a common center. It also feeds upon other algae.
In keeping with the small size of this species, the com
monest animal foods taken are mites and small crustaceans,
which are never as important in terms of volume of food
eaten than the algae, when both are found together.
Eriphia scabricula feeds upon a variety of plant and
animal foods; the alga most commonly eaten by crabs on the
reef flat at Eniwetok station 4 was Jania, while on algal
mounds at station 2 blue-green algae, 4 to 16/v in diameter,
were taken. Crabs from beachrock at station 7 were found
to have eaten a great variety of algae. Small Crustacea
305
are more often found in their stomachs than larger ones,
although this species is undoubtedly capable of killing
small crabs; in one case, crab remains identifiable as
those of Zozymodes biunguis were found. Insect larvae,
adult insects, polychaetes, and sipunculids are other com
monly eaten animal foods.
Pseudozius caystrus feeds upon various algae as well
as animals. The algae include Polysiphonia, Jania, and
blue-green algae; the latter include finer types from 4 to
8^ in diameter, and a larger yellow form from 22 to 26yu/in
diameter. It is probable that some of the algal material
(Jania) washes in from the reef flat and becomes lodged
under the rocks where P. caystrus hides; however, some of
the blue-green algae at least occur at the same level as
the crabs. Remains of larger crustaceans often occur in
the stomachs of these animals; sometimes these could be
identified as those of Pachygrapsus sp. and in one case
they belonged to a small Pseudozius caystrus. Polychaete
remains also commonly are present in the stomachs of crabs
from rubble beaches. Foraminifers occur more often in this
species than any other; as many as 11 foraminifers were
observed in the otherwise empty digestive tracts of some
individuals.
Lydia annulipes eats chiefly animals of various kinds.
Crabs collected on the algal mounds at Eniwetok station 2
were found to have eaten mainly Siphonaria, although in
various conglomerate areas this species feeds mainly upon
sipunculids. Other animal foods such as polychaetes, in
sects, or foraminifera are also taken, but seldom in great
quantities (except in the case of one individual crab which
had eaten 75 foraminifers, which half-filled the stomach).
Algae are of less importance in the diet of this crab than
they are to the seven preceding species; however, the sto
machs of four or five L. annulipes from the highest part of
the rock groin at Uliga Island (station 47) were full of a
yellow-colored, blue-green form, 22yw in diameter, and a
finer green alga, 10y in diameter, which form a thick turf
that blackens much of the rock surface. By comparison,
crabs from lower parts of the groin had eaten mainly sipun
culids. Hence, in some cases even this species will feed
heavily upon algae; in fact, more crabs were found to have
eaten algae alone than had consumed algae in addition to
some animal food, which may indicate that it becomes her-
biverous when its preferred food is scarce.
Categories of Feeding Behavior
Subjectively, these crabs can be divided into three
groups with regard to feeding habits. Herbivorous species
can be defined as those in which at least 50 percent of the
individuals examined have eaten algae and less than 50 per
cent have eaten animal foods; omnivorous crabs are those
which more than 50 percent have eaten algae and over 50
percent have eaten animal foods; and carnivorous crabs
307
are those in which less than 50 percent have eaten algae
and more than 50 percent have eaten animal foods. Hence,
Zozymodes biunguis, Xantho leptodon, X. gracilis, X. san
guineus , and Dacryopilumnus rathbunae can be classed as
herbivores; Eriphia scabricula and Pseudozius caystrus as
omnivores; and Lydia annulipes as a carnivore (Table 32).
Claw morphology correlates very well with these groupings:
crabs of the first group have relatively spatulate claws,
adapting them to an algae diet; those of the second have
claws of generalized morphology, enabling them to deal with
a variety of foods. In Lydia annulipes the large claw
probably functions in crushing mollusks, while the small
claw is evidently used for capturing sipunculids (or occa
sionally other foods).
The above scheme, while providing us with a useful
approximation of the trophic functions of individual spe
cies , does not demonstrate ecological separation according
to the statistical criteria used in previous chapters. In
the case of studies on feeding habits, unlike habitat com
parisons where each crab of a given species lives in one
or another habitat, we are faced with the problem that
crabs are not limited to one food or another, but rather a
mixture of several food types is often found in the stomach
of one individual. It is, therefore, necessary to make
comparisons between species using individual food items or
classes of food items. Owing to the great variety of foods
TABLE 32
PERCENTAGE OF CRABS CONTAINING ALGAE AND ANIMAL FOODS
Species
Algae
Only
Algae and
Animals
Animals
Only
Total..
Algae
Total
Animals
Siphonaria and/
or Sipunculids
No.
CrabsJ
Dacryopilumnus
rathbunae
67.9 24.7 7.4 92.6 32.1 1.3 76
Eriphia
scabricula
28.2 50.0 21.8 78.2 71.8 8.5 92
Lydia
annulipes
15.9 11.6 72.5 27.5 84.2 56.5 70
Pseudozius
caystrus
26.8 63.0 10.2 89. 8 73.2 3.9 127
Xantho
gracilis
89.0 9.8 1.2 98.8 11.0 0 82
Xantho
leptodon
85.5 13.1 1.4 98.6 14.5 0 84
^Total Algae = Algae Only + Algae and Animals.
O
Total Animals = Algae and Animals + Animals Only.
3
with food in their stomachs
308
TABLE 32 - Continued
Species
Algae
Only
Algae and
Animals
Animals
Only
Total
Algae
Total
Animals
Siphonaria and/
or Sipunculids
No.
Crabs
Xantho
sanguxneus
70.5 25.0 4.5 95.5 29.5 0 92
Zozymodes
biunguis
89.1 10.9 0 100.0 10.9 0 112
309
310
eaten by the crabs, it is most practical in this case to
compare each study species with every other species twice,
for total algae and for total animal food (Table 3 3). In
each species-pair comparison, a percentage of separation
was calculated by subtracting the percentage of individuals
of the species which had eaten the food item in question
less frequently from the percentage of individuals of the
species which had eaten it more frequently. In cases where
more than 50 percent separation was found (this being the
cut-off point for valid ecological separation), Chi2 values
were calculated (all were found to be significant at .001).
. o
In the Chi analyses, presence or absence of a given type
of food was compared for each pair of species.
The results of this analysis are as follows. In some
instances, an unequivocal separation in feeding habits can
be established, where there is greater than 50% separation
for both algae and animal foods. Thus, it can be shown
that Lydia annulipes is predominately carnivorous, and that
Zozymodes biunguis, Xantho leptodon, X. gracilis, X. san
guineus , and Dacryopilumnus rathbunae are predominately
herbivorous. In other instances, only a partial separation
can be established (over 50 percent separation for algae or
for animal food, but not both). In this way, Zozymodes
biunguis, Xantho leptodon, and X. gracilis, which are her
bivorous can be distinguished from Eriphia scabricula and
Pseudozius caystrus which are omnivorous; and the latter
311
TABLE 33
ECOLOGICAL SEPARATIONS WITH RESPECT TO
DIFFERENCES IN UTILIZATION OF ALGAE AND ANIMAL FOODS
Separation Due to
Utilization of
Algae
Separation Due to
Utilization of
Animal Foods
Species
Compared
Percent of
Separation Chi^ Sig.
Percent of
Separation Chi2 Sig.
Dacryopilumnus
rathbunae,
Eriphia
scabricula
Dacryopilumnus
rathbunae,
Lydia annulipes
Dacryopilumnus
rathbunae,
Pseudozius
caystrus
Dacryopilumnus
rathbunae,
Xantho gracilis
Dacryopilumnus
rathbunae,
Xantho leptodon
Dacryopilumnus
rathbunae,
Xantho
sanguineus
Dacryopilumnus
rathbunae,
Zozymodes
biunguis
Eriphia
scabricula,
Lydia annulipes
14.4
65.1
2.8
6.2
6.0
2.9
7.4
50.7
58.2 .001
30.7 .001
39.7
52.1
41.1
21.1
17.6
2.6
21.2
13.4
35.6 .001
312
TABLE 33 - Continued
Separation Due to
Utilization of
Algae
Separation Due
Utilization of
Animal Foods
to
Species
Compared
Percent of
Separation Chi2 Sig.
Percent of
Separation Chi2 Sig.
Eriphia
scabricula,
Pseudozius
caystrus
9.6 1.4
Eriphia
scabricula,
Xantho gracilis
20.6 60.8 65.3 .001
Eriphia
scabricula,
Xantho leptodon
20.4 57.3 57.0 .001
Eriphia
scabricula,
Xantho
sanguineus
17.3 42.3
Eriphia
scabricula,
Zozymodes
biunguis
21.8 60.9 82.6 .001
Lydia
annulipes,
Pseudozius
caystrus
62.3 64.1 .001 11.0
Lydia
annulipes,
Xantho gracilis
71.3 72.8 .001 73.2 70.0 .001
Lydia
annulipes,
Xantho leptodon
71.1 73.7 .001 69.7 61.2 .001
Lydia
annulipes,
Xantho sanguineu
68.0
s
71.8 .001 64.7 39.7 .001
313
TABLE 33 - Continued
Separation Due to
Utilization of
Algae
Separation Due to
Utilization of
Animal Foods
Species
Compared
Percent of
Separation Chi2 Sig.
Percent of
Separation Chi2 Sig.
Lydia annulipes,
Zozymodes
biunguis
72.5 93.3 .001 73.3 86.2 .001
Pseudozius
caystrus,
Xantho gracilis
9.0 62.2 77.4 .001
Pseudozius
caystrus,
Xantho leptodon
8.8 58.7 68.4 .001
Pseudozius
caystrus,
Xantho
sanguineus
5.7 43.7
Pseudozius
caystrus,
Zozymodes
biunguis
10.2 62.3 95.1 .001
Xantho gracilis,
Xantho leptodon
0.2 3.5
Xantho gracilis,
Xantho
sanguineus
3.3 18.5
Xantho gracilis,
Zozymodes
biunguis
1.2 0.1
Xantho leptodon,
Xantho
sanguineus
3.1 15.0
Xantho leptodon,
Zozymodes
biunguis
1.4 3.6
TABLE 33 ~ Continued
314
Separation Due to
Utilization of
Algae
Separation Due to
Utilization of
Animal Foods
Species
Compared
Percent of
Separation Chi^ Sig.
Percent of
Separation Chi2 Sig.
Xantho
sanguineus,
Zozymodes
biunguis
4.5 18.6
Chi^ Data
Number of Crabs Number of Crabs
Containing
Algae
Containing
Animal Foods
Species +
- - +
Dacryopilumnus
rathbunae
72 4 54 22
Eriphia
scabricula
72 20 26 66
Lydia annulipes 19 39 11 47
Pseudozius 114 13 34 93
caystrus
Xantho gracilis 81 1 73 9
Xantho leptodon 83 1 72 13
Xantho 89 3 66 26
sanguineus
Zozymodes
biunguis
112 0 101 11
315
species can be distinguished from Lydia annulipes which is
carnivorous. Xantho sanguineus and Daeryopilumnus rath
bunae cannot be distinguished by these criteria from the
omnivorous species.
As the carnivorous habits of Lydia annulipes are
rather specialized, it is worthwhile making a further set
of comparisons (Table 34) between this species and the om
nivorous forms (as well as the two herbivorous species that
most frequently eat animal foods). A significant differ
ence (greater than 50 percent separation, supported by Chi^
tests, significant at .001) can be shown between Lydia an
nulipes and Xantho sanguineus, Dacryopilumnus rathbunae,
and Pseudozius caystrus on the basis of utilization of Si-
phonaria and sipunculids.
While it is clear from the above analyses that the
herbivorous and carnivorous feeding categories can be dis
tinguished, the distinctness of the third category of om
nivorous crabs is questionable, as it seems to overlap the
other two categories. However, stomach-content analyses
may not provide a complete picture of the ecological separ
ation of crabs which feed upon algae, as those species
which are able to scrape a thin film (up to 1mm thick) of
algae from the rock surface are probably capable of utili
zing a food resource unavailable to those which cannot
graze upon the algae until (or unless) it becomes thicker
or higher. Therefore, variations in the thickness of algae
316
TABLE 34
ECOLOGICAL SEPARATIONS WITH RESPECT TO
DIFFERENCES IN UTILIZATION OF
SIPHONARIA AND SIPUNCULIDS
Species Compared
Percent of
Separation Chi2 Sig.
Dacryopilumnus rathbunae,
Lydia annulipes
55.2 68.5 .001
Eriphia scabricula,
Lydia annulipes
48.0
Lydia annulipes,
Pseudozius caystrus
52.6 86.9 .001
Lydia annulipes,
Xantho sanguineus
56.5 83.5 .001
CHI2 DATA
Number of Crabs
Containing
Siphonaria and/
or Sipunculids
Species
- +
Dacryopilumnus rathbunae 75 1
Eriphia scabricula 75 17
Lydia annulipes 19 39
Pseudozius caystrus 122 5
Xantho sanguineus 92 0
317
may allow partial ecological separation between herbivorous
and omnivorous species, even if they are feeding on the
same kinds of algae.
It is also possible that certain foods may be of
greater importance to the crabs than their frequency of oc
currence in the stomach contents may indicate; and that
some types of animal food may be of greater nutritive value
to omnivorous crabs than an equivalent volume of algae.
It lies beyond the scope of this work to investigate com
petition for food between these species or their compara
tive energetics. It will suffice for us to know whether
there is a marked difference in manner of feeding and food
eaten between those species which often co-occur on the
reef, and whether feeding habits may restrict certain spe
cies to particular habitats.
Ecological Separations Due
to Feeding Habits
In some cases, the crabs occur in the same or similar
habitats and have largely the same feeding habits. This
is true of Zozymodes biunguis, Xantho leptodon, and X. san
guineus on reef flats, and of Xantho leptodon, X. sanguin
eus , and X. gracilis on rubble beaches. Likewise, Dacryo
pilumnus rathbunae and Zozymodes biunguis appear to have
the same feeding habits when they occur on a cyanophyte-
covered rock surface.
In other cases, the crabs have different feeding hab
318
its where they co-occur, but other factors are more impor
tant in determining their distribution. Eriphia scabricula
feeds more on animal foods than do Zozymodes biunguis or
Dacryopilumnus rathbunae, and is unable to scrape the rock
closely as can the latter species; however, E. scabricula
does not occur at all where there are no holes large enough
to provide it refuge, regardless of apparent food availa
bility .
On the other hand, where Lydia annulipes and Pseudo
zius caystrus live together under conglomerate slabs, there
is a marked degree of separation in feeding habits which
may be of some importance, as both species use the same
crevices for cover. Lydia annulipes in these places feeds
mainly on sipunculids (and also takes some insects), while
Pseudozius caystrus eats larger crustaceans and various
algae. Where P. caystrus and Xantho gracilis occur togeth
er on beaches, feeding differences may also be important;
the former relies much more on animal foods, and apparently
much less on finer blue-green algae than does the latter.
At Eniwetok station 2, the vertical zonation differ
ence between Eriphia scabricula and Lydia annulipes appears
to coincide with a difference in feeding habits; E. scab
ricula fed upon algae and sipunculids, while Lydia annu
lipes at a higher level fed upon Siphonaria. However, at
Uliga Island (station 47), L. annulipes on the higher rock
groin fed upon sipunculids whereas E. scabricula on the
319
reef flat below also ate some sipunculids, indicating that
vertical zonation is not determined by type of food eaten.
In some localities, as at Eniwetok stations 3 and 4, Si
phonaria is common on higher parts of the reef flat; des
pite this, Lydia annulipes does not live in holes at this
level (though it may come down at night to forage across
the inner reef).
