Citation
Farming of cordgrass, Spartina alterniflora Loisel., by fiddler crabs, Uca rapax (Smith) (Decapoda: ocypodidae)

Material Information

Title:
Farming of cordgrass, Spartina alterniflora Loisel., by fiddler crabs, Uca rapax (Smith) (Decapoda: ocypodidae)
Creator:
Genoni, Giulio Piero, 1957-
Publisher:
[s.n.]
Publication Date:
Language:
English
Physical Description:
v, 61 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Burrowing ( jstor )
Crabs ( jstor )
Farms ( jstor )
Female animals ( jstor )
Food ( jstor )
Food availability ( jstor )
Salt marshes ( jstor )
Sediments ( jstor )
Shelters ( jstor )
Time management ( jstor )
Dissertations, Academic -- Zoology -- UF
Fiddler crabs -- Behavior ( lcsh )
Spartina ( lcsh )
Zoology thesis Ph.D
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1987.
Bibliography:
Bibliography: leaves 54-60.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Giulio Piero Genoni.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
022408839 ( ALEPH )
18540021 ( OCLC )

Downloads

This item has the following downloads:


Full Text
















FARMING OF CORDGRASS, SPARTINA ALTERNIFLORA LOISEL.,
BY FIDDLER CRABS, UCA RAPAX (SMITH)
(DECAPODA: OCYPODIDAE)










BY


GIULIO PIERO GENONI



















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA 1987


















ACKNOWLEDGEMENTS




I wish to express my gratitude to my graduate committee, Drs. F.J.S. Maturo, Jr., C.L. Montague, F.G. Nordlie, W.J. Lindberg, and M.G. Wheatly, for their generous assistance and guidance. I am especially indebted to Dr. Maturo, who provided much support with facilities, supplies and advice, to Dr. Montague, who stimulated my interest in this study, and to Dr. Lindberg, for helpful discussion.

I also thank Drs. M.D. Bertness, W.F. Herrnkind,

C.S. Hopkinson and L.W. Powers for their helpful comments and G. Nemes, V. Orcel and A.M. Fiore for help in the lab and in the field. D. Harrison prepared the illustrations.























ii















TABLe OF CONTENTS


PAGE

ACKNOWLEDGEMENTS .................................. ii

ABSTRACT ................................... ....... iv

INTRODUCTION ........................................ 1

METHODS o............................................ 4

Burrow/crab Ratios in the Salt Marsh .......... 4 Effects of Burrow Density and Food Availability 6 RESULTS ............................................ 11

Burrow/crab Ratios in the Salt Marsh .......... 11 Effects of Burrow Density and Food Availability 12 DISCUSSION ................................. .... 34

burrow/crao Ratios in the Salt Marsh .......... 34 Effects of Burrow Density and Food Availability 36 General Considerations ........................ 43

APPENDIX ............ .............................. 46

LITERATURE CITED ................................... 54

BIOGRAPHICAL SKETCH ................................ 61

















Abstract of Dissertation Presented to the Graduate Scnool of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



FiiRMIG OF CORDGRASS, SPARTI-A ALIERNIFLORA LUISEL.
BY FIDDLER CRABS, UCA RAPAX (SMITH) (DECAPODA: OCYPODIDAE)


BY


Giulio Piero Genoni


December 1987


Chairman: Frank J.S. Maturo, Jr. Major Department: Zoology





Tute fiddler crab Uca rapax is a food limited detritivore that feeds on decaying cordgrass (Spartina alterniflora) in the salt marsh. Crab burrowing activity stimulates cordgrass growth, and thereby indirectly enhances their food supply. Fiddler crabs may dig more burrows than they require for their protective and physiological needs. There may be a selective advantage in digging "excess" burrows, because of the indirect benefit of enhancing food supply and decreasing the competition for food. The hypothesis that fiddler crabs "farm" cordgrass was examined by (a) comparing burrow and crab density in a salt marsh,

(b) manipulating burrow density to test whether fiddler crabs dig burrows in excess of the ones available, and (c) .anipulating

iv









food availability to test whether crabs adjust their burrowing activity in response to food. Results showed that (a) there were more burrows than fiddler crabs, (b) crabs dug new burrows despite the presence of unoccupied burrows, and (c) their burrowing activity varied inversely with food availability. Food availability affected the rate of burrowing by females and males equally, but had a stronger effect on burrowing by small crabs than on that of large crabs. Thus, small crabs may be more sensitive to food limitation, and may be more efficient in "farming" cordgrass.








































v

















INTRODUCTION




Organisms may regulate or control other tropnic levels by establishing interactions with each other and feedback relationships (Kitchell et al., 1979; Montague et al., 1981; Patten and Odum, 1961). In salt marshes, fiddler crabs have a large impact on cordgrass, Spartina alterniflora: their burrowing activities, bioturbation and fecal pellet production increase the growth of Spartina, resulting in an increased food supply in the form of microorganisms that colonize the grass when it dies and decays (Montague, 1980a, 1980b; Katz, 1980; Bertness, 1985). These crabs are food limited in the salt marsh (Genoni, 1985). Selection may favor activities and by-products that would decrease food limitation, such as, perhaps, burrowing activity. Fiddler crabs -ray be said, then, to "farm" salt marsh grass, like some earthworms "farm" plants in their habitat (Darwin, 1881), an indirect effect sensu Wilson (1980).

Burrows have a number of direct advantages for fiddler crabs: they function as shelters from predators and environmental extremes; provide water for physiological needs; and are sites for reproduction (Crane, 1975; Hyatt and Salmon, 1977; Young and Ambrose, 1978; Ringold, 1979). They may serve other presently unidentified functions. If fiddler crabs dig more burrows than they require for these needs, then the benefit

1





2




of an increased food supply may not be merely a by-product of their burrowing activity (Montague, 1980a).

The hypothesis that the fiddler crab Uca rapax (Smith) farms cordgrass was examined by (a) measuring and comparing natural burrow density and crab density; (b) manipulating burrow density to test whether fiddler crabs tend to dig burrows in excess of the ones available; and (c) manipulating food levels to test whether such excess burrowing may represent farming.

The prediction was made that the burrow/crab ratio would be >1. Since fiddler crabs are food limited, and since burrows enhance their food supply, they may tend to excavate burrows or burrow branches in excess of a ratio of one burrow available per crab. If they do so in spite of the comparatively high energy investment required to burrow in a muddy substrate, one explanation may be that they farm Spartina. Fiddler crabs sometimes forage in parts of the marsh with no burrows (Montague, 1980a), suggesting that excess burrows are not only for predator escape. The possibility that excess burrows are only for reproduction is equally unlikely, because males may not be able to defend several burrows (where females incubate their eggs; Crane, 1975). To test against these and other possibilities (e.g. parasite avoidance, reduced time constraints, physiological stress, etc.), food levels were manipulated. If fiddler crabs are capable of adjusting their burrowing activity to food availability, burrowing activity was predicted to be lower if food is abundant and higher if food is scarce.






3




The species used in this study is Uca rapax. Work on Uca pugnax will be referred to because of its similarity to U. rapax.
















METHODS




Burrow/crab Ratios in the Salt Marsh


A sampling of Uca rapax population density, sex ratio, size class distribution, and burrow density was done during Summer 1985 in an intertidal salt marsh at Matanzas Inlet, St. Johns County, Florida (29045'50" N, 81017'20 W). The vertical tidal range is .1.9 m, and the distribution of vegetative zones is fairly distinct. The site is dominated by a medium growth form (.50 cm tall) of cordgrass, Spartina alterniflora Loisel. (-140 shoots/m2). The sediment is fine mud with a small sandy fraction.

To identify burrow/crab ratios, I used sampling techniques as described by Frey et al. (1973), Wolf et al. (1975), and Ringold (1979). An area of 12 by 12 m, which was apparently homogeneous in Spartina height and density, Uca density, and physical factors, was chosen. It was divided into 0.5 by 0.5 m quadrats and marked permanently with dowels. Between 9 June and 6 September, 17 quadrats were chosen by the simple random method (Snedecor and Cochran, 1980) for measurement of Uca and burrow density. Each quadrat was enclosed with corrugated fiberglass panels while water still covered the substrate and most crabs were in their burrows (Wolf et al., 1975; Genoni, 1985). At low tide, grass shoots within the quadrat were clipped near the

4





5



substrate. Shoots were counted and later dried 48 h at 700 C and weighed to the nearest 0.1 g with an Ohaus trip balance. To determine the characteristics of each burrow and identify resident crabs, burrow openings were marked with numbered sticks, and burrow casts were taken with polyester resin (Dembowski, 1926; Shinn, 1968). The resin caused resident crabs to escape. Their sex was noted and their carapace breadth was measured to the nearest 0.05 mm with Vernier calipers. Because some burrows are kept plugged by resident crabs at any time (Crane, 1975), this procedure was repeated on the next two days with newly opened burrows. Association of burrows with structural elements in the substrate (grass stems or mussel shells, Geukensia demissa) was estimated by scoring burrows as either occurring on bare substrate (>2 cm from a stem or mussel) or structure-associated (<2 cm) (Bertness and Miller, 1984).

On the third day, burrow casts were removed and their depth, angle from the substrate surface, diameter and configuration were noted. To sample root mat density, a 20-cm deep sediment core was taken with a PVC tube (5.2 cm in diameter). To sample water content and organic content, two 7-cm deep cores were taken with a PVC tube (3.0 cm in diameter). In the laboratory, the larger sediment core was sieved through a 0.5 mm mesh, and the roots were dried 48 h at 700 C and weighed to the nearest 0.01 g with a Mettler H8 analytic balance. One smaller core was weighed wet, dried 48 h at 1100 C and weighed again to determine the water content. The upper 0.5 cm from the other core was weighed wet, dried 48 h at 700 C, then weighed dry and left for 4 h in a Sybron Corp. Thermolyne 10500 muffle furnace at 5000 C. It was





6



then rehydrated, dried again, and weighed. The ash-free dry weight was determined and the organic content was estimated as the weight of the fraction lost in the ashing (Paine, 1971).

Total counts of burrow and crab density, and Spartina

biomass and density, were pooled for all 17 quadrats, whereas data from subsamples were averaged over all quadrats.



Effects of Burrow Density and Food Availability


In a manipulative experiment in the laboratory, the effect of food availability and burrow density on burrowing activity of U. rapax was tested. Two sets of four circular arenas were built, each consisting of a plastic pail on top of which a sheet of metal was riveted to make a funnel with walls inclined at a 450 angle. The metal was lined with plastic and topped with a 10-cm strip of aluminium flashing that crabs could not climb. The arenas were filled with sandy mud (collected from a creek in the salt marsh) 45 cm deep, with an upper diameter of 52 cm and a
2
surface area of 0.21 m Results from the field study were used to determine the initial conditions used in the arenas. At the beginning of each trial, complete randomization (Snedecor and Cochran, 1980) was used in assigning each arena to one of four treatments: (1) low food abundance and low burrow density, (2) low food and high density, (3) high food and low density, and (4) high food and high density. This 22 factorial experiment was replicated in 17 trials, between April 7 and September 14, 1987. The following responses were predicted: burrowing activity should be highest in treatment (1), both to stimulate food and





7




for other needs (protective, reproductive and physiological); intermediate in (2), to stimulate food; and in (3), to meet other needs; and lowest in (4).

Food treatments consisted of 0 g and 11 g, respectively, of Purina Fly Larvae Medium added to each arena. For a 7-day trial, the high value corresponds to 19 kcal/(m2x day), i.e. about twice the highest estimate of average energy consumption of fiddler crabs calculated by Cammen et al. (1980) (3522 kcal/(m2x yr) or 9.7 kcal/(m2x day)). The arenas where food was not added contained some organic matter mixed with the sediment, as evidenced by the feeding response of crabs. This behavior is mediated by chemoreceptors in the minor chelae (Robertson et al., 1980, 1981). To measure the organic content of the upper 0.5 cm of sediment, cores (2.7 cm in diameter) were taken from each arena before each trial, and ash-free dry weight was determined as outlined above.

Initial burrows (43 and 64 per arena, for high and low

density treatments) were made by pushing dowels 15 cm into the substrate, at angles of 60O-80 Thus, burrow densities were 205/m2 and 302/m2. Burrow openings were marked with Spartina stems cut to 10 cm, on which numbered tags were stapled. Additional stems were placed in the low-density arenas to achieve similar stem densities.

Uca rapax were captured at the Matanzas Inlet site and

43 individuals were distributed into each arena. The sex ratio and size class distribution in each arena were similar to those observed in the sampling study, with the exception that no crabs <6 mm in carapace breadth were used, due to the difficulty of





8



surveying their burrows. The following size classes were used, as determined by carapace breadth: 6-8, 8-11, 11-15, and >15 mm. Because competition for burrows of adequate size may be stronger for larger crabs (Bertness and Miller, 1984), the diameter of initial burrows was varied to match crab carapace length (crabs enter their burrows sideways). Crabs were left in the arenas

4.0-7.5 days, at 250C, in a laboratory equipped with windows and with fluorescent lights set on a timer simulating the sunrise-sunset rhythm. Thus, the natural photoperiod was supplemented by an artificial photoperiod. Arenas were separated by blinds from the rest of the laboratory. The sediment was maintained wet by additions of 50 % seawater.

The following parameters were used to describe fiddler crab burrowing activity: the number of new unbranched and branched burrows, the amount of excavated sediment, and the percent of time allocated to burrowing (Katz, 1980; Montague, 1980a). At the end of each trial, excavated material was collected, dried 48 h at 700 C, and weighed to the nearest 0.01 g. Newly excavated burrows were marked. I noted the presence of half-domes of mud (or "shelters"; Zucker, 1974) constructed by some individuals on burrow openings. To determine the characteristics of each burrow and identify resident crabs, burrow casts were made with latex concrete (Christy, 1982). The concrete caused resident crabs to escape. Their sex and carapace breadth were noted. This procedure was repeated with newly opened burrows on the next two days. Burrow casts were removed and their depth, diameter and configuration were noted.





9



Association of burrows with structural elements (grass stems) was estimated as described for the field study.

Counts of new unbranched and branched burrows and

measurements of excavated sediment were divided by the duration of each trial in days. These variables, burrow depths, and branching ratios (defined as the ratio of the number of branches to the number of burrows) were analysed by 22 factorial ANOVAs for tests of the effects of food and density treatments and their interaction. The assumption of normality was tested by two-tailed t-tests on skewness and kurtosis; the assumption of homogeneity of variances was tested by Levene's one-way ANOVA. If these assumptions were not met, factorial ANOVAs were done on log-transformed data (the assumptions were tested again). Comparisons between treatments of the percent of unbranched burrows were done by X2 tests of independence.

Comparisons of responses to initial burrow density and to food availability were also made for each sex and each size class. Crabs of each sex or size class may not be equally sensitive to food supply, because of different foraging efficiencies or food assimilation, or because dominance by one sex or by larger crabs may limit food availability to other crabs (Hyman, 1922; Crane, 1975). Comparisons of shelter building between sexes and size classes was done by X2 tests of independence.

To investigate time allocation to different activities,

including burrowing, crabs were marked individually with numbered plastic tags affixed with cyanoacrylate glue to their carapace. For every arena, activities of individual crabs were recorded





10



from circular visual scans. Successive observations were made every 10 s for 15 min. This procedure was repeated 1 to 4 times, on different days, during 7 of the trials. The percent of the total time allocated to various activities was estimated from the frequencies of their occurrence among the observations (scan sampling; Altmann, 1974). Activities were grouped as follows: standing still (motionless), walKing, eating, cleaning, agonism (combat and chasing), display and courtship, and burrow construction and maintenance. General comparisons between treatments of the fraction of time allocated to various activities were done by X2 tests of independence across all treatments, for all crabs, and for each sex and size class. Comparisons of the fraction of time allocated to certain activities (e.g. burrowing) were done on arcsine-transformed data by factorial ANOVAs. An index of relative activity of crabs of each sex or size class was calculated from the frequency of sightings observed and expected from the sex ratios and size class distribution in the arenas. Activity levels were compared by X2 tests of indepedence.

Statistical analyses were done on an IBM 3081-D32/3090-200 mainframe computer at the North Eastern Regional Data Center of the State University System, Gainesville, FL, using the Statistical Analysis System (SAS Institute Inc., 1985a, 1985b) and the Biomedical Data Program P-Series (Dixon, 1985).
















RESULTS




Burrow/crab Ratios in the Salt Marsh


At the study site, the sediment was sandy mud, and sediment water was 64.59 % of wet weight (s.d.=3.16, N=17). Organic matter in the upper 0.5 cm was 9.05 % (s.d.=2.64, N=12). Spartina shoot density was 138.1/m2 (N=17) and standing stock was 269.52 dry g/m2 (N=15). The root mat (roots and rhizomes), between 0 and 20 cm, was 0.11 dry g/ml (s.d.=0.04, N=15).

There were 212 burrows/m2 (N=17) and 156 fiddler crabs/m2 (N=17). Thus the burrow/crab ratio was 1.36. Sex ratio (M/F) was 0.79 (N=565 crabs). Size class distribution was 31.0 % (carapace breadth <6 mm), 20.6 % (6-8), 17.0 %(8-11), 23.2 % (11-15) and 8.2 o (>15) (N=548 crabs).

Burrows were significantly associated with structural

elements of the substrate; using a value of 70.0 % available bare substrate (Bertness and Miller, 1984), a conservative estimate considering the shoot density at the study site, burrows occurred near stems or mussels more frequently than expected by chance alone (37.0% of all burrows, X2=29.90, df=1, p<0.0001, N=1284). Burrow angles from the substrate ranged from 300 to 900 but most (64.3 %, N=213 burrows) were at an angle between 600 and 800. Most burrows (89.0 %, N=427) were unbranched and J-shaped, their main shaft being usually straight but sometimes comprising 11





12


one or more turns. Other configurations (11.0 % of the burrows) reflected branching or interconnection of burrows, and might be grouped under the designations of Y, U, and H configurations (Allen and Curran, 1974). They had (in decreasing frequency of occurrence) 2, 3, or 4 branches, most of which led to an opening. Burrow depth (projected against an imaginary vertical plane, i.e. not taking into account burrow angle) ranged from 1 to 11 cm (N=94).

These data, to the extent that they were used to set up experimental arenas, are summarized in Table I.



Effects of Burrow Density and Food Availability


Fiddler crabs occupied and remodeled the artificial burrows, and dug additional burrows and branches. Few were seen walking along the walls or burrowing against them. Fiddler crabs dug, on average, 34.99 + 13.30 new burrows per trial (N=68). Of these, 19.88 + 10.52 were unbranched burrows. Thus, the branching ratio for all burrows was 1.25 + 0.58 (N=3340). Another useful variable is the count of all burrows occupied or maintained to the end of each trial, excluding abandoned burrows, which usually collapsed: there were 8.86/day + 1.28 maintained burrows, 11.33/day + 1.68 new branches and 2.44/day + 1.08 branches added to initial burrows. The burrowing activity removed 70.31 dry g/day of sediment (s.d.=17.82), or an average of 1.64 dry g/crab/day. The fraction of time allocated to the various activities is presented in Table IIa. The fraction allocated to burrowing was 5.9 % (N=4628 observations).





13




Table I. (a) Summary of environmental and population characteristics observed at the salt marsh site and used as initial conditions in the experimental arenas; (b) Matching of initial burrow diameter with crab size. CB=carapace breadth.

----------------------------------------------------------------(a)
Variable Field study Arenas
---------------------------------------------------------Sediment type Sandy mud Sandy mud

Sediment water 64.6 % 55.4 %

Burrow density 212/m2 High: 302/m (64/arena) Low: 205/m (43/arena)

Uca density 156/m2 205/m2 (43/arena)

Burrow/crab ratios 1.36 High: 1.5 Low: 1.0

Burrow angles 64.8 % 60o-80 60o-80

Burrow configurations 89.0 % J-shaped Straight

Uca sex ratio (M/F) 0.79 0.79

Uca size-class distribution, by sex CB 6 8 mm 18.7% F, 13.7% M 18.6% F, 14.0% M
CB 8 11 mm 13.7 10.1 14.0 9.3 CB 11 15 mm 17.3 11.5 16.3 11.6 CB >15 mm 6.5 8.6 9.3 7.0

Food available to Uca 9.7 kcal/(m2x yr)(1) High: 19.0 kcal/(m2x yr) Low: 2.6 kcal/(m x yr)

(b)
Size class Carapace length (2) Burrow diameter

CB 6 8 mm 5.44 mm 5.50 mm CB 8 11 mm 7.34 mm 7.40 mm CB 11 15 mm 9.87 mm 9.90 mm CB > 15 mm 11.46 mm 12.20 mm

(1) Cammen et al. (1980).
(2) CL was predicted on the basis of a regression of carapace length against carapace breadth. The morphometric analysis of carapace and chela measurements is presented in the Appendix.









Table II. Time allocation (%) to various activities by Uca rapax; (a) all crabs, (b) each sex and (c) each size class.


---- --------------------------------------------------------------------------------------------Still Eat Walk Burrow Clean Agonism Display N Test df Significance
---- --------------------------------------------------------------------------------------------(a) 36.7 36.0 11.7 5.9 5.3 3.0 1.3 4628
---- --------------------------------------------------------------------------------------------(b) 4493 X2 =131.69 6 p<0.0001 Females 35.5 38.3 12.8 8.3 4.5 0.7 1722 Males 37.2 35.2 10.9 4.1 6.0 4.4 2.2 2771
---- --------------------------------------------------------------------------------------------(c) 4536 X2=168.53 6 p<0.0001 Size class 1 23.4 55.2 5.2 8.4 6.8 0.9 440
2 39.3 37.4 10.9 7.0 3.3 1.4 0.6 905 3 36.9 35.6 12.6 5.0 5.5 2.8 1.6 1777 4 39.8 29.8 12.5 5.3 5.5 4.7 1.8 1414
--- ----------------------------------------------------------------------------------------------





15




Configurations were of the J, Y, U and H types, and sometimes more complex. Most shafts were straight, even those of newly dug burrows, and walls were smooth. Burrow depth was 5.63 + 2.71 cm (N=2597, max.=20.0 cm). There were only 1.84 shelters per arena, on average (N=125). Burrows were significantly associated with structural elements (stems): using a calculated value of 75 % available substrate, burrows were adjacent to stems more frequently than expected by chance alone (28 % of burrows, X2=5.05, df=1, p>0.025, N=634).

Water content of the sediment in arenas was 55.35 %

(s.d.=5.43, N=27), which was lower than that of the salt marsn sediment. However this was the highest water content that allowed construction of artificial burrows. The organic matter present was 7.969 % (s.d.=1.690, N=55). Using a conversion of 0.323 kcal/(m2x yr) for an organic fraction of 1 % calculated from data of Cammen et al. (1980), this value for low food treatments corresponds to 2.62 kcal/(m2x yr). Effect of initial burrow density


The interaction between initial burrow density and food availability was not significant for all variables examined by ANOVAs.

Some of the variables describing burrowing activity were affected by initial burrow density. Results are summarized in Tables III and IVa and Figure 1. The data shown in Table III are collapsed for the two levels of the burrow density treatment, but the value of F is for the main effect of burrow density in the factorial ANOVA. There were no differences between treatments in















Table III. Effect of initial burrow density on burrowing activity of Uca rapax (all crabs).


Variable Low density High density Skewness Kurtosis Levene N Test df Significance

All new burrows 39.63+11.76 30.06+13.24 NS NS NS 17 F=11.13 1 p<0.O014 New unbranched burrows 24.74+ 9.41 14.73+ 9.19 NS NS NS 17 F=19.43 1 p burrows (3) 2.41+ 1.06 2.48+ 1.11 p<0.01 p<0.001 NS 16 F= 0.03 1 NS (1) Maintained branches (3) 11.09+ 1.57 11.58+ 1.79 p<0.001 p<0.001 NS 17 F= 1.50 1 NS (1) Branching ratio 1.27+ 0.58 1.22+ 0.58 p
(1) Log-transformed data.
(2) Arcsine-transformed data.
(3) Divided by the duration of each trial in days.








Table IV. Response to initial burrow density of time allocation to various activities (%) by Uca rapax;
(a) all crabs, (b) each sex and (c) each size class.