Finally, there is little reason to believe that reef
species such as Zozymodes biunguis, Dacryopilumnus rath
bunae , or Eriphia scabricula are excluded from loose rubble
habitats by lack of food, or that Xantho gracilis is ex
cluded from consolidated rock for the same reason, though
it is possible that Lydia annulipes is uncommon in loose
rubble in part because of an absence of its preferred
foods. In general, feeding habits appear to be of impor
tance in separating intertidal xanthid species in only a
few cases.
Conclusions
In consolidated rock situations, five of the study
species can be placed into three categories of behavior,
with respect to when and where they are active: those
which feed in the open at low tide at night (Zozymodes bi
unguis and Lydia annulipes), those which feed in the open
at low tide, day and night ( . Dacryopilumnus rathbunae and
Eriphia s'cabricu'la) , and those which appear never to come
out from under cover (Pseudozius caystrus). In loose rub
320
ble, the last-named species behaves in the same way. Xan
tho leptodon and X. sanguineus appear to straddle all three
categories, depending on the habitat; they remain hidden
under rocks on beaches but come out to feed at night, and
sometimes during the day, in tide pools. Xantho gracilis
behaves in the same way as the other Xantho species in the
aquarium as well as on rubble beaches in nature. It is
doubtful whether the above temporal differences in activity
are of great importance in ecological separation, although
a spatial difference between Lydia annulipes (which feeds
in the open) and Pseudozius caystrus (which feeds under
cover) is probably significant in separating these species
in conglomerate.
Stomach-content analyses and feeding observations in
dicate that these crabs can be placed in three categories
with respect to feeding habits: herbivores, which can
scrape a thin coating of algae from the rocks (Zozymodes
biunguis, Xantho leptodon, X. sanguineus, X. gracilis, and
Dacryopilumnus rathbunae); omnivores, which eat a wide
variety of plant and animal foods (Eriphia scabricula and
Pseudozius caystrus); and carnivores (Lydia annulipes)
which feed on sipunculids and mollusks. These categories
are closely correlated with the claw morphology of the
crabs. Statistically, however, it is uncertain whether
more than two categories (carnivores and herbivores) can be
321
established. Data on feeding habits and behavior are sum
marized in Table 35.
TABLE 35
SUMMARY OF BEHAVIORAL DATA AND FEEDING
HABITS FOR THE EIGHT STUDY SPECIES
Species
Temporal-Spatial
Activity Pattern^ Type of Food
Dacryopilumnus
rathbunae
1. Diurnal-Nocturnal, Low Tide
On Drying Rock Surfaces
Various Algae, Especially Blue-
Greens (Often Scraped from Rocks);
Occasionally Eat Mites, Small
Crustaceans
Eriphia
scabricula
1. Diurnal-Nocturnal, Low Tide
In Pools and on Drying Rock
Surfaces
Various Algae; Also Foraminifers,
Polychaetes, Sipunculids, Small
Crustaceans, Crabs, Adult and Larval
Insects
Lydia
annulipes
1. Nocturnal, Low Tide
On Drying Rock Surfaces
Mostly Sipunculids or Siphonaria;
Also Other Gastropods, Bivalves.
Occasionally Foraminifers,
Polychaetes, Small Crustaceans,
Crabs, Mites, Adult Insects, or
Blue-Green Algae
Pseudozius
caystrus
1.
2.
Probably More Nocturnal than
Diurnal, Always Remaining in
Crevices at Low Tide
Always Under Cover of Rocks
Various Algae; Also Foraminifers,
Polychaetes, and Crabs
^Habitats: 1. Consolidated Rock. 2. Loose Rubble.
322
TABLE 35 - Continued
Species
Temporal-Spatial
Activity Pattern^ Type of Food
Xantho
gracilis
2. Always Under Cover of Rocks Various Algae (Often Scraped from
Rocks or Eaten from Pebbles); Occa
sionally Eat Crabs or Scavenge
Xantho
leptodon
1.
2.
Nocturnal, Low Tide
In Tide Pools
Always Under Cover of Rocks
Except in Pooled Places
Various Algae (Often Scraped from
Rocks or Eaten from Pebbles); Occa
sionally Eat Foraminifers, Small
Crustaceans or Crabs, or Scavenge
Xantho
sanguineus
1.
2.
Nocturnal, Low Tide
In Tide Pools
Always Under Cover of Rocks
Except in Pooled Places
Various Algae (Often Scraped from
Rocks); Also Crabs and Occasionally
Foraminifers
Zozymodes
biunguis
1. Nocturnal, Low Tide
In Tide Pools
Various Algae (Often Scraped from
Rocks); Occasionally Foraminifers
323
CHAPTER VII
CONCLUSIONS
Ecological Characteristics of the
Eight Study Species
The habitats and habits of the individual species
can be described briefly as follows:
Zozymodes biunguis lives mainly in small holes on
reef flats, and can also be found in turf-covered areas
on the margins of conglomerate platforms, algal ridge
remnants, and similar features, and in low crevices in
cemented beachrock. This form lives mostly in the low
intertidal zone, feeds mainly upon various kinds of
algae, including a film of blue-greens which is scraped
from the surface of the rock, and is active in the water
of tide pools during low tide at night.
Xantho leptodon lives mainly under small rocks on
rubble beaches and flats, although it is also abundant
in larger holes on reef flats which have a considerable
amount of sand bound in an algal turf. This is a low
intertidal species, though its vertical zone may be
elevated in pooled places and at the reef flat/beach
interface. Xantho leptodon feeds mainly upon various
algae, including a thin film which is scraped from rocks
or removed from the surface of pebbles; it may also
324
325
occasionally eat crustaceans, or scavenge. These crabs
usually remain under rock on rubble beaches, but come out
to feed in the open, especially during low tide at night,
where they live in tide pools.
Xantho gracilis lives under rocks of various sizes
(including loose beachrock slabs) on rubble beaches and
flats, where the underlying sediment is sand or sand
mixed with some gravel. This species occurs mainly in
the high intertidal, ranging from a little below mean
sea level to the level reached by the highest tides
during neaps; feeding habits are the same as those of X.
leptodon. These crabs usually remain hidden under rocks
on the rubble beaches they inhabit.
Xantho sanguineus lives under large rocks (including
beachrock slabs) on rubble beaches, and is the dominant
crab in firmly wedged heaps of small rocks on rubble
flats. It is also found in larger holes on reef flats
which have many overhanging ledges around the borders of
tide pools (especially where there is only a thin sandy
turf) and lives under rocks in pothole-like depressions
that occur on certain reef flats; it is most abundant
in the low intertidal zone, although it occasionally
occurs in the high intertidal on beaches. Xantho
sanguineus feeds mainly upon a variety of algae, and
also eats larger crustaceans such as crabs; its behavior
326
is otherwise similar to that of X. leptodon.
Dacryopilumnus rathbunae lives in small sipunculid
holes (and perhaps also those of boring clams) on reef
features of irregular topography such as algal ridge
remnants, eroding margins of conglomerate platforms, reef
blocks and boulders, and narrow rims of eroding beachrock;
it is limited to places where there is no retention of
water in tide pools, and occurs mainly where the rock sur
face lacks a sandy algal turf. This form is most common in
the low intertidal zone, though it is found in the high
intertidal in much lower population densities. Dacryopi
lumnus rathbunae feeds mainly upon blue-green algae which
form a thin film on the rock surface where it lives, though
it also eats small crustaceans and mites; it is both day-
and night-active, at low tide when the water has receded.
Eriphia scabricula lives in larger holes on reef
flats without either a thick sandy turf or large ledges;
being also abundant on algal ridge remnants, eroding
conglomerate rims (in Lithotrya holes), large boulders,
and exposed lagoon beachrock. This is a low intertidal
form, which is markedly omnivorous, eating a variety of
algae, small crustaceans, other crabs, adult and larval
insects, polychaetes, and sipunculids. Eriphia scabricula
is day-and night-active, wandering widely over the
327
reef at low tide.
Lydia annulipes lives in large holes in higher
conglomerate (especially along the eroding margins of
conglomerate platforms), on algal mounds, and in beachrock,
also occuring under large, cemented slabs on conglomerate
flats and in beach conglomerate. This is a high
intertidal species, ranging up to the highest spring
tide levels; it feeds mainly upon sipunculids and the
gastropod Siphonaria, though it will probably attack
other mollusks which have shells small enough for it
to crush. Lydia annulipes also occasionally eats
foraminiferans, small crustaceans, insects, polychaetes,
and blue-green algae, and is mainly night-active,
wandering widely over the rock surface at low tide.
Pseudozius caystrus lives mainly under large rocks
(including beachrock slabs) on rubble beaches, especially
where the underlying sediment is gravel (though it can
also burrow in sand); it is the commonest crab under
large, cemented slabs on conglomerate flats and in
beach conglomerates, and also occurs in the margins of
conglomerate platforms where the rock is very porous.
Pseudozius caystrus is mainly a high intertidal crab,
ranging up to the highest spring-tide levels. This
species is markedly omnivorous, feeding upon a variety
of algae, as well as polychaetes, other crabs, and
328
foraminiferans; it remains hidden under rocks on rubble
beaches; whereas in conglomerate, it appears never to
wander out from under the ledges beneath which it may
be observed at low tide at night.
The above information is concisely tabulated in
table 36; details such as the precise dimensions of
holes or rocks, tidal ranges of each species, and so
forth, have been given in preceding chapters and will not
be repeated here. For the less common intertidal species
found both at Eniwetok and in the Southern Marshall
Islands, and which are likely to be encountered with the
eight study species, Zozymodes pumilis appears to live
in the same situations as Dacryopilumnus rathbunae,
while Pseudozius pacificus occurs in the high intertidal
in small holes in the margins of conglomerate platforms.
Both species appear to live in low crevices but are not
restricted to them; neither has been observed out in
the open, though no particular effort was made to learn
about their behavior. Eriphia sebana, the largest
intertidal species, has usually been collected near mean
sea level, and lives in very large holes around the
bases of rock groins, under large slabs on the reef, and
under large boulders on rubble flats. It wanders widely
across the reef, at low tide during the day or at night,
like its smaller cousin Eriphia scabricula.
TABLE 36
SUMMARY OF HABITATS AND HABITATS OF THE EIGHT STUDY SPECIES
Species Substrate Vertical
Zonation
Feeding
Habits
Behavior
Dacryopilumnus
rathbunae
Eriphia
scabricula
Lydia
annulipes
Pseudozius
caystrus
Small Holes (Usually
Sipunculid Borings) in
Eroding Algal Rock, Con
glomerate , or Beachrock
with a Thin Film of Blue-
Green Algae
Large Holes in Eroding
Algal Rock, Conglomerate
Beachrock, and Reef Flats
with Little Sand
Large Holes in Eroding
Algal Rock, Conglomerate
and Beachrock; and Under
Cemented Conglomerate
Slabs
Under Large Rocks,
Especially in Gravelly
Places on Rubble Beaches
and Flats; and Under
Cemented Conglomerate
Slabs
Mostly Low
Intertidal
Low
Intertidal
High
Intertidal
High
Intertidal
Especially
Above High
Water Neaps
Herbivorous:
Feeds Upon Algae
and Occasionally
Small Arthropods
Omnivorous;
Feeds Upon Algae,
Small Arthropods
and Other Animals
Carnivorous;
Feeds Mainly on
Sipunculids and
Siphonaria
Omnivorous;
Feeds Upon Algae,
Foraminifers,
Polychaetes, and
Crabs
Diurnal-
Nocturnal
Diurnal-
Nocturnal
Nocturnal
Remains
Under Cover
329
TABLE 36- Continued
Species Substrate Vertical
Zonation
Feeding
Habits
Behavior
Xantho
gracilis
Under Small and Large
Rocks, Especially in
Sandy Places on Rubble
Beaches and Flats
High
Intertidal,
Especially
Below High
Water Neaps
Herbivorous;
Feeds on Algae
Remains
Under Cover
Xantho
leptodon
Under Small Rocks on
Rubble Beaches and Flats,
or in Larger Holes on
Reef Flats with a Thick
Sandy Turf or Much Loose
Sand
Low
Intertidal
Herbivorous;
Feeds on Algae
Nocturnal
(Reef Flats);
Remains
Under Cover
(Beaches)
Xantho
sanguineus
Under Large Rocks or in
Semi-Consolidated Heaps
of Small Rocks on Rubble
Beaches and Flats; or
Under Ledges or Small
Rocks on Reef Flats
Mostly Low
Intertidal
Herbivorous;
Feeds Upon Algae
and Occasionally
Crabs
Nocturnal
(Reef Flats);
Remains
Under Cover
(Beaches)
Zozymodes
biunguis
Small Holes on Reef Flats
or in Turf-Covered
Eroding Algal Rock,
Conglomerate, or Beachrock
Mostly Low
Intertidal
Herbivorous;
Feeds Upon Algae
Nocturnal
u>
331
Comparisons with Previous Literature on
Coral Reef Xanthids
An attempt has been made to characterize as
accurately as possible the habits of the eight study
species in the Marshall Islands. As has been shown
previously, it is possible for the niche of a species to
change in different parts of its geographical range.
Although no dramatic change in the ecology of the study
species was expected from one atoll to the next within
the Marshalls group, three atolls were visited with the
intent of obtaining a broader perspective on the reef
morphology and associated crab faunas that would result
from a study of just one atoll. Though certain
differences were found between the reefs in these
places (e.g.. porous reef flats, where high densities of
certain larger xanthid species are found, are more
characteristic of Majuro Atoll than Eniwetok), the
ecology of the eight species was relatively consistent
from one atoll to another, However, as reports on the
same species from other parts of the world are considered,
it must be borne in mind that their niches may change in
response to changes in reef morphology and tidal range,
or due to the presence or absence of various competing
species.
Ekman (1953) considers the Indian Ocean and the
Western and Central Pacific to be parts of one continuous
332
marine faunal province. Reference to Chapter 3 will
reveal that six of the eight study species occur in at
least parts of the Indian Ocean as well as the Pacific.
Hence, it is appropriate to refer to studies which include
observations on crab ecology from the east coast of
Africa to Hawaii and the Tuamotu Archipelago.
Morrison (1954) provides information on the
zonation of intertidal xanthids at Raroia Atoll, in the
Tuamotus. Prior to discussing the ecology of crabs
at Raroia, however, it should be noted that there are
certain differences between the reefs of the Tuamotus
and those of the Marshall Islands. Raroia appears to
be especially aberrant as compared to other atolls,
because the seaward reef flat is a continuous pooled
zone, often deepening towards shore (Doty & Morrison,
1954). By comparison, reef flats in the Marshalls
which corresponds in width to this Tuamotuan reef slope
downward towards the sea, at least where they are backed
by islands.
At Raroia, the landward intertidal zone above the
reef flat is only 22cm high, forming a rather narrow
sloping rock surface, and the higher conglomerate
platform extending from this point to the island shore
is 0.5-to lm above mean high tide (Newell, 1956).
Although this conglomerate corresponds in position and
333
(judging from descriptions in the literature) appearance
to the shoreward conglomerate examined in the Marshall
Islands, it evidently lies in the supralittoral zone at
Raroia, while it is usually in the high intertidal in the
Marshalls.
There is a strong similarity between the algal ridge
crab fauna found at Raroia (Morrison, 1954) and that of
the Marshall Islands. For the inner reef flat, however,
Pilodius areolatus, Xanthias lamarcki, Libia tesselata,
Eriphia sebana, and Lydia annulipes were listed; the
former three species occur in the infratidal fringe in
the Marshalls, and were probably collected from the
inner reef at Raroia because of the permanently-pooled
conditions there. Eriphia sebana was recorded as being
active on that part of the shoreward reef that is exposed
at low tide (as expected); the fact that Lydia annulipes
was not noted from the higher conglomerate platform at
Raroia may relate to the fact that these deposits lie
above the high tide level.
For the shoreward lagoon reef, Eriphia scabricula,
E. sebana, Lydia annulipes, Xantho "exaratus" (X.
leptodon?), and X. gracilis were said to occur at or just
below low tide (Morrison, 1954). This is questionable,
as these species are all found above the low spring-tide
level at Eniwetok, and two of them are high intertidal
334
forms. It is possible that the poor delineation of tidal
zonation in these crabs at Raroia is due to the fact
that there is a drop of only 1 foot (about 0.33m) on
the lagoon shore of that atoll (Doty & Morrison, 1954).
It is noteworthy that Pseudozius caystrus was recorded
from lagoon shores where beach conglomerate is exposed
(Morrison, 1954), as this species is found in the same
type of deposit at Eniwetok.
Banner and Randall (1952) , writing of Onotoa
Atoll in the Gilbert Islands, indicated that the "red
eye crab" (Eriphia sebana) was found under beach rock in
the high intertidal, windward and leeward. They also
noted that this species occurred together with Uca
(a burrowing ocypodid crab) in long, shallow burrows
under incipient beach rock (though no indication was
given that the crabs constructed these burrows).
Edmondson (1925) reported that on Johnston Island,
an aberrant atoll southwest of Hawaii, Zozymodes
biunguis was very common in dead coral blocks and
coralline algal heads, near shore and on the outer reef.
For Hawaii, Edmondson (1946) noted Z_. biunguis to be
common in shallow water, and that Xantho sanguineus was
abundant under stones at the shore. He also stated that
Pseudozius caystrus was common under stones between the
tides, with P. inornatus (Dana) and Ozius hawaiiensis
335
(Rathbun, two species not encountered at Eniwetok.
In 1970, I found Zozymodes biunguis and Xantho
sanguineus to be very common in holes and crevices on
intertidal reefs and on limestone benches; in addition,
X. sanguineus was taken from under rocks. Both these
forms were common in holes in near-vertical walls of
limestone rock on the shoreward parts of the benches;
in this habitat, Xantho sanguineus occupied large
holes which would, in similar topography in the Marshall
Islands, have sheltered Eriphia scabricula or Lydia
annulipes (which were not encountered while I was in
Hawaii). It appears that the niche of X. sanguineus is
"expanded" in the absence of certain competitors in
Hawaii, though otherwise it is found in similar habitats
to those in which it occurs in the Marshall Islands.
Ward (1933) has provided a wealth of information on
the zonation and habitats of xanthid crabs at Heron
Island, at the southern end of the Great Barrier Reef.
Heron Island is a platform reef, and probably corresponds
in exposure to some of the leeward reefs seen in the
Marshalls. At Heron Island, a reef crest or rampart
composed of dead corals and reef blocks separates the
outer and inner reef flats, where live corals flourish.
Next to the island there is a "conglomerate zone" of
local origin (presumably beachrock). Eriphia sebana
336
occurred both on the outer rampart and in the conglomerate
zone, and Xantho sanguineus only in the latter; both
therefore belong in the intertidal zone proper, and
not in the intervening "lagoon zone" which probably
corresponds to the infratidal fringe.
Farther north on the barrier reef, several studies
refer to the distribution of crabs at Low Isles, another
platform reef. T.A. Stephenson et al. (1931) reported
Xantho exaratus (this species or a related one?) from
the boulder tract and ramparts of the outer reef, and
McNeill (196 8), in discussing collections made by the
above authors on the expedition many years before, noted
that this form was the most abundant decapod there,
being characteristic of marginal regions with shingle
and boulders (that is, the higher parts of the reef
above the level of living corals). He also noted that
Lydia annulipes was taken from under rocks among beach-
rock of the cay. Collections made at a later time at
Low Isles (W. Stephenson el: al. ,1958) revealed that X.
exaratus occurred at about mean sea level on the shingle
ramparts, where the rocks are brown, with gritty sand 1.5"
down; it occupied a vertical level above that of Actaea
tomentosa (which occurs at a lower level than X. leptodon
on the reef at Majuro). Xantho exaratus occurred in both
loose and consolidated shingle, under loose rocks on the
337
sand flat, arid with Lydia annulipes under rocks among
beachrock.
The intertidal crab fauna of Lindeman Island, a high
island along the mainland of Queensland, was reported by
Ward (1965). The xanthid crabs Actaea scabra (Fabr.)
and Eurupellia granulosa A. Milne Edwards were found
several feet above low tide in oyster beds, a habitat
not encountered in the Marshall Islands. On the north
side of the island, Epixanthus frontalis H. Milne
Edwards was the common xanthid under stones on sand
in the high-tide zone, while Xantho exaratus occupied
the zone below it, being commoner on the lower half of
the shingle beach. At the eastern edge of the area,
Xantho australis and Etisodes australis Ward replaced
exaratus. In an earlier paper, Ward (1936) noted
that Xantho exaratus occurred from just below the high
tide level to the limit of the low neap tides, whereas
australis inhabited a narrow belt of rocks at the low
neap-tide level.
None of the Australian data is at great variance
with the findings of the present study: Lydia annulipes,
Xantho sanguineus, and Eriphia sebana were all reported
from beachrock, and the last-mentioned also from rubble,
there as in the Marshall Islands. Species related to
Xantho leptodon, and probably replacing it on the Great
338
Barrier Reef, also appear to have similar habits; thus
Xantho australis was reported from approximately the
same relative level on rubble beaches, and X. exaratus
from both loose and consolidated rubble, corresponding
to the range of substrate type (rubble beaches, porous
reef flats) with which X. leptodon is associated in
the Marshalls group.
For the western Pacific Ocean, Tokioka (1953) has
provided some notes on the zonation of crabs on the
raised reef at Takarazima Island, in the Tokara group.
Eriphia scabricula and E. sebana were reported to be
common in the "purple zone" of the erosional intertidal
bench. This habitat has been described as a fairly
smooth platform with pothole-like depressions, cut by
channels and covered by Porolithon and zoanthids;
while this suggests a fairly exposed situation, the
"purple zone" is not the most seaward part of the bench,
and probably corresponds in exposure and relative level
to the outer reef flat at Eniwetok station 4, where
Eriphia scabricula is also common.
Tweedie (1950) has published a considerable amount
of information on the crab fauna of the Cocos-Keeling
Islands, based on notes by C.A. Gibson-Hill. Judging
from Darwin's (1851) account and others, the main atoll
of Cocos-Keeling is similar in reef morphology to the
339
atolls of the Marshall Islands. According to Tweedie,
Dacryooilumnus rathbunae was plentiful in abandoned worm
or mollusk tunnels in soft coral rock, just below high
tide, in situations where it may be dry for much of the
day; and Zozymodes pumilis was also said to be common
between the tides in the same sort of habitat. This
corresponds well with the findings of the present study,
although these species were most abundant in the low
intertidal in the Marshalls. Lydia annulipes was reported
to be common along with Ozius tuberculosus H. Milne
Edwards in holes and crevices below the high-tide level,
which is the type of habitat in which the former species
is most common in the Marshall Islands. Xantho gracilis
was said to occur with X. sanguineus along the landward
border of the reef, although Gibson-Hill1s (1947) notes
on the habitat of the latter species were not given.
Pseudozius caystrus was reportedly found under stones in
shallow pools on the landward border of the reef; this
form was seldom found in pools in the Marshall Islands.
Gibson-Hill (1947) described in detail the crab
fauna of Christmas Island (in the Indian Ocean); this
is a high island, with most of its coastline made up
of cliffs of raised reef rock. Around part of the coast,
a narrow, 15' wide bench lies at the base of the cliff,
at sea level; where this is cut in hard limestone, no
340
xanthid crabs were found, but where it is formed from
softer, cemented coral rock, tunneled by worms, the crabs
Eriphia sebana, Pilodius harmsi Balss, and a species of
Pilumnus were taken. Eriphia smithi MacLeay was found in
the same places, clinging to the cliff 2 or 3 feet above
the water, rather than living in pools on the bench.
In places along the coast of Christmas Island there
are reefs with beaches to shoreward; at several beaches
on the sheltered north coast, where a fringing reef runs
out from shore, Eriphia sebana was found in pools, and
Dacryopilumnus rathbunae, D. emerita, Pilodius harmsi,
and several forms of Pilumnus occurred in tunnels in
softer rock. Along the northeast coast of the island,
Xantho sanguineus was found in shallow, slightly sandy
pools on the intertidal reef flat. And at Flying Fish
Cove, facing northwest, Eriphia sebana, Pseudozius
caystrus, and Xantho gracilis were reported to live under
rounded boulders which had fallen onto the beach from the
cliff above. Thus, the Christmas Island xanthids which
are included in the present study have been reported
from habitats similar to those they inhabit in the
Marshall Islands, but differing to some extent in origin.
Henderson (1892) has provided some notes on the
ecology of crabs along the coast of India; he found
341
Xantho exaratus and Eriphia sebana to be common in rubble
in the intertidal zone at Rameswaram Island. J.D.
Taylor (196 8) listed Xantho sanguineus from the low tide
zone, on higher parts of the "algal ridge" on the Mahe
fringing reef in the Seychelles, Western Indian Ocean.
It should be noted that the algal ridge he describes
is not the marginal zone of the reef, to which the term
usually applies, but an area farther back where coralline
algae is often found in the form of loose nodules,
indicative of quieter conditions.
Finally, J.D. Taylor (1971), writing of the inter
tidal zone at Aldabra, a raised atoll in the western
Indian Ocean, stated that Xantho sanguineus and Xanthias
lamarcki were taken from the lowest levels on beach rock,
and considered Eriphia sebana a mid-intertidal crab.
Lydia annulipes was said to occur in the high intertidal
zone in crevices, though the type of deposit was not
specified. The latter species was observed feeding upon
juveniles of the chiton Acanthopleura brevispinosa,
indicating that it feeds on mollusks at Aldabra, as
in the Marshall Islands.
For the most part, the descriptive notes on crab
ecology reported by previous workers would seem to
support rather than contradict the present findings.
Although there may well be geographical variations in
342
the niches of the study species, due to differences in
reef morphology or tidal range, it is unfortunately
difficult to establish the extent of such variations
due to the often sketchy nature of earlier records.
Only in one case (Xantho sanguineus in Hawaii) is it
certain that the habitat preference of a species is
broader, but it should be noted that Hawaii is a marginal
area where certain of the study species are absent and
others apparently very rare.
Although six of the eight study species are wide
spread in distribution, two are not; of these Xantho
leptodon is probably restricted to the Oceanic parts of
the central and western Pacific Ocean, but replaced
ecologically by two closely related forms, X . australis
and X. exaratus farther west. Zozymodes biunguis has
been thus far reported only from the northern Pacific;
it is not known what species may have similar niches in
the southern Pacific, Australian region, and Indian
Ocean. Whereas eleven xanthid species which occur in
the high and low intertidal zones at Eniwetok undoubtedly
constitute an important part of the intertidal crab
fauna in the Indo-Pacific, judging from the literature,
other intertidal species are also important in regions
other than the northern Marshall Islands. Ward (1965)
was able to distinguish an island fauna from a continental
343
one in Australia, based in part on the distribution of
xanthid crabs; but there is not enough information
available at present to allow a high island-low island
distinction to be made for other parts of the Indo-
Pacific, nor to state how the study species may figure
in this.
Previous Records of Hole-related Crab Behavior
Although much has been written of crabs which
excavate their own holes in sand or mud, there is
little information on the behavior of crabs associated
with holes on the reef, nor of species specificity for
holes of different kinds. Tweedie (1947), speaking of
crabs at Christmas Island, noted that Dacryopilumnus
rathbunae lives in holes of such a size that one side of
the carapace can, with the claw on that side, block
the opening; this was not observed at Eniwetok, where
most of the holes inhabited by this species were so
deep that the crabs were not visible when hidden. For
non-xanthid crabs, the following information is available:
Cloud (1952) and Banner and Randall (1952) observed
holes of the fiddler crab Uca in fairly solid beachrock
on Onotoa, which Cloud suggested were made by the crabs
(which ordinarily burrow in loose sediment) prior to
the induration of the rock. Morrison (1954), writing of
Raroia, found holes of Uca tetragonon going through a
344
thin crust of conglomerate into the sand beneath; the
crabs could not be dug out when found in this kind of
situation. Aside from these cases, in which the holes
may have originated as burrows that antedated the
formation of the rock, there is no indication that
intertidal crabs may form their own holes in limestone.
Stephens et al. (1970) found that two species of
California blennies are ecologically separated by the
size of the holes of the boring clams (pholadids and
Lithophaga) they inhabit and that fish size could be
predicted from tube diameter. Caldwell (personal
communication, 1972) believes hole size and shape to
be important factors in the ecological separation of
stomatopod crustaceans at Eniwetok; no doubt similar
phenomena will be reported as the ecology of other groups
becomes better known.
The Relative Importance of Factors Responsible
for Ecological Separations between
Intertidal Xanthid Crabs
Four factors considered to be of potential importance
in the ecological separation of the study species were
investigated: substrate, vertical zonation, feeding
habits, and temporal-spatial differences in activity.
The first two factors determine the habitats of the
animals, the latter two determine their roles in these
habitats.
345
As has been shown previously, the maximum number of
substrate categories in any vertical zone is five (for
the low intertidal): (1) smaller rocks on rubble flats
and beaches, or larger holes on reef flats with a thick
sandy turf, (2) larger rocks and beachrock slabs on
rubble flats and beaches, semi-consolidated heaps of
smaller rocks on rubble flats, rounded rocks in pothole
like depressions on reef flats, or larger holes on reef
flats with overhanging ledges around pool rims, (3)
larger holes in reef flats other than those previously
described, and especially where there is a thick mat
of algae but little sediment; or larger holes in beach
rock, algal ridge remnants, or eroding conglomerate,
(4) smaller holes on reef flats, or in algal rock or
conglomerate where there is a sandy algal turf, and (5)
smaller sipunculid holes in algal ridge remnants, eroding
conglomerate or beachrock, and reef blocks and large
boulders.
The maximum number of vertical zones in any given
substrate type is three (for loose rubble lying over
sand or gravelly sand): (1) from the level of low
neap tides to just below mean sea level, (2) from just
below mean sea level to the level of high neap tides,
and (3) from the high neap tide level to the highest
levels reached by spring tides.
346
For feeding habits, at least two categories can
be recognized: (1) herbivorous crabs which scrape
algae from the substratum, and (2) carnivorous species
which eat mollusks and sipunculids; another category,
that of (3) omnivorous forms which feed upon a thicker
coating of algae and on various animal foods, may also
be recognized, although it broadly overlaps the first
two.
For temporal-spatial activity patterns, at least
two categories can be discerned, those of (1) forms
which are active only under rocks or slabs, and those
which feed out in the open; the latter may be subdivided
into (2) those which are night-active and (3) those
which are both day-and night-active, although (2) is
completely overlapped by (3).
According to this criterion, then, substrate is the
most important factor in determining ecological
separations. Substrate and vertical zonation together
constitute the habitats (and microhabitats) of these
crabs, and feeding habits and temporal-spatial activity
patterns define their behavior in these habitats. It
should be noted that each of the eight species has its
own distinct habitat or microhabitat, whereas at most
there are five behavioral categories: (1) night-active
herbivorous crabs, (2) night-active carnivorous crabs,
347
(3) day-and night-active herbivorous crabs, (4) day-and
night-active omnivorous crabs, and (5) omnivorous crabs
which are active only under cover.
The second method of analysis of the importance of
these ecological factors was also carried out; each
species was compared with every other species, based on
the findings of preceding chapters (table 37). Note
is made of cases in which behavioral factors can be
excluded from consideration due to little or no habitat
overlap; cases in which the behavior of the crabs in
question is essentially the same, whether or not there
is habitat overlap, are also indicated. In instances
where there is little question but that substrate or
vertical zonation is responsible for the ecological
separation, no further discussion is needed; cases in
which there is question as to which of several factors
is more important are discussed below.