Still Eat Walk Burrow Clean Agonism Display N Test df Significance

(a) 4628 X2=37.79 5 p<0.0001 Low density 34.8 39.1 10.3 6.8 4.7 3.3 1.1 High density 38.7 32.9 13.2 5.1 6.0 2.7 1.5
---- ------------------------------------------------------------------------------------------
(b) 2 Females 1722 X =29.96 5 p<0.OOO1 Low density 32.6 42.9 11.3 9.1 3.3 0.8 High density 38.7 33.2 14.4 7.4 5.7 0.6 2 Males 2771 X =18.24 6 p<0.006 Low density 35.4 37.5 9.4 4.9 5.9 5.0 1.9 High density 38.8 33.0 12.3 3.4 6.2 3.9 2.5
---- ------------------------------------------------------------------------------------------
(c) 2 Size class 1 440 X =8.23 5 NS Low density 19.3 56.3 5.3 9.1 8.0 1.5 High density 28.8 53.7 5.1 7.3 5.1 - 2 Size class 2 905 X =14.93 6 p<0.021 Low density 38.2 38.0 9.6 9.2 2.5 1.8 0.8 high density 40.8 36.7 12.8 4.1 4.3 1.0 0.3 2 Size class 3 1777 X =35.55 6 p<0.0001 Low density 33.4 41.8 12.0 4.3 5.2 2.3 0.5 High density 39.9 30.4 13.2 5.6 5.8 2.6 2.6 2 Size class 4 1414 X =17.24 6 p ----------------------------------------------------------------------------------------





18



the number of maintained burrows, the number of branches added to initial burrows and branches of new burrows, the amount of excavated sediment, or burrow depth. However, in arenas with a lower initial burrow density, there were significantly more newly excavated burrows and the branching ratio was significantly higher; there was a trend for an increase in the fraction of time allocated to burrowing.

Differences between the reported fraction of time allocated to burrowing in Tables III and IVa (and in comparisons presented below) are the consequence of averaging across all treatments and averaging by factorial univariate analysis, respectively.


Effect of food availability


Some of the variables describing burrowing activity were

affected by the food treatment. Results are summarized in Tables V and VIa and Figure 1. The data shown in Table V are collapsed for the two levels of the food treatment, but the value of F is for the main effect of food in the factorial ANOVA. There was no difference between treatments in the number of newly excavated burrows, the total number of maintained burrows, the amount of excavated sediment or the fraction of time allocated to burrowing. However, in arenas receiving no food additions, there were significantly more branches added to initial burrows and branches of new burrows, the branching ratio was higher, and average burrow depth was lower. In addition, time allocation to various activities was significantly affected by the treatment: in low-food arenas, crabs spent less time eating, wandered more












Table V. Effect of food availability on burrowing activity of Uca rapax (all crabs).


---- -------------------------------------------------------------------------------------------------- -Variable Low food High food Skewness Kurtosis Levene N Test df Significance
---- ---------------------------------------------------------------------------------------------------All new burrows 39.26+13.57 30.71+11.73 NS NS NS 17 F= 8.91 1 p<0.O04 New unbranched burrows 20.88+10.20 18.88+10.89 NS NS NS 17 F= 0.78 1 NS Maintained burrows (3) 8.92+ 1.31 8.79+ 1.27 p<0.001 NS NS 16 F= 0.15 1 NS (1) New branches to initial
burrows (3) 2.96+ 1.10 1.92+ 0.78 p<0.01 p<0.001 NS 16 F=17.68 1 p<0.0001 (1) Maintained branches (3) 11.90+ 1.78 10.75+ 1.38 p ---- ------------------------------------------------------------------------------------------------------(1) Log-transformed data.
(2) Arcsine-transformed data.
(3) Divided by the duration of each trial in days.








Table VI. Response to food availability of time allocation to various activities (%) by Uca rapax;
(a) all crabs, (b) each sex and (c) each size class.

---- --------------------------------------------------------------------------------------------Still Eat Walk Burrow Clean Agonism Display N Test df Significance
-------------------------------------------------------------------------------------------(a) 4628 X2=134.58 6 p<0.0001 Low food 40.6 29.5 13.7 6.0 4.4 4.0 1.8 High food 32.4 43.1 9.5 5.9 6.4 1.8 0.8
--- ---------------------------------------------------------------------------------------------(b)
Females 1722 X2=35.43 5 p --- ---------------------------------------------------------------------------------------------(c)
Size class 1 440 X2=25.01 5 p<0.0001 Low food 32.7 47.1 5.3 7.2 5.8 1.9 High food 15.1 62.5 5.2 9.5 7.8 Size class 2 905 X2=26.29 6 p<0.0001 Low food 37.4 34.9 13.4 8.9 2.1 2.6 0.6 High food 41.4 40.2 8.3 4.8 4.6 0.2 0.5 Size class 3 1777 X2=55.83 6 p<0.0001 Low food 40.8 29.1 14.3 4.4 4.8 4.1 2.6 High food 33.1 42.0 11.0 5.6 6.3 1.5 0.7 Size class 4 1414 X2 =63.50 6 p --- ----------------------------------------------------------------------------------------------































Figure 1. Effect of initial burrow density and food availability on burrowing activity of Uca rapax (all crabs) (N=17,. Bars represent 1.96 standard error (95 % confidence intervals).







Total new burrowsNew branched burrows New unbranched burrows All maintained branches All maintained burrows w Branches of initial burrows 3o 40 o Z
Sa: 30C3O
mo 20
z
0 10



W 60

< 40
x
W J20



D 1.5Z o
I
S 1-- 1.0 z <
Cr 0.50
6-
SW
m o



(D 8.



a: 4
I 2


LOW LOW HIGH HIGH FOOD
LOW HIGH LOW HIGH DENSITY





23



(presumably in search of food), and agonistic encounters were more frequent than in high-food arenas.


Comparison of burrowing by crabs of each sex


There was no overall difference between females and males in the percent of unbranched burrows or branching ratio of burrows. However, females were less active at the surface, allocated different amounts of time to various activities, spent significantly more time burrowing, and burrowed less deeply (Tables VII and IIb). Males constructed more shelters than did females (Table VIII), and allocated significantly more time to agonism (F=27.47, df=1, p<0.0001, N=53) and to display (F=14.56, df=1, p<0.0004, N=53) (Table IIb).


Response of crabs of each sex to initial burrow density


The three-level interaction between food, burrow density, and sex, as well as the two-level interactions between pairs of these factors, were not significant for all variables examined by ANOVAs.

The effect of initial burrow density did not differ between males and females with respect to burrow configuration, branching ratio, depth, or time allocation to different activities, or to burrowing in particular. This is suggested by a qualitative comparison between the two sexes in Tables IX and IVb, and by the non-significant interaction between sex and burrow density for each variable examined. However, in arenas with a lower initial burrow density, the trend for allocating a higher fraction of time to burrowing was stronger for females.












Table VII. Comparison of burrowing activity by each sex of Uca rapax.


-------- ---------------------------------------------------------------------Variable Females Males N Test df Significance
-------------------------------------------------------------------% unbranched burrows 64.94 % 66.53 % 1391 X =0.13 1 NS Branching ratio 1.38+ 0.58 1.37+ 0.59 1385 F= 0.12 1 NS (1) Burrow depth (cm) 6.04+ 2.71 6.60+ 2.83 1142 F212.46 1 p<0.0004 (1) Activity at surface 0.69 1.40 4493 X =560.88 1 p ------- ----------------------------------------------------------------------(1) Log-transformed data.
(2) Arcsine-transformed data.





25









Table VIII. Fraction of the total numbers of shelters constructed by Uca rapax, for each sex and each size class. The interaction between sex and size was not significant.



% N Test df Significance

64 X2=36.04 1 p<0.0001 Females 18.8 Males 81.3

62 X2=51.48 3 p 2 9.7 3 29.0 4 48.4












Table IX. Response of burrowing activity by each sex of Uca rapax to initial burrow density.

------- --------------------------------------------------------------------Variable Low density High density N Test df Significance
----------------------------------------------------------------------------------(a) Females 2 % unbranched burrows 64.62 % 65.85 % 653 X =0.11 1 NS Branching ratio 1.40+ 0.57 1.37+ 0.54 653 F= 0.27 1 NS (1) Burrow depth (cm) 6.03+ 2.76 6.07+ 2.64 532 F= 0.21 1 NS (1) % time burrowing 12.727% 5.79 % 26 F= 3.27 1 NS (2)
(b) Males 2 % unbranched burrows 65.05 % 68.82 % 732 X =1.17 1 NS Branching ratio 1.40+ 0.59 1.35+ 0.55 732 F= 1.63 1 NS (1) Burrow depth (cm) 6.44+ 2.62 6.82+ 3.06 610 F= 2.30 1 NS (1) % time burrowing 5.33 % 3.09 % 27 F= 2.02 1 NS (2)
------- --------------------------------------------------------------------(1) Log-transformed data.
(2) Arcsine-transformed data.





27



Response of crabs of each sex to food availability


Crabs of the two sexes were not affected differently by food availability, with respect to percent of unbranched burrows, branching ratio, depth, time allocation to different activities, or fraction of time allocated to burrowing (Tables X and VIb). This is suggested by a qualitative comparison between the two sexes in Tables X and VIb, and by the non-significant interaction between sex and food for each variable examined. Comparison of burrowing by crabs of different size


There was no overall difference between crabs of different sizes in the percent of unbranched burrows or branching ratio of burrows (Table XI). Smaller crabs were less active at the surface, allocated different amounts of time to various activities and burrowed less deeply (Tables IX and IIc). There was a strong trend for a higher time investment in burrowing by smaller crabs. Larger crabs constructed more shelters than did smaller crabs (Table VIII), and allocated significantly more time to agonism (F=9.24, df=3, p<0.0001, N=108) and to display (F=4.51, df=3, p<0.005, N=108). There was a strong trend for a higher time expenditure in foraging for the smaller size classes.










Table X. Response of burrowing activity by each sex of Uca rapax to food availability.


Variable Low food High food N Test df Significance

(a) Females
% unbranched burrows 57.93 % 71.80 % 653 X =13.82 1 p<0.0001 Branching ratio 1.46+ 0.57 1.32+ 0.54 653 F=11.15 1 p (b) Males
% unbranched burrows 60.72 % 72.65 % 732 X =11.74 1 p
(1) Log-transformed data.
(2) Arcsine-transformed data.












Table XI. Comparison of burrowing activity by Uca rapax of different sizes.


---- ----------------------------------------------------------------------------------------------------Variable Size class 1 Size class 2 Size class 3 Size class 4 N Test df Significance
---- --------------------------------------------------------------------------------------------------$ unbranched burrows 68.73 % 61.47 % 65.70 % 69.41 % 1399 X =6.17 3 NS Branching ratio 1.31+ 0.54 1.43+ 0.58 1.39+ 0.60 1.35+ 0.59 1393 F= 2.21 3 NS (1) Burrow depth (cm) 5.07+ 2.03 5.48+ 2.15 6.87+ 2.79 8.06+ 3.18 1149 F263.99 3 p<0.0001 (1) Activity at surface 0.10 0.20 0.39 0.31 4536 X =1573.78 3 p ---- ----------------------------------------------------------------------------------------------------(1) Log-transformed data.
(2) Arcsine-transformed data.





30



Response of crabs of different size to initial burrow density


The three-level interaction between food, burrow density,

and size, as well as the two-level interactions between pairs of these factors, were not significant for most variables examined by ANOVAs. It was significant (F=2.11, df=3, p<0.1, N=255) for the interaction of size with initial burrow density for the percent of unbranched burrows.

Crabs of different sizes were affected differently by

initial burrow density (Tables XII and IVc). This is suggested by a qualitative comparison between size classes in Tables XII and IVc, and by the significant interaction between size and burrow density for the percent of unbrancned burrows. In arenas with a lower initial burrow density, for crabs of the smallest size class, branching ratio and burrow depth were significantly lower; the time allocation to burrowing and other activities was not affected by the treatment. For some of the larger crabs, time allocation to various activities was significantly affected by the treatment; the fraction of time allocated to burrowing was higher in arenas with a lower initial burrow density. No differences occurred for branching ratio and depth. Response of crabs of different size to food availability


Crabs of different sizes were affected differently by food availability (Tables XIII and VIc). This is suggested by a qualitative comparison between size classes in Tables XIII and VIc. However, variance was high, and the interaction between size and food availability was not significant for any variable











Table XII. Response of burrowing activity by Uca pugnax of different sizes to initial burrow density.


Variable Low density High density N Test df Significance

(a) Size class 1
% unbranched burrows 64.00 % 74.68 % 304 X =4.08 1 p<0.0043 Branching ratio 1.39+ 0.54 1.40+ 0.55 304 F= 3.88 1 p<0.0497 (1) Burrow depth (cm) 4.72+ 1.92 5.39+ 2.05 226 F= 6.63 1 p<0.0106 (1) % time burrowing 11.487% 13.72 % 27 F= 0.01 1 NS (2)
(b) Size class 2 2 % unbranched burrows 60.28 % 62.98 % 422 X =0.32 1 NS Branching ratio 1.46+ 0.61 1.40+ 0.55 422 F= 0.87 1 NS (1) Burrow depth (cm) 5.59+ 2.32 5.36+ 1.96 345 F= 0.17 1 NS (1) % time burrowing 9.63-% 2.74 % 27 F= 5.13 1 p<0.03 (2)
(c) Size class 3
% unbranched burrows 66.07 % 65.61 % 413 X =0.01 1 NS Branching ratio 1.40+ 0.60 1.38+ 0.56 413 F= 0.04 1 NS (1) Burrow depth (cm) 6.71+ 2.56 7.10+ 3.00 350 F= 1.00 1 NS (1) % time burrowing 3.12 % 4.60-% 27 F= 0.02 1 NS (2)
(d) Size class 4
% unbranched burrows 70.59 % 68.64 % 254 X =0.11 1 NS Branching ratio 1.33+ 0.55 1.37+ 0.60 254 F= 0.28 1 NS (1) Burrow depth (cm) 7.93+ 2.88 8.26+ 3.47 225 F= 0.18 1 NS (1) % time burrowing 7.21--% 3.55 % 27 F= 1.46 1 NS (2)

(1) Log-transformed data.
(2) Arcsine-transformed data.











Table XIII. Response of burrowing by Uca rapax of different sizes to food availability.


Variable Low food High food N Test df Significance ^--- -- -- -- -- -- -- -- -- -- -- --- --4 -- -- -- -- -- -- -- ------ -----
(a) Size class 1 2 % unbranched burrows 57.14 % 79.88 % 304 X =18.39 1 p<0.O001 Branching ratio 1.46+ 0.55 1.22+ 0.46 304 F=17.76 1 p<0.001 (1) Burrow depth (cm) 4.61+ 1.82 5.42+ 2.10 226 F= 8.60 1 p<0.0037 (1) % time burrowing 5.01 % 20.69 % 27 F= 2.82 1 NS (2)
(b) Size class 2 2 % unbranched burrows 54.71 % 69.35 % 422 X =9.53 1 p<0.002 Branching ratio 1.51+ 0.60 1.34+ 0.54 422 F= 9.16 1 p<0.0026 (1) Burrow depth (cm) 5.27+ 2.05 5.75+ 2.24 346 F= 3.55 1 NS (1) % time burrowing 8.94 % 5.21 % 21 F= 0.04 1 NS (2)
(c) Size class 3 2 % unbranched burrows 62.83 % 68.47 % 413 X =1.45 1 NS Branching ratio 1.42+ 0.58 1.36+ 0.58 413 F= 1.04 1 NS (1) Burrow depth (cm) 6.47+ 2.47 7.24+ 2.96 350 F= 4.99 1 p<0.0261 (1) % time burrowing 3.94 % 3.72 % 27 F= 0.04 1 NS (2)
(d) Size class 4 2 % unbranched burrows 66.38 % 72.46 % 254 X =1.10 1 NS Branching ratio 1.41+ 0.62 1.30+ 0.52 254 F= 1.79 1 NS (1) Burrow depth (cm) 7.38+ 2.40 8.66+ 3.58 226 F= 6.20 1 NS (1) % time burrowing 4.61 % 6.07 % 27 F= 0.10 1 NS (2)

(1) Log-transformed data.
(2) Arcsine-transformed data.





33



examined. In arenas with a lower food availability, for size classes 1 and 2, branching ratio was higher, and burrow depth was lower (although the difference was not significant in size class 2). For size class 3, burrow depth was also significantly lower. No differences occurred between treatments for the largest size class.
















DISCUSSION




Burrow/crab Ratios in the Salt Marsh


The burrow/crab ratio was found to be higher than 1; thus, there were excess burrows. Burrow/crab ratios higher than one have been found in previous studies. Both male and female crabs dig new burrows with each tidal cycle. Burrows are relatively stable in a muddy substrate with a root mat, particularly in the medium Spartina zone. When abandoned, they persist through several tidal cycles (Powers, 1973; Crane, 1975; Basan and Frey, 1977; Hyatt and Salmon, 1977; Katz, 1980; Bertness and Miller, 1984).

Spartina shoot density and biomass, and sediment organic content were reasonably high though comparable to values for other medium Spartina marshes (Kurz and Wagner, 1957; Basan and Frey, 1977; Cammen et al., 1980; Gosselink et al., 1984) and root mat density was relatively low (C.S. Hopkinson, pers. comm.). Sediment water content was very similar to that found in a salt marsh in Georgia, and indicative of a relatively firm substrate and high burrow holding capacity (Teal and Kanwiser, 1961; Kraeuter and Wolf, 1974; Bertness and Miller, 1984). All these characteristics, together with a low degree of burrow intersection (discussed below; see also Basan and Frey, 1977) are "typical" of medium Spartina marshes. These characteristics 34





35



should allow variations in crab burrowing activity in response to food supply, because they help maintain a high enough Uca population density to have a significant impact on Spartina. Furthermore, burrows in this type of substrate may last long enough to help maintain a burrow/crab ratio >1 and help enhance Spartina growth.

Uca rapax density was reasonably high compared to reported densities (Allen and Curran, 1974; Wolf et al., 1975; Aspey, 1978; Katz, 1980; Montague et al., 1981; Bertness and Miller, 1984; Cammen et al., 1984). The sex ratio (M/F) was 0.79. Sex ratios higher than 1 are often observed (Wolf et al., 1975; Bertness and Miller, 1984; Genoni, 1985), but this may be due partly to sampling biases, as females often burrow more deeply than males and move closer to the water edge in the days before the release of larvae (Schwartz and Safir, 1915). There are also discrete portions of the marsh where males display (display areas; Greenspan, 1980); Bertness and Miller (1984), however, noted that sex ratios may be lower in a medium Spartina zone than in other areas of the same marsh, presumably because the difference in burrowing ability between females and males is broader in the substrate of this zone. As for size class distribution, it was more skewed toward small sizes than in the study by these authors. However, population structure can vary remarkably between seasons and years in a salt marsh (Shanholtzer, 1973; Cammen et al., 1984; Genoni, 1985).

Maximum burrow depth was found to be 11 cm, which might be an underestimate due to the limitations of the casting technique: the presence of water in burrows sometimes prevented the resin





36


from filling them completely. Katz (1980) found burrows as deep as 15 cm, and Bertness (1985) 20 cm, a value also attained in the arenas in this study. Basan and Frey (1977) found burrows 25 cm deep. Fiddler crabs dig even more deeply in winter (Crane, 1975). The significant association of burrows with structural elements of the substrate is probably related to the relatively low root mat density, which helps reduce the probability of burrow collapse (Bertness and Miller, 1984). Burrow configurations reflected little intersection and branching, and most burrows were J-shaped, with a straight shaft, an enlarged terminal chamber and irregular walls. Other burrows were branched or interconnected. This low degree of branching and interconnection may be related to the relatively high burrow holding capacity of the substrate (Bertness and Miller, 1984), and the irregularity of walls to the protrusion of plant roots. Allen and Curran (1974), Basan and Frey (1977) and Katz (1980) give similar descriptions of configurations. These include nearly straight (J-shaped), highly sinuous shafts, Y- or U-shaped systems and more complex systems with straight shafts or U-shaped components; the complexity decreases from the marsh edge to the higher marsh. Most burrows have an enlarged lower end, where the resident crab presumably remains during the high tide or when molting or mating.



Effects of Burrow Density and Food Availability


Uca rapax also dug excess burrows. In all treatments,

fiddler crabs dug new burrows despite the presence of preexisting





37



unoccupied burrows. Additionally, they adjusted their burrowing activity to burrow density and food availability. These crabs are known to adjust their burrowing activity to a variety of conditions, such as stem density, root mat density, substrate water, ground temperature, tidal and diurnal zeitgebers, reproductive activity, threat by potential predators, season, and male display activity, itself a function of many of these factors (Zucker, 1974; Allen and Curran, 1974; Ringold, 1979; Bertness, 1985; L.W. Powers, pers. comm.). The results presented here suggest that the digging of excess burrows and burrow branches varies inversely with food availability. Thus, part of the benefit of burrowing may be to farm Spartina.

Organic content in the surface layer of untreated arenas was lower (88 %) than the average organic content in the marsh, presumably because of the lack of macroscopic detritus. Yet this value is within the range observed in the marsh. Since fiddler crabs are food limited in the medium marsh (Genoni, 1985), this value was presumed to indeed reflect a low food availability.

Average and maximum burrow depth in arenas were greater than those found in the salt marsh, but this may simply reflect a more successful casting due to a lower water content. In fact, depths observed in the arenas are comparable to the ones found in the salt marsh studies cited above. The initial depth of artificial burrows (15 cm) was presumably adequate, since crabs did not remodel to greater depths.

The significant association of burrows with structural

elements is probably related to the lack of a root mat, which makes the need of structural support for burrows particularly





38



critical. Burrow configurations were similar to those observed in the marsh, and could be assigned to the same categories described above. However, burrow walls were smoother, due to the lack of a root mat. The lack of a root mat would presumably reduce burrow longevity. However, longevity in bare sediment was found by Bertness and Miller (1984) to be in excess of 1 wk. Since each trial was ended at or before 7.5 d, results should not be significantly biased by the collapse of abandoned burrows.

Fiddler crabs allocated 5.9 % of their activity to

burrowing, which is probably an underestimate due to the visual sampling technique, since burrowing activity will continue below the surface. The display activity seemed low, perhaps due to the lack of display areas in the confinement of the arenas; to insufficient time to establish display activity; to the reduced light intensity in the laboratory; or to the phase of the tidal cycle (Greenspan, 1980). Although season affects burrowing activity (Crane, 1975), this did not complicate the present study since field and laboratory experiments were conducted during the same season.


Effect of initial burrow density


Fiddler crabs dug new burrows and branches even in arenas with a high initial burrow density. A lower initial burrow density, however, resulted in a higher number of branched and unbranched burrows. There was a trend for an increase in time spent burrowing and for less deep burrows, suggesting that time for burrowing was allocated to digging new burrows and branches rather than to making existing burrows deeper. Burrows were,





39



then, abandoned more often in low density arenas than in high density arenas, and burrow turnover was higher. This sensitivity to the existing burrow density may reflect a tendency to dig burrows wherever possible. New burrows in a substrate with a low density may be less likely to collapse, thus making excavation of new burrows beneficial to the crabs. A second reason may be a stronger competition for territories under higher burrow density. Uca are responsive to the presence of other crabs, and are more likely to colonize or burrow in areas with lower apparent crab densities, such as areas with vegetation cover (Bertness and Miller, 1984; Vliet, 1981). Some species, including Uca rapax, construct half-domes or shelters. One function of these may be to reduce their visual contact with other males, thereby reducing aggression (Zucker, 1974, 1981).


Effect of food availability


Fiddler crabs dug more burrow branches, but average burrow depth was lower, in arenas receiving no food additions. Thus, time investment in burrowing was similar, but it was allocated to digging new burrow branches rather than digging new burrows or making existing burrows deeper. Digging burrow branches near the surface may better stimulate Spartina growth. The physico-chemical and nutrient requirements in the upper substrate layers may be most critical to Spartina growth. Indeed, in the upper layer there is a higher density of roots and rhizomes (Valiela et al., 1976; Howes et al., 1981; Hopkinson and Schubauer, 1985), and nutrients may be more easily extracted from sediment (DeLaune and Patrick, 1980). Furthermore, in the upper





40



layer, oxygenation by Spartina (metabolic oxidation and passive 02 release), nutrient inputs, or burrows have a stronger effect on these parameters (Howes et al., 1981; Bertness, 1985). In addition, digging branches near the surface may better enable crabs to keep control of the original burrow, thus reducing the risk of losing it to another crab, at the same time reducing the risk of being without a burrow should the new burrow collapse or encounter some construction obstacle. These constraints may be particularly critical where food is limiting. In contrast, in the low density situation, both unbranched and branched burrows are excavated.