Cases in which low intertidal species are compared
against each other will be considered first. Eriphia
scabricula and Zozymodes biunguis show 72.6 percent
separation with respect to hole size (on reef flats),
whereas there is only 21.8 percent separation in the
utilization of algae and 60.9 percent separation in the
utilization of animal foods. There is great overlap
in the behavior of these species, as both are active
348
TABLE 37
RELATIVE IMPORTANCE OF DIFFERENT
ECOLOGICAL FACTORS IN
ECOLOGICAL SEPARATION
Ecological Separation Due Primarily to:
Species
Compared
Type of
Substrate
Vertical
Zonation
Type of
Food
Temporal-
Spatial
Activity
Dacryopilumnus
rathbunae,
Eriphia
scabricula
M?
p
s
Dacryopilumnus
rathbunae,
Lydia annulipes
M? H?
p p
Dacryopilumnus
rathbunae,
Pseudozius
caystrus
H n n
Dacryopilumnus
rathbunae,
Xantho gracilis
H s, n n
Dacryopilumnus
rathbunae,
Xantho leptodon
H s, n n
Explanation of Symbols Used:
H: Habitat is of Primary Importance.
M: Microhabitat is of Primary Importance.
B: Behavior is of Primary Importance.
?: Of Questionable Importance,
n: Little or No Habitat Overlap,
s: Behaviorally Similar.
349
TABLE 37- Continued
Ecological Separation Due Primarily to:
Species
Compared
Type of
Substrate
Vertical
Zonation
Type of
Food
Temporal-
Spatial
Activity
Dacryopilumnus
rathbunae,
Xantho
sanguineus
H s, n n
Dacryopilumnus
rathbunae,
Zozymodes
biunguis
H s, n n
Eriphia
scabricula,
Lydia annulipes
H n n
Eriphia
scabricula,
Pseudozius
caystrus
H? s, n n
Eriphia
scabricula,
Xantho gracilis
H n n
Eriphia
scabricula,
Xantho leptodon
H n n
Eriphia
scabricula,
Xantho
sanguineus
H n n
Eriphia
scabricula,
Zozymodes
biunguis
M?
p ?
Lydia annulipes,
Pseudozius
caystrus
H? B? B?
350
TABLE 37- Continued
Ecological Separation Due Primarily to:
Species
Compared
Type of
Substrate
Vertical
Zonation
Type of
Food
Temporal-
Spatial
Activity
Lydia annulipes,
Xantho gracilis
H? n n
Lydia annulipes,
Xantho leptodon
H? n n
Lydia annulipes,
Xantho
sanguineus
H? n n
Lydia annulipes,
Zozymodes
biunguis
H?
p
s
Pseudozius
caystrus,
Xantho gracilis
H? H?
p
s
Pseudozius
caystrus,
Xantho leptodon
H n s, n
Pseudozius
caystrus,
Xantho
sanguineus
H s s
Pseudozius
caystrus,
Zozymodes
biunguis
H? H? n n
Xantho gracilis,
Xantho leptodon
H s s
351
TABLE 37- Continued
Ecological Separation Due Primarily to:
Species
Compared
Type of
Substrate
Vertical
Zonation
Type of
Food
Temporal-
Spatial
Activity
Xantho gracilis,
Xantho
sanguineus
H s s
Xantho gracilis,
Zozymodes
biunguis
H? H? s, n n
Xantho leptodon,
Xantho
sanguineus
M s s
Xantho leptodon,
Zozymodes
biunguis
M s s
Xantho
sanguineus,
Zozymodes
biunguis
M s s
Total Cases 11-19 5-13 0-7 0-3
352
at night. In the case of Dacryopilumnus rathbunae and
Eriphia scabricula, there is 76.2 percent separation in
hole size, as compared to 14.4 percent separation in the
utilization of algae and 39.7 percent separation in the
utilization of animal foods. In both instances,
microhabitat type appears to be the primary factor in
ecological separation. It should be remembered in cases
such as these in which the crabs often occur in the
same habitats but differ in microhabitat, that behavioral
factors can only be important when several micro
habitats (e.g. holes of different sizes) are randomly
spaced in a given habitat (e.g. reef flat). It is common
to find areas of reef or conglomerate where the holes
are predominately of sizes that favor only one species.
In this sense, therefore, microhabitat differences may be
more important relative to behavioral differences than
actual measurements of percentage spearation may indicate.
In cases where the species under comparison live
mainly at different vertical levels but also show a
considerable substrate separation, it is especially
difficult to establish which factor is more important.
Zozymodes biunguis and Lydia annulipes show 67.1 percent
separation (22.9 percent overlap) in hole size; Zozymodes
biunguis is 20.2 percent as abundant as Lydia annulipes
in the high intertidal, while the latter species is
353
rare in the low intertidal on conglomerate. In flat
areas, where the bulk of the Zozymodes biunguis
population lives, there is no overlap in vertical
zonation, and there is 65.4 percent separation in hole
size between Z_. biunguis on reef flats and L. annulipes
on conglomerate. Thus vertical zonation appears more
important than microhabitat type in determining the
ecological separation between these crabs. Although
there is considerable separation in feeding habits
(72.5 percent separation in the utilization of algae
and 73.3 percent separation in the use of animal foods),
this appears to affect only a small percent of the
population of these species.
Dacryopilumnus rathbunae and Lydia annulipes show
71.8 percent separation (28.2 percent overlap) in hole
size; the former is 37.1 percent as abundant as the
latter in the high intertidal while the latter is rare
in the low intertidal. There is 65,1 percent separation
in the utilization of algae and 52.1 percent separation
in the utilization of animal foods; there is considerable
overlap in temporal behavior patterns, as both forms
are night active. In this case it is very difficult
to assess whether vertical zonation or substrate is
more important, because there are a number of localities,
especially on Majuro Atoll, where the low intertidal
354
habitats (e.cf., algal ridge remnants) of D. rathbunae
do not occur but where high intertidal conglomerate
harboring both species is present. For this reason, a
larger proportion of the D. rathbunae population may live
in the high intertidal than is suggested by the relative
abundance figures, and the microhabitat difference
may be relatively more important than the statistical
comparisons would suggest.
Dacryopilumnus rathbunae does not overlap either
Pseudozius caystrus or Xantho gracilis in substrate
type. On the other hand, the former species ranges up
to a relative tidal level of .6 51 or higher, while P.
caystrus is abundant at this level and probably lower
on conglomerate and extends downward to a relative level
of .46 2 in loose rubble. Xantho gracilis is abundant
as low as a relative level of .462 on beaches. In
these cases it is clear that substrate is the primary
factor in ecological separation, as there is a certain
degree of overlap in vertical zonation.
Zozymodes biunguis extends up to a relative level
of .647 or higher on conglomerate, but as has just been
indicated, it is probable that only a very small
proportion of the total biunguis population occurs
above reef-flat levels. While there is some a certain
amount of overlap in vertical zonation between this
355
species and Xantho gracilis, there is also a slight
degree of overlap in substrate type, since Z_. biunguis
is found (though uncommonly) on rubble beaches, where
it occurs below the level of X. gracilis. There is even
more difference between Zozymodes biunguis and Pseudozius
caystrus with regard to vertical zonation, as the latter
occurs mainly at a higher level on beaches than does
Xantho gracilis; but in this case there seems to be more
overlap in substrate type, as both Z_. biunguis and P.
caystrus will live in the space beneath slabs of
conglomerate or beachrock (though it is not certain
whether the height of the crevices is the same). In
both these cases, it is difficult to discern whether
substrate or vertical zonation is more important in
ecological separation.
Eriphia scabricula and Xantho gracilis are completely
separated by difference in substrate type. Eriphia
scabricula extends up to a relative level of .467 on
conglomerate and as high as a relative level of .594
on beachrock. Xantho gracilis on the other hand is
abundant as low as a relative level of .462 on rubble
beaches. It is evident in this case that substrate is
the primary factor isolating these forms, as there is
some overlap in vertical zonation. In comparing Eriphia
scabricula with Pseudozius caystrus, both forms occur
356
in consolidated rock situations where they are completely
separated by vertical zonation. As both species are
found in the crevices under slabs of beachrock, it is
probable that substrate is less important than vertical
zonation in separating these forms, although it is true
that Pseudozius caystrus is abundant at various levels
on rubble beaches, a habitat in which Eriphia scabricula
does not occur.
Lydia annulipes is similarly separated from Xantho
leptodon and Xantho sanguineus by vertical zonation in
consolidated rock habitats, whereas there is probably
some overlap in hole size and other substrate require
ments between these spe -ies in such places. Vertical
zonation is probably the most important factor separating
Lydia annulipes from these low intertidal species,
although the former is rare on rubble beaches where
the Xantho species are common.
The remaining cases involve comparisons between
high intertidal species. There is considerable overlap
in vertical zonation between Xantho gracilis and
Pseudozius caystrus (P_. caystrus is 48.9 percent as
abundant as X. gracilis at a relative level of .467 to
.757), as well as in substrate type (P. caystrus is
41.3 percent as abundant as X. gracilis under rocks
on sand). It is difficult in this case to evaluate
357
which factor is more important. It is possible that in
the vertical zone and on the substrate where the greatest
overlap occurs, feeding habits may be important in
separating these species. However, there is only nine
percent separation in the utilization of algae, and 62.2
percent separation in the utilization of animal foods.
There seems therefore to be even more overlap in feeding
habits than in habitat type.
Xantho gracilis and Lydia annulipes are almost
completely separated by type of substrate. While it
is possible that the latter species is restricted to
consolidated rock deposits because that is where its
preferred foods occur, this does not explain why X.
gracilis is limited to rubble beaches. Therefore,
substrate is probably the most important factor separating
these species.
Finally, in the case of Pseudozius caystrus and
Lydia annulipes, there is considerable separation in
substrate type; the former is rare on the less porous
forms of eroding conglomerate, while the latter is rare
on rubble beaches. The greatest amount of habitat over
lap occurs under slabs on conglomerate flats, where L.
annulipes is 25.3 percent as abundant as P. caystrus.
It is possible that in this habitat, the specialized
feeding habits of L. annulipes may be responsible for
358
its lower abundance; there is 52.6 percent separation in
the utilization of sipunculids and Siphonaria, and 62.3
percent separation in the utilization of algae. It is also
likely that L. annulipes forages in the open while P. cays
trus feeds under cover. It is probable that in this case
behavioral factors are of considerable importance in deter
mining the ecological separation* along with substrate
type.
The results of this analysis indicate that in at least
11 and probably 14 cases* substrate is the most important
factor in separating intertidal xanthid species; in at
least 5 and probably 9 cases vertical zonation is most im
portant. In 4 cases it is very difficult to assess whether
substrate or vertical zonation is more important* and in
one remaining cases substrate* feeding habits* and tem
poral-spatial behavior patterns all appear to be important.
It is strongly suggested* then* that substrate is the most
important factor in determining ecological separation. Tak
ing substrate and vertical zonation together* habitat dif
ferences are of primary importance in at least 21* and
probably 27* of the 28 cases. Therefore* there is no doubt
that habitat is of much greater importance than behavior in
maintaining the diversity of this group of species.
This method of analysis probes to have the drawback
that it is difficult to assess the relative importance
359
of several factors when several different statistical
methods must be used, especially in cases in which there
appears to be nearly the same amount of overlap with
respect to several factors. While it would theoretically
be desirable to use the same method of assessing the
extent of overlap throughout the study, this is difficult
to do when certain species occupy multiple habitats, with
the result that the percentage of the population of these
forms in a given habitat will vary depending upon the
prevalence of the habitat in the study area. It is also
difficult to assess the importance of behavioral factors in
separating species which occur in the same habitats, with
out a complex series of experiments.
In applying this method to the present study, it
has proven useful in determining the minimum number of
cases in which a given factor is of primary importance,
and has enabled the relative importance of habitat and
behavioral differences to be assessed. It has not,
however, allowed the relative importance of different
habitat factors (substrate and vertical zonation) to be
determined with complete certainty. As a research tool,
it would be more easily applied and would yield more
precise and positive results in studies of groups whose
members live only in one habitat and whose feeding habits
are more easily quantified.
360
The Role of Intertidal Xanthid Crabs in
the Reef Ecosystem
Xanthid crabs are eaten by numerous fishes, and
probably a number of other reef organisms, as will be
shown in the following section; because of their great
abundance, they must be important links in the food chain
on coral reefs. It would be difficult to estimate the
extent of predation on intertidal species on the basis
of our present knowledge, however.
The finding that five of the eight study species are
adapted for scraping a thin film of algae from the rocks
is noteworthy, in view of the fact that several important
groups of tropical reef fishes (acanthurids, scarids,
siganids) have members which feed upon the same kind of
food (W. Stephenson and Searles, 1960, Bakus, 1967B,
1969, Chartock, 1970). The function of the claws of
these crabs in removing this algal material is analogous
to that of a parrotfish's beak which is composed of
fused teeth.
Although reef fishes belonging to the above-
mentioned groups are evidently responsible for the low
standing crops of algae in certain reef areas, and along
with this the barren appearance of the substratum, it is
uncertain whether xanthid crabs have as drastic an effect
on the algae. Grazing fish are capable of wandering
over large areas of the reef at high tide, whereas the
361
crabs can only influence algae in close proximity to
their refuges. It is doubtful whether crabs that live
on rubble beaches feed to any extent upon the algae of
exposed rock surfaces, although by grazing on the algal
coating of under-rock pebbles they may utilize a food
source unavailable to many other organisms.
Herbivorous reef crabs appear to graze most heavily
on algae which are adjacent to the apertures of their
holes; therefore, they may feed in concavities in the
reef surface where the algal growths are inaccessible
to larger grazing fishes. Omnivorous and carnivorous
crabs, however, often occupy crevices where there is
a fairly rich algal turf; these animals certainly feed
upon algae, but do not scrape it close to the rock
surface. Hence, larger crevices in certain habitats
(e.£. algal mounds, beachrock) support a much richer
flora than do adjacent exposed surfaces (in the high
intertidal, however, this may be due to higher
humidity in the crevices rather than to the effects of
grazing by fishes on exposed surfaces).
At Eniwetok Island, xanthid crabs are relatively
uncommon on the reef flat at station 2, apparently due
to a paucity of holes on the reef surface, whereas great
numbers of fish of several species graze there, leaving
a multitude of tooth marks on the reef rock, which is
362
covered by a film of blue-green algae. At station 4, on
the other hand, grazing fishes are not nearly as abundant,
while there are large populations of xanthid crabs
living in holes and crevices. There may be a causal
relationship here. While it is doubtful that intensive
grazing by the surgeonfish Acanthurus guttatus on the
inner reef at station 2 has planed off the reef surface,
removing most of whatever rugosity may have originally
been there, and with it many crab holes. It may be
significant that many of the holes in this and other
highly-grazed reef flats appear to be the borings of
sipunculids, that is, holes which may be formed on the
reef at the present day.
It is unlikely that herbivorous crabs are responsible
for as much erosion of the reef rock as is caused by
certain scarids and acanthurids, except perhaps for the
area immediately around the apertures of their holes;
the xanthids which feed by means of spatulate claw-tips
were never observed to scrape through the algae and
expose bare rock in one stroke, which grazing fishes are
evidently able to do, judging from examinations of their
tooth-marks.
On the other hand, in some places the smaller crabs
evidently scrape the film of blue-green algae close enough
to permit encrusting coralline algae to become established
363
on the rock adjacent to their holes. At station 4, the
encrusting coralline alga Goniolithon was seen growing
over a thin film of blue-greens, while the knobby centers
of the Goniolithon colonies were themselves overgrown
by larger filamentous algae such as Jania. It would be
of interest to know whether the small xanthid crab
Zozymodes biunguis, which is extremely abundant in the
crevices provided by this encrusting coralline, may help
to maintain the latter by grazing on its competitors.