Comparison of burrowing by crabs of each sex


Females spent approximately half as much time outside their burrows as did males. This observation may be an artifact of the visual sampling technique, as females are actually more involved in burrowing, removing them from view. Also they are less involved in agonistic and courtship activities. Additionally, gravid females need to protect their brood. A higher burrow turnover rate may be expected for females, because (a) they have an increased food requirement (see below); (b) they may be dislodged by large males, which are less able to burrow due to their large claw (Pearse, 1914; Hyatt and Salmon, 1977; Bertness and Miller, 1984); (c) they allocate less time to agonistic and courtship activities; and (d) non-receptive females dig more if forced to interact with displaying males (L.W. Powers, pers. comm.), an artifact of the arenas.





41



Males burrowed more deeply and constructed more shelters, which agrees with findings of Crane (1975), Zucker (1971, 1974) and Bertness and Miller (1984). Response of crabs of each sex to initial burrow density


In arenas with a lower initial burrow density, females dug more burrows and remained more in their burrows. Because of their higher burrowing ability, females may be more likely to take advantage of a low burrow density by digging new burrows. Females were already spending more time burrowing than males but tended to increase their burrowing time further. Males defend display territories and breeding burrows (Greenspan, 1980), and may be also sensitive to perceived population density. Indeed, most shelters were constructed by males. Because females are less involved in agonistic and courtship activities, they may respond to a high population density by digging fewer burrows and remaining more in their burrows rather than by constructing shelters as males do.


Response of crabs of each sex to food availability


There were no differences between sexes in the effect of food availability on various parameters of burrowing activity. Thus, food availability seemed to affect the two sexes equally. Comparison of burrowing by crabs of different size


Smaller crabs were generally less active outside their burrows, which may be due partly to an artifact of the visual sampling technique, as small crabs tended to be more involved in





42



burrowing, thus being visible less frequently, and to their being less involved in agonistic and courtship activities. A higher burrow turnover rate may be expected for smaller crabs, because

(a) they have an increased food requirement, (see below),

(b) they lose their burrows more often to larger, competitively dominant, crabs (Pearse, 1914; Hyatt and Salmon, 1977; Bertness and Miller, 1984); (c) they do not allocate as much time (or any at all in the case of the smallest ones) to agonistic and courtship activities; (d) burrow longevity is lower for smaller burrows when there is no root mat (Bertness and Miller, 1984); and (e) small males may dig more if forced to interact with displaying males (L.W. Powers, pers. comm.), an artifact of the arenas.

Larger crabs burrowed more deeply and constructed more shelters, which agrees with findings of Crane (1975), Zucker (1971, 1974) and Bertness and Miller (1984). Smaller crabs, even though they tended to allocate more time to burrowing, can excavate only small amounts of sediment at a time. Response of crabs of different size to initial burrow density


Time allocation to burrowing was increased for larger crabs in arenas with a low initial burrow density, but not by smaller crabs. Smaller crabs already tend to spend more time burrowing than larger crabs and may not be able to increase their burrowing further. Rather, smaller crabs respond by burrowing less deeply and allocating their burrowing time to new burrows. In addition, sensitivity to burrow density may be particularly high in small crabs because of their relatively high population density.





43



Because small crabs are less involved in agonistic and courtship activities, they may respond to a high population density by digging more unbranched and branched burrows and remaining more in their burrows, rather than by reducing their time allocation to burrowing and constructing shelters as larger crabs do.


Response of crabs of different size to food availability


Smaller crabs burrowed more under low food availability, but again dug more branches rather than increasing their time allocated to burrowing. Small crabs may be particularly sensitive to food availability because they may stand more chance of deriving benefit from their investment during their lifetime; because they can invest more time in burrowing as opposed to displaying, etc.; because their better burrowing ability may allow them to better adjust their burrowing to food availability; and because their need for food may be greater for growth. Early stages are particularly sensitive to food availability: megalops larvae may select a settling substrate partly on the basis of its organic content (Crane, 1975) and early crab stages may be more limited by food supply (Genoni, 1985). Given these reasons, smaller crabs may be more efficient at farming Spartina. Furthermore, they constitute a large fraction of the population.



General Considerations


As with many complex biological interactions, the observed results might be attributable to secondary interactions, or to indirect cause-and-effect relationships. For example, burrowing





44



activity might have been affected by relative differences in physical conditions or in microtopography caused by the density of initial burrows or the addition of food flakes. However, in light of what is presently known about Uca rapax burrow dynamics, the effects of their burrows on Spartina growth, in addition to the fact that they are food limited, the results would collectively suggest a farming relationship.

Although fiddler crabs are presumably capable of responding to Spartina detritus availability, the food treatment used in this study consisted, instead, of artificial food. An attempt was made at using Spartina detritus, derived from standing dead Spartina that was collected, dried, thin-cut in a hammer mill, and aged in pens in the salt marsh for 8-12 weeks. Microorganism load was measured as the gross primary production (GPP) of the microbial population: the release of CO2 was monitored with an Infra-Red Analyzer, and GPP was calculated by adding the slopes of CO2 concentration in the dark (community respiration) and in the light (net community production) (Soeder and Talling, 1969). Microorganism load was extremely variable (by two orders of magnitude) between samples, and therefore detritus was deemed unsuitable as a controlled food source. This also assumes that Spartina detritus, albeit limiting in quantity, provides a suitable substrate for microorganisms, which constitute the food of Uca rapax. Conversion of detritus to microbial biomass is relatively high (Gosselink and Kirby, 1974; Haines and Hanson, 1979). Indeed, it constitutes a significant portion of the diet of these crabs (Shanholtzer, 1973; Haines, 1976; Montague, 1980b). Thus, the benefit of obtaining an increased detritus





45



supply may not be merely a by-product of their burrowing activity (Montague, 1980a). The increased investment in burrowing under low food supply may be cost-effective because of the indirect benefit accruing to the crabs.

It would be interesting to compare responses to food

availability of crabs of different sizes, in the salt marsh. If small crabs are more sensitive to food limitation, and respond to low food availability by increasing their burrowing, then their burrowing activity should be inversely proportional to their organic content. Additionally, if smaller crabs are more efficient at farming Spartina, selective removal of large or small crabs should have different effects on Spartina production. The sand fiddler crab, Uca pugilator (Bosc) would provide a useful comparison, as it often occurs sympatrically with Uca rapax or Uca pugnax in salt marshes. Burrows of Uca pugilator are excavated in sandy substrates, and have usually one opening. They include short temporary burrows (that last one tidal cycle) and deep breeding burrows, the latter resembling, in their configuration, those of Uca rapax (Basan and Frey, 1977; Christy, 1982). These characteristics (short life of one kind, use for breeding of the other kind) suggest that a flexibility in adjusting time allocation to burrowing, burrow density, branching and depth may not be as high as in Uca rapax. Moreover, since this species feeds mainly on benthic algae (Miller, 1961), it presumably does not farm grass, and although it does dig excess burrows (L.W. Powers, pers. comm.), it may be predicted not to do so in response to food limitation.
















APPENDIX
MORPHOMETRIC ANALYSIS OF CARAPACE AND CHELA PROPORTIONS IN UCA RAPAX




Morphometric analyses in decapods have generated much interest because of their relevance to many facets of life history and their taxonomic diagnostic value (Teissier, 1960; Barnes 1968; Hartnoll, 1982). Allometric constraints affect growth and reproductive output as well as many aspects of basic physiology (e.g. Miller, 1971; Savage and Sullivan, 1978; Hines, 1982). In burrowing animals such as Uca, they may affect burrowing behavior and mating strategies: they allow predictions of minimal burrow diameter (Bertness and Miller, 1984) or largest mate (Uca rapax breed in the male's burrow; Greenspan, 1980).

This morphometric analysis correlates the basic

measurements, carapace breadth (or width) and carapace length (or depth). Carapace breadth is often used as the reference dimension for morphometric studies in brachyurans (see e.g. Barnes, 1968; Hines, 1982; Davidson and Marsden, 1987). However, in the case of Uca, carapace length is more satisfactory (Miller, 1971; Crane, 1975). I have used both of these dimensions. The development of the major cheliped in males and its role in territorial defence, combat, display and courtship, are of special interest in fiddler crabs (Crane, 1975). Indeed it is a useful measurement for associating male size to mate size and for predicting mating success of males (Miller, 1971; 46





47



Greenspan, 1980). Regressions of propodus length and dactylus length against carapace breadth and length were calculated. The slopes and elevations of regression lines were compared between the two sexes.

The analysis includes 330 males and 308 female Uca rapax, ranging from 4.50 to 21.50 mm in carapace breadth. Carapace breadth (CB), carapace length (CL), propodus length (PL) and dactylus length (DL) were measured with vernier calipers to the nearest 0.05 mm. CB was measured across the long axis of the carapace at its maximum breadth; CL was measured medially along the minor axis of the carapace, from the front to the posterior margin; PL was measured along its lower margin, from the articulation with the carpus to the tip of the immovable digit; and DL from the articulation point on the propodus to the tip (Crane, 1975). Linear regressions are good approximations of correlations between dimensions of similar magnitude (Hartnoll, 1982). Linear regressions of the form y=ax+b, where y is the dependent variable, x the reference variable, and a and b are constants (Sokal and Rohlf, 1969) were done for CB vs. CL and CL vs. CB for females and males. Because the growth of the major cheliped in males is likely to follow a power function of the form y=axb (Hartnoll, 1982), regressions of log(PL) and of corresponding power function was derived. The adjusted coefficient of determination (r 2) and the standard error on the estimate of the slope were calculated. Sexual differences in slope and elevation between regression lines were compared by t-tests (Zar, 1974).










The relationships between carapace breadth and length

approximate closely straight lines (Figures 2 and 3) while the ratio of male chela measurements to carapace breadth and length follows a logarithmic straight line (Figures 4 and 5), as in some other Ocypodidae (Barnes, 1968) (Table XIV). Similarly, the carapace breadth is proportionally larger in larger individuals. It can also be noted from comparisons of the standard errors on the estimates of slopes that for any given carapace breadth, the size of the male chela shows a greater variation than carapace length.

Regressions for carapace measurements of females and males were significantly different in slope and elevation (t=4.947 and

5.945 respectively, p<0.001, df=635).

Comparison of the standard errors on the estimates of the slopes suggest that the measurement for the major cheliped exhibit more variation than carapace measurements. This may be due entirely or in part to smaller, regenerating chelipeds in some individuals, and has been observed in other Ocypodidae (Barnes, 1968). However, the range and rate of change of these morphometric proportions may have taxonomic or biogeographical significance (Barnes, 1968).





49







1.6



1.4 F- -ALL 'M

1.2
E

1.0

-j
w 0.8
a

< 0.6


0.4 M

ALL~ F
0.2.



0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2

CARAPACE BREADTH (cm)


Fi ure 2. Regression of carapace length against carapace breadth for Uca rapax. F=females, M-=males, ALL=All crabs.




50







2.2 M
2.2 -ALL

2.0 F

1.8


E 1.6

I 1.4




1.0 a: 0.8


0 .4


04ALL- M

0.2



0.2 0.4 0.6 0.8 I.0 1.2 1.4 1.6 CARAPACE LENGTH (cm)


Figure 3. Regression of carapa-e breadth against carapac length
for Uca rapax. F=females, M--males, ALL=all crabs.




51




0.5 / Log (PL) Log (D L)

0.3 0.1I




-0.1




-0.3




-0.5







-0.2 0.0 0.2 0.4 0.6 Loa(CARAPACE BREADTH)
Figure 4. Regression o g (PL) (propodus lengtn) and lo (DLj (dactylus length) against log (CB) (carapace breadth) for Uca rapax.




52




0.5 Log(PL) Log(PL)


0.3




0.I




-0.1




-0.3




-0.5




-0.7

-0.4 -0.2 0.0 0.2 0.4 Log(CARAPACE LENGTH)

Figure 5. Regression of log (PL) (propoa us length) and lol (DL) (dactylus length) against log (CL) (carapace length) for Uca rapax.





53


Table XIV. Summary of allometric relationships for carapace breadth
(CB), carapace length (CL), propodus length (PL) and dactylus length
(DL) of Uca rapax. s.e.=standard error of estimate on the slope.



Type of 2
Sex Variables regression Growth equation s.e. r N

F CL vs. CB linear y=0.6504x+0.0245 0.007 0.9691 308 M y=0.6253x+0.0418 0.005 0.9783 330 All y=0.6330x+0.0378 0.004 0.9739 638

F CB vs. CL linear y=1.4901x-0.0030 0.015 0.9690 308 M y=1.5645x-0.0394 0.013 0.9783 330 All y=1.5386x-0.0282 0.010 0.9739 638
-0.0451
M PL vs. CB power =1.8416x041 0.055 0.9339 79 M PL vs. CL power y=1.9680x2840 0.063 0.9236 79 M DL vs. CB power y=2.1156x 0.064 0.9360 79 y=2.56x0.1172
M DL vs. CL power y=2.2596x 0.073 0.9247 79
















LITERATURE CITED





Altmann, J., 1974. Observational sampling of behavior: sampling
methods. Behav., Vol. 49, pp. 227-265.


Allen, E.A. and H.A. Curran, 1974. Biogenic sedimentary
structures produced by crabs in lagoon margins and salt
marsh environments near Beaufort, North Carolina.
J. Sedim. Petrol., Vol. 44, pp. 538-548.


Aspey, P., 1978. Fiddler crab burrowing ecology: burrow density
in Uca pugnax (Smith) and Uca pugilator (Bosc).
Crustaceana, Vol. 34, pp. 235-244.


Barnes, R.S.K., 1968. Relative carapace and chela proportions in
some ocypodid crabs (Brachyura, Ocypodidae). Crustaceana,
Vol. 14, pp. 131-136.


Basan, P.B. and R.W. Frey, 1977. Actual paleontology and
neoichnology of salt marshes near Sapelo Island, Georgia.
In, Trace fossils 2, Geological J. Special Issue #9, edited by T.P. Crimes and J.C. Harper, Seel House Press, Liverpool,
pp. 41-70.


Bertness, M.D., 1985. Fiddler crab regulation of Spartina
alterniflora production on a New England salt marsh. Ecol.,
Vol. 66, pp. 1042-1055.


Bertness, M.D. and T. Miller, 1984. The distribution and
dynamics of Uca pugnax (Smith) burrows in a New England salt
marsh. J. Exp. Mar. Biol. Ecol., Vol. 83, pp. 211-237.


Cammen, L.M., E.D. Seneca and L.M. Stroud, 1980. Energy flow
through the fiddler crabs Uca pugnax and Uca minax and the
marsh periwinkle Littorina irrorata in a North Carolina salt
marsh. Am. Midl. Nat., Vol. 103, pp. 687-694.



54





55



Cammen, L.M., E.D. Seneca and L.M. Stroud, 1984. Long-term
variations of fiddler crab populations in North Carolina
salt marshes. Estuaries, Vol. 7, pp. 171-175.


Christy, J.H., 1982. Burrow structure and use in the sand
fiddler crab, Uca pugilator (Bosc). Anim. Behav., Vol. 30,
pp. 687-694.


Crane, J., 1975. Fiddler crabs of the world (Ocypodidae:
Genus Uca). Princeton Univ. Press, Princeton, N.J.
736 pp. + xxiii.


Darwin, C., 1881. The formation of vegetable mould through
the action of worms, with observations on their habits.
Edited by J. Murray, D. Appleton and Co., London.
326 pp. + xxix.


Davidson, R.J. and I.D. Marsden, 1987. Size relationships and
relative growth of the New Zealand swimming crab Ovalipes
catharus (White, 1843). J. Crust. Biol., Vol. 17,
pp. 308-317.


DeLaune, R.D. and W.H. Patrick, 1980. Nitrogen and phosphorus
cycling in a Gulf Coast salt marsh. In,
Estuarine perspectives, edited by' V.S. Kennedy, Academic
Press, New York, pp. 143-151.


Dembowski, J., 1926. Notes on the behavior of the fiddler crab.
Biol. Bull. (Woods Hole), Vol. 50, pp. 179-201.


Dixon, W.J., 1985. BMDP statistical software manual.
Univ. California Press, Berkeley. 734 PP.


Frey, R.W., P.B. Basan and R.M. Scott, 1973. Techniques for
sampling salt marsh benthos and burrows. Am. Midl. Nat.,
Vol. 89, pp. 228-234.


Genoni, G.P., 1985. Food limitation in salt marsh fiddler crabs
Uca rapax (Smith, 1870) (Decapoda, Ocypodidae). J. Exp.
Mar. Biol. Ecol., Vol. 87, pp. 97-110.


Gosselink, J.G., R. Hatton and C.S. Hopkinson, 1984.
Relationships of organic carbon and mineral content to bulk density in Louisiana salt marsh soils. Soil Sci., Vol. 137,
pp. 177-181.









Gosselink, J.G. and C.J. Kirby, 1974. Decomposition of salt
marsh grass, Spartina alterniflora Loisel. Limnol.
Oceanogr., Vol. 19, pp. 825-832.


Greenspan, B.N., 1980. Male size and reproductive success in the
communal courtship system of the fiddler crab Uca rapax.
Anim. Behav., Vol. 28, pp. 387-392.


Haines, E.B., 1976. Relation betwen the stable carbon isotope
composition of fiddler crabs, plants and soils in a salt
marsh. Est. Coast. Mar. Sci., Vol. 4, pp. 609-616.


Haines, E.B. and R.B. Hanson, 1979. Experimental degradation of
detritus made from the salt marsh plants Spartina
alterniflora Loisel., Spartina virginiana L. and Juncus roemerianus Scheele. J. Exp. Mar. Biol. Ecol., Vol. 40,
pp. 27-40


Hartnoll, R.G., 1982. Growth. In, The biology of Crustacea,
Vol. 2, edited by L.G. Abele, Academic Press, New York,
pp. 111-196.


Hines, A.H., 1982. Allometric constraints and variables of
reproductive effort in brachyuran crabs. Mar. Biol.,
Vol. 69, pp. 309-320.


Hopkinson, C.S. and J.P. Schubauer, 1984. Static and dynamic
aspects of nitrogen cycling in the salt marsh graminoid
Spartina alterniflora. Ecol., Vol. 65, pp. 961-969.


Howes, B.L., R.W. Howarth, J.M. Teal and I. Valiela, 1981.
Oxidation-reduction potentials in a salt marsh: spatial
patterns and interaction with primary production.
Limnol. Oceanogr., Vol. 26, pp. 350-360.


Hyatt, G.W. and M. Salmon, 1977. Combat in the fiddler crabs Uca
pugilator and Uca pugnax: a quantitative analysis.
Behaviour, Vol. 65, pp. 182-211.


Hyman, 0.W., 1922. Adventures in the life of a fiddler crab.
Ann. Rep. Smiths. Insn. for 1920, pp. 443-459.


Katz, L.C., 1980. Effects of burrowing by the fiddler crab, Uca
pugnax (Smith). Estuar. Coast. Mar. Sci., Vol. 11,
pp. 233-237.





57



Kitchell, J.F., R.V. O'Neill, D. Webb, G.W. Gallepp,
S.M. Bartell, J.F. Koonce, and B.S. Ausmus, 1979. Consumer
regulation of nutrient cycling. Bioscience, Vol. 29,
pp. 28-34.


Kraeuter, J.N. and P.L. Wolf, 1974. The relationship of marine
macroinvertebrates to salt marsh plants. In, Ecology of
halophytes, edited by R.J. Reimold and W.H. Queen, Academic
Press, New York, pp. 449-462.


Kurz, H. and K. Wagner, 1957. Tidal marshes of the Gulf and
Atlantic coasts of northern Florida and Charleston, South Carolina. Florida State Univ. Series in Biology, No. 24,
168 pp.


Miller, D.C., 1961. The feeding mechanism of fiddler crabs, with
ecological considerations of feeding adaptations.
Zoologica, Vol. 46, pp. 89-100.


Miller, D.C., 1971. Growth in Uca. 1. Ontogeny of asymmetry in
Uca pugilator (Bosc) (Decapoda, Ocypodidae). Crustaceana,
Vol. 24, pp. 119-131.


Montague, C.L., 1980a. The net influence of the fiddler crab,
Uca pugnax, on carbon flow through a Georgia salt marsh:
the importance of work by macroorganisms to the metabolism of ecosystems. Ph.D. Dissertation, University of Georgia,
Athens.


Montague, C.L., 1980b. A natural history of temperate Western
Atlantic fiddler crabs (genus Uca) with reference to their
impact on the salt marsh. Contrib. Mar. Sci., Vol. 23,
pp. 25-55.


Montague, C.L., S.M. Bunker, E.B. Haines, M.L. Pace and
R.L. Wetzel, 1981. Aquatic macroconsumers. In, The ecology of a salt marsh, edited by L.R. Pomeroy and R.G. Wiegert, Springer-Verlag, New York, pp. 69-85.


Paine, R.T., 1971. The measurement and application of the
calorie to ecological problems. Annu. Rev. Ecol. Syst.,
Vol. 2, pp. 145-164.


Patten, B.C. and E.P. Odum, 1981. The cybernetic nature of
ecosystems. Am. Nat., Vol. 118, pp. 886-895.





58



Pearse, A.S., 1914. Habits of fiddler crabs. Rep. Smiths.
Insn., pp. 415-428.


Powers, L.W., 1973. Ecological aspects of burrows and fiddler
crab behavior. Am. Zool., Vol. 13, pp. 1271.


Ringold, P., 1979. Burrowing, root mat density, and the
distribution of fiddler crabs in the Eastern United States.
J. Exp. Mar. Biol. Ecol., Vol. 3, pp. 11-21.


Robertson, J.R., K. Bancroft, G.K. Vermeer and K. Plaisier, 1980.
Experimental studies on the foraging behavior of the sand
fiddler crab Uca pugilator (Bosc). J. Exp. Mar. Biol.
Ecol., Vol. 44, pp. 67-83.


Robertson, J.R., J.A. Fudge and G.K. Vermeer, 1981. Chemical and
live feeding stimulants of the sand fiddler crab, Uca pugilator (Bosc). J. Exp. Mar. Biol. Ecol., Vol. 53,
PP. 47-64.


SAS Institute Inc., 1985a. SAS user's guide. Basics. Version 5
edition. Cary, N.C.: SAS Institute Inc., 1290 pp.


SAS Institute Inc., 1985b. SAS user's guide. Statistics.
Version 5 edition. Cary, N.C.: SAS Institute Inc., 956 pp.


Savage, T. and J.R. Sullivan, 1978. Growth and claw regeneration
of the stone crab, Menippe mercenaria. Florida Mar. Res.
Publ., No. 32, 23 pp.


Schwartz, B. and S.R. Safir, 1915. The natural history and
behavior of the fiddler crabs. Cold Spring Harbor Monogr.,
Vol. 8, pp. 1-24.


Shanholtzer, S.F., 1973. Energy flow, food habits, and
population dynamics of Uca pugnax in a salt marsh system.
Ph.D. Dissertation, University of Georgia, Athens. 91 pp.


Shinn, E.A., 1968. Burrowing in recent lime sediments of Florida
and the Bahamas. J. Paleontol., Vol. 43, pp. 879-894.


Snedecor, G.W. and W.G. Cochran, 1980. Statistical methods.
7th edition. Iowa State Univ. Press, Ames, Iowa. 507 pp.





59




Soeder, C.J., and J.F. Talling, 1969. Measurements (in situ) on
isolated samples of natural communities. In, A manual on methods for measuring primary production in aquatic
environments, IBP handbook No. 12, edited by
R.A. Vollenweider, Blackwell Scientific Publications,
Oxford, pp. 62-73.


Sokal, R.R. and F.J. Rohlf, 1969. Biometry. Freeman, San
Francisco, CA. 776 pp.