Similarly, Bakus (1966) has suggested that the activities
of reef fishes which graze on fleshy algae may have
been an important factor in the evolution of calcareous
algae, such as coralline algae, in the tropics.
Predation on Xanthid Crabs
Why should habitat type be of such importance in
separating these species? It has been previously
shown that reef-dwelling crabs are highly dependent upon
holes and crevices in the rock for hiding places and are
very difficult to dislodge once they have gained entry
to these. Crabs which must emerge from under cover to
feed are active mainly at low tide; some of these carry
out most of their activity at night. Day-active species
are very wary and difficult to approach. The commonest
intertidal species Zozymodes biunguis is very seldom
seen in the open during the day, and appears reluctant to
364
leave the vicinity of its hole at any time. Crabs that
live under rubble seldom come out at all, and exhibit
several characteristic modes of escape behavior when
dislodged.
These behavior patterns suggest that they are subject
to heavy predation pressure, which forces them to take
refuge in holes or under rocks. Hiatt and Strasburg's
(1960) study of the feeding habits of fishes in the
Marshall Islands indicates that of 233 species of
fishes sampled, at least 79 (33.9 percent) had preyed
on crabs, and at least 42 (18 percent) of these on
xanthid crabs. No doubt an even higher percentage of
fish species eat xanthid crabs, for these authors had
only a few specimens of certain species available for
stomach-content analyses. Among the groups of fish which
commonly eat crabs in the Marshall Islands are apogonids,
balistids, holocentrids, labrids, lutjanids, mullids,
muraenids, serranids, and tetradontids. Bakus (1967B)
was impressed by the voracity with which certain fish will
eat crabs extracted from corals and dropped in shallow
inshore waters on an incoming tide, and noted that
feeding experiments with brachyuran crabs suggest that
their main enemies are predatory fishes (Bakus, 1969).
Heavy predation on crabs by fishes is probably
characteristic of all tropical shallow-water areas.
Randall's (1967) investigation of the feeding habits of
365
reef fishes in the West Indies reveals that crabs
(including xanthids) are often eaten by fishes. Some
114 (53.8 percent) of 212 fish species studied had
eaten some crabs; 25 species were found to have 30
percent or more of the bulk of material in their stomachs
consisting of crab remains. Among the fish species
which had fed most heavily upon crabs were a muraenid,
and several holocentrids and serranids.
Hobson's (1968) study indicates that in the Gulf of
California, most of the prey of nocturnal fishes
consists of small, motile invertebrates, especially
crustaceans; he suggests that the feeding habits of
fishes which are active among rocks at night reflects
an increase in the availability of prey organisms. In
this area, day-active predatory fishes tend to eat other
fishes or sessile invertebrates. Hobson notes that the
effectiveness of measures taken by small organisms to
evade capture during the day is reflected in the special
techniques used by fishes to capture hidden organisms at
this time. His data suggest that although crustaceans
are eaten by fishes mainly at night, they remain
hidden during the daytime because they run a much greater
risk of being eaten if they reveal themselves then.
366
Though it may be unwise to extrapolate from
Hobson's findings that this pattern is typical of all
tropical regions, it should be noted that on Eniwetok,
subtidal crabs in the large quarry pool on Eniwetok
Island (including several portinids and a number of
xanthid species) were observed to be much more active
at night. Although they were seldom seen far from their
crevices in the rubble at night, they were not visible
at all in the same areas during the daytime (many of
these crabs were red in color, which may in some way
be correlated with their cryptic habits).
In the intertidal zone, the situation is far more
complicated. A variety of fishes can be seen in the
surge channels of the algal ridge and in the lagoon
close to beachrock deposits; it is probably the danger
of predation by fishes that accounts at night in the low
intertidal, or at high tide in general. On reef flats
and other areas removed some distance from subtidal reefs,
predatory fishes may move in with the rising tide; Hiatt
and Strasburgh (1960) noted that mullids (which eat
crabs) behave in this manner. According to data
collected by Dr. Gerald Bakus (unpublished), the squirrel
fish, Holocentrus lacteoguttatus, occurs on the reef
flat at Eniwetok station 3 at night; and I have found
groupers at least 30cm long in shallow tide pools next
367
to shore at station 4, at night (probably stranded with
the fall of the tide). These fish may well be important
nocturnal predators on crabs at high tide.
Certain smaller fishes which are commonly found
in tide pools on the reef may also prey on crabs to a
certain extent: the gobiid Bathygobius fuscus, the
pomacentrid, Abudefduf sordidus, and the pseudochromid,
Plesiops nigricans (Hiatt & Strasburg, 1960). These
fishes may influence the behavior of the smaller reef
crabs. The muraenid, Gymothorax pictus, is a major
predator on xanthid crabs; this species is active on
the reef flat in the day but apparently less active
at night. It has been observed to leave the water to
chase grapsoid crabs, and will forage along the beach
at high tide (Bakus, 1964, Chave & Randall, 1970).
Knudsen (19 68) suggests that predation by G. pictus, in
and out of water, on the large grapsoid crab, Grapsus
grapsus tenuicrustatus, is responsible for the
terrestrial molting behavior of the latter. I have
observed this eel to be very abundant on the reef
flat at Runit Island and in other places on Eniwetok
Atoll; in the daytime, it will slither out of water very
readily when moving from pool to pool. It has been
observed foraging under rocks on reef and rubble flats,
and as it will attack the large Grapsus grapsus, it is
368
no doubt capable of dealing with smaller crabs most
effectively.
Octopuses were not observed in most of the habitats
where intertidal xanthids were collected. Small octopi
were sometimes seen on the algal ridge and around the
rock quarry at Eniwetok Island; it is likely that their
role as predators on crabs is mainly restricted to sub
tidal reefs. Small stomatopods on the other hand are
abundant on some reef flats where intertidal xanthids
live; they are found in holes like the crabs, and may
possibly be included among the predators on smaller crabs.
Birds are often seen on the reef flat at low tide
in the day; I have watched white reef herons stalking
small fish in tide pools at Eniwetok, but did not observe
them in the act of capturing xanthid crabs. Gardiner
(1909) noted that a crab plover feeds upon fiddler crabs
(Uca) at Diego Garcia Atoll (Indian Ocean); and Bakus
(1967A) observed the bristle-thighed curlew feeding upon
Uca annulipes at Fanning Island (central Pacific).
A part of the general picture of predation is no
doubt played by the xanthid crabs themselves. Eriphia
scabricula and its larger cousin E. sebana are both day-
active, wandering widely across the reef at low tide;
both are known to eat other crabs. Another potential
369
predator on xanthid crabs is a large and vigorous green
portunid, Tha 1-amita sp., which occurs in large tide
pools, in beachrock, and in piles of loose rubble in
the low intertidal and shallow subtidal waters.
Predation and Diversity of Xanthid Crabs
There is little doubt but that predation pressure
is responsible for the cryptic habits of reef-dwelling
xanthids. As habitat and microhabitat differences (of
which substrate is a major component) are most responsible
for the ecological separations between species, it
follows that predation by fishes and other organisms
is probably of major importance in maintaining the
high diversity of this group.
The actual maintence of ecological separations
between crab species must depend of course on the ability
of each species to maintain itself in its preferred
habitat. The length and tortuous nature of many of
the holes in which xanthid crabs live must prevent
access to many larger predators; it is possible that
ecological separations due to differences in the diameter
of refuge holes are due to larger crabs evicting or
preying upon smaller ones inside larger holes. Digging
ability and tolerance of fine sediment in the incurrent
flow of water to the gills must be important to species
living in sandy habitats; if predation pressure forces
370
these species to take cover much of the time, they may
be more subject to the adverse effects of loose sediment
than would be the case if they were allowed to remain
above the surface.
Predation may relate to vertical zonation as well
as substrate preferance. The anomuran crab Emerita
analoga, a burrower on sandy beaches, migrates up and
down with the tide; these animals allow the waves of a
rising or falling tide to carry them across the beach,
although occasionally a part of the population burrows
in instead of migrating (Macginitie & Macginitie, 1949).
Unlike Emerita, the xanthid crabs included in this study
do not display any discernible daily vertical movement.
If it is true that predation pressure is responsible
for the secretive behavior and relative lack of mobility
exhibited by certain of the species, then predation,
combined with high crab population densities which would
limit a migrant's ability to find an unoccupied refuge,
could be a factor in confining these crabs to their
respective vertical zones. Although this is pure
speculation, it is interesting to note that there are
more vertical zones on rubble beaches (where there are
three zones) than in consolidated rock situations (where
there are only two); crabs in loose rubble are more
secretive (appearing to remain continually hidden) than
371
those which inhabit reefs (some of which are very mobile).
Actual zonation, of course, must be due to the differential
ability of various species to tolerate environmental
severity, which presumably increases (for a marine
species) with the increase in tidal level.
It is also possible that the adaptation of
Dacryopilumnus rathbunae for life in old sipunculid
holes is the result of an indirect effect of predation.
The sipunculids themselves fall prey to xanthid predators
such as Lydia annulipes and Eriphia scabricula, and
presumably their burrowing habit gives them some
measure of protection against these crabs; as D. rathbunae
is probably also preyed upon by E. scabricula, it is
possible that both the small crab and the organism
which creates its holes may be taking refuge (in part)
from the same predators.
Interpretation of the Xanthid Crab Findings in
the Light of Previous Work on Crab
Ecology and on the Diversity of
Reef Organisms
The finding that habitat type, of which substrate is
a major component, is of primary importance in separating
intertidal xanthid species fits in well with previous
studies which indicate that habitat differences are
largely responsible for separating species of grapsoid
crabs and crayfish. Although it may be true
372
that high population densities of reef-dwelling xanthids
are permitted by the abundant cover provided by crevices
and loose rubble in some reef habitats, it should be
noted that the Delaware Bay mud crab data (Tweed, 1973)
show that temperate xanthids may attain population
densities as high as those of tropical species. Though
the elucidation of the causes of lowered diversity in
temperate xanthids was not the object of this study, it
is evident that amount of cover per se is unlikely to
be a factor.
The results of the present investigation confirm
previous indications that the feeding habits of sympatric
xanthid crabs exhibit broad overlap. Although the
Delaware Bay mud crabs appear to be for the most part
carnivorous (Tweed, 1973), the Eniwetok species proved
to rely very heavily upon algae. However, temperate
zone xanthids on the California coast (Knudsen, 1960) are
largely herbivorous; this suggests that more work must
be done before the latitudinal differences, if any, in
the feeding habits of xanthid crabs can be characterized.
Kohn's (1959, 1971) findings, taken together with
the results of this study, suggest that diversity may
be controlled by different factors in different groups
of reef invertebrates. In cone snails, feeding habits
are apparently more important than habitat or microhabitat
373
type in determining niches. I have concluded that
habitat type (substrate and vertical zonation) is far
more important than feeding habits in determining the
ecological separations between sympatric xanthid crabs
in the intertidal zone.
Of the various groups which occur in the tropical
intertidal, some of these appear to lack truly intertidal
members. For example, portunid crabs at Eniwetok
include certain species which are abundant in the low
intertidal but also in the infratidal fringe as well;
these should probably be considered as hardy shallow-
water forms which range up into the intertidal zone. On
the other hand, the family Xanthidae includes species
whose maximum abundance is above the low spring tide
level and which are apparently replaced by competing
forms in the infratidal fringe and subtidal. Cone
snails appear to conform to the former pattern, as all
the species found on intertidal benches also occur
on shallow reefs (Kohn, 1971); this must be taken into
account in comparing these animals with truly intertidal
forms. The fact that vertical zonation is a major factor
contributing to the ecological separation between
intertidal xanthid species provides a partial explanation
for greater importance of habitat differences in this
group. Intertidal xanthid crabs find abundant refuge
374
in holes and crevices in the intertidal, and perhaps
have more potential to exploit the spatial heterogeneity
of intertidal reefs or benches than do cone snails.
Finally, it would be interesting to know whether cones,
which are venomous, are as subject to predation as xanthid
crabs; if predation pressure on these snails is less, they
may be more free to wander in search of specific prey
organisms, allowing more feeding specialization.
Six of the eight xanthid species discussed here
are relatively sedentary in their habits, not wandering
far from cover; this may well be a factor which relates
to the relatively non-specific feeding habits these
crabs exhibit. In this regard they are similar in
behavior to the ophiuroids of Chartock's (197 2) study;
I have often observed the latter in the same habitats as
xanthid crabs, in crevices of a similar nature. Of the
two xanthid crabs which are wide-ranging in their
movements, Lydia annulipes has specialized carnivorous
habits and Eriphia scabricula is omnivorous; and it
should be noted that feeding habits appear to be more
important in separating L. annulipes from Pseudozius
caystrus, the other omnivorous form. The latter finding
may lend support to Kohn's hypothesis that feeding
habits may be more important in separating carnivorous
species.
375
It is possible that subtidal xanthids (living in
the infratidal fringe and on shallow reefs) will be
found to have more specific feeding habits than do the
species upon which this study is focussed. Certainly
the mucous-feeding behavior described by Knudsen (19 67)
for coral-inhabiting xanthids is an adaptive pathway not
open to intertidal forms which occur above the level of
living corals. However, Kohn's finding that there is less
food specificity in cone snails on highly patchy,
heterogeneous subtidal reefs, together with the results
of Chartock's study of (mainly subtidal) brittle stars,
suggests that spatial heterogeneity may be found to be
at least as important in the ecology of subtidal crabs
as it is for intertidal forms, especially if there is
even heavier predation pressure. These questions will of
course remain for future workers to answer. Xanthid
crabs are an ideal group for comparisons between the
subtidal and the intertidal; they are very diverse in
the former and a number of species occur in the latter.
Investigations of the ecology of subtidal xanthid crabs
should prove very fruitful in their bearing on questions
of tropical species diversity, especially when considered
in the light of the present findings.
CHAPTER VIII
SUMMARY
Xanthid crabs comprise one of the most diverse and
abundant groups of organisms on coral reefs, but have
hitherto been little studied. Eight species of inter
tidal xanthids common in the jMarshall Islands were se
lected for investigation, with the aim of determining
which of four factors (substrate, vertical zonation,
feeding habits, or temporal-spatial differences in
activity) is most important in maintaining the ecolog
ical separations between species. It was hypothesized
that substrate would prove to be the most important
factor. A large number of comparisons between species
were made, employing several statistical techniques
(analysis of percentage separation, relative abundance,
and Chi^ tests) were employed to quantify data on eco
logical separations.
Brief discussions of the niche concept and of the spe
cies diversity problem are included. There is evidence
that substrate and other habitat factors are important
in the ecology of other decapod crustacean groups.
There is also reason to believe that spatial hetero-
geneity combined, with predation is of great importance
in maintaining the high diversity of coral reef commun
ities. However, previous work on cone snails, which
have been well studied, suggests that differences in
feeding habits are more important in determining niches
in that group.
Investigations were carried out on three atolls: Eni
wetok, Kwajalein, and Majuro; the reef morphology and
climate of the Marshall Islands are described. De
tailed descriptions are provided for 11 major habitat
types in which xanthid crabs occur. Although reefs in
the northern and southern Marshalls and their associ
ated crab faunas are generally similar, several of the
intertidal species are more abundant in the southern
Marshalls, and there are several important study spe
cies on Majuro which are not found on Eniwetok.
Clarification is made of the taxonomic status of the
study species and other intertidal forms; details of
morphology, biometrical data, and color notes are also
included. Population density data were taken in a
variety of habitats; a maximum density of 242 crabs/m^
was found for the most abundant species.
The tidal regime is described for Eniwetok and the
other Marshallese atolls visited. Three zones are
recognized: the infratidal fringe, the low intertidal,
and the high intertidal; true intertidal crabs inhabit
the latter two zones. Tidal zonation is complicated
by reef morphology and exposure. Vertical zonation is
important in the ecological separation of intertidal
xanthids both on consolidated rock (reef-flats, con
glomerate, and beachrock) and in loose rubble, espec
ially in the latter where the high intertidal is divi
ded into two sub-zones.
Refuge holes are of considerable importance to crabs
living in consolidated rock; various kinds of hole-
related defensive behavior were observed. Analyses of
paired crab and hole data obtained by poisoning indi
vidual crabs from their holes and obtaining putty im
pressions of the apertures reveal that crab size and
shape correlate positively with hole size and shape.
Differences in hole size contribute to the ecological
separation of a number of species; hole shape differen
ces are much less important. Topography, extent and
kind of algal cover, and loose sediment on the reef are
also of importance in determining ecological separa
tions in consolidated rock substrates.
Crabs in loose rubble are very secretive, seldom leav
ing cover; these crabs burrow readily and display other
forms of escape behavior. The size of rocks under
which the crabs hide, the type of loose sediment under
rocks, and into which the animals burrow, and the
looseness of heaped rocks are all factors in the eco-
379
logical separation of xanthids in unconsolidated sedi
ments. Finally, it should be noted that certain spe
cies are restricted to either loose rubble or consoli
dated rock habitats.
8. Field observations revealed that the study species can
be divided into three groups in terms of temporal-
spatial activity patterns: diurnal and nocturnal
crabs, nocturnal forms, and those which are active only
under cover. It is doubtful that these differences are
of much importance in the ecological separation of very
many species, however.
9. Observations were made on the feeding habits of crabs
in the field and in the laboratory, and stomach-content
analyses were used to determine what food was actually
eaten. Three categories can be distinguished with re
spect to feeding habits: herbivorous crabs, omnivorous
crabs, and one carnivorous form which feeds upon mol-
lusks and sipunculids. Claw morphology can be cor
related with the feeding habits of the various species.
Algae are major food items in the stomachs of most spe
cies; five of the eight species, including the most
abundant species, are herbivorous. A wide selection of
animal foods is eaten by intertidal xanthids. There is
great overlap in the type of algal food eaten, especi
ally when herbivorous and omnivorous crabs are com
pared. In general, feeding habits appear to play a
380
relatively minor role in determining ecological sep
arations .
IQ. Comparisons with previous records in the literature
suggest that the study species play similar roles in
reef ecology in other parts of the Indo-Pacific. With
respect to substrate, the study species can be as
signed to five categories; there are three categories
with respect to vertical zonation, three of temporal-
spatial behavior patterns, and three of feeding be
havior. There are eight habitat categories, as com
pared to five behavioral categories. Comparisons of
each crab species with every other species suggest
but do not conclusively prove that substrate is the
most important factor in determining ecological separ
ations. However, it can be shown that habitat dif
ferences (substrate and vertical zonation together)
are responsible for the ecological separation in 21
(and probably 27) of 28 cases. It is concluded that
habitat is much more important than intraspecific be
havior in the ecological isolation of intertidal xan
thid crabs.
11. The role of intertidal xanthids in the reef ecosystem
is discussed; they are important links in the food
chain, eating algae and smaller animals and being
eaten in turn by many other organisms. It is believed
that predation, largely by fishes, on xanthid crabs
forces the latter to take refuge in holes or under
rocks, but that interactions between the various crab
species and different substrate types determine where
individual species live. Predation is considered to
be a major factor responsible for the high diversity
of xanthid crabs, working in concert with spatial
heterogeneity. The present findings taken together
with Kohn's work on cone snails indicate that the di
versity of various groups of reef organisms may be
controlled by different factors.
WORKS CITED
Adams, A. and A. White, 1848. Zoology of the Voyage of
H.M.S. Samarang. Crustacea. London. 66 p.
Alcock, A., 1898. "Materials for a Carcinological Fauna of
India. No. 3. The Brachyura Cyclometopa. Pt. 1.
The Family Xanthidae." J. Asiatic Soc. Bengal 67 (2):
67-233.
Allee, W. , 0. Park, A. Emerson, T. Park, K. Schmidt, 1949.
Principles of Animal Ecology. W. B« Saunders Co.,
Philadelphia. 837 p.
Bakus, G. J., 1964. "The Effects of Fish-Grazing on Inver
tebrate Evolution in Shallow Tropical Waters." Allan
Hancock Found. Occas. Pap. 27: 1-2 9.
_., 1966. "Some Relationships of Fishes to Benthic
Organisms on Coral Reefs." Nature 210 (5033): 280-
284.
_________. , 1967A. "Changes in the Avifauna of Fanning Is
land, Central Pacific, Between 1924 and 1963." Condor
69 (2): 207-209.
_________. , 1967B. "The Feeding Habits of Fishes and Pri
mary Production at Eniwetok, Marshall Islands." Mi-
cronesica 3 (2): 135-149.
_________. , 1969. "Feeding and Energetics in Shallow Marine
Waters." Int. Rev. Gen. and Exper. Zool. 4: 2 75-369.
Balss, H., 1932. "Uber Einige Systematische Interessante
Xanthidae der Harmsshen Reise nach dem Sundaarchipel."
Zeitschrift fur Wissenschaftlische Zoologie (Liepzig)
142: 510-519.
., 1938. "Die Dekapoda Brachyura von Dr. Sixten
Bocks Pazifik=Expedition 1917-1918." Goteborgs Kun-
gliga - Vetenskaps=och Witterhets=Samhalles Handlin-
ger. Femte Foljden, Ser. B, 5 (7): 1-85.
382
383
Banner, A. H., and J. E. Randall, 1952. "Preliminary Re
port on Marine Biological Study of Onotoa Atoll, Gil
bert Islands." Atoll Res. Bull. 13: 1-6 3.
Borradaile, L. A., 1903. "Marine Crustaceans. III. The
Xanthidae and Some Other Crabs," p. 237-271 In:
J. S. Gardiner, The Fauna & Geography of the Maldive
and Laccadive Archipelagoes. Vol. 1. C. J. Clay &
Sons. Cambridge.
Cannon, H. G., 1935. "On the Rock-Boring Barnacle, Litho-
trya valentina." Great Barrier Reef Expedition 192 8-
1929. Scientific Reports 5 (1): 1-17.
Chave, E. H., and H. A. Randall, 1970. "Feeding Behavior
of the Moray Eel, Gymnothorax pictus." Fanning Island
Expedition, January 1970. Hawaii Inst. Geophysics,
U. of Hawaii. HIG-70-23: 173-178.
Clausade, M., 1970. "Importance et Variations du Peuplement
Mobile des Cavites au sein Formations Epirecifales,
et Modalites d'Enchantillonage en Vue de son Evalua
tion." Recueil des Travaux de la Station Marine
d'Endoume Fascicule Hors Serie, Supplement 10: 107-
109.
Cloud, P. F., 1952. "Preliminary Report on Geology and
Marine Environments of Onotoa Atoll, Gilbert Islands."
Atoll Res. Bull. 12: 1-73.
Crane, J., 1941. "Eastern Pacific Expeditions of the New
York Zoological Society. XXVI. Crabs of the Genus
Uca from the West Coast of Central America. Zoologica
267 Pt. 3 (19): 145-208.
Dana, J. D., 1852. Crustacea, 1. U. S. Exploring Expedi
tion in the Years 1838, 1839, 1840, 1841, 1842. Under
the Command of Charles Wilkes, U. S. N. Pt. 1 (13):
1-685. Atlas, 1855; 1-27, 96 pi.
Darwin, C., 1851. Geological Observations on Coral Reefs,
Volcanic Islands, and on South America. Reprinted as:
The Structure and Distribution of Coral Reefs. U. of
Cal. Press, Berkeley, Cal., 1962. 214 p.
., 1859. The Origin of Species, 6th Ed., D. Apple
ton and Co., N. Y., 1884. 458 p.
Dawson, E. Y., 1954. "Marine Plants in the Vicinity of the
Institut Oce'anographique de Nha Trang, Vi£t Nam."
Pacific Science 8 (4): 372-481.
384
Dawson, E. Y., 1956. "Marine Algae of the Southern Mar
shalls." Pacif. Sci. 10 (1): 25-66.
., 1957. "Annotated List of Marine Algae from Eni
wetok." Pacific Science 11 (1): 92-132.
De Man, J. G., 1888. "Bericht Uber die in Indischen Archi-
pel von Herrn Dr. Brock Gesammelten Dekapoden und
Stomatopoden." Archiv fur Naturgeschichte 5 3 (1):
215-600.
Doty, M. S., and J. P. E. Morrison, 1954. "Interrelation
ships of the Organisms on Raroia Aside .from Man."
Atoll Res. Bull. 35: 1-61.
Edmondson, C. H., 1923. "Crustacea of the Tropical Central
Pacific." Bernice P. Bishop Mus. Bull. 5: 1-43.
_________., 1925. "Crustacea of the Tropical Central Paci
fic." Bernice P. Bishop Mus. Bull. 27: 3-62.
_________., 1935. "New and Rare Polynesian Crustacea."
Bernice P. Bishop Mus. Occas. Pap. 10 (24): 1-40.
_________., 1946. Reef and Shore Fauna of Hawaii. Bernice
P. Bishop Mus. Spec. Pub. 22. Honolulu. 381 p.
Ekman, S., 1953. Zoogeography of the Sea. Sidgwick and
Jackson, Ltd. Dublin. 1967 Ed. 417 p.
Elton C., 1927. Animal Ecology. William, Clovis and Sons
Ltd. 1966. 207 p.
Emery, K. O., J. I. Tracey, and H. S. Ladd, 1954. "Geology
of Bikini and Nearby Atolls. Pt. 1. Geology." U. S.
Geological Survey Prof. Pap. 260-A: 1-265.
Fisz, M., 1963. Probability Theory and Mathematical Sta
tistics. John Wiley & Sons, Inc. N. Y. 677 p.
Forest, J., and D- Guinot, 1961. "Crustaces Decapodes
Brachyoures de Tahiti et des Tuamotu." Editions de
la Fondation Singer-Polignac. Paris 196 p.
_________. , and _________ ., 1962. "Remarques Biogeogra-
phiques sur les Crabes des Archipels de la Societe et
des Tuamotu." Cahiers du Pacifique 4: 41-75.
385
Fosberg, F. R., 1956. "Military Geography of the Northern
Marshalls." U. S. Army. Chief of Engineers. Engi
neering Intelligence Dossier. Strategic Study Mar
shalls. Subfile 19. Analysis of the Natural Environ
ment . 320 p.
Fryer, G., 1959. "The Trophic Interrelationships and Ecol
ogy of Some Littoral Communities of Lake Nyasa with
Especial Reference to Fishes, and a Discussion of the
Evolution of a Group of Rock-Frequenting Cichlidae."
Proc. Zool. Soc. London 132: 153-281.
_________. , 1960. "Some Controversial Aspects of Speciation
of African Cichlid Fishes." Proc. Zool. Soc. London,
135: 569-578.
Gardiner, J. S., 1909. "The Percy Sladen Trust Expedition
to the Indian Ocean in 1905. No. 1. Description of
the Expedition." Transac. Linn. Soc. London, 2nd
Ser., Zoology 12: 1-56.
Garth, J. S., 1964. "The Crustacea Decapoda (Brachyura and
Anomura) of Eniwetok Atoll, Marshall Islands, with
Special Reference to the Obligate Commensals of Bran
ching Corals." Micronesica 1 (1): 137-144.
Gibson-Hill, C. A., 1947. "Christmas Island - The Nature
of the Coast." Raffles Mus. Bull. 18: 8-17.
Green, J., 19 68. The Biology of Estuarine Animals. Univ.
of Washington, Seattle, London. 185 p.
Grinnel, J., 1917. "Field Tests of Theories Concerning
Distributional Control." Amer. Nat. 51 (602): 115-128.
., 1924. "Geography and Evolution." Ecology 5
JJT: 225-229.
Guinot, D., 1964. "Crustaces Decapodes Brachyures (Xan-
thidae) des Campagnes de la Calypso en Mer Rouge
(1952), dans le Golfe Persique et a l'lle Aldabra
(1954)." Memoires Museum National d'Histoire Natu-
relle. Nouv. Ser, S^r A Zoologie 32 (1): 1-108.
Havens, A. D., 1966. "Ecology of Crayfish in the Vicinity
of Yellow Springs, Ohio." Unpublished.
Heller, C., 1861. "Beitrage zur Crustaceen-Fauna des Ro-
then Meeres." Sitzungsberichte Akademie der Wissen-
schaftliche Wien 43 (.4) : 297-374.
386
Henderson, J. R., 1892. "A Contribution to Indian Carcin-
ology." Transac. Linn. Soc. London, Ser. 2, Zoology
5 (10): 325-458.
Hiatt, R. W. , and D. W. Strasburg, 1960. "Ecological Re
lationships of the Fish Fauna on Coral Reefs of the
Marshall Islands." Ecol. Monogr. 30 (1): 65-127.
Hobbs, H. H., P. C. Hold, and M. Walton, 1967. "The Cray
fishes and Their Epizootic Ostracod and Branchiobdel-
lid Associates of the Mountain Lake, Virginia Region."
Proc. U. S. Nat. Mus. 123 (3602): 1-84.
________., and L. Marchand, 194 3. "A contribution Toward
a Knowledge of the Crayfishes of the Reelfoot Lake
Area." Jour. Tenn. Acad. Sci. 18 (1): 6-35.
Hobson, E. S., 1968. "Predatory Behavior of Some Shore
Fishes in the Gulf of California." U. S. Dept. In
terior. Fish and Wildlife Serv. Bur. of Sport Fish,
and Wildlife Res. Rept. 73: 1-92.
Hollenberg, G. J., 1968. "An Account of the Species of
Polysiphonia of the Central and Western Tropical Paci
fic Ocean." Pacific Science 22 (1): 56-98.
Holthuis, L. B., 1953. "Enumeration of the Decapod and
stomatopod Crustacea from Pacific Coral Islands."
Atoll Res. Bull. 24: 1-66.
________. , 1954. "Proposed Addition to the 'Official List
of Generic Names in Zoology' of the Generic Name
'Xantho' Leach, 1814, (Class Crustacea, Order Deca-
poda)." Bull. Zool. Nomencl. 9 (9): 270-271.
Hutchinson, G. E., 1944. "Limnological Studies in Connec-
ticutt. VII. A Critical Examination of the Supposed
Relationship Between Phytoplankton Periodicity and
Chemical Changes in Lake Waters." Ecology 25 (1):
3-25.
________. , 1957. "Concluding Remarks." Cold Spring Harbor
Symp. on Quant. Biol, for 1957. pp. 415-427.
________. , and E. S. Deevey, 1949. "Ecological Studies on
Population." Surv. Biol. Progr. 1: 325-359.
Jacquinot, H., 1852. Voyage au Pole Sud et dans l'Oceania
sur les Corvettes 1'Astrolabe et la Zele'e. Zoologie,
Atlas Crustaces, Pi. 1-9.
387
Jacquinot, H., and H. Lucas, 1853. Voyage au Pole Sud et
dans 1'Oceania sur les Corvettes 1'Astrolabe et la
Z^lee. Zoologie. Crustaces., 1-107 p.
Knudsen, J. W., 1960. "Aspects of the Ecology of the
California Pebble Crabs (Crustacea: Xanthidae)."
Ecol. Monogr. 30 (2); 80-84.
______., 1967. "Trapezia and Tetralia (Decapoda, Brachy-
ura, Xanthidae") as Obligate Ectoparasites of Pocil-
loporid and Acroporid Corals." Pacific Science 21:
51-57.