Teal, J.M. and J. Kanwisher, 1961. Gas exchange in a salt marsh.
Limnol. Oceanogr., Vol. 6, pp. 388-399.


Teissier, G. 1960. Relative growth. In,
The physiology of Crustacea, Vol. 1, edited by
T.H. Waterman, Academic Press, New York, pp. 537-560.


Valiela, I., J.M. Teal and N.Y. Persson, 1976. Production and
dynamics of experimentally enriched salt marsh vegetation:
belowground biomass. Limnol. Oceanogr., Vol. 21,
pp. 245-252.


Vliet, K.A., 1981. Population density and size-class
distribution in a population of the sand fiddler crab,
Uca pugilator. Univ. Florida, unpubl.


Wilson, D.S., 1980. The natural selection of populations and
communities. Benjamin/Cummings Publ., Menlo Park, CA.
186 pp.


Wolf, P.L., S.F. Shanholtzer and R.J. Reimold, 1975. Population
estimates for Uca pugnax (Smith, 1870) on the Duplin Estuary
marsh, Georgia, U.S.A. (Decapoda, Brachyura, Ocypodidae).
Crustaceana, Vol. 29, pp. 79-91.


Young, D.Y. and H.W. Ambrose, III, 1978. Underwater orientation
in the fiddler crab, Uca pugilator. Biol. Bull., Vol. 155,
pp. 246-258.


Zar, J.H., 1974. Biostatistical analysis. Prentice Hall,
Princeton, N.J., 620 pp.


Zucker, N., 1974. Shelter building as a means of reducing
territory size in the fiddler crab, Uca terpsichores
(Crustacea: Ocypodidae). Am. Midl. Nat., Vol. 91, p. 224.





60



Zucker, N., 1981. The role of hood-building in defining
territories and limiting combat in fiddler crabs.
Anim. Behav., Vol. 29, pp. 387-395.

















BIOGRAPHICAL SKETCH




Giulio Piero Genoni was born in Milano, Italy, in 1957. He received the Licence es Sciences Naturelles (equivalent to the degree of Bachelor of Science in biology) from the University of Lausanne, Switzerland, in 1979. He worked at the Institute of Biochemistry of the Zuerich Institute of Technology during 1980. He took a course in oceanography at the University of Aix-Marseille, France, during the 1980-1981 year. In the fall of 1981 he enrolled in the Department of Zoology of the University of Florida, and he obtained the Master of Science degree in 1984. His doctoral research was supported by research assistanships from the Department of Fisheries and Aquaculture, teaching assistantships from the Department of Zoology, the University of Florida Marine Laboratory, and a grant-in-aid of research of the Sigma Xi Research Society.



















61
















I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Dr. Frank J.S Maturot Jr., Chairr Professor of oology




I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Dr. Frank G. Wordlie -rProfessor of Zoology




I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Dr. Cl L. Montague
Assis nt Professor Environmental Engineering Scie ces









I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Dr. Michele G. Wheatly
Assistant Professor of ogy


I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Dr. William J. indberg / Assistant Professor of Fisheri s d Aquaculture


This dissertation was submitted to the Graduate Faculty of the Department of Zoology in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.

December 1987

Dean, Graduate School




Full Text
2
of an increased food supply may not be merely a by-product of
their burrowing activity (Montague, 1980a).
The hypothesis that the fiddler crab Uca rapax (Smith) farms
cordgrass was examined by (a) measuring and comparing natural
burrow density and crab density; (b) manipulating burrow density
to test whether fiddler crabs tend to dig burrows in excess of
the ones available; and (c) manipulating food levels to test
whether such excess burrowing may represent farming.
The prediction was made that the burrow/crab ratio would be
>1. Since fiddler crabs are food limited, and since burrows
enhance their food supply, they may tend to excavate burrows or
burrow branches in excess of a ratio of one burrow available per
crab. If they do so in spite of the comparatively high energy
investment required to burrow in a muddy substrate, one
explanation may be that they farm Spartina. Fiddler crabs
sometimes forage in parts of the marsh with no burrows (Montague,
1980a), suggesting that excess burrows are not only for predator
escape. The possibility that excess burrows are only for
reproduction is equally unlikely, because males may not be able
to defend several burrows (where females incubate their eggs;
Crane, 1975) To test against these and other possibilities
(e.g. parasite avoidance, reduced time constraints, physiological
stress, etc.), food levels were manipulated. If fiddler crabs
are capable of adjusting their burrowing activity to food
availability, burrowing activity was predicted to be lower if
food is abundant and higher if food is scarce.


3
The species used in this study is Uca rapax. Work on
Uca pugnax will be referred to because of its similarity to
U. rapax.


Abstract of Dissertation Presented to the Graduate Scnool
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
Farming of cordgrass, sparti.,a alierniflora LuISel.
BY FIDDLER CRABS, UCA RAPAX (SMITH)
(DECAPODA: OCYPODIDAE)
BY
Giulio Piero Genoni
December 1987
Chairman: Frank J.S. Maturo, Jr.
Major Department: Zoology
Tue fiddler crab Uca rapax is a food limited detritivore
that feeds on decaying cordgrass (Spartina alterniflora) in the
salt marsh. Crab burrowing activity stimulates cordgrass growth,
and thereby indirectly enhances their food supply. Fiddler crabs
may dig more burrows than they require for their protective and
physiological needs. There may be a selective advantage in
digging "excess" burrows, because of the indirect benefit of
enhancing food supply and decreasing the competition for food.
The hypothesis that fiddler crabs "farm" cordgrass was examined
by (a) comparing burrow and crab density in a salt marsh,
(b) manipulating burrow density to test whether fiddler crabs dig
burrows in excess of the ones available, and \c) manipulating
IV


Table V. Effect of food availability on burrowing activity of Uca rapax (all crabs)
Variable
Low food
High food
Skewness
Kurtosis
Levene
N
Test
df
Significance
All new burrows
39.26+13.57
30.71+11.73
NS
NS
NS
17
F= 8.91
1
p<0.004
New unbranched burrows
20.88+10.20
18.88+10.89
NS
NS
NS
17
F= 0.78
1
NS
Maintained burrows (3)
New branches to initial
8.92+ 1.31
Q.79 1.27
p<0.001
NS
NS
16
F= 0.15
1
NS
(1)
burrows (3)
2.96+ 1.10
1.92+ 0.78
p<0.01
p<0.001
NS
16
F=17.68
1
p<0.0001
(1)
Maintained branches (3)
11.90+ 1.78
10.75+ 1.38
p<0.001
p<0.001
NS
17
F= 8.36
1
p<0.005
(1)
Branching ratio
1.31+ 0.62
1.19+ 0.53
p<0.001
p<0.001
p<0.0001
3340
F=25.53
1
p<0.0001
(1)
Burrow depth (cm)
5.19+ 2.38
6.08+ 2.95
p<0.001
p<0.001
p<0.0001
2596
F=51.53
1
p<0.0001
(1)
g excavated sediment (3)
67.92+19.06
72.70+16.43
NS
NS
NS
17
F= 1.19
1
NS
(2)
% time burrowing
5.49 %
5.72J6 w
-
-
27
F= 0.25
1
NS
(1) Log-transformed data.
(2) Arcsine-transformed data.
(3) Divided by the duration of each trial in days.


42
burrowing, thus being visible less frequently, and to their being
less involved in agonistic and courtship activities. A higher
burrow turnover rate may be expected for smaller crabs, because
(a) they have an increased food requirement, (see below),
(b) they lose their burrows more often to larger, competitively
dominant, crabs (Pearse, 1914 Hyatt and Salmon, 1977; Bertness
and Miller, 1984); (c) they do not allocate as much time (or any
at all in the case of the smallest ones) to agonistic and
courtship activities; (d) burrow longevity is lower for smaller
burrows when there is no root mat (Bertness and Miller, 1984);
and (e) small males may dig more if forced to interact with
displaying males (L.W. Powers, pers. comm.), an artifact of the
arenas.
Larger crabs burrowed more deeply and constructed more
shelters, which agrees with findings of Crane (1975), Zucker
(1971, 1974) and Bertness and Miller (1984). Smaller crabs, even
though they tended to allocate more time to burrowing, can
excavate only small amounts of sediment at a time.
Response of crabs of different size to initial burrow density
Time allocation to burrowing was increased for larger crabs
in arenas with a low initial burrow density, but not by smaller
crabs. Smaller crabs already tend to spend more time burrowing
than larger crabs and may not be able to increase their burrowing
further. Rather, smaller crabs respond by burrowing less deeply
and allocating their burrowing time to new burrows. In addition,
sensitivity to burrow density may be particularly high in small
crabs because of their relatively high population density.


TABL OF CONTENTS
PAGE
ACKNOWLEDGEMENTS ii
ABSTRACT iv
INTRODUCTION 1
METHODS 4
Burrow/crab Ratios in the Salt Marsh 4
Effects of Burrow Density and Food Availability 6
RESULTS 11
Burrow/crab Ratios in the Salt Marsh 11
Effects of Burrow Density and Food Availability 12
DISCUSSION 34
burrow/crao Ratios in the Salt Marsh 34
Effects of Burrow Density and Food Availability 36
General Considerations 43
APPENDIX 46
LITERATURE CITED 34
BIOGRAPHICAL SKETCH 61
iii


57
Kitchell, J.F., R.V. O'Neill, D. Webb, G.W. Gallepp,
S.M. Barbell, J.F. Koonce, and B.S. Ausmus, 1979. Consumer
regulation of nutrient cycling. Bioscience, Vol. 29,
pp. 28-34*
Kraeuter, J.N. and P.L. Wolf, 1974* The relationship of marine
macroinvertebrates to salt marsh plants. In, Ecology of
halophytes, edited by R.J. Reimold and W.H. Queen, Academic
Press, New York, pp. 449-462.
Kurz, H. and K. Wagner, 1957. Tidal marshes of the Gulf and
Atlantic coasts of northern Florida and Charleston, South
Carolina. Florida State Univ. Series in Biology, No. 24,
168 pp.
Miller, D.C., 1961. The feeding mechanism of fiddler crabs, with
ecological considerations of feeding adaptations.
Zoolgica, Vol. 46, pp. 89-100.
Miller, D.C., 1971. Growth in Uca. 1. Ontogeny of asymmetry in
Uca pugilator (Bose) (Decapoda, Ocypodidae). Crustaceana,
Vol. 24, pp. 119-131.
Montague, C.L., 1980a. The net influence of the fiddler crab,
Uca pugnax, on carbon flow through a Georgia salt marsh:
the importance of work by macroorganisms to the metabolism
of ecosystems. Ph.D. Dissertation, University of Georgia,
Athens.
Montague, C.L., 1980b. A natural history of temperate Western
Atlantic fiddler crabs (genus Uca) with reference to their
impact on the salt marsh. Contrib. Mar. Sci., Vol. 23,
pp. 25-55.
Montague, C.L., S.M. Bunker, E.B. Haines, M.L. Pace and
R.L. Wetzel, 1981. Aquatic macroconsumers. In, The
ecology of a salt marsh, edited by L.R. Pomeroy and
R.G. Wiegert, Springer-Verlag, New York, pp. 69-85.
Paine, R.T., 1971. The measurement and application of the
calorie to ecological problems. Annu. Rev. Ecol, Syst.,
Vol. 2, pp. 145-164.
Patten, B.C. and E.P. Odum, 1981. The cybernetic nature of
ecosystems. Am. Nat., Vol. 118, pp. 886-895.


15
Configurations were of the J, Y, U and H types, and sometimes
more complex. Most shafts were straight, even those of newly dug
burrows, and walls were smooth. Burrow depth was 5.63 +_ 2.71 cm
(N=2597, max.=20.0 cm). There were only 1.84 shelters per arena,
on average (N=125). Burrows were significantly associated with
structural elements (stems): using a calculated value of 75 %
available substrate, burrows were adjacent to stems more
frequently than expected by chance alone (28 % of burrows,
X2=5.05, df=1, p>0.025, N=634).
Water content of the sediment in arenas was 55*35 %
(s.d.=5*43, N=27), which was lower than that of the salt marsh
sediment. However this was the highest water content that
allowed construction of artificial burrows. The organic matter
present was 7*969 % (s.d.=1.690, N=55). Using a conversion of
2
0.323 kcal/(m x yr) for an organic fraction of 1 % calculated
from data of Cammen et al. (1980), this value for low food
2
treatments corresponds to 2.62 kcal/(m x yr).
Effect of initial burrow density
The interaction between initial burrow density and food
availability was not significant for all variables examined by
ANOVAs.
Some of the variables describing burrowing activity were
affected by initial burrow density. Results are summarized in
Tables III and IVa and Figure 1. The data shown in Table III are
collapsed for the two levels of the burrow density treatment, but
the value of F is for the main effect of burrow density in the
factorial ANOVA. There were no differences between treatments in


Table XI. Comparison of burrowing activity by Uca rapax of different sizes.
Variable
Size class 1
Size class 2
Size class 3
Size class 4
N
Test
df
Significance
% unbranched burrows
68.73 %
61.47 %
65.70 %
69.41 %
1399
X2=6.17
3
NS
Branching ratio
1.31+ 0.54
1.43+ 0.58
1.39+ 0.60
1.35+ 0.59
1393
F= 2.21
3
NS
(1)
Burrow depth (era)
5.07+ 2.03
5.48+ 2.15
6.87+ 2.79
8.06+ 3.18
1149
F=63.99
3
p<0.0001
(1)
Activity at surface
0.10
0.20
0.39
0.31
4536
X =1573.
C0
p<0.0001
% time burrowing
12.56 %
6.31 %
3.83 %
5.31 %
108
F= 1.44
3
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.


38
Pearse, A.S., 1914* Habits of fiddler crabs. Rep. Smiths.
Insn., pp. 415-428.
Powers, L.W., 1973. Ecological aspects of burrows and fiddler
crab behavior. Am. Zool,, Vol. 13 pp. 1271.
Ringold, P., 1979. Burrowing, root mat density, and the
distribution of fiddler crabs in the Eastern United States.
J. Exp. Mar. Biol. Ecol., Vol. 3 pp. 11-21.
Robertson, J.R., K. Bancroft, G.K. Vermeer and K. Plaisier, 1980.
Experimental studies on the foraging behavior of the sand
fiddler crab Uca pugilator (Bose). J. Exp. Mar. Biol.
Ecol., Vol. 44, pp. 67-83.
Robertson, J.R., J.A. Fudge and G.K. Vermeer, 1981. Chemical and
live feeding stimulants of the sand fiddler crab, Uca
pugilator (Bose). J. Exp, Mar. Biol. Ecol., Vol. 53,
pp. 47-64.
SAS Institute Inc., 1985a. SAS user's guide. Basics. Version 5
edition. Cary, N.C.: SAS Institute Inc., 1290 pp.
SAS Institute Inc., 1985b. SAS user's guide. Statistics.
Version 5 edition. Cary, N.C.: SAS Institute Inc., 956 pp.
Savage, T. and J.R. Sullivan, 1978. Growth and claw regeneration
of the stone crab, Menippe mercenaria. Florida Mar. Res.
Publ., No. 32, 23 pp.
Schwartz, B. and S.R. Safir, 1915* The natural history and
behavior of the fiddler crabs. Cold Spring Harbor Monogr.,
Vol. 8, pp. 1-24.
Shanholtzer, S.F., 1973. Energy flow, food habits, and
population dynamics of Uca pugnax in a salt marsh system.
Ph.D. Dissertation, University of Georgia, Athens. 91 pp.
Shinn, E.A., 1968. Burrowing in recent lime sediments of Florida
and the Bahamas. J. Paleontol., Vol. 43, pp. 879-894.
Snedecor, G.W. and W.G. Cochran, 1980. Statistical methods.
7th edition. Iowa State Univ. Press, Ames, Iowa. 507 pp.


12
one or more turns. Other configurations (11.0 % of the burrows)
reflected branching or interconnection of burrows, and might be
grouped under the designations of Y, U, and H configurations
(Allen and Curran, 1974)* They had (in decreasing frequency of
occurrence) 2, 3, or 4 branches, most of which led to an opening.
Burrow depth (projected against an imaginary vertical plane,
i.e. not taking into account burrow angle) ranged from 1 to 11 cm
(N=94).
These data, to the extent that they were used to set up
experimental arenas, are summarized in Table I.
Effects of Burrow Density and Food Availability
Fiddler crabs occupied and remodeled the artificial burrows,
and dug additional burrows and branches. Few were seen walking
along the walls or burrowing against them. Fiddler crabs dug, on
average, 34*99 +_ 13-30 new burrows per trial (N=68). Of these,
19-88 _+ 10.52 were unbranched burrows. Thus, the branching ratio
for all burrows was 1.25 _+ 0.58 (N=3340). Another useful
variable is the count of all burrows occupied or maintained to
the end of each trial, excluding abandoned burrows, which usually
collapsed: there were 8.86/day _+ 1.28 maintained burrows,
11.33/day _+ 1.63 new branches and 2.44/day + 1.08 branches added
to initial burrows. The burrowing activity removed
70.31 dry g/day of sediment (s.d.=17-82), or an average of
1.64 dry g/crab/day. The fraction of time allocated to the
various activities is presented in Table Ila. The fraction
allocated to burrowing was 5-9 % (N=4628 observations).


FARMING OF C0RDGRAS3, SPARTINA ALTERNIFLORA LOISEL.,
EY FIDDLER CRABS, (JCA RAPAX (SMITH)
(DECAPODA: OCYPODIDAE)
BY
GIULIO PIERO GENONI
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
19S7

ACKNOWLEDGEMENTS
I wish to express my gratitude to my graduate committee,
Drs. F.J.S. Maturo, Jr., C.L. Montague, F.G. Nordlie,
W.J. Lindberg, and M.G. Wheatly, for their generous assistance
and guidance. I am especially indebted to Dr. Maturo, who
provided much support with facilities, supplies and advice, to
Dr. Montague, who stimulated my interest in this study, and to
Dr. Lindberg, for helpful discussion.
I also thank Drs. M.D. Bertness, W.F. Hermkind,
C.S. Hopkinson and L.W. Powers for their helpful comments and
G. Nemes, V. Orcel and A.M. Fiore for help in the lab and in the
field. D. Harrison prepared the illustrations.
ii

TABL OF CONTENTS
PAGE
ACKNOWLEDGEMENTS ii
ABSTRACT iv
INTRODUCTION 1
METHODS 4
Burrow/crab Ratios in the Salt Marsh 4
Effects of Burrow Density and Food Availability 6
RESULTS 11
Burrow/crab Ratios in the Salt Marsh 11
Effects of Burrow Density and Food Availability 12
DISCUSSION 34
burrow/crao Ratios in the Salt Marsh 34
Effects of Burrow Density and Food Availability 36
General Considerations 43
APPENDIX 46
LITERATURE CITED 34
BIOGRAPHICAL SKETCH 61
iii

Abstract of Dissertation Presented to the Graduate Scnool
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
Farming of cordgrass, sparti.,a alierniflora LuISel.
BY FIDDLER CRABS, UCA RAPAX (SMITH)
(DECAPODA: OCYPODIDAE)
BY
Giulio Piero Genoni
December 1987
Chairman: Frank J.S. Maturo, Jr.
Major Department: Zoology
Tue fiddler crab Uca rapax is a food limited detritivore
that feeds on decaying cordgrass (Spartina alterniflora) in the
salt marsh. Crab burrowing activity stimulates cordgrass growth,
and thereby indirectly enhances their food supply. Fiddler crabs
may dig more burrows than they require for their protective and
physiological needs. There may be a selective advantage in
digging "excess" burrows, because of the indirect benefit of
enhancing food supply and decreasing the competition for food.
The hypothesis that fiddler crabs "farm" cordgrass was examined
by (a) comparing burrow and crab density in a salt marsh,
(b) manipulating burrow density to test whether fiddler crabs dig
burrows in excess of the ones available, and \c) manipulating
IV

food availability to test whether crabs adjust their burrowing
activity in response to food. Results showed that (a) there were
more burrows than fiddler crabs, (b) crabs dug new burrows
despite the presence of unoccupied burrows, and (c) their
burrowing activity varied inversely with food availability. Food
availability affected the rate of burrowing by females and males
equally, but had a stronger effect on burrowing by small crabs
than on that of large crabs. Thus, small crabs may be more
sensitive to food limitation, and may be more efficient in
"farming" cordgrass.
v

INTRODUCTION
Organisms may regulate or control other tropnic levels by
establishing interactions with each other and feedback
relationships (Kitchell et al., 1979; Montague et al., 1961;
Patten and Odum, 1951). In salt marshes, fiddler crabs have a
large impact on cordgrass, Spartina alterniflora: their
burrowing activities, bioturbation and fecal pellet production
increase the growth of Spartina, resulting in an increased food
supply in the form of microorganisms that colonize the grass when
it dies and decays (Montague, 1980a, 1980b; Katz, 1980; bertness,
1985). These crabs are food limited in the salt marsh (Genoni,
1985). Selection may favor activities and by-products that would
decrease food limitation, such as, perhaps, burrowing activity.
Fiddler crabs may be said, then, to "farm" salt marsh grass, like
some earthworms "farm" plants in their habitat (Darwin, 1881), an
indirect effect sensu Wilson (1980).
Burrows have a number of direct advantages for fiddler
crabs: they function as shelters from predators and
environmental extremes; provide water for physiological needs;
and are sites for reproduction (Crane, 1975; Hyatt and Salmon,
1977; Young and Ambrose, 1978; Ringold, 1979). They may serve
other presently unidentified functions. If fiddler crabs dig
more burrows than they require for these needs, then the benefit
1

2
of an increased food supply may not be merely a by-product of
their burrowing activity (Montague, 1980a).
The hypothesis that the fiddler crab Uca rapax (Smith) farms
cordgrass was examined by (a) measuring and comparing natural
burrow density and crab density; (b) manipulating burrow density
to test whether fiddler crabs tend to dig burrows in excess of
the ones available; and (c) manipulating food levels to test
whether such excess burrowing may represent farming.
The prediction was made that the burrow/crab ratio would be
>1. Since fiddler crabs are food limited, and since burrows
enhance their food supply, they may tend to excavate burrows or
burrow branches in excess of a ratio of one burrow available per
crab. If they do so in spite of the comparatively high energy
investment required to burrow in a muddy substrate, one
explanation may be that they farm Spartina. Fiddler crabs
sometimes forage in parts of the marsh with no burrows (Montague,
1980a), suggesting that excess burrows are not only for predator
escape. The possibility that excess burrows are only for
reproduction is equally unlikely, because males may not be able
to defend several burrows (where females incubate their eggs;
Crane, 1975) To test against these and other possibilities
(e.g. parasite avoidance, reduced time constraints, physiological
stress, etc.), food levels were manipulated. If fiddler crabs
are capable of adjusting their burrowing activity to food
availability, burrowing activity was predicted to be lower if
food is abundant and higher if food is scarce.

3
The species used in this study is Uca rapax. Work on
Uca pugnax will be referred to because of its similarity to
U. rapax.