_________. , 1968. "The Terrestrial Molting Behavior of the
Crab Grapsus grapsus tenuicrustatus." Bull. So. Cal.
Acad. Sci. §7 (2) : §"0-84.
Kohn, A. J., 1959. "The Ecology of Conus in Hawaii."
Ecol. Monogr. 29 (1): 47-90.
________ ., 1967. "Environmental Complexity and Species Di
versity in the Gastropod Genus Conus on Indo-West
Pacific Reef Platforms." Amer. Nat. 101 (919): 251-
259.
_________., 1968. "Microhabitats, Abundance, and Food of
Conus on Atoll Reefs in the Maldive and Chagos Is
lands." Ecology 49 (6): 1046-1062.
_________., 1971. "Diversity, Utilization of Resources, and
Adaptive Radiation in Shallow Water Marine Inverte
brates of Tropical Oceanic Islands." Limnol. and
Oceanog. 16 (2): 332-348.
Krumbein, W. C., and F. A. Graybill, 1965. An Introduction
to Statistical Models in Geology. McGraw Hill, N. Y.
475 p.
Lack, D., 1944. "Ecological Aspects of Species Formation
in Passerine Birds." Ibis 86: 260-286.
MacArthur, R. H., 1964. "Environmental Factors Affecting
Bird Species Diversity." Amer. Nat. 98: 387-398.
MacGinitie, G. E. and N., 1949. Natural History of Marine
Animals. McGraw-Hill Book Co., Ltd., N. Y. 473 p.
MacNae, W., 1963. "Mangrove Swamps in South Africa." J.
Ecol. 51: 1-28.
388
MacNae, W., and M. Kalk, 1962. "The Ecology of the Man
grove Swamps in Inhaca Island, Mozambique." J.
Ecol. 50, 19-34.
McKee, E. D., and G. W. Weir, 1953. "Terminology for
Stratification and Cross-Stratification in Sedimentary
Rocks." Bull. Geol. Soc. Amer. 64: 381-390.
McNeill, F. A., 1968. "Crustacea, Decapoda and Stomato-
poda." Great Barrier Reef Expedition 1928-1929. Sci.
Repts. 7 (1): 1-98.
Milne Edwards, A., 1863. "Monographie des Crustaces
Fossiles de la Familie des Canceriens." Annales des
Sciences Naturelles, 4e Serie, Zoologie 20 (5): 273-
324.
Milne Edwards, H., 1834. Histoire Naturelle des Crustaces,
T. 1. Paris. 468 p.
Miyake, S., 1939. "Notes on Crustacea Brachyura Collected
by Prof. Teiso Esaki's Micronesia Expeditions 19 37-
1938. II. A Checklist of Micronesian Brachyura."
Rec. Oceanogr. Works in Japan 10 (2): 168-247.
Morrison, J. P. E., 1954. "Ecological Notes on the Mol-
lusks and Other Animals of Raroia." Atoll Res. Bull.
34 (1): 1-18.
Newell, N. D., 1956. "Geological Reconnaissance of Raroia
(Kon Tiki) Atoll, Tuamotu Archipelago." Bull. Amer.
Mus. Nat. Hist. 109 (3): 315-372.
Nobili, G., 1907. "Richerche sui Crostacei Della Polyne
sia." Memorie Accademia della Scienze di Torino,
Serie 2, 57: 351-430.
Odhner, T., 1925. "Monographierte Gattungen der Krabben-
familie Xanthidae." Kungliga Vetenskaps=och Vitter-
hers=Samhalles Handlinger. Fjarde -Foljden 29 (1): 1-
92.
Otter, G. W., 1937. "Rock-Destroying Organisms in Relation
to Coral Reefs." Great Barrier Reef Expedition 1928-
1929. Sci. Repts. 1: 323-352.
Paine, R. T., 1966. "Food Web Complexity and Species Di
versity." Amer. Nat. 100 (910): 65-75.
Pianka, E. R., 1966. "Latitudinal Gradients in Species Di
versity." Amer. Nat. 100: 33-46.
389
Randall, J. E., 1963. "An Analysis of the Fish Populations
of Artificial and Natural Reefs in the Virgin Is
lands." Carib, J. Sci. 3 (1): 31-47.
____. 1967. "Food Habits of Reef Fishes of the West
Indies." Studies in Trop. Ocean. 5: 6 65-847.
Rathbun, M. J., 1906. "The Brachyura and Macrura of the
Hawaiian Islands." Bull. U. S. Fish Comm. 23 (3)
for 1903: 827-930.
_________., 1907. "Reports on the Scientific Results of the
Expedition to the Tropical Pacific, in Charge of
Alexander Agassiz, by the U. S. Fish Commission
Steamer 'Albatross,1 from August, 1899, to March, 1900,
Commander Jefferson F. Moser, U.S.N., Commanding. IX,
X. The Brachyura." Mem. Mus. Comp. Zool. Cambridge,
35 (2): 23-74.
_________. , 1911. "Brachyura of the Percy Sladen Expedition
to the Indian Ocean in 1905." Transac. Linn. Soc.
London, Zool. 14: 191-261.
_________. , 1930. "The Cancroid Crabs of America of the
Families Euryalidae, Portunidae, Atelecyclidae, Can-
cridae, and Xanthidae." U. S. Nat. Mus. Bull. 152:
1-593.
Rhoades, R., 1962. "Further Studies of Ohio Crayfishes.
Cases of Sympatry of Stream Species in Southern Ohio."
Ohio J. Sci. 62 (1): 27-33.
Ryan, E. R., 1956. "Observations on the Life Histories and
the Distribution of the Xanthidae (Mud Crabs) of Ches
apeake Bay." Amer. Midland Nat. 56 (1): 138-162.
Saito, Y., 1969. "The Algal Genus Laurencia from Hawaii
and the Philiopines." Pacific Science 23 (2): 148-
160.
Sakai, T., 1939. Studies on the Crabs of Japan. IV.
Brachygnatha, Brachyrhyncha. Tokyo. 741 p.
________., 1965. Crabs of Sagami Bay. Maruzen Co., Ltd.,
Tokyo. 232 p.
Shaw, G., and F. P. Nodder, 1809. Naturalist's Miscellany
15, PI. 589-612.
Simpson, G. G., 1964. "Species Density of North American
Mammals." Syst. Zool. 13: 57-73.
390
Stephens, J. S., R. K. Johnson, G. S. Key, and J. E. McCas-
ker, 1970. "The Comparative Ecology of Three Sympat-
ric Species of California Blennies of the Genus Hyp-
soblennius Gill (Teleostomi, Bleniidae) .1 1 Ecol.
Monogr. 40: 213-233.
Stephenson, T. A. and A., 1949. "The Universal Features of
Zonation Between the Tide-Marks on Rocky Coasts."
J. Ecol. 37: 289-305.
________., ________., and G. Tandy, 1931. "Ecology of Low
Isles.1 1 Great Barrier Reef Expedition 1928-1929.
Sci. Repts. 3 (2): 35-79.
Stephenson, W., R. Endean, and I. Bennett, 1958. "An Eco
logical Survey of the Marine Fauna of Low Isles,
Queensland." Austr. Jour. Marine and Fresh Water Res.
9 (2): 261-318.
_________., and R. B. Searles, 1960. "Experimental Studies
on the Ecology of Intertidal Environments at Heron
Island." Austr. Jour. Marine and Fresh Water Res.
11 (2): 241-267.
Talbot, F. H., 1965. "A Description of the Coral Structure
of Tutia Reef (Tanganyika Territory, East Africa) and
its Structure." Proc. Zool. Soc. London 145 (4):
431-470.
Taylor, J. D., 1968. "Coral Reef and Associated Inverte
brate Communities (Mainly Molluscan) Around Mahe, Sey
chelles." Phil. Trans. Royal Soc. London, Ser. B, 254
(793): 129-206.
________., 1971. "Intertidal Zonation at Aldabra Atoll."
Phil. Trans. Royal Soc. London, Ser. B, 260: 173-213.
Taylor, W. R., 1950. Plants of Bikini. U. of Mich. Press
Sci. Ser. 18. 227 p.
Takeda, M., and S. Miyake, 1968. "Pilumnoid Crabs of the
Family Xanthidae from West Pacific. 1. Twenty-Three
Species of the Genus Pilumnus, with Description of
Four New Species." Ohmu. Occas. Pap. Zool. Lab. Fac.
of Agr. Kyushu Univ. 1 (1): 1-60.
Tokioka, T., 1953. "Invertebrate Fauna of the Intertidal
Zone of the Tokara Islands. 1. Introductory Notes with
the Outline of the Shore and the Fauna." Pub. Seto
Mar. Biol. Lab. 3 (2): 123-138.
391
Tweedie, M. W. F. , 1947. "On the Brachyura of Christinas
Island." Raffles Mus. Bull. 18; 27-42.
. , 1950. "The Fauna of the Cocos-Keeling Islands,
Brachyura and Stomatopoda." Raffles Mus. Bull. 22;
105-148.
U. S. Dept. Commerce, 1970. Tide Tables 1970. Central and
Western Pacific Ocean and Indian Ocean. 386 p.
Ward, M., 1933. "True Crabs of the Capricorn Group,
Queensland. Part 1. Xanthidae." Austr. Jour. Zool.
7; 237-255.
. , 1936. "Crustacea Brachyura from the Coasts of
Queensland." Mem. Queensland Mus. 11 (1): 1-13.
. , 1965. "Notes on Marine Ecology of Lindeman Is
land, Queensland, with Special Reference to the Brach-
yuran Crabs." Austr. Zool. 13; 127-134.
Wells, J. W., 1951. "Coral Reefs of Arno Atoll, Marshall
Islands." Atoll Res. Bull. 9; 1-13.
Williams, J. B. (Undated). Olivetti Underwood Programma
101. Statistical Analysis. Systems Group, Mktg.
Serv. Div., Olivetti Underwood Corp. n.p.
392
APPENDIX I
STATION LIST
Station Island Location Habitats Sampled
Eniwetok Atoll
1 Eniwetok South End of
Island
Algal Ridge, Reef
Flat
2
II
Off Airfield,
South Half of
Island
II
3
II
Mid-point of
Island
II
4
II
North Half of
Island
Reef Flat
5A
II
Pond at South
End of Quarry
Rubble Beach and
Pool
5B
II
Beach off Mid
point of Quarry
II
5C
II
Beach at North
End of Quarry
II
5D
II
Main Quarry
Pool
Reef Flat, Dead
Corals, Rubble
Heaps, Cracks
6
II
*Extreme North
End of Island
Rubble Beaches
7A
II
*01d EMGL, Mid
point of Island
Beachrock
7B
II
South of
Fishing Pier
II
7C
II
*North of
Fishing Pier
It
7D
It
*North of
Tradewinds Bar
II
*Lagoon-side Stations.
APPENDIX I - Continued
393
Station Island Location Habitats Sampled
7E Eniwetok ♦Off New EMBL Beachrock
8A
fl
Channel to
Sand Island
Algal Ridge
8B
II
Channel to
Sand Island
Reef Flat
8C
tl
Channel to
Sand Island
*Porites Beds
9A Sand South Side of
Island
Beachrock, Rubble
on Reef
9B
It
Ocean Side of
Island
Reef Flat,
Conglomerate
10A Reef Between Sand and
Parry Islands
Beachrock, Rubble
Flat
10B
It
Rubble Flat
IOC
II
♦Beachrock
10D
II II
11A Parry Quarry, South
End of Island
Rubble Heaps
11B
11
South Half of
Island
Reef Flat, Rubble
12A Japtan West End of
Deep Channel
Eroding Rims of
Conglomerate
12B
11
East End of
Deep Channel
Conglomerate,
Rubble Flat
12C
II
Ocean Side of
Island
Reef Flat, Rubble
Flats and Beach
13A Aniyaanii South of
Pemphis-covered
Islet
Rock Groin, Reef
Flat, Rubble Flat
Porites Beds
394
APPENDIX I - Continued
Station Island Location Habitats Sampled
13B Aniyaanii North of
Pemphis-covered
Islet
Algal Ridge, Reef
Flat
13C
II
North Shore of
Main Island
Rubble Flat
13D
II
*South Half of
Island
Beachrock
14A Run it Mid-point of
Island
Algal Ridge, Reef
Flat
14B
IT
North Shore of
Southern Embay-
ment on Reef
Conglomerate,
Beachrock, Rubble
Beach
14C
It
South Shore of
Embayment
Rock Groin, Rubble
Beach
14D
II
*South Half of
Island
Beachrock
15 Piiraai Eastern Shore
of Island
Reef Flat
16A Ro joa Northern Shore
of Island
Beachrock
16B Biijiri Eastern Shore
of Island
Conglomerate
Beachrock, Corals
in Channel
16C
II
Reef off North
east Tip of
Island
Reef Flat, Algal
Ridge
17A Bogallua Middle of
Island
Porites Beds
17B
II
West of Island Coral Heads,
Gravel Bar
17C
II II
*Porites Reef
17D
II
Middle of
Island
*Rubble on Sand
Flat
APPENDIX I - Continued
395
Station Island Location Habitats Sampled
18 Rigili South Shore
of Island
Conglomerate
19A Grinem North Half of
Island
Conglomerate
19B
11 II
*Rubble Beach
20 Boganegan West of Island Algal Ridge
21 Igurin Mid-point of
Island
Reef Block,
Rubble Flat and
Pool
Kwajaleirl Atoll
31 Kwajalein End of Runway
to Fishing Pier
Reef Flat
32
II
North Half of
Island
Reef Flat,
Conglomerate
33
Majuro At
II
:oll
Interisland
Reef North of
Kwajalein Island
Boulders on Reef
41 Small Islet West of Arniel
Island
Conglomerate
42 Interisland Reef, Just
West of Arniel Island
Reef Flat, Rubble
Train
43 Arniel Island and Reef
Just to East of Island
Algal Ridge, Reef
Flat, Rubble on
Reef and Beach
44 Ejit Algal Ridge, Reef
Flat, Corals in
Channel, Rubble
Train
45 Small Islet West of Darrit
Island
Rubble Beach
46 Darrit West End of
Island
Reef Flat,
Conglomerate
APPENDIX I - Continued
396
Station Island Location Habitats Sampled
47 Uliga West End of
Island
Algal Ridge, Reef
Flat, Conglomerate
48A Dalap West End of
Island
Reef Flat, Rubble
Pool, Conglomerate
48B
II If
*Reef Flat, Rubble
Beach, Conglomerate
49A Ajurotake East of Island Reef Flat,
Conglomerate
49B
II II
*Reef Flat, Beach
rock, Rubble Flat
and Beach
50 Uotjaa (approximate local) *Rubble Flat
51 Rotoin (approximate local) Conglomerate
52 Majuro West End of
Island
Reef Flat,
Boulders on Reef
397
APPENDIX
REEF AND BEACH
II
TRANSECTS
Station Habitat Zone Width (m)
2 Reef Flat off Runway
High algal Ridge and 4 5.5
Backridge Trough
(Maximum Length of (36)
Surge of Channels)
Inner Reef, Orange Zone_______30. 5
Total 76
Reef Flat off Old EMBL
High Algal Ridge 19
(Buttresses (10)
(Add itiona.l. Leng th of ( 9)
Surge Channels)
Backridge Trough 19
(Dead Coral Zone) ( 9)
(Smooth Reef Flat) (10)
Outer Reef Flat 21
Secondary Trough 10
Inner Reef Flat 35
(Pooled Shore Zone) ( 5)
Total 104
Reef Flat off New EMBL
Low Algal Ridge 8
Smooth Backridge Trough 12
Outer Reef Flat 15
Secondary Trough 10
Inner Reef Flat 40
(Coralline-Vermetid Zone) (27)
(Zone of large Tide Pools (13)
at Shore)
Total 85
Beach at North End of Quarry, 6.7
0.34 to 0.9m Level
Beach at North End of Island
Lower Beach with Cobbles 24
0.22 to 1.1m Level
Upper Beach with Boulders 12
1.1 to 1.9m Level
Total 36
APPENDIX II - Continued
398
Station Habitat Zone Width (m)
7B Lagoon Beachrock, North of Pier
Coralline-Vermetid Zone 3
Vermetid Zone 1.5
Black Algal Zone 6
Parched Grey Upper Zone 3
Total 13.5
9B Seaward Side of Sand Island
Orange Inner Reef with
Pothole-like Tide Pools
14.6
Orange Pitted Zone 5.5
Exfoliating Conglomerate 9.2
Total 29.3
10A Beachrock and Rubble on Interisland
Reef
Rubble Flat 33.6
Blackened Beachrock 6.4
Sand-Bottomed Moat 4.3
Lagoon Sand Bar 15.2
IOC
Beachrock Dam 3.1
Lagoon Abrasion Ramp 4.6
12A
Eroding Conglomerate Rims 4.6-9.2
12B
Rubble Zone on Inner Reef
East of Rubble Field
8.5
Eroding Conglomerate Rims 30
Blackened Margin of Shore
Conglomerate
1.5
Abrasion Ramp 1.5
Gravel Rampart 1.5
13A North-South Transect across Rock Groin
Eroding Margin of Reef Flat 0.3-0.9
Reef Flat, Covered with
Pink Pasty Material
1.8
Eroding Margin of
Conglomerate
0.9-0.9
Exfoliating Conglomerate
Platform
24
APPENDIX II - Continued
399
Station Habitat Zone Width (m)
Rubble Field South of
Groin 12
14D Lagoon Beachrock
Eroding Lower Rim of
Beachrock
Pools and Basins Behind
Rim
Sloping Beachrock with
Sand
0.3-0.9
1-2
33
31 Reef Just North of Pier, Kwajalein
Island
Inner Reef with Pothole
like Tide Pools
27.5
1Total 128(approx.