METHODS
Burrow/crab Ratios in the Salt Marsh
A sampling of Uca rapax population density, sex ratio, size
class distribution, and burrow density was done during Summer
1985 in an intertidal salt marsh at Matanzas Inlet, St. Johns
County, Florida (2945'50" N, 8117'20" W). The vertical tidal
range is ~1.9 m, and the distribution of vegetative zones is
fairly distinct. The site is dominated by a medium growth form
(50 cm tall) of cordgrass, Spartina alterniflora Loisel.
(140 shoots/m ). The sediment is fine mud with a small sandy
fraction.
To identify burrow/crab ratios, I used sampling techniques
as described by Frey et al. (1973), Wolf et al. (1975) and
Ringold (1979). An area of 12 by 12 m, which was apparently
homogeneous in Spartina height and density, Uca density, and
physical factors, was chosen. It was divided into 0.5 by 0.5 m
quadrats and marked permanently with dowels. Between 9 June and
6 September, 17 quadrats were chosen by the simple random method
(Snedecor and Cochran, 1980) for measurement of Uca and burrow
density. Each quadrat was enclosed with corrugated fiberglass
panels while water still covered the substrate and most crabs
were in their burrows (Wolf et al., 1975; Genoni, 1985). At low
tide, grass shoots within the quadrat were clipped near the
4

5
substrate. Shoots were counted and later dried 48 h at 70 C
and weighed to the nearest 0.1 g with an Ohaus trip balance. To
determine the characteristics of each burrow and identify
resident crabs, burrow openings were marked with numbered sticks,
and burrow casts were taken with polyester resin (Dembowski,
1926; Shinn, 1968). The resin caused resident crabs to escape.
Their sex was noted and their carapace breadth was measured to
the nearest 0.05 mm with Vernier calipers. Because some burrows
are kept plugged by resident crabs at any time (Crane, 1975),
this procedure was repeated on the next two days with newly
opened burrows. Association of burrows with structural elements
in the substrate (grass stems or mussel shells,
Geukensia demissa) was estimated by scoring burrows as either
occurring on bare substrate (>2 cm from a stem or mussel) or
structure-associated (<2 cm) (Bertness and Miller, 1984).
On the third day, burrow casts were removed and their depth,
angle from the substrate surface, diameter and configuration were
noted. To sample root mat density, a 20-cm deep sediment core
was taken with a PVC tube (5*2 cm in diameter). To sample water
content and organic content, two 7-cm deep cores were taken with
a PVC tube (3*0 cm in diameter). In the laboratory, the larger
sediment core was sieved through a 0.5 mm mesh, and the roots
were dried 48 h at 70 C and weighed to the nearest 0.01 g with
a Mettler H8 analytic balance. One smaller core was weighed wet,
dried 48 h at 110 C and weighed again to determine the water
content. The upper 0.5 cm from the other core was weighed wet,
dried 48 h at 70 C, then weighed dry and left for 4 h in a
Sybron Corp. Thermolyne 10500 muffle furnace at 500 C. It was

6
then rehydrated, dried again, and weighed. The ash-free dry
weight was determined and the organic content was estimated as
the weight of the fraction lost in the ashing (Paine, 1971).
Total counts of burrow and crab density, and Spartina
biomass and density, were pooled for all 17 quadrats, whereas
data from subsamples were averaged over all quadrats.
Effects of Burrow Density and Food Availability
In a manipulative experiment in the laboratory, the effect
of food availability and burrow density on burrowing activity of
U. rapax was tested. Two sets of four circular arenas were
built, each consisting of a plastic pail on top of which a sheet
of metal was riveted to make a funnel with walls inclined at a
o
45 angle. The metal was lined with plastic and topped with a
10-cm strip of aluminium flashing that crabs could not climb.
The arenas were filled with sandy mud (collected from a creek in
the salt marsh) 45 cm deep, with an upper diameter of 52 cm and a
2
suriace area of 0.21 m Results from the field study were used
to determine the initial conditions used in the arenas. At the
beginning of each trial, complete randomization (Snedecor and
Cochran, 1980) was used in assigning each arena to one of four
treatments: (1) low food abundance and low burrow density, (2)
low food and high density, (3) high food and low density, and (4)
2
high food and high density. This 2 factorial experiment was
replicated in 17 trials, between April 7 and September 14, 1987.
The following responses were predicted: burrowing activity
should be highest in treatment (1), both to stimulate food and

7
for other needs (protective, reproductive and physiological);
intermediate in (2), to stimulate food; and in (3), to meet other
needs; and lowest in (4).
Food treatments consisted of 0 g and 11 g, respectively, of
Purina Fly Larvae Medium added to each arena. For a 7-day trial,
2
the high value corresponds to 19 kcal/(m x day), i.e. about
twice the highest estimate of average energy consumption of
fiddler crabs calculated by Cammen et al. (1980)
2 2
(3522 kcal/(m x yr) or 9*7 kcal/(m x day)). The arenas where
food was not added contained some organic matter mixed with the
sediment, as evidenced by the feeding response of crabs. This
behavior is mediated by chemoreceptors in the minor chelae
(Robertson et al., 1980, 1981). To measure the organic content
of the upper 0.5 cm of sediment, cores (2.7 cm in diameter) were
taken from each arena before each trial, and ash-free dry weight
was determined as outlined above.
Initial burrows (43 and 64 per arena, for high and low
density treatments) were made by pushing dowels 15 cm into the
substrate, at angles of 60-80. Thus, burrow densities were
9 2
205/m and 302/m Burrow openings were marked with Spartina
stems cut to 10 cm, on which numbered tags were stapled.
Additional stems were placed in the low-density arenas to achieve
similar stem densities.
Uca rapax were captured at the Matanzas Inlet site and
43 individuals were distributed into each arena. The sex ratio
and size class distribution in each arena were similar to those
observed in the sampling study, with the exception that no crabs
<6 mm in carapace breadth were used, due to the difficulty of

8
surveying their burrows. The following size classes were used,
as determined by carapace breadth: 6-8, 8-11, 11-15 and >15 mm.
Because competition for burrows of adequate size may be stronger
for larger crabs (Bertness and Miller, 1984) the diameter of
initial burrows was varied to match crab carapace length (crabs
enter their burrows sideways). Crabs were left in the arenas
4.0-7.5 days, at 25C, in a laboratory equipped with windows and
with fluorescent lights set on a timer simulating the
sunrise-sunset rhythm. Thus, the natural photoperiod was
supplemented by an artificial photoperiod. Arenas were separated
by blinds from the rest of the laboratory. The sediment was
maintained wet by additions of 50 % seawater.
The following parameters were used to describe fiddler crab
burrowing activity: the number of new unbranched and branched
burrows, the amount of excavated sediment, and the percent of
time allocated to burrowing (Katz, 1980; Montague, 1980a). At
the end of each trial, excavated material was collected, dried
48 h at 70 C, and weighed to the nearest 0.01 g. Newly
excavated burrows were marked. I noted the presence of
half-domes of mud (or "shelters"; Zucker, 1974) constructed by
some individuals on burrow openings. To determine the
characteristics of each burrow and identify resident crabs,
burrow casts were made with latex concrete (Christy, 1982). The
concrete caused resident crabs to escape. Their sex and carapace
breadth were noted. This procedure was repeated with newly
opened burrows on the next two days. Burrow casts were removed
and their depth, diameter and configuration were noted.

9
Association of burrows with structural elements (grass stems) was
estimated as described for the field study.
Counts of new unbranched and branched burrows and
measurements of excavated sediment were divided by the duration
of each trial in days. These variables, burrow depths, and
branching ratios (defined as the ratio of the number of branches
2
to the number of burrows) were analysed by 2 factorial ANOVAs
for tests of the effects of food and density treatments and their
interaction. The assumption of normality was tested by
two-tailed t-tests on skewness and kurtosis; the assumption of
homogeneity of variances was tested by Levene's one-way ANOVA.
If these assumptions were not met, factorial ANOVAs were done on
log-transformed data (the assumptions were tested again).
Comparisons between treatments of the percent of unbranched
2
burrows were done by X tests of independence.
Comparisons of responses to initial burrow density and to
food availability were also made for each sex and each size
class. Crabs of each sex or size class may not be equally
sensitive to food supply, because of different foraging
efficiencies or food assimilation, or because dominance by one
sex or by larger crabs may limit food availability to other crabs
(Hyman, 1922; Crane, 1975). Comparisons of shelter building
2
between sexes and size classes was done by X tests of
independence.
To investigate time allocation to different activities,
including burrowing, crabs were marked individually with numbered
plastic tags affixed with cyanoacrylate glue to their carapace.
For every arena, activities of individual crabs were recorded

10
from circular visual scans. Successive observations were made
every 10 s for 15 min. This procedure was repeated 1 to 4 times,
on different days, during 7 of the trials. The percent of the
total time allocated to various activities was estimated from the
frequencies of their occurrence among the observations (scan
sampling; Altmann, 1974)* Activities were grouped as follows:
standing still (motionless), walKing, eating, cleaning, agonism
(combat and chasing), display and courtship, and burrow
construction and maintenance. General comparisons between
treatments of the fraction of time allocated to various
2
activities were done by X tests of independence across all
treatments, for all crabs, and for each sex and size class.
Comparisons of the fraction of time allocated to certain
activities (e.g. burrowing) were done on arcsine-transformed data
by factorial ANOVAs. An index of relative activity of crabs of
each sex or size class was calculated from the frequency of
sightings observed and expected from the sex ratios and size
class distribution in the arenas. Activity levels were compared
2
by X tests of indepedence.
Statistical analyses were done on an IBM 5081-D32/3090-200
mainframe computer at the North Eastern Regional Data Center of
the State University System, Gainesville, FL, using the
Statistical Analysis System (SAS Institute Inc., 1985a, 1985b)
and the Biomedical Data Program P-Series (Dixon, 1985)

RESULTS
Burrow/crab Ratios in the Salt Marsh
At the study site, the sediment was sandy mud, and sediment
water was 64*59 % of wet weight (s.d.=3*l6, N=17). Organic
matter in the upper 0*5 cm was 9*05 % (s.d.=2.64, N=12).
2
Spartina shoot density was 138.1/m (N=17) and standing stock
2
was 269*52 dry g/m (N=15)* The root mat (roots and rhizomes),
between 0 and 20 cm, was 0.11 dry g/ml (s.d.=0.04 N=15).
2 2
There were 212 burrows/m (N=17) and 156 fiddler crabs/m
(N=17). Thus the burrow/crab ratio was 1.36. Sex ratio (M/F)
was 0.79 (N=565 crabs). Size class distribution was 31*0 %
(carapace breadth <6 mm), 20.6 % (6-8), 17.0 #(8-11), 23.2 %
(11-15) and 8.2 % (>15) (N=548 crabs).
Burrows were significantly associated with structural
elements of the substrate; using a value of 70.0 % available bare
substrate (Bertness and Miller, 1984) a conservative estimate
considering the shoot density at the study site, burrows occurred
near stems or mussels more frequently than expected by chance
alone (37.0% of all burrows, X2=29.90, df=1, p<0.0001, N=1284).
Burrow angles from the substrate ranged from 30 to 90 but
most (64.3 %, N=213 burrows) were at an angle between 60 and
80 Most burrows (89.0 %, N=427) were unbranched and J-shaped,
their main shaft being usually straight but sometimes comprising
11

12
one or more turns. Other configurations (11.0 % of the burrows)
reflected branching or interconnection of burrows, and might be
grouped under the designations of Y, U, and H configurations
(Allen and Curran, 1974)* They had (in decreasing frequency of
occurrence) 2, 3, or 4 branches, most of which led to an opening.
Burrow depth (projected against an imaginary vertical plane,
i.e. not taking into account burrow angle) ranged from 1 to 11 cm
(N=94).
These data, to the extent that they were used to set up
experimental arenas, are summarized in Table I.
Effects of Burrow Density and Food Availability
Fiddler crabs occupied and remodeled the artificial burrows,
and dug additional burrows and branches. Few were seen walking
along the walls or burrowing against them. Fiddler crabs dug, on
average, 34*99 +_ 13-30 new burrows per trial (N=68). Of these,
19-88 _+ 10.52 were unbranched burrows. Thus, the branching ratio
for all burrows was 1.25 _+ 0.58 (N=3340). Another useful
variable is the count of all burrows occupied or maintained to
the end of each trial, excluding abandoned burrows, which usually
collapsed: there were 8.86/day _+ 1.28 maintained burrows,
11.33/day _+ 1.63 new branches and 2.44/day + 1.08 branches added
to initial burrows. The burrowing activity removed
70.31 dry g/day of sediment (s.d.=17-82), or an average of
1.64 dry g/crab/day. The fraction of time allocated to the
various activities is presented in Table Ila. The fraction
allocated to burrowing was 5-9 % (N=4628 observations).

13
Table I. (a) Summary of environmental and population characteristics
observed at the salt marsh site and used as initial conditions in the
experimental arenas; (b) Matching of initial burrow diameter with crab
size. CB=carapace breadth.
(a)
Variable
Field study
Arenas
Sediment type
Sandy mud
Sandy mud
Sediment water
64.6 %
55.4 %
Burrow density
212/m2
High: 302/m2 (64/arena)
Low: 205/m (43/arena)
Uca density
156/m2
p
205/m (43/arena)
Burrow/crab ratios
1.36
High: 1.5
Low: 1.0
Burrow angles
64.8 % 60-80
60-80
Burrow configurations
89.0 % J-shaped
Straight
Uca sex ratio (M/F)
0.79
0.79
Uca size-class distribution, by sex
CB 6 8 mm 18.7# F, 13.7% M
CB 8 11 mm 13.7 10.1
CB 11 15 mm 17.3 11.5
CB >15 mm 6.5 8.6
18.6% F, 14.0% M
14.0 9.3
16.3 11.6
9.3 7.0
Food available to Uca
9.7 kcal/(m2x yr)(l)
High: 19.0 kcal/(m2x yr)
Low: 2.6 kcal/(m x yr)
(b)
Size class
Carapace length (2)
Burrow diameter
CB 6 8 mm
CB 8 11 mm
CB 11-15 mm
CB > 15 mm
5.44 mm
7*34 mm
9.87 mm
11.46 mm
5.50 mm
7.40 mm
9.90 mm
12.20 mm
(1) Cammen et al. (1980).
(2) CL was predicted on the basis of a regression of carapace length
against carapace breadth. The morphometric analysis of carapace and
chela measurements is presented in the Appendix.

Table II. Time allocation (%) to various activities by Uca rapax; (a) all crabs, (b) each sex
and (c) each size class.
Still
Eat
Walk
Burrow
Clean
Agonism
Display
N
Test
df
Significance
(a)
36.7
36.0
11.7
5.9
5.3
3.0
1.3
4628
(b)
4493
X2=131.69
6
p<0.0001
Females
35.5
38.3
12.8
8.3
4.5
0.7
-
1722
Males
37.2
35.2
10.9
4-1
6.0
4*4
2.2
2771
(c)
4536
X2=168.53
6
p<0.0001
Size class
1
23.4
55.2
5.2
8.4
6.8
0.9
-
440
2
39.3
37.4
10.9
7.0
3.3
1.4
0.6
905
3
36.9
35.6
12.6
5.0
5.5
2.8
1.6
1777
4
39.8
29.8
12.5
5.3
5.5
4.7
1.8
1414
&

15
Configurations were of the J, Y, U and H types, and sometimes
more complex. Most shafts were straight, even those of newly dug
burrows, and walls were smooth. Burrow depth was 5.63 +_ 2.71 cm
(N=2597, max.=20.0 cm). There were only 1.84 shelters per arena,
on average (N=125). Burrows were significantly associated with
structural elements (stems): using a calculated value of 75 %
available substrate, burrows were adjacent to stems more
frequently than expected by chance alone (28 % of burrows,
X2=5.05, df=1, p>0.025, N=634).
Water content of the sediment in arenas was 55*35 %
(s.d.=5*43, N=27), which was lower than that of the salt marsh
sediment. However this was the highest water content that
allowed construction of artificial burrows. The organic matter
present was 7*969 % (s.d.=1.690, N=55). Using a conversion of
2
0.323 kcal/(m x yr) for an organic fraction of 1 % calculated
from data of Cammen et al. (1980), this value for low food
2
treatments corresponds to 2.62 kcal/(m x yr).
Effect of initial burrow density
The interaction between initial burrow density and food
availability was not significant for all variables examined by
ANOVAs.
Some of the variables describing burrowing activity were
affected by initial burrow density. Results are summarized in
Tables III and IVa and Figure 1. The data shown in Table III are
collapsed for the two levels of the burrow density treatment, but
the value of F is for the main effect of burrow density in the
factorial ANOVA. There were no differences between treatments in

Table III. Effect of initial burrow density on burrowing activity of Uca rapax (all crabs)
Variable
Low density
High density
Skewness
Kurtosis
Levene
N
Test
df
Significance
All new burrows
39.63+11.76
30.06+13.24
NS
NS
NS
17
F=11.13
1
p<0.0014
New unbranched burrows
24.74+ 9.41
14-73+ 9.19
NS
NS
NS
17
F=19.43
1
pCO.0001
Maintained burrows (3)
New branches to initial
8.62+ 1.23
9.11 j+ 1.31
p<0.001
NS
NS
16
F= 2.41
1
NS
(1)
burrows (3)
2.41+ 1.06
2.48+ 1.11
p<0.01
p<0.001
NS
16
F= 0.03
1
NS
(1)
Maintained branches (3)
11.09+ 1.57
11.58+ 1.79
p<0.001
p<0.001
NS
17
F= 1.50
1
NS
(1)
Branching ratio
1.27+ 0.58
1.22+ 0.58
p<0.0001
p<0.001
p<0.0001
3340
F= 9.21
1
p<0.0024
(1)
Burrow depth (cm)
5.57+ 2.66
5.68+ 2.76
p<0.0001
p<0.001
pCO.0001
2596
F= 1.25
1
NS
(1)
g excavated sediment (3)
70.91+19-.23
69.68+16.48
NS
NS
NS
17
F= 0.08
1
NS
% time burrowing
6.56 %
4-56 %
-
-
-
26
F= 1.72
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.
(3) Divided by the duration of each trial in days.

Table IV. Response to initial burrow density of time allocation to various activities (%) by Uca rapax;
(a) all crabs, (b) each sex and (c) each size class.
Still
Eat
Walk
Burrow
Clean
Agonism
Display
N
Test
df
Significance
(a)
4628
X2=37.79
5
p<0.0001
Low density
34.8
39.1
10.3
6.8
4.7
3.3
1.1
High density
38.7
32.9
13.2
5.1
6.0
2.7
1.5
(b)
Females
1722
X2=29-96
5
p<0.0001
Low density
32.6
42.9
11.3
9.1
3.3
0.8
-
High density 38.7
33.2
14.4
7.4
5.7
0.6
-
O
Males
2771
X =18.24
6
p<0.006
Low density
35-4
37.5
9.4
4-9
5.9
5.0
1.9
High density
38.8
33.0
12.3
3.4
6.2
3.9
2.5
(c)
Size class 1
440
X2=8.23
5
NS
Low density
19.3
56.3
5.3
9.1
8.0
1.5
-
High density
28.8
53.7
5.1
7.3
5.1
-
-
O
Size class 2
905
X =14.93
6
p<0.021
Low density
38.2
38.0
9.6
9.2
2.5
1.8
0.8
High density
40.8
36.7
12.8
4.1
4-3
1.0
0.3
P
Size class 3
1777
X =35.55
6
p<0.0001
Low density
33-4
41.8
12.0
4.3
5.2
2.3
0.5
High density
39.9
30.4
13.2
5.6
5.8
2.6
2.6
p
Size class 4
1414
x =17.24
6
p<0.008
Low density
40.0
30.8
10.3
6.7
4.8
5.1
2.5
High density
39.6
28.8
14.5
4.0
7.3
4-4
1.3

- 18
the number of maintained burrows, the number of branches added to
initial burrows and branches of new burrows, the amount of
excavated sediment, or burrow depth. However, in arenas with a
lower initial burrow density, there were significantly more newly
excavated burrows and the branching ratio was significantly
higher; there was a trend for an increase in the fraction of time
allocated to burrowing.
Differences between the reported fraction of time allocated
to burrowing in Tables III and IVa (and in comparisons presented
below) are the consequence of averaging across all treatments and
averaging by factorial univariate analysis, respectively.
Effect of food availability
Some of the variables describing burrowing activity were
affected by the food treatment. Results are summarized in Tables
V and Via and Figure 1. The data shown in Table V are collapsed
for the two levels of the food treatment, but the value of F is
for the main effect of food in the factorial ANOVA. There was no
difference between treatments in the number of newly excavated
burrows, the total number of maintained burrows, the amount of
excavated sediment or the fraction of time allocated to
burrowing. However, in arenas receiving no food additions, there
were significantly more branches added to initial burrows and
branches of new burrows, the branching ratio was higher, and
average burrow depth was lower. In addition, time allocation to
various activities was significantly affected by the treatment:
in low-food arenas, crabs spent less time eating, wandered more

Table V. Effect of food availability on burrowing activity of Uca rapax (all crabs)
Variable
Low food
High food
Skewness
Kurtosis
Levene
N
Test
df
Significance
All new burrows
39.26+13.57
30.71+11.73
NS
NS
NS
17
F= 8.91
1
p<0.004
New unbranched burrows
20.88+10.20
18.88+10.89
NS
NS
NS
17
F= 0.78
1
NS
Maintained burrows (3)
New branches to initial
8.92+ 1.31
Q.79 1.27
p<0.001
NS
NS
16
F= 0.15
1
NS
(1)
burrows (3)
2.96+ 1.10
1.92+ 0.78
p<0.01
p<0.001
NS
16
F=17.68
1
p<0.0001
(1)
Maintained branches (3)
11.90+ 1.78
10.75+ 1.38
p<0.001
p<0.001
NS
17
F= 8.36
1
p<0.005
(1)
Branching ratio
1.31+ 0.62
1.19+ 0.53
p<0.001
p<0.001
p<0.0001
3340
F=25.53
1
p<0.0001
(1)
Burrow depth (cm)
5.19+ 2.38
6.08+ 2.95
p<0.001
p<0.001
p<0.0001
2596
F=51.53
1
p<0.0001
(1)
g excavated sediment (3)
67.92+19.06
72.70+16.43
NS
NS
NS
17
F= 1.19
1
NS
(2)
% time burrowing
5.49 %
5.72J6 w
-
-
27
F= 0.25
1
NS
(1) Log-transformed data.
(2) Arcsine-transformed data.
(3) Divided by the duration of each trial in days.

Table VI. Response to food availability of time allocation to various activities {%) by Uca rapax;
(a) all crabs, (b) each sex and (c) each size class.
Still
Eat
Walk
Burrow
Clean
Agonism
Display
N
Test
df
Significance
(a)
4628
X2=134.58
6
p<0.0001
Low food
40.6
29.5
13.7
6.0
4.4
4.0
1.8
High food
32.4
43.1
9.5
5.9
6.4
1.8
0.8
(b)
Females
1722
X2=3543
5
p<0.0001
Low food
39.4
31.9
14.2
9.2
4.0
1.2
-
High food
31.7
44-4
11.4
7.4
4.9
0.2
-
Males
2771
X =103.53
6
p<0.0001
Low food
41-4
28.4
13.1
3.9
4.6
5.7
2.9
High food
32.2
43.2
8.3
4-3
7.7
3.0
1.3
(c)
Size class
1
440
X2=25.01
5
p<0.0001
Low food
32.7
47.1
5.3
7.2
5.8
1.9
-
High food
15.1
62.5
5.2
9.5
7.8
-
-
Size class
2
905
X =26.29
6
pCO.0001
Low food
37-4
34-9
13-4
8.9
2.1
2.6
0.6
High food
41-4
40.2
8.3
4.8
4.6
0.2
0.5
Size class
3
1777
X =55.83
6
p<0.0001
Low food
40.8
29.1
14-3
4-4
4.8
4.1
2.6
High food
33.1
42.0
11.0
5.6
6.3
1.5
0.7
Size class
4
1414
X =63.50
6
p<0.0001
Low food
45.5
22.3
14*4
5.6
4-9
5.1
2.2
High food
32.3
39.7
9.8
4-9
7.7
4.3
1.3

rigure 1. Effect of initial burrow density and food availability
on burrowing activity of Uca rapax (all crabs) (N=17). Bars
represent 1.96 standard error (95 % confidence intervals).

% TIME BURROW BRANCHING EXCAVATED BURROWS
BURROWING DEPTH RATIO SEDIMENT (g) AND BRANCHES

23
(presumably in search of food), and agonistic encounters were
more frequent than in high-food arenas.
Comparison of burrowing by crabs of each sex
There was no overall difference between females and males in
the percent of unbranched burrows or branching ratio of burrows.
However, females were less active at the surface, allocated
different amounts of time to various activities, spent
significantly more time burrowing, and burrowed less deeply
(Tables VII and lib). Males constructed more shelters than did
females (Table VIII), and allocated significantly more time to
agonism (F=27.47, df=1, p<0.0001, N=53) and to display (F=14*56,
df=1, p<0.0004, N=53) (Table lib).
Response of crabs of each sex to initial burrow density
The three-level interaction between food, burrow density,
and sex, as well as the two-level interactions between pairs of
these factors, were not significant for all variables examined by
ANOVAs.
The effect of initial burrow density did not differ between
males and females with respect to burrow configuration, branching
ratio, depth, or time allocation to different activities, or to
burrowing in particular. This is suggested by a qualitative
comparison between the two sexes in Tables IX and IVb, and by the
non-significant interaction between sex and burrow density for
each variable examined. However, in arenas with a lower initial
burrow density, the trend for allocating a higher fraction of
time to burrowing was stronger for females.