32 Reef off North Half of Kwajalein
Island
Slick, Brown Inner Reef
Smooth, Shallow Pool Zone
Pitted Dark Green Zone
Grey Conglomerate Flat
17. 8
6.4
4.3
10. 7
41 Small Islet West of Arniel
Eroding Conglomerate Flat 30
43 Beach at East Side of Arniel Island
0.3 to 1.83m Level
10.1
46 East-West Transect across Channel
Scour Zone, Mid-Channel
Vermetid Zone
Jania-Vermetid Zone
15.5
7.3
23. 8
Reef off West End of Darrit Island
Algal Ridge and Backridge
Trough
Outer Reef Flat
Shallow Secondary Trough
Inner Reef Flat
Dark Green Pitted Zone
Conglomerate Flat
32.6
16.5
19.2
32
3
12.8
Total 120.1
APPENDIX II - Continued
400
Station Habitat Zone Width (m)
47 Transect Through Rock Groin,
Uliga Island
Algal Ridge Buttresses 13.7
Backridge Trough (with
Buttress Remnants)
8.2
Outer Reef Flat 22
Shallow Secondary Trough 10
Dark Green Inner Reef 29.3
Dark Green Pitted
Conglomerate
5.5
Blackened Zone of Groin 13.7
Transition Zone 13.7
Zone of Peeling Brown Turf 21
Pooled, Sandy Zone Behind
Conglomerate
36.6
Rubble between Pool and
Beach
16.5
Total
Transect Just East of Groin
194.2
Algal Ridge to Inner Reef (as above)
Dark Green Inner Reef 51
Inner Reef with Thick
Sand-Algal Turf
30
48A Transect across Seaward and
Lagoon Reefs, Dalap Island
Algal Ridge and Seaward
Reef Flat
44
Dark Green Pitted Zone 3
Black Conglomerate Flat 6.4
Pooled Zone with Sand and
Rubble
6.4
Road Embankment 13.7
48B Pooled Zone with
Conglomerate Islets
45.7
Conglomerate Flat 7.3
Porous Margin of
Conglomerate
3
Smooth "Visor" above Reef
Flat
1.2
APPENDIX II - Continued
401
Station Habitat Zone Width (m)
Lagoon Reef Flat 16.5
Total 94.9
402
APPENDIX III
TEMPERATURE MEASUREMENTS IN VARIOUS
INTERTIDAL HABITATS IN 1970
Habitat
June 25-
27 (e)
July 3-
9 (E)
July 20-
26 (E)
Aug. 3-
6 (E)
Aug. 26-
31 (K,M)
Submerged
Reef Flats
28. 8,
29. 6
31.5,
31.8
30,
30
Intertidal
Reef Flats
Tide Pools
Exposed
Reef Rock
31. 5-
32.1
27. 5-
27.7,
32.9,
34. 9,
35.5
31. 5
34.5 33,
34.1
35
35.5,
35.5
36,
37.5
36,
36,
36.5
Conglomerate
Margins
Exposed Roc
Within Hole
k
30.4
s
29. 3-
29.5
30. 6
32.5
34. 8
31.7
34.5
30
36.8,
39.8
Comglomerate
Flats
Tide Pools
Exposed Roc
Under Slabs
1
32.1
k 35.5
28.4,
29
36.5
38.5
(E) = Eniwetok Atoll.
(K,M) = Kwajalein and Majuro Atolls.
APPENDIX III - Continued
403
Habitat
June 25-
27 (E)
July 3-
9 (E)
July 20-
26 (E)
Aug. 3-
6 (E)
Aug. 26-
31 (K,M)
Lagoon
Beachrock
Exposed Rock
Within Holes
38.5
32
Rubble Flats,
Beaches
Tide Pools
Exposed Rock
Under Rocks
28.5
27.3- 29.1,
27.4, 29.1,
29 29.1-
30.4,
30.1
32.7,
38
39
31
32
Note: Temperature measurements connected by a hyphen indi
cate a range of data collected on the same day and
in the same locality.
404
APPENDIX IV
MEAN CRAB SIZE AND HOLE SIZE COMPARED
FOR 5 SPECIES OF REEF CRABS
Species
Hole
Width
(mm)
Crab
Length
(mm)
Crab Length/
Hole Width
Crab
Width
(mm)
Dacryopilumnus
rathbunae
6.7-1.6 4.6*1.1 . 630 6.7*1.5
Eriphia
scabricula
15.5-5. 7 10.1*3.3 .648 12.8*4.2
Lydia
annulipes
14.7*6.0 9.0*3.7 .610 12.8*5.2
Xantho
leptodon
15.3*4.2 9.5*1.5 . 622 12.3*2.0
Zozymodes
biunguis
7.3*3.2 3.9*0.9 .531 5.4*1.3
Species
Hole
Height
(mm)
Crab
Height
(mm)
Crab Height/
Hole Height
No.
Crabs,
Holes
Dacryopilumnus
rathbunae
5.3*1.2 3.5*0.8 .653 52
Eriphia
scabricula
9.2*3.6 5.9*1.9 . 650 59
Lydia
annulipes
8.7*3.6 5.3*2.2 .603 46
Xantho
leptodon
9.1*2.6 5.0*0.8 .547 21
Zozymodes
biunguis
4.5*1.3 2.4*0.6 .537 69
405
APPENDIX V
FORMULA FOR CORRELATION COEFFICIENT
(CORRELATION OF PAIRED XY DATA)
Pearson's product-moment correlation
coefficient was used in all calculations
of the coefficient of correlation; the
calculations were made on an Olivetti
Underwood Programma 101. The following
formulas were employed (Williams, undated):
Program Code 2.14
ZX2-XZX = Z(X-2)2 = IX2
IY2-NXY = Z(Y-?) 2 = 3EY2
ZXY-ZZY = 2(X-X) (Y-Y) = JEXY
Program Code 2.15
2 = (2XY)2
ZX2£Y2
r
Significance of the correlations
was determined with the aid of confidence
belts given by Krumbein and Graybill (1965)
406
APPENDIX VI
CORRELATION BETWEEN CRAB SIZE
AND HOLE SIZE
Correlations of Individual Crab and Hole
Data Pairs, for Each Species
Species
Lcr: Wh
Corr. Sig.
No. Data
Pairs
Regression
Equations
Dacryopilumnus
rathbunae
.848 .05 52 Lcr=0.756+.568Wh
Eriphia
scabricula
. 807 .05 59 Lcr=2.785+.469Wh
Lydia
annulipes
.634 . 05 46 Lcr=3.220+.390Wh
Xantho
leptodon
.575 .05 21 Lcr=6.494+.198Wh
Zozymodes
biunguis
.630 .05 69 Lcr=2.523+.185Wh
Species
Her: Hh
Corr. Sig.
No. Data
Pairs
Regression
Equations
Dacryopilumnus
rathbunae
.879 . 05 52 Hcr=0.392+.580Hh
Eriphia
scabricula
.777 .05 59 Hcr=2.19 3+.410Hh
Lydia
annulipes
.674 . 05 46 Hcr=l.686+.411Hh
Xantho
leptodon
.497 . 05 21 Hcr=3.583+.154Hh
Zozymodes
biunguis
.637 .05 69 Hcr=l.264+.258Hh
Lcr = Length of Crab. Her = Height of Crab.
Wh = Width of Hole. Hh = Height of Hole.
APPENDIX VI - Continued
Overall Correlation Between Crab Size and
Hole Size, Using Means of Data for Each
Species.
Lcr: Wh
Corr. Sig.
No. Data
Pairs*
Regression
Equation
.988 .05 5 Lcr = -.440+.659Wh
Her: Hh
Corr. Sig.
No. Data
Pairs*
Regression
Equation
.962 .05 5 Her = -.484+.661Hh
* Mean crab length and hole width,
and crab and hole Height for the
five species, given in Appendix IV.
408
APPENDIX VII
MEAN CRAB SHAPE AND HOLE SHAPE COMPARED
FOR 5 SPECIES OF REEF CRABS, AND
CORRELATION BETWEEN CRAB SHAPE
AND HOLE SHAPE
Mean Crab and Hole Shape
Species Hole H/W Crab H/L
No. Crabs,
Holes
Dacryopilumnus
rathbunae
.805-096 .769-017 52
Eriphia
scabricula
.605-156 .596-018 59
Lydia
annulipes
.643-171 . 587-012 46
Xantho
leptodon
.608-119 .522*011 21
Zozymodes
biunguis
.670-184 . 630*025 69
Overall Correlation Between Crab Shape and
Hole Shape, Using Means of Data for Each
Species (above).
Crab H/L: No. Data Regression
Hole H/W Corr. Sig. Pairs Equation
.952 .05 5 H/Lcr = .571+.025H/Wh
Crab H/L, H/Lcr = Crab Height/ Crab Length.
Hole H/L, H/Lh = Hole Height/ Hole Width.
409
APPENDIX VIII
RELATIONSHIP BETWEEN CRAB SHAPE
AND HABITAT TYPE
Species
Crab
H/L
Shape
H/W Habitat
No.
Crabs
Dacryopilumnus
rathbunae
.769 . 524 Holes in Consolidated
Rock
52
Eriphia
scabricula
.596 .467 Mostly in Holes in
Consolidated Rock;
Also Under Cemented
Slabs
59
Zozymodes
biunguis
. 630 .461 Mostly in Holes in
Consolidated Rock:
Also Under Cemented
Slabs, Rare in Rubble
69
Lydia
annulipes
.587 .412
I I
46
Xantho
leptodon
.522 .407 Mostly in Rubble; May
Be Abundant in Holes
in Consolidated Rock,
But in Few Localities
21
Xantho
sanguineus
.498 . 366
I I
21
Pseudozius
caystrus
.498 . 356 Mostly in Rubble;
Also Under Cemented
Slabs
21
Xantho
gracilis
.504 .368 Rubble 21
INFORMATION TO USERS
This material was produced from a microfilm copy of the original document. While
the most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the original
submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1.The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. !f necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific pages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print. Filmed as
received.
Xerox University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED.
VL61
SN3AYH SVIDflOa NVTV
Aq q-uBxaAdoo 0
'l o q i y UU V ' A u e d iu o o \0 ^ 3 X V 's u ij ij o io ij^ A jtsie A iu jq
A6oLOLg
1?Z6L “ O'Md ‘ BLUJOJ.LLBQ ua9qq.nos 10 /fyLsaaALUf)
•saiw isi twhshvw
3H1 NI 3V0IH1NVX AHWVJ 3H1 JO SaW O
!Vaiia31NI JO S3ID3dS 1H9I3 JO A9010D3 3H1
-£fr6l *S8L6noa ublv *S N 3 A V H
6 K ‘ZL-W
I
1
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Ecology of toxicity and antibiosis in marine sponges
PDF
Evolution in the oceanic midwaters: comparative osteology and relationships of the lanternfishes (family Myctophidae)
PDF
The ecology of the mid-water Amphipoda in the waters over the Santa Catalina and the San Pedro Basins off the coast of southern California
PDF
Correlation analysis of the responses of mammalian muscle sense organs to single motor unit stimulation
PDF
Aspects of feeding on the bryozoan Zoobotryon verticillatum (Delle Chiaje)
PDF
Biosystematic studies on the frog genus, Leptodactylus
PDF
An Analysis Of The Distribution And Feeding Habits Of The Heteropod Mollusk, Carinaria Japonica, In Waters Off Southern California
PDF
A biogeographic analysis of the freshwater fishes of Honduras
PDF
The Australian gekkonid lizard genus Diplodactylus Gray: an evolutionary and zoogeographical study
PDF
Comparative morphology and evolution of the neotropical frog genera Atelopus, Dendrophryniscus, Melanophryniscus, Oreophrynella and Brachycephalus
PDF
Distribution And Ecology Of Two Families Of Natant Decapod Crustacea -- Oplophoridae And Pasiphaeidae -- In Waters Off Southern California
PDF
The ceratioid anglerfishes of the genus Oneirodes (family Oneirodidae): comparative osteology, systematics, distribution, and biology
PDF
Serology of certain marine trematodes
PDF
A morphological study of the relationship of certain hormone imbalances to the pineal gland
PDF
Reproduction, life history and ecology of the round stingray, Urolophus halleri Cooper
PDF
On the ultrastructure and morphogenesis of a marine chonotrichous ciliateprotozoan, Lobochona prorates
PDF
The Role Of Detritus In A Tropical Marine Ecosystem: Niche Separation In Congeneric Ophiuroids, Food Partitioning In Cryptic Invertebrates, And Herbivore Detritus Production At Eniwetok, Marshal...
PDF
Cultured aortic smooth muscle cells: a model for the study of atherogenesis
PDF
The Cumacea (Crustacea, Peracarida) Of California: Systematics, Ecology And Distribution
PDF
Adrenal Cortical Steroid Involvement In The Activation Of Phosphorylase By Epinephrine And Cyclic-3', 5'-Adenosinemonophosphate
Asset Metadata
Creator
Havens, Alan Douglas, 1943-
(author)
Core Title
The ecology of eight species of intertidal crabs of the family xanthidae in the Marshall Islands
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Biology
Degree Conferral Date
1974-02
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
biology, ecology,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Garth, John S. (
committee chair
), Abbott, Bernard C. (
committee member
), Bakus, Gerald J. (
committee member
), Easton, William H. (
committee member
), Nafpaktitis, Basil George (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-686002
Unique identifier
UC11356029
Identifier
7417349.pdf (filename),usctheses-c18-686002 (legacy record id)
Legacy Identifier
7417349.pdf
Dmrecord
686002
Document Type
Dissertation
Rights
Havens, Alan Douglas
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
biology, ecology