Table VII. Comparison of burrowing activity by each sex of Uca rapax
Variable
Females
Males
N
Test
df
Significance
% unbranched burrows
64.94 %
66.53 %
1391
X2=0.13
1
NS
Branching ratio
1.38+ 0.58
1.37+ 0.59
1385
F= 0.12
1
NS
(1)
Burrow depth (cm)
6.04+ 2.71
6.60+ 2.83
1142
Fs12.46
1
0
0
0

0
V
a.
(1)
Activity at surface
0.69
1.40
4493
X =560.88 1
p<0.001
% time burrowing
9.78 %
4-33 %
53
F= 4.33
1
P<0.04
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.

25
Table VIII. Fraction of the total numbers of shelters
constructed by Uca rapax, for each sex
and each size class. The interaction between sex
and size was not significant.
%
N
Test
df Significance
64
X2=36.04
1
p<0.0001
Females
18.8
Males
81.3
62
X2=51.48
3
pCO.0001
Size class
1
12.9
2
9.7
3
29.0
4
43.4

Table IX. Response of burrowing activity by each sex of Uca rapax to initial burrow density
Variable
Low density
High density N
Test
df Significance
(a) Females
2
% unbranched burrows
64.62 %
65.85 %
653
X =0.11
1
NS
Branching ratio
1.40+ 0.57
1.37+ 0.54
653
F= 0.27
1
NS
(1)
Burrow depth (cm)
6.03+ 2.76
6.07+ 2.64
532
F= 0.21
1
NS
(1)
% time burrowing
12.72 %
5.79 %
26
F= 3.27
1
NS
(2)
(b) Males
2
% unbranched burrows
65.05 %
68.82 %
732
x =1.17
1
NS
Branching ratio
1.40+ 0.59
1.35+ 0.55
732
F= 1.63
1
NS
(1)
Burrow depth (cm)
6.44+ 2.62
6.82+ 3.06
610
F= 2.30
1
NS
(1)
% time burrowing
5.33 %
3.09 %
27
F= 2.02
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.

27
Response of crabs of each sex to food availability
Crabs of the two sexes were not affected differently by food
availability, with respect to percent of unbranched burrows,
branching ratio, depth, time allocation to different activities,
or fraction of time allocated to burrowing (Tables X and VIb).
This is suggested by a qualitative comparison between the two
sexes in Tables X and VIb, and by the non-significant interaction
between sex and food for each variable examined.
Comparison of burrowing by crabs of different size
There was no overall difference between crabs of different
sizes in the percent of unbranched burrows or branching ratio of
burrows (Table XI). Smaller crabs were less active at the
surface, allocated different amounts of time to various
activities and burrowed less deeply (Tables IX and lie). There
was a strong trend for a higher time investment in burrowing by
smaller crabs. Larger crabs constructed more shelters than did
smaller crabs (Table VIII), and allocated significantly more time
to agonism (F=9.24> df=3, p<0.0001, N=108) and to display
(F=451, df=3, p<0.005, N=108). There was a strong trend for a
higher time expenditure in foraging for the smaller size classes.

Table X. Response of burrowing activity by each sex of Uca rapax to food availability
Variable
Low food
High food
N
Test
df
Significance
(a) Females
% unbranched burrows
57.93 %
71.80 %
653
X =13.82
1
pCO.0001
Branching ratio
1.46+ 0.57
1.32+ 0.54
653
F=11.15
1
p<0.0009
(1)
Burrow depth (cm)
5.68+ 2.44
6.38+ 2.87
532
F= 7.34
1
p<0.007
(1)
% time burrowing
10.43 %
9.24 %
26
F= 0.15
1
NS
(2)
(b) Males
% unbranched burrows
60.72 %
72.65 %
732
X =11.74
1
p<0.001
Branching ratio
1.45+ 0.61
CM

0
o'

732
F=13.06
1
p<0.0003
(1)
Burrow depth (cm)
6.07+ 2.37
7.11+ 3.12
610
F=15.70
1
p<0.0001
(1)
% time burrowing
4.18 %
4.48 %
27
F= 0.30
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.

Table XI. Comparison of burrowing activity by Uca rapax of different sizes.
Variable
Size class 1
Size class 2
Size class 3
Size class 4
N
Test
df
Significance
% unbranched burrows
68.73 %
61.47 %
65.70 %
69.41 %
1399
X2=6.17
3
NS
Branching ratio
1.31+ 0.54
1.43+ 0.58
1.39+ 0.60
1.35+ 0.59
1393
F= 2.21
3
NS
(1)
Burrow depth (era)
5.07+ 2.03
5.48+ 2.15
6.87+ 2.79
8.06+ 3.18
1149
F=63.99
3
p<0.0001
(1)
Activity at surface
0.10
0.20
0.39
0.31
4536
X =1573.
C0
p<0.0001
% time burrowing
12.56 %
6.31 %
3.83 %
5.31 %
108
F= 1.44
3
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.

30
Response of crabs of different size to initial burrow density
The three-level interaction between food, burrow density,
and size, as well as the two-level interactions between pairs of
these factors, were not significant for most variables examined
by ANOVAs. It was significant (F=2.11, df=3, p<0.1, N=255) for
the interaction of size with initial burrow density for the
percent of unbranched burrows.
Crabs of different sizes were affected differently by
initial burrow density (Tables XII and IVc). This is suggested
by a qualitative comparison between size classes in Tables XII
and IVc, and by the significant interaction between size and
burrow density for the percent of unbranched burrows. In arenas
with a lower initial burrow density, for crabs of the smallest
size class, branching ratio and burrow depth were significantly
lower; the time allocation to burrowing and other activities was
not affected by the treatment. For some of the larger crabs,
time allocation to various activities was significantly affected
by the treatment; the fraction of time allocated to burrowing was
higher in arenas with a lower initial burrow density. No
differences occurred for branching ratio and depth.
Response of crabs of different size to food availability
Crabs of different sizes were affected differently by food
availability (Tables XIII and Vic). This is suggested by a
qualitative comparison between size classes in Tables XIII and
Vic. However, variance was high, and the interaction between
size and food availability was not significant for any variable

Table XII. Response of burrowing activity by Uca pugnax of different sizes to
initial burrow density.
Variable Low density High density N Test df Significance
(a) Size class 1
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
(b) Size class 2
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
(c) Size class 3
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
(d) Size class 4
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
64.00 %
74-68 %
304
1.39+ 0.54
1.40+ 0.55
304
4.72+ 1.92
5.39+ 2.05
226
11.48 %
13.72 %
27
60.28 %
62.98 %
422
1.46+ 0.61
1.40+ 0.55
422
5.59+ 2.32
5.36+ 1.96
345
9.63 %
2.74 %
27
66.07 %
65.61 %
413
1.40+ 0.60
1.38+ 0.56
413
6.71+ 2.56
7.10+ 3.00
350
3.12 %
4.60 %
27
70.59 %
66.64 %
254
1.33+ 0.55
1.37+ 0.60
254
7.93+ 2.88
8.26+ 3-47
225
7.21 %
3.55 %
27
X2=4.08
1
p<0.0043
F= 3.88
1
p<0.0497
(1)
F= 6.63
1
p<0.0106
(1)
F= 0.01
1
NS
(2)
X2=0.32
1
NS
F= 0.87
1
NS
(1)
F= 0.17
1
NS
(1)
F= 5.13
1
p<0.03
(2)
X2=0.01
1
NS
F= 0.04
1
NS
(1)
F= 1.00
1
NS
(1)
F= 0.02
1
NS
(2)
X2=0.11
1
NS
F= 0.28
1
NS
(1)
F= 0.18
1
NS
(1)
F= 1.46
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.

Table XIII. Response of burrowing by Uca rapax of different sizes to food availability
Variable
Low food
High food
N
Test
df
Significance
(a) Size class 1
% unbranched burrows
57.14 %
79.88 %
304
X =18.39
1
P<0.0001
Branching ratio
1.46+ 0.55
1.22+ 0.46
304
F=17.76
1
pCO.0001
(1)
Burrow depth (cm)
4.61+ 1.82
5.42+ 2.10
226
F= 8.60
1
p<0.0037
(1)
% time burrowing
5.01 %
20.69 %
27
F= 2.82
1
NS
(2)
(b) Size class 2
% unbranched burrows
54-71 %
69.35 %
422
X =9.53
1
p<0.002
Branching ratio
1.51+ 0.60
1.34+ 0.54
422
F= 9.16
1
p<0.0026
(1)
Burrow depth (cm)
5.27+ 2.05
5.75+ 2.24
346
F= 3.55
1
NS
(1)
% time burrowing
8.94 %
5.21 %
21
F= 0.04
1
NS
(2)
(c) Size class 3
% unbranched burrows
62.83 %
68.47 %
413
X =1.45
1
NS
Branching ratio
1.42+ 0.58
1.36+ 0.58
413
F= 1.04
1
NS
(1)
Burrow depth (cm)
6.47+ 2.47
7.24+ 2.96
350
F= 4.99
1
P<0.0261
(1)
% time burrowing
3.94 %
3.72 %
27
F= 0.04
1
NS
(2)
(d) Size class 4
% unbranched burrows
66.38 %
72.46 %
254
X =1.10
1
NS
Branching ratio
1.41+ 0.62
1.30+ 0.52
254
F= 1.79
1
NS
(1)
Burrow depth (cm)
7.38+ 2.40
8.66+ 3.58
226
F= 6.20
1
NS
(1)
% time burrowing
4.61 %
6.07 %
27
F= 0.10
1
NS
(2)
(1) Log-transorraed data.
(2) Arcsine-transformed data.

33
examined. In arenas with a lower food availability, for size
classes 1 and 2, branching ratio was higher, and burrow depth was
lower (although the difference was not significant in size class
2). For size class 3 burrow depth was also significantly lower.
No differences occurred between treatments for the largest size
class.

DISCUSSION
Burrow/crab Ratios in the Salt Marsh
The burrow/crab ratio was found to be higher than 1; thus,
there were excess burrows. Burrow/crab ratios higher than one
have been found in previous studies. Both male and female crabs
dig new burrows with each tidal cycle. Burrows are relatively
stable in a muddy substrate with a root mat, particularly in the
medium Spartina zone. When abandoned, they persist through
several tidal cycles (Powers, 1973; Crane, 1975; Basan and Frey,
1977; Hyatt and Salmon, 1977; Katz, 1980; Bertness and Miller,
1984).
Spartina shoot density and biomass, and sediment organic
content were reasonably high though comparable to values for
other medium Spartina marshes (Kurz and Wagner, 1957; Basan and
Frey, 1977; Cammen et al., 1980; Gosselink et al., 1984) and root
mat density was relatively low (C.S. Hopkinson, pers. comm.).
Sediment water content was very similar to that found in a salt
marsh in Georgia, and indicative of a relatively firm substrate
and high burrow holding capacity (Teal and Kanwisner, 1961;
Kraeuter and Wolf, 1974; Bertness and Miller, 1984). All these
characteristics, together with a low degree of burrow
intersection (discussed below; see also Basan and Frey, 1977) are
"typical" of medium Spartina marshes. These characteristics
34

35
should allow variations in crab burrowing activity in response to
food supply, because they help maintain a high enough Uca
population density to have a significant impact on Spartina.
Furthermore, burrows in this type of substrate may last long
enough to help maintain a burrow/crab ratio >1 and help enhance
Spartina growth.
Uca rapax density was reasonably high compared to reported
densities (Allen and Curran, 1974 Wolf et al., 1975; Aspey,
1978; Katz, 1980; Montague et al., 1981; Bertness and Miller,
1984; Cammen et al., 1984). The sex ratio (M/F) was 0.79. Sex
ratios higher than 1 are often observed (Wolf et al., 1975;
Bertness and Miller, 1984; Genoni, 1985), but this may be due
partly to sampling biases, as females often burrow more deeply
than males and move closer to the water edge in the days before
the release of larvae (Schwartz and Safir, 1915). There are also
discrete portions of the marsh where males display (display
areas; Greenspan, 1980); Bertness and Miller (1984) however,
noted that sex ratios may be lower in a medium Spartina zone than
in other areas of the same marsh, presumably because the
difference in burrowing ability between females and males is
broader in the substrate of this zone. As for size class
distribution, it was more skewed toward small sizes than in the
study by these authors. However, population structure can vary
remarkably between seasons and years in a salt marsh
(Shanholtzer, 1975; Cammen et al., 1984; Genoni, 1985).
_Maximum burrow depth was found to be 11 cm, which might be
an underestimate due to the limitations of the casting technique:
the presence of water in burrows sometimes prevented the resin

36
from filling them completely. Katz (1980) found burrows as deep
as 15 cm, and Bertness (1985) 20 cm, a value also attained in the
arenas in this study. Basan and Frey (1977) found burrows 25 cm
deep. Fiddler crabs dig even more deeply in winter (Crane,
1975). The significant association of burrows with structural
elements of the substrate is probably related to the relatively
low root mat density, which helps reduce the probability of
burrow collapse (Bertness and Miller, 1984) Burrow
configurations reflected little intersection and branching, and
most burrows were J-shaped, with a straight shaft, an enlarged
terminal chamber and irregular walls. Other burrows were
branched or interconnected. This low degree of branching and
interconnection may be related to the relatively high burrow
holding capacity of the substrate (Bertness and Miller, 1984),
and the irregularity of walls to the protrusion of plant roots.
Allen and Curran (1974), Basan and Frey (1977) and Katz (1980)
give similar descriptions of configurations. These include
nearly straight (J-shaped), highly sinuous shafts, Y- or U-shaped
systems and more complex systems with straight shafts or U-shaped
components; the complexity decreases from the marsh edge to the
higher marsh. Most burrows have an enlarged lower end, where the
resident crab presumably remains during the high tide or when
molting or mating.
Effects of Burrow Density and Food Availability
Uca rapax also dug excess burrows. In all treatments,
fiddler crabs dug new burrows despite the presence of preexisting

37
unoccupied burrows. Additionally, they adjusted their burrowing
activity to burrow density and food availability. These crabs
are known to adjust their burrowing activity to a variety of
conditions, such as stem density, root mat density, substrate
water, ground temperature, tidal and diurnal zeitgebers,
reproductive activity, threat by potential predators, season, and
male display activity, itself a function of many of these factors
(Zucker, 1974; Allen and Curran, 1974; Ringold, 1979; Bertness,
1985; L.W. Powers, pers. comm.). The results presented here
suggest that the digging of excess burrows and burrow branches
varies inversely with food availability. Thus, part of the
benefit of burrowing may be to farm Spartina.
Organic content in the surface layer of untreated arenas was
lower (88 %) than the average organic content in the marsh,
presumably because of the lack of macroscopic detritus. Yet this
value is within the range observed in the marsh. Since fiddler
crabs are food limited in the medium marsh (Genoni, 1985), this
value was presumed to indeed reflect a low food availability.
Average and maximum burrow depth in arenas were greater than
those found in the salt marsh, but this may simply reflect a more
successful casting due to a lower water content. In fact, depths
observed in the arenas are comparable to the ones found in the
salt marsh studies cited above. The initial depth of artificial
burrows (15 cm) was presumably adequate, since crabs did not
remodel to greater depths.
The significant association of burrows with structural
elements is probably related to the lack of a root mat, which
makes the need of structural support for burrows particularly

38
critical. Burrow configurations were similar to those observed
in the marsh, and could be assigned to the same categories
described above. However, burrow walls were smoother, due to the
lack of a root mat. The lack of a root mat would presumably
reduce burrow longevity. However, longevity in bare sediment was
found by Bertness and Miller (1984) to be in excess of 1 wk.
Since each trial was ended at or before 7.5 d, results should not
be significantly biased by the collapse of abandoned burrows.
Fiddler crabs allocated 5*9 % of their activity to
burrowing, which is probably an underestimate due to the visual
sampling technique, since burrowing activity will continue below
the surface. The display activity seemed low, perhaps due to the
lack of display areas in the confinement of the arenas; to
insufficient time to establish display activity; to the reduced
light intensity in the laboratory; or to the phase of the tidal
cycle (Greenspan, 1980). Although season affects burrowing
activity (Crane, 1975) this did not complicate the present study
since field and laboratory experiments were conducted during the
same season.
Effect of initial burrow density
Fiddler crabs dug new burrows and branches even in arenas
with a high initial burrow density. A lower initial burrow
density, however, resulted in a higher number of branched and
unbranched burrows. There was a trend for an increase in time
spent burrowing and for less deep burrows, suggesting that time
for burrowing was allocated to digging new burrows and branches
rather than to making existing burrows deeper. Burrows were,

39
then, abandoned more often in low density arenas than in high
density arenas, and burrow turnover was higher. This sensitivity
to the existing burrow density may reflect a tendency to dig
burrows wherever possible. New burrows in a substrate with a low
density may be less likely to collapse, thus making excavation of
new burrows beneficial to the crabs. A second reason may be a
stronger competition for territories under higher burrow density.
Uca are responsive to the presence of otner crabs, and are more
likely to colonize or burrow in areas with lower apparent crab
densities, such as areas with vegetation cover (Bertness and
Miller, 1984; Vliet, 1981). Some species, including Uca rapax,
construct half-domes or shelters. One function of these may be
to reduce their visual contact with other males, thereby reducing
aggression (Zucker, 1974 1981).
Effect of food availability
Fiddler crabs dug more burrow branches, but average burrow
depth was lower, in arenas receiving no food additions. Thus,
time investment in burrowing was similar, but it was allocated to
digging new burrow branches rather than digging new burrows or
making existing burrows deeper. Digging burrow branches near the
surface may better stimulate Spartina growth. The
physico-chemical and nutrient requirements in the upper substrate
layers may be most critical to Spartina growth. Indeed, in the
upper layer there is a higher density of roots and rhizomes
(Valiela et al., 1976; Howes et al., 1981; Hopkinson and
Schubauer, 1985), and nutrients may be more easily extracted from
sediment (DeLaune and Patrick, 1980). Furthermore, in the upper

40
layer, oxygenation by Spartina (metabolic oxidation and passive
0^ release), nutrient inputs, or burrows have a stronger effect
on these parameters (Howes et al., 1981; Bertness, 1985)* In
addition, digging branches near the surface may better enable
crabs to keep control of the original burrow, thus reducing the
risk of losing it to another crab, at the same time reducing the
risk of being without a burrow should the new burrow collapse or
encounter some construction obstacle. These constraints may be
particularly critical where food is limiting. In contrast, in
the low density situation, both unbranched and branched burrows
are excavated.
Comparison of burrowing by crabs of each sex
Females spent approximately half as much time outside their
burrows as did males. This observation may be an artifact of the
visual sampling technique, as females are actually more involved
in burrowing, removing them from view. Also they are less
involved in agonistic and courtship activities. Additionally,
gravid females need to protect their brood. A higher burrow
turnover rate may be expected for females, because (a) they have
an increased food requirement (see below); (b) they may be
dislodged by large males, which are less able to burrow due to
their large claw (Pearse, 1914; Hyatt and Salmon, 1977; Bertness
and Miller, 1984); (c) they allocate less time to agonistic and
courtship activities; and (d) non-receptive females dig more if
forced to interact with displaying males (L.W. Powers,
pers. comm.), an artifact of the arenas.

41
Males burrowed more deeply and constructed more shelters,
which agrees with findings of Crane (1975) Zucker (1971 1974)
and Bertness and Miller (1984)*
Response of crabs of each sex to initial burrow density
In arenas with a lower initial burrow density, females dug
more burrows and remained more in their burrows. Because of
their higher burrowing ability, females may be more likely to
take advantage of a low burrow density by digging new burrows.
Females were already spending more time burrowing than males but
tended to increase their burrowing time further. Males defend
display territories and breeding burrows (Greenspan, 1980), and
may be also sensitive to perceived population density. Indeed,
most shelters were constructed by males. Because females are
less involved in agonistic and courtship activities, they may
respond to a high population density by digging fewer burrows and
remaining more in their burrows rather than by constructing
shelters as males do.
Response of crabs of each sex to food availability
There were no differences between sexes in the effect of
food availability on various parameters of burrowing activity.
Thus, food availability seemed to affect the two sexes equally.
Comparison of burrowing by crabs of different size
Smaller crabs were generally less active outside their
burrows, which may be due partly to an artifact of the visual
sampling technique, as small crabs tended to be more involved in

42
burrowing, thus being visible less frequently, and to their being
less involved in agonistic and courtship activities. A higher
burrow turnover rate may be expected for smaller crabs, because
(a) they have an increased food requirement, (see below),
(b) they lose their burrows more often to larger, competitively
dominant, crabs (Pearse, 1914 Hyatt and Salmon, 1977; Bertness
and Miller, 1984); (c) they do not allocate as much time (or any
at all in the case of the smallest ones) to agonistic and
courtship activities; (d) burrow longevity is lower for smaller
burrows when there is no root mat (Bertness and Miller, 1984);
and (e) small males may dig more if forced to interact with
displaying males (L.W. Powers, pers. comm.), an artifact of the
arenas.
Larger crabs burrowed more deeply and constructed more
shelters, which agrees with findings of Crane (1975), Zucker
(1971, 1974) and Bertness and Miller (1984). Smaller crabs, even
though they tended to allocate more time to burrowing, can
excavate only small amounts of sediment at a time.
Response of crabs of different size to initial burrow density
Time allocation to burrowing was increased for larger crabs
in arenas with a low initial burrow density, but not by smaller
crabs. Smaller crabs already tend to spend more time burrowing
than larger crabs and may not be able to increase their burrowing
further. Rather, smaller crabs respond by burrowing less deeply
and allocating their burrowing time to new burrows. In addition,
sensitivity to burrow density may be particularly high in small
crabs because of their relatively high population density.

43
Because small crabs are less involved in agonistic and courtship
activities, they may respond to a high population density by
digging more unbranched and branched burrows and remaining more
in their burrows, rather than by reducing their time allocation
to burrowing and constructing shelters as larger crabs do.
Response of crabs of different size to food availability
Smaller crabs burrowed more under low food availability, but
again dug more branches rather than increasing their time
allocated to burrowing. Small crabs may be particularly
sensitive to food availability because they may stand more chance
of deriving benefit from their investment during their lifetime;
because they can invest more time in burrowing as opposed to
displaying, etc.; because their better burrowing ability may
allow them to better adjust their burrowing to food availability;
and because their need for food may be greater for growth. Early
stages are particularly sensitive to food availability: megalops
larvae may select a settling substrate partly on the basis of its
organic content (Crane, 1975) and early crab stages may be more
limited by food supply (Genoni, 1985). Given these reasons,
smaller crabs may be more efficient at farming Spartina.
Furthermore, they constitute a large fraction of the population.
General Considerations
As with many complex biological interactions, the observed
results might be attributable to secondary interactions, or to
indirect cause-and-effect relationships. For example, burrowing

44
activity might have been affected by relative differences in
physical conditions or in microtopography caused by the density
of initial burrows or the addition of food flakes. However, in
light of what is presently known about Uca rapax burrow dynamics,
the effects of their burrows on Spartina growth, in addition to
the fact that they are food limited, the results would
collectively suggest a farming relationship.
Although fiddler crabs are presumably capable of responding
to Spartina detritus availability, the food treatment used in
this study consisted, instead, of artificial food. An attempt
was made at using Spartina detritus, derived from standing dead
Spartina that was collected, dried, thin-cut in a hammer mill,
and aged in pens in the salt marsh for 8-12 weeks. Microorganism
load was measured as the gross primary production (GPP) of the
microbial population: the release of CO^ was monitored with an
Infra-Red Analyzer, and GPP was calculated by adding the slopes
of CO^ concentration in the dark (community respiration) and in
the light (net community production) (Soeder and Tailing, 1969).
Microorganism load was extremely variable (by two orders of
magnitude) between samples, and therefore detritus was deemed
unsuitable as a controlled food source. This also assumes that
Spartina detritus, albeit limiting in quantity, provides a
suitable substrate for microorganisms, which constitute the food
of Uca rapax. Conversion of detritus to microbial biomass is
relatively high (Gosselink and Kirby, 1974 Haines and Hanson,
1979). Indeed, it constitutes a significant portion of the diet
of these crabs (Shanholtzer, 1973; Haines, 1976; Montague,
1980b). Thus, the benefit of obtaining an increased detritus

45
supply may not be merely a by-product of their burrowing activity
(Montague, 1980a). The increased investment in burrowing under
low food supply may be cost-effective because of the indirect
benefit accruing to the crabs.
It would be interesting to compare responses to food
availability of crabs of different sizes, in the salt marsh. If
small crabs are more sensitive to food limitation, and respond to
low food availability by increasing their burrowing, then their
burrowing activity should be inversely proportional to their
organic content. Additionally, if smaller crabs are more
efficient at farming Spartina, selective removal of large or
small crabs should have different effects on Spartina production.
The sand fiddler crab, Uca pugilator (Bose) would provide a
useful comparison, as it often occurs sympatrically with
Uca rapax or Uca pugnax in salt marshes. Burrows of
Uca pugilator are excavated in sandy substrates, and have usually
one opening. They include short temporary burrows (that last one
tidal cycle) and deep breeding burrows, the latter resembling, in
their configuration, those of Uca rapax (Basan and Frey, 1977;
Christy, 1982). These characteristics (short life of one kind,
use for breeding of the other kind) suggest that a flexibility in
adjusting time allocation to burrowing, burrow density, branching
and depth may not be as high as in Uca rapax. Moreover, since
this species feeds mainly on benthic algae (Miller, 1961), it
presumably does not farm grass, and although it does dig excess
burrows (L.W. Powers, pers. comm.), it may be predicted not to do
so in response to food limitation.

APPENDIX
MORPHOMETRIC ANALYSIS OF CARAPACE AND CHELA PROPORTIONS
IN UCA RAPAX
Morphometric analyses in decapods have generated much
interest because of their relevance to many facets of life
history and their taxonomic diagnostic value (Teissier, 1960;
Barnes 1968; Hartnoll, 1982). Allometric constraints affect
growth and reproductive output as well as many aspects of basic
physiology (e.g. Miller, 1971; Savage and Sullivan, 1978; Hines,
1982). In burrowing animals such as Uca, they may affect
burrowing behavior and mating strategies: they allow predictions
of minimal burrow diameter (Bertness and Miller, 1984) or largest
mate (Uca rapax breed in the male's burrow; Greenspan, 1980).
This morphometric analysis correlates the basic
measurements, carapace breadth (or width) and carapace length (or
depth). Carapace breadth is often used as the reference
dimension for morphometric studies in brachyurans (see
e.g. Barnes, 1968; Hines, 1982; Davidson and Marsden, 1987).
However, in the case of Uca, carapace length is more satisfactory
(Miller, 1971; Crane, 1975). I have used both of these
dimensions. The development of the major cheliped in males and
its role in territorial defence, combat, display and courtship,
are of special interest in fiddler crabs (Crane, 1975). Indeed
it is a useful measurement for associating male size to mate size
and for predicting mating success of males (Miller, 1971;
46

47,
Greenspan, 1980). Regressions of propodus length and dactylus
length against carapace breadth and length were calculated. The
slopes and elevations of regression lines were compared between
the two sexes.
The analysis includes 330 males and 308 female Uca rapax,
ranging from 4.50 to 21.50 mm in carapace breadth. Carapace
breadth (CB), carapace length (CL), propodus length (PL) and
dactylus length (DL) were measured with vernier calipers to the
nearest 0.05 mm. CB was measured across the long axis of the
carapace at its maximum breadth; CL was measured medially along
the minor axis of the carapace, from the front to the posterior
margin; PL was measured along its lower margin, from the
articulation with the carpus to the tip of the immovable digit;
and DL from the articulation point on the propodus to the tip
(Crane, 1975). Linear regressions are good approximations of
correlations between dimensions of similar magnitude (Hartnoll,
1982). Linear regressions of the form y=ax+b, where y is the
dependent variable, x the reference variable, and a and b are
constants (Sokal and Rohlf, 1969) were done for CB vs. CL and CL
vs. CB for females and males. Because the growth of the major
cheliped in males is likely to follow a power function of the
form y=ax^ (Hartnoll, 1982), regressions of log(PL) and of
corresponding power function was derived. The adjusted
coefficient of determination (r ) and the standard error on the
estimate of the slope were calculated. Sexual differences in
slope and elevation between regression lines were compared by
t-tests (Zar, 1974)-

48
The relationships between carapace breadth and length
approximate closely straight lines (Figures 2 and 3) while the
ratio of male chela measurements to carapace breadth and length
follows a logarithmic straight line (Figures 4 and 5)> as in some
other Ocypodidae (Barnes, 1968) (Table XIV). Similarly, the
carapace breadth is proportionally larger in larger individuals.
It can also be noted from comparisons of the standard errors on
the estimates of slopes that for any given carapace breadth, the
size of the male chela shows a greater variation than carapace
length.
Regressions for carapace measurements of females and males
were significantly different in slope and elevation (t=4947 and
5-945 respectively, p<0.001, df=635).
Comparison of the standard errors on the estimates of the
slopes suggest that the measurement for the major cheliped
exhibit more variation than carapace measurements. This may be
due entirely or in part to smaller, regenerating chelipeds in
some individuals, and has been observed in other Ocypodidae
(Barnes, 1968). However, the range and rate of change of these
morphometric proportions may have taxonomic or biogeographical
significance (Barnes, 1968).

CARAPACE LENGTH (cm)
49
Figure 2. Regression of carapace length against carapace breadth
for Uca rapax. F=females, M=males, ALL=A11 crabs.

CARAPACE BREADTH (cm)
50
Figure 3. Regression of carapa.e breadth against carapac
for Uca rapax. F=females, M=males, ALL=all crabs.
length

51
Log (CARAPACE BREADTH) ,
Figure 4* Regression of log (PL) (propodus length) and log (DL;
(dactylus length) against log (CB) (carapace breadth) for
Uca rapax.

52
1
Log(CARAPACE LENGTH)
Figure 5. Regression of log (PL) (propoaus length) and los (DL)
(dactylus length) against log (CL) (carapace length) for
Uca rapax.

53
Table XIV. Summary of allometrie relationships for carapace breadth
(CB), carapace length (CL), propodus length (PL) and dactylus length
(DL) of Uca rapax. s.e.=standard error of estimate on the slope.
Type of
p
Sex
Variables
regression
Growth equation
s.e.
r
N
F
CL
vs.
CB
linear
y=0.6504x+0.0245
0.007
0.9691
308
M
y=0.6253x+0.0418
0.005
0.9783
330
All
y=0.6330x+0.0378
0.004
0.9739
638
F
CB
vs.
CL
linear
y=1.4901x-0.0030
0.015
0.9690
308
M
y=1.5645x-0.0394
0.015
0.9783
330
All
y=1.5386x-0.0282
0.010
0.9739
638
M
PL
vs.
CB
power
y:!-?x-o.2607
y=2!25S6x0-1172
0.055
0.9339
79
M
PL
vs.
CL
power
0.063
0.9236
79
M
DL
vs.
CB
power
0.064
0.9360
79
M
DL
vs.
CL
power
0.073
0.9247
79

LITERATURE CITED
Altmann, J., 1974* Observational sampling of behavior: sampling
methods. Behav., Vol. 49 pp. 227-265.
Allen, E.A. and H.A. Curran, 1974 Biogenic sedimentary-
structures produced by crabs in lagoon margins and salt
marsh environments near Beaufort, North Carolina.
J. Sedim. Petrol., Vol. 44 PP* 538-548.
Aspey, P., 1978. Fiddler crab burrowing ecology: burrow density
in Uca pugnax (Smith) and Uca pugilator (Bose).
Crustaceans, Vol. 34 PP* 235-244*
Barnes, R.S.K., 1968. Relative carapace and chela proportions in
some ocypodid crabs (Brachyura, Ocypodidae). Crustaceans,
Vol. 14, pp. 131-136.
Basan, P.B. and R.W. Frey, 1977. Actual paleontology and
neoichnology of salt marshes near Sapelo Island, Georgia.
In, Trace fossils 2, Geological J. Special Issue #9, edited
by T.P. Crimes and J.C. Harper, Seel House Press, Liverpool,
pp. 41-70.
Bertness, M.D., 1985. Fiddler crab regulation of Spartina
alterniflora production on a New England salt marsh. Ecol.,
Vol. 66, pp. 1042-1055.
Bertness, M.D. and T. Miller, 1984* The distribution and
dynamics of Uca pugnax (Smith) burrows in a New England salt
marsh. J. Exp. Mar. Biol. Ecol., Vol. 83 pp. 211-237.
Cammen, L.M., E.D. Seneca and L.M. Stroud, 1980. Energy flow
through the fiddler crabs Uca pugnax and Uca minax and the
marsh periwinkle Littorina irrorata in a North Carolina salt
marsh. Am. Midi. Nat., Vol. 103 pp. 687-694*
54

55
Caminen, L.M., E.D. Seneca and L.M. Stroud, 1984. Long-term
variations of fiddler crab populations in North Carolina
salt marshes. Estuaries, Vol. 7 PP 171-175.
Christy, J.H., 1982. Burrow structure and use in the sand
fiddler crab, Uca pugilator (Bose). Anim. Behav., Vol. 30,
pp. 687-694*
Crane, J., 1975. Fiddler crabs of the world (Ocypodidae:
Genus Uca). Princeton Univ. Press, Princeton, N.J.
736 pp. + xxiii.
Darwin, C., 1881. The formation of vegetable mould through
the action of worms, with observations on their habits.
Edited by J. Hurray, D. Appleton and Co., London.
326 pp. + xxix.
Davidson, R.J. and I.D. Marsden, 1987. Size relationships and
relative growth of the New Zealand swimming crab Ovalipes
catharus (White, 1845). J. Crust. Biol., Vol. 17,
pp. 308-317.
DeLaune, R.D. and W.H. Patrick, 1980. Nitrogen and phosphorus
cycling in a Gulf Coast salt marsh. In,
Estuarine perspectives, edited by7 V.S. Kennedy, Academic
Press, New York, pp. 143-151.
Dembowski, J., 1926. Notes on the behavior of the fiddler crab.
Biol. Bull. (Woods Hole), Vol. 50, pp. 179-201.
Dixon, W.J., 1985. BMDP statistical software manual.
Univ. California Press, Berkeley. 734 PP*
Frey, R.W., P.B. Basan and R.M. Scott, 1973. Techniques for
sampling salt marsh benthos and burrows. Am. Midi. Nat.,
Vol. 89, pp. 228-234-
Genoni, G.P., 1985. Food limitation in salt marsh fiddler crabs
Uca rapax (Smith, 1870) (Decapoda, Ocypodidae). J. Exp.
Mar. Biol. Ecol., Vol. 87, pp. 97-110.
Gosselink, J.G., R. Hatton and C.S. Hopkinson, 1984
Relationships of organic carbon and mineral content to bulk
density in Louisiana salt marsh soils. Soil Sci., Vol. 137,
pp. 177-181.

56-
Go sse link, J.G. and C.J. Kirby, 1974 Decomposition of salt
marsh grass, Spartina alterniflora Loisel. Limnol
Qceanogr., Vol. 19, pp. 825-832.
Greenspan, B.N., 1980. Male size and reproductive success in the
communal courtship system of the fiddler crab Uca rapax.
Anim. Behav., Vol. 28, pp. 387-392.
Haines, E.B., 1976. Relation betwen the stable carbon isotope
composition of fiddler crabs, plants and soils in a salt
marsh. Est. Coast. Mar. Sci., Vol. 4 PP 609-616.
Haines, E.B. and R.B. Hanson, 1979. Experimental degradation of
detritus made from the salt marsh plants Spartina
alterniflora Loisel., Spartina virginiana L. and Juncus
roemerianus Scheele. J. Exp. Mar. Biol. Ecol., Vol. 40,
pp. 27-40
Hartnoll, R.G., 1982. Growth. In, The biology of Crustacea,
Vol. 2, edited by L.G. Abele, Academic Press, New York,
pp. 111-196.
Hines, A.H., 1982. Allometric constraints and variables of
reproductive effort in brachyuran crabs. Mar. Biol.,
Vol. 69, pp. 309-320.
Hopkinson, C.S. and J.P. Schubauer, 1984. Static and dynamic
aspects of nitrogen cycling in the salt marsh graminoid
Spartina alterniflora. Ecol., Vol. 65, pp. 961-969.
Howes, B.L., R.W. Howarth, J.M. Teal and I. Valiela, 1981.
Oxidation-reduction potentials in a salt marsh: spatial
patterns and interaction with primary production.
Limnol. Qceanogr., Vol. 26, pp. 350-360.
Hyatt, G.W. and M. Salmon, 1977. Combat in the fiddler crabs Uca
pugilator and Uca pugnax: a quantitative analysis.
Behaviour, Vol. 65, pp. 182-211.
Hyman, O.W., 1922. Adventures in the life of a fiddler crab.
Ann. Rep. Smiths. Insn. for 1920, pp. 443-459.
Katz, L.C., 1980. Effects of burrowing by the fiddler crab, Uca
pugnax (Smith). Estuar. Coast. Mar. Sci., Vol. 11,
pp. 233-237.

57
Kitchell, J.F., R.V. O'Neill, D. Webb, G.W. Gallepp,
S.M. Barbell, J.F. Koonce, and B.S. Ausmus, 1979. Consumer
regulation of nutrient cycling. Bioscience, Vol. 29,
pp. 28-34*
Kraeuter, J.N. and P.L. Wolf, 1974* The relationship of marine
macroinvertebrates to salt marsh plants. In, Ecology of
halophytes, edited by R.J. Reimold and W.H. Queen, Academic
Press, New York, pp. 449-462.
Kurz, H. and K. Wagner, 1957. Tidal marshes of the Gulf and
Atlantic coasts of northern Florida and Charleston, South
Carolina. Florida State Univ. Series in Biology, No. 24,
168 pp.
Miller, D.C., 1961. The feeding mechanism of fiddler crabs, with
ecological considerations of feeding adaptations.
Zoolgica, Vol. 46, pp. 89-100.
Miller, D.C., 1971. Growth in Uca. 1. Ontogeny of asymmetry in
Uca pugilator (Bose) (Decapoda, Ocypodidae). Crustaceana,
Vol. 24, pp. 119-131.
Montague, C.L., 1980a. The net influence of the fiddler crab,
Uca pugnax, on carbon flow through a Georgia salt marsh:
the importance of work by macroorganisms to the metabolism
of ecosystems. Ph.D. Dissertation, University of Georgia,
Athens.
Montague, C.L., 1980b. A natural history of temperate Western
Atlantic fiddler crabs (genus Uca) with reference to their
impact on the salt marsh. Contrib. Mar. Sci., Vol. 23,
pp. 25-55.
Montague, C.L., S.M. Bunker, E.B. Haines, M.L. Pace and
R.L. Wetzel, 1981. Aquatic macroconsumers. In, The
ecology of a salt marsh, edited by L.R. Pomeroy and
R.G. Wiegert, Springer-Verlag, New York, pp. 69-85.
Paine, R.T., 1971. The measurement and application of the
calorie to ecological problems. Annu. Rev. Ecol, Syst.,
Vol. 2, pp. 145-164.
Patten, B.C. and E.P. Odum, 1981. The cybernetic nature of
ecosystems. Am. Nat., Vol. 118, pp. 886-895.

38
Pearse, A.S., 1914* Habits of fiddler crabs. Rep. Smiths.
Insn., pp. 415-428.
Powers, L.W., 1973. Ecological aspects of burrows and fiddler
crab behavior. Am. Zool,, Vol. 13 pp. 1271.
Ringold, P., 1979. Burrowing, root mat density, and the
distribution of fiddler crabs in the Eastern United States.
J. Exp. Mar. Biol. Ecol., Vol. 3 pp. 11-21.
Robertson, J.R., K. Bancroft, G.K. Vermeer and K. Plaisier, 1980.
Experimental studies on the foraging behavior of the sand
fiddler crab Uca pugilator (Bose). J. Exp. Mar. Biol.
Ecol., Vol. 44, pp. 67-83.
Robertson, J.R., J.A. Fudge and G.K. Vermeer, 1981. Chemical and
live feeding stimulants of the sand fiddler crab, Uca
pugilator (Bose). J. Exp, Mar. Biol. Ecol., Vol. 53,
pp. 47-64.
SAS Institute Inc., 1985a. SAS user's guide. Basics. Version 5
edition. Cary, N.C.: SAS Institute Inc., 1290 pp.
SAS Institute Inc., 1985b. SAS user's guide. Statistics.
Version 5 edition. Cary, N.C.: SAS Institute Inc., 956 pp.
Savage, T. and J.R. Sullivan, 1978. Growth and claw regeneration
of the stone crab, Menippe mercenaria. Florida Mar. Res.
Publ., No. 32, 23 pp.
Schwartz, B. and S.R. Safir, 1915* The natural history and
behavior of the fiddler crabs. Cold Spring Harbor Monogr.,
Vol. 8, pp. 1-24.
Shanholtzer, S.F., 1973. Energy flow, food habits, and
population dynamics of Uca pugnax in a salt marsh system.
Ph.D. Dissertation, University of Georgia, Athens. 91 pp.
Shinn, E.A., 1968. Burrowing in recent lime sediments of Florida
and the Bahamas. J. Paleontol., Vol. 43, pp. 879-894.
Snedecor, G.W. and W.G. Cochran, 1980. Statistical methods.
7th edition. Iowa State Univ. Press, Ames, Iowa. 507 pp.

59
Soeder, C.J., and J.F. Tailing, 1969. Measurements (in situ) on
isolated samples of natural communities. In, A manual
on methods for measuring primary production in aquatic
environments, IBP handbook No. 12, edited by
R.A. Vollenweider, Blackwell Scientific Publications,
Oxford, pp. 62-73.
Sokal, R.R. and F.J. Rohlf, 1969. Biometry. Freeman, San
Francisco, CA. 776 pp.
Teal, J.M. and J. Kanwisher, 1961. Gas exchange in a salt marsh.
Limnol. Oceanogr., Vol. 6, pp. 388-399.
Teissier, G. 1960. Relative growth. In,
The physiology of Crustacea, Vol. 1, edited by
T.H. Waterman, Academic Press, New York, pp. 537-560.
Valiela, I., J.M. Teal and N.Y. Persson, 1976. Production and
dynamics of experimentally enriched salt marsh vegetation:
belowground biomass. Limnol. Oceanogr., Vol. 21,
pp. 245-252.
Vliet, K.A., 1981. Population density and size-class
distribution in a population of the sand fiddler crab,
Uca pugilator. Univ. Florida, unpubl.
Wilson, D.S., 1980. The natural selection of populations and
communities. Benjamin/Cummings Publ., Menlo Park, CA.
186 pp.
Wolf, P.L., S.F. Shanholtzer and R.J. Reimold, 1975. Population
estimates for Uca pugnax (Smith, 1870) on the Duplin Estuary
marsh, Georgia, U.S.A. (becapoda, Brachyura, Ocypodidae).
Crustaceana, Vol. 29, pp. 79-91.
Young, D.Y. and H.W. Ambrose, III, 1978. Underwater orientation
in the fiddler crab, Uca pugilator. Biol. Bull., Vol. 155,
pp. 246-258.
Zar, J.H., 1974* Biostatistical analysis. Prentice Hall,
Princeton, N.J., 620 pp.
Zucker, N., 1974* Shelter building as a means of reducing
territory size in the fiddler crab, Uca terpsichores
(Crustacea: Ocypodidae). Am. Midi. Nat., Vol. 91, p. 224.

60
Zucker, N., 1981. The role of hood-building in defining
territories and limiting combat in fiddler crabs.
Anim. Behav., Vol. 29 pp. 387-395.

BIOGRAPHICAL SKETCH
Giulio Piero Genoni was born in Milano, Italy, in 1957. He
received the Licence es Sciences Naturelies (equivalent to the
degree of Bachelor of Science in biology) from the University of
Lausanne, Switzerland, in 1979. He worked at the Institute of
Biochemistry of the Zuerich Institute of Technology during 1980.
He took a course in oceanography at the University of
Aix-Marseille, France, during the 1980-1981 year. In the fall of
1981 he enrolled in the Department of Zoology of the University
of Florida, and he obtained the Master of Science degree in 1984*
His doctoral research was supported by research assistanships
from the Department of Fisheries and Aquaculture, teaching
assistantships from the Department of Zoology, the University of
Florida Marine Laboratory, and a grant-in-aid of research of the
Sigma Xi Research Society.
61

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Dr. Frank J.SVMaturo
Professor of Zoology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Professor of Zoology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Environmental Engineering

I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Dr. Michele G. Wheatly
Assistant Professor of ZjQpiogy
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Dr. William J. ^¡indberg ^
Assistant Professor of Fisheries qnd
Aquaculture
This dissertation was submitted to the Graduate Faculty of the
Department of Zoology in the College of Liberal Arts and Sciences
and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of
Philosophy.
December 1987
Dean, Graduate School



INTRODUCTION
Organisms may regulate or control other tropnic levels by
establishing interactions with each other and feedback
relationships (Kitchell et al., 1979; Montague et al., 1961;
Patten and Odum, 1951). In salt marshes, fiddler crabs have a
large impact on cordgrass, Spartina alterniflora: their
burrowing activities, bioturbation and fecal pellet production
increase the growth of Spartina, resulting in an increased food
supply in the form of microorganisms that colonize the grass when
it dies and decays (Montague, 1980a, 1980b; Katz, 1980; bertness,
1985). These crabs are food limited in the salt marsh (Genoni,
1985). Selection may favor activities and by-products that would
decrease food limitation, such as, perhaps, burrowing activity.
Fiddler crabs may be said, then, to "farm" salt marsh grass, like
some earthworms "farm" plants in their habitat (Darwin, 1881), an
indirect effect sensu Wilson (1980).
Burrows have a number of direct advantages for fiddler
crabs: they function as shelters from predators and
environmental extremes; provide water for physiological needs;
and are sites for reproduction (Crane, 1975; Hyatt and Salmon,
1977; Young and Ambrose, 1978; Ringold, 1979). They may serve
other presently unidentified functions. If fiddler crabs dig
more burrows than they require for these needs, then the benefit
1


9
Association of burrows with structural elements (grass stems) was
estimated as described for the field study.
Counts of new unbranched and branched burrows and
measurements of excavated sediment were divided by the duration
of each trial in days. These variables, burrow depths, and
branching ratios (defined as the ratio of the number of branches
2
to the number of burrows) were analysed by 2 factorial ANOVAs
for tests of the effects of food and density treatments and their
interaction. The assumption of normality was tested by
two-tailed t-tests on skewness and kurtosis; the assumption of
homogeneity of variances was tested by Levene's one-way ANOVA.
If these assumptions were not met, factorial ANOVAs were done on
log-transformed data (the assumptions were tested again).
Comparisons between treatments of the percent of unbranched
2
burrows were done by X tests of independence.
Comparisons of responses to initial burrow density and to
food availability were also made for each sex and each size
class. Crabs of each sex or size class may not be equally
sensitive to food supply, because of different foraging
efficiencies or food assimilation, or because dominance by one
sex or by larger crabs may limit food availability to other crabs
(Hyman, 1922; Crane, 1975). Comparisons of shelter building
2
between sexes and size classes was done by X tests of
independence.
To investigate time allocation to different activities,
including burrowing, crabs were marked individually with numbered
plastic tags affixed with cyanoacrylate glue to their carapace.
For every arena, activities of individual crabs were recorded


13
Table I. (a) Summary of environmental and population characteristics
observed at the salt marsh site and used as initial conditions in the
experimental arenas; (b) Matching of initial burrow diameter with crab
size. CB=carapace breadth.
(a)
Variable
Field study
Arenas
Sediment type
Sandy mud
Sandy mud
Sediment water
64.6 %
55.4 %
Burrow density
212/m2
High: 302/m2 (64/arena)
Low: 205/m (43/arena)
Uca density
156/m2
p
205/m (43/arena)
Burrow/crab ratios
1.36
High: 1.5
Low: 1.0
Burrow angles
64.8 % 60-80
60-80
Burrow configurations
89.0 % J-shaped
Straight
Uca sex ratio (M/F)
0.79
0.79
Uca size-class distribution, by sex
CB 6 8 mm 18.7# F, 13.7% M
CB 8 11 mm 13.7 10.1
CB 11 15 mm 17.3 11.5
CB >15 mm 6.5 8.6
18.6% F, 14.0% M
14.0 9.3
16.3 11.6
9.3 7.0
Food available to Uca
9.7 kcal/(m2x yr)(l)
High: 19.0 kcal/(m2x yr)
Low: 2.6 kcal/(m x yr)
(b)
Size class
Carapace length (2)
Burrow diameter
CB 6 8 mm
CB 8 11 mm
CB 11-15 mm
CB > 15 mm
5.44 mm
7*34 mm
9.87 mm
11.46 mm
5.50 mm
7.40 mm
9.90 mm
12.20 mm
(1) Cammen et al. (1980).
(2) CL was predicted on the basis of a regression of carapace length
against carapace breadth. The morphometric analysis of carapace and
chela measurements is presented in the Appendix.


CARAPACE LENGTH (cm)
49
Figure 2. Regression of carapace length against carapace breadth
for Uca rapax. F=females, M=males, ALL=A11 crabs.


10
from circular visual scans. Successive observations were made
every 10 s for 15 min. This procedure was repeated 1 to 4 times,
on different days, during 7 of the trials. The percent of the
total time allocated to various activities was estimated from the
frequencies of their occurrence among the observations (scan
sampling; Altmann, 1974)* Activities were grouped as follows:
standing still (motionless), walKing, eating, cleaning, agonism
(combat and chasing), display and courtship, and burrow
construction and maintenance. General comparisons between
treatments of the fraction of time allocated to various
2
activities were done by X tests of independence across all
treatments, for all crabs, and for each sex and size class.
Comparisons of the fraction of time allocated to certain
activities (e.g. burrowing) were done on arcsine-transformed data
by factorial ANOVAs. An index of relative activity of crabs of
each sex or size class was calculated from the frequency of
sightings observed and expected from the sex ratios and size
class distribution in the arenas. Activity levels were compared
2
by X tests of indepedence.
Statistical analyses were done on an IBM 5081-D32/3090-200
mainframe computer at the North Eastern Regional Data Center of
the State University System, Gainesville, FL, using the
Statistical Analysis System (SAS Institute Inc., 1985a, 1985b)
and the Biomedical Data Program P-Series (Dixon, 1985)


food availability to test whether crabs adjust their burrowing
activity in response to food. Results showed that (a) there were
more burrows than fiddler crabs, (b) crabs dug new burrows
despite the presence of unoccupied burrows, and (c) their
burrowing activity varied inversely with food availability. Food
availability affected the rate of burrowing by females and males
equally, but had a stronger effect on burrowing by small crabs
than on that of large crabs. Thus, small crabs may be more
sensitive to food limitation, and may be more efficient in
"farming" cordgrass.
v


59
Soeder, C.J., and J.F. Tailing, 1969. Measurements (in situ) on
isolated samples of natural communities. In, A manual
on methods for measuring primary production in aquatic
environments, IBP handbook No. 12, edited by
R.A. Vollenweider, Blackwell Scientific Publications,
Oxford, pp. 62-73.
Sokal, R.R. and F.J. Rohlf, 1969. Biometry. Freeman, San
Francisco, CA. 776 pp.
Teal, J.M. and J. Kanwisher, 1961. Gas exchange in a salt marsh.
Limnol. Oceanogr., Vol. 6, pp. 388-399.
Teissier, G. 1960. Relative growth. In,
The physiology of Crustacea, Vol. 1, edited by
T.H. Waterman, Academic Press, New York, pp. 537-560.
Valiela, I., J.M. Teal and N.Y. Persson, 1976. Production and
dynamics of experimentally enriched salt marsh vegetation:
belowground biomass. Limnol. Oceanogr., Vol. 21,
pp. 245-252.
Vliet, K.A., 1981. Population density and size-class
distribution in a population of the sand fiddler crab,
Uca pugilator. Univ. Florida, unpubl.
Wilson, D.S., 1980. The natural selection of populations and
communities. Benjamin/Cummings Publ., Menlo Park, CA.
186 pp.
Wolf, P.L., S.F. Shanholtzer and R.J. Reimold, 1975. Population
estimates for Uca pugnax (Smith, 1870) on the Duplin Estuary
marsh, Georgia, U.S.A. (becapoda, Brachyura, Ocypodidae).
Crustaceana, Vol. 29, pp. 79-91.
Young, D.Y. and H.W. Ambrose, III, 1978. Underwater orientation
in the fiddler crab, Uca pugilator. Biol. Bull., Vol. 155,
pp. 246-258.
Zar, J.H., 1974* Biostatistical analysis. Prentice Hall,
Princeton, N.J., 620 pp.
Zucker, N., 1974* Shelter building as a means of reducing
territory size in the fiddler crab, Uca terpsichores
(Crustacea: Ocypodidae). Am. Midi. Nat., Vol. 91, p. 224.


rigure 1. Effect of initial burrow density and food availability
on burrowing activity of Uca rapax (all crabs) (N=17). Bars
represent 1.96 standard error (95 % confidence intervals).


41
Males burrowed more deeply and constructed more shelters,
which agrees with findings of Crane (1975) Zucker (1971 1974)
and Bertness and Miller (1984)*
Response of crabs of each sex to initial burrow density
In arenas with a lower initial burrow density, females dug
more burrows and remained more in their burrows. Because of
their higher burrowing ability, females may be more likely to
take advantage of a low burrow density by digging new burrows.
Females were already spending more time burrowing than males but
tended to increase their burrowing time further. Males defend
display territories and breeding burrows (Greenspan, 1980), and
may be also sensitive to perceived population density. Indeed,
most shelters were constructed by males. Because females are
less involved in agonistic and courtship activities, they may
respond to a high population density by digging fewer burrows and
remaining more in their burrows rather than by constructing
shelters as males do.
Response of crabs of each sex to food availability
There were no differences between sexes in the effect of
food availability on various parameters of burrowing activity.
Thus, food availability seemed to affect the two sexes equally.
Comparison of burrowing by crabs of different size
Smaller crabs were generally less active outside their
burrows, which may be due partly to an artifact of the visual
sampling technique, as small crabs tended to be more involved in


8
surveying their burrows. The following size classes were used,
as determined by carapace breadth: 6-8, 8-11, 11-15 and >15 mm.
Because competition for burrows of adequate size may be stronger
for larger crabs (Bertness and Miller, 1984) the diameter of
initial burrows was varied to match crab carapace length (crabs
enter their burrows sideways). Crabs were left in the arenas
4.0-7.5 days, at 25C, in a laboratory equipped with windows and
with fluorescent lights set on a timer simulating the
sunrise-sunset rhythm. Thus, the natural photoperiod was
supplemented by an artificial photoperiod. Arenas were separated
by blinds from the rest of the laboratory. The sediment was
maintained wet by additions of 50 % seawater.
The following parameters were used to describe fiddler crab
burrowing activity: the number of new unbranched and branched
burrows, the amount of excavated sediment, and the percent of
time allocated to burrowing (Katz, 1980; Montague, 1980a). At
the end of each trial, excavated material was collected, dried
48 h at 70 C, and weighed to the nearest 0.01 g. Newly
excavated burrows were marked. I noted the presence of
half-domes of mud (or "shelters"; Zucker, 1974) constructed by
some individuals on burrow openings. To determine the
characteristics of each burrow and identify resident crabs,
burrow casts were made with latex concrete (Christy, 1982). The
concrete caused resident crabs to escape. Their sex and carapace
breadth were noted. This procedure was repeated with newly
opened burrows on the next two days. Burrow casts were removed
and their depth, diameter and configuration were noted.


27
Response of crabs of each sex to food availability
Crabs of the two sexes were not affected differently by food
availability, with respect to percent of unbranched burrows,
branching ratio, depth, time allocation to different activities,
or fraction of time allocated to burrowing (Tables X and VIb).
This is suggested by a qualitative comparison between the two
sexes in Tables X and VIb, and by the non-significant interaction
between sex and food for each variable examined.
Comparison of burrowing by crabs of different size
There was no overall difference between crabs of different
sizes in the percent of unbranched burrows or branching ratio of
burrows (Table XI). Smaller crabs were less active at the
surface, allocated different amounts of time to various
activities and burrowed less deeply (Tables IX and lie). There
was a strong trend for a higher time investment in burrowing by
smaller crabs. Larger crabs constructed more shelters than did
smaller crabs (Table VIII), and allocated significantly more time
to agonism (F=9.24> df=3, p<0.0001, N=108) and to display
(F=451, df=3, p<0.005, N=108). There was a strong trend for a
higher time expenditure in foraging for the smaller size classes.


FARMING OF C0RDGRAS3, SPARTINA ALTERNIFLORA LOISEL.,
EY FIDDLER CRABS, (JCA RAPAX (SMITH)
(DECAPODA: OCYPODIDAE)
BY
GIULIO PIERO GENONI
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
19S7


7
for other needs (protective, reproductive and physiological);
intermediate in (2), to stimulate food; and in (3), to meet other
needs; and lowest in (4).
Food treatments consisted of 0 g and 11 g, respectively, of
Purina Fly Larvae Medium added to each arena. For a 7-day trial,
2
the high value corresponds to 19 kcal/(m x day), i.e. about
twice the highest estimate of average energy consumption of
fiddler crabs calculated by Cammen et al. (1980)
2 2
(3522 kcal/(m x yr) or 9*7 kcal/(m x day)). The arenas where
food was not added contained some organic matter mixed with the
sediment, as evidenced by the feeding response of crabs. This
behavior is mediated by chemoreceptors in the minor chelae
(Robertson et al., 1980, 1981). To measure the organic content
of the upper 0.5 cm of sediment, cores (2.7 cm in diameter) were
taken from each arena before each trial, and ash-free dry weight
was determined as outlined above.
Initial burrows (43 and 64 per arena, for high and low
density treatments) were made by pushing dowels 15 cm into the
substrate, at angles of 60-80. Thus, burrow densities were
9 2
205/m and 302/m Burrow openings were marked with Spartina
stems cut to 10 cm, on which numbered tags were stapled.
Additional stems were placed in the low-density arenas to achieve
similar stem densities.
Uca rapax were captured at the Matanzas Inlet site and
43 individuals were distributed into each arena. The sex ratio
and size class distribution in each arena were similar to those
observed in the sampling study, with the exception that no crabs
<6 mm in carapace breadth were used, due to the difficulty of


44
activity might have been affected by relative differences in
physical conditions or in microtopography caused by the density
of initial burrows or the addition of food flakes. However, in
light of what is presently known about Uca rapax burrow dynamics,
the effects of their burrows on Spartina growth, in addition to
the fact that they are food limited, the results would
collectively suggest a farming relationship.
Although fiddler crabs are presumably capable of responding
to Spartina detritus availability, the food treatment used in
this study consisted, instead, of artificial food. An attempt
was made at using Spartina detritus, derived from standing dead
Spartina that was collected, dried, thin-cut in a hammer mill,
and aged in pens in the salt marsh for 8-12 weeks. Microorganism
load was measured as the gross primary production (GPP) of the
microbial population: the release of CO^ was monitored with an
Infra-Red Analyzer, and GPP was calculated by adding the slopes
of CO^ concentration in the dark (community respiration) and in
the light (net community production) (Soeder and Tailing, 1969).
Microorganism load was extremely variable (by two orders of
magnitude) between samples, and therefore detritus was deemed
unsuitable as a controlled food source. This also assumes that
Spartina detritus, albeit limiting in quantity, provides a
suitable substrate for microorganisms, which constitute the food
of Uca rapax. Conversion of detritus to microbial biomass is
relatively high (Gosselink and Kirby, 1974 Haines and Hanson,
1979). Indeed, it constitutes a significant portion of the diet
of these crabs (Shanholtzer, 1973; Haines, 1976; Montague,
1980b). Thus, the benefit of obtaining an increased detritus


RESULTS
Burrow/crab Ratios in the Salt Marsh
At the study site, the sediment was sandy mud, and sediment
water was 64*59 % of wet weight (s.d.=3*l6, N=17). Organic
matter in the upper 0*5 cm was 9*05 % (s.d.=2.64, N=12).
2
Spartina shoot density was 138.1/m (N=17) and standing stock
2
was 269*52 dry g/m (N=15)* The root mat (roots and rhizomes),
between 0 and 20 cm, was 0.11 dry g/ml (s.d.=0.04 N=15).
2 2
There were 212 burrows/m (N=17) and 156 fiddler crabs/m
(N=17). Thus the burrow/crab ratio was 1.36. Sex ratio (M/F)
was 0.79 (N=565 crabs). Size class distribution was 31*0 %
(carapace breadth <6 mm), 20.6 % (6-8), 17.0 #(8-11), 23.2 %
(11-15) and 8.2 % (>15) (N=548 crabs).
Burrows were significantly associated with structural
elements of the substrate; using a value of 70.0 % available bare
substrate (Bertness and Miller, 1984) a conservative estimate
considering the shoot density at the study site, burrows occurred
near stems or mussels more frequently than expected by chance
alone (37.0% of all burrows, X2=29.90, df=1, p<0.0001, N=1284).
Burrow angles from the substrate ranged from 30 to 90 but
most (64.3 %, N=213 burrows) were at an angle between 60 and
80 Most burrows (89.0 %, N=427) were unbranched and J-shaped,
their main shaft being usually straight but sometimes comprising
11


60
Zucker, N., 1981. The role of hood-building in defining
territories and limiting combat in fiddler crabs.
Anim. Behav., Vol. 29 pp. 387-395.


52
1
Log(CARAPACE LENGTH)
Figure 5. Regression of log (PL) (propoaus length) and los (DL)
(dactylus length) against log (CL) (carapace length) for
Uca rapax.


55
Caminen, L.M., E.D. Seneca and L.M. Stroud, 1984. Long-term
variations of fiddler crab populations in North Carolina
salt marshes. Estuaries, Vol. 7 PP 171-175.
Christy, J.H., 1982. Burrow structure and use in the sand
fiddler crab, Uca pugilator (Bose). Anim. Behav., Vol. 30,
pp. 687-694*
Crane, J., 1975. Fiddler crabs of the world (Ocypodidae:
Genus Uca). Princeton Univ. Press, Princeton, N.J.
736 pp. + xxiii.
Darwin, C., 1881. The formation of vegetable mould through
the action of worms, with observations on their habits.
Edited by J. Hurray, D. Appleton and Co., London.
326 pp. + xxix.
Davidson, R.J. and I.D. Marsden, 1987. Size relationships and
relative growth of the New Zealand swimming crab Ovalipes
catharus (White, 1845). J. Crust. Biol., Vol. 17,
pp. 308-317.
DeLaune, R.D. and W.H. Patrick, 1980. Nitrogen and phosphorus
cycling in a Gulf Coast salt marsh. In,
Estuarine perspectives, edited by7 V.S. Kennedy, Academic
Press, New York, pp. 143-151.
Dembowski, J., 1926. Notes on the behavior of the fiddler crab.
Biol. Bull. (Woods Hole), Vol. 50, pp. 179-201.
Dixon, W.J., 1985. BMDP statistical software manual.
Univ. California Press, Berkeley. 734 PP*
Frey, R.W., P.B. Basan and R.M. Scott, 1973. Techniques for
sampling salt marsh benthos and burrows. Am. Midi. Nat.,
Vol. 89, pp. 228-234-
Genoni, G.P., 1985. Food limitation in salt marsh fiddler crabs
Uca rapax (Smith, 1870) (Decapoda, Ocypodidae). J. Exp.
Mar. Biol. Ecol., Vol. 87, pp. 97-110.
Gosselink, J.G., R. Hatton and C.S. Hopkinson, 1984
Relationships of organic carbon and mineral content to bulk
density in Louisiana salt marsh soils. Soil Sci., Vol. 137,
pp. 177-181.


33
examined. In arenas with a lower food availability, for size
classes 1 and 2, branching ratio was higher, and burrow depth was
lower (although the difference was not significant in size class
2). For size class 3 burrow depth was also significantly lower.
No differences occurred between treatments for the largest size
class.


Table II. Time allocation (%) to various activities by Uca rapax; (a) all crabs, (b) each sex
and (c) each size class.
Still
Eat
Walk
Burrow
Clean
Agonism
Display
N
Test
df
Significance
(a)
36.7
36.0
11.7
5.9
5.3
3.0
1.3
4628
(b)
4493
X2=131.69
6
p<0.0001
Females
35.5
38.3
12.8
8.3
4.5
0.7
-
1722
Males
37.2
35.2
10.9
4-1
6.0
4*4
2.2
2771
(c)
4536
X2=168.53
6
p<0.0001
Size class
1
23.4
55.2
5.2
8.4
6.8
0.9
-
440
2
39.3
37.4
10.9
7.0
3.3
1.4
0.6
905
3
36.9
35.6
12.6
5.0
5.5
2.8
1.6
1777
4
39.8
29.8
12.5
5.3
5.5
4.7
1.8
1414
&


30
Response of crabs of different size to initial burrow density
The three-level interaction between food, burrow density,
and size, as well as the two-level interactions between pairs of
these factors, were not significant for most variables examined
by ANOVAs. It was significant (F=2.11, df=3, p<0.1, N=255) for
the interaction of size with initial burrow density for the
percent of unbranched burrows.
Crabs of different sizes were affected differently by
initial burrow density (Tables XII and IVc). This is suggested
by a qualitative comparison between size classes in Tables XII
and IVc, and by the significant interaction between size and
burrow density for the percent of unbranched burrows. In arenas
with a lower initial burrow density, for crabs of the smallest
size class, branching ratio and burrow depth were significantly
lower; the time allocation to burrowing and other activities was
not affected by the treatment. For some of the larger crabs,
time allocation to various activities was significantly affected
by the treatment; the fraction of time allocated to burrowing was
higher in arenas with a lower initial burrow density. No
differences occurred for branching ratio and depth.
Response of crabs of different size to food availability
Crabs of different sizes were affected differently by food
availability (Tables XIII and Vic). This is suggested by a
qualitative comparison between size classes in Tables XIII and
Vic. However, variance was high, and the interaction between
size and food availability was not significant for any variable


Table VI. Response to food availability of time allocation to various activities {%) by Uca rapax;
(a) all crabs, (b) each sex and (c) each size class.
Still
Eat
Walk
Burrow
Clean
Agonism
Display
N
Test
df
Significance
(a)
4628
X2=134.58
6
p<0.0001
Low food
40.6
29.5
13.7
6.0
4.4
4.0
1.8
High food
32.4
43.1
9.5
5.9
6.4
1.8
0.8
(b)
Females
1722
X2=3543
5
p<0.0001
Low food
39.4
31.9
14.2
9.2
4.0
1.2
-
High food
31.7
44-4
11.4
7.4
4.9
0.2
-
Males
2771
X =103.53
6
p<0.0001
Low food
41-4
28.4
13.1
3.9
4.6
5.7
2.9
High food
32.2
43.2
8.3
4-3
7.7
3.0
1.3
(c)
Size class
1
440
X2=25.01
5
p<0.0001
Low food
32.7
47.1
5.3
7.2
5.8
1.9
-
High food
15.1
62.5
5.2
9.5
7.8
-
-
Size class
2
905
X =26.29
6
pCO.0001
Low food
37-4
34-9
13-4
8.9
2.1
2.6
0.6
High food
41-4
40.2
8.3
4.8
4.6
0.2
0.5
Size class
3
1777
X =55.83
6
p<0.0001
Low food
40.8
29.1
14-3
4-4
4.8
4.1
2.6
High food
33.1
42.0
11.0
5.6
6.3
1.5
0.7
Size class
4
1414
X =63.50
6
p<0.0001
Low food
45.5
22.3
14*4
5.6
4-9
5.1
2.2
High food
32.3
39.7
9.8
4-9
7.7
4.3
1.3


Table IV. Response to initial burrow density of time allocation to various activities (%) by Uca rapax;
(a) all crabs, (b) each sex and (c) each size class.
Still
Eat
Walk
Burrow
Clean
Agonism
Display
N
Test
df
Significance
(a)
4628
X2=37.79
5
p<0.0001
Low density
34.8
39.1
10.3
6.8
4.7
3.3
1.1
High density
38.7
32.9
13.2
5.1
6.0
2.7
1.5
(b)
Females
1722
X2=29-96
5
p<0.0001
Low density
32.6
42.9
11.3
9.1
3.3
0.8
-
High density 38.7
33.2
14.4
7.4
5.7
0.6
-
O
Males
2771
X =18.24
6
p<0.006
Low density
35-4
37.5
9.4
4-9
5.9
5.0
1.9
High density
38.8
33.0
12.3
3.4
6.2
3.9
2.5
(c)
Size class 1
440
X2=8.23
5
NS
Low density
19.3
56.3
5.3
9.1
8.0
1.5
-
High density
28.8
53.7
5.1
7.3
5.1
-
-
O
Size class 2
905
X =14.93
6
p<0.021
Low density
38.2
38.0
9.6
9.2
2.5
1.8
0.8
High density
40.8
36.7
12.8
4.1
4-3
1.0
0.3
P
Size class 3
1777
X =35.55
6
p<0.0001
Low density
33-4
41.8
12.0
4.3
5.2
2.3
0.5
High density
39.9
30.4
13.2
5.6
5.8
2.6
2.6
p
Size class 4
1414
x =17.24
6
p<0.008
Low density
40.0
30.8
10.3
6.7
4.8
5.1
2.5
High density
39.6
28.8
14.5
4.0
7.3
4-4
1.3


LITERATURE CITED
Altmann, J., 1974* Observational sampling of behavior: sampling
methods. Behav., Vol. 49 pp. 227-265.
Allen, E.A. and H.A. Curran, 1974 Biogenic sedimentary-
structures produced by crabs in lagoon margins and salt
marsh environments near Beaufort, North Carolina.
J. Sedim. Petrol., Vol. 44 PP* 538-548.
Aspey, P., 1978. Fiddler crab burrowing ecology: burrow density
in Uca pugnax (Smith) and Uca pugilator (Bose).
Crustaceans, Vol. 34 PP* 235-244*
Barnes, R.S.K., 1968. Relative carapace and chela proportions in
some ocypodid crabs (Brachyura, Ocypodidae). Crustaceans,
Vol. 14, pp. 131-136.
Basan, P.B. and R.W. Frey, 1977. Actual paleontology and
neoichnology of salt marshes near Sapelo Island, Georgia.
In, Trace fossils 2, Geological J. Special Issue #9, edited
by T.P. Crimes and J.C. Harper, Seel House Press, Liverpool,
pp. 41-70.
Bertness, M.D., 1985. Fiddler crab regulation of Spartina
alterniflora production on a New England salt marsh. Ecol.,
Vol. 66, pp. 1042-1055.
Bertness, M.D. and T. Miller, 1984* The distribution and
dynamics of Uca pugnax (Smith) burrows in a New England salt
marsh. J. Exp. Mar. Biol. Ecol., Vol. 83 pp. 211-237.
Cammen, L.M., E.D. Seneca and L.M. Stroud, 1980. Energy flow
through the fiddler crabs Uca pugnax and Uca minax and the
marsh periwinkle Littorina irrorata in a North Carolina salt
marsh. Am. Midi. Nat., Vol. 103 pp. 687-694*
54


I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Dr. Frank J.SVMaturo
Professor of Zoology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Professor of Zoology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Environmental Engineering


37
unoccupied burrows. Additionally, they adjusted their burrowing
activity to burrow density and food availability. These crabs
are known to adjust their burrowing activity to a variety of
conditions, such as stem density, root mat density, substrate
water, ground temperature, tidal and diurnal zeitgebers,
reproductive activity, threat by potential predators, season, and
male display activity, itself a function of many of these factors
(Zucker, 1974; Allen and Curran, 1974; Ringold, 1979; Bertness,
1985; L.W. Powers, pers. comm.). The results presented here
suggest that the digging of excess burrows and burrow branches
varies inversely with food availability. Thus, part of the
benefit of burrowing may be to farm Spartina.
Organic content in the surface layer of untreated arenas was
lower (88 %) than the average organic content in the marsh,
presumably because of the lack of macroscopic detritus. Yet this
value is within the range observed in the marsh. Since fiddler
crabs are food limited in the medium marsh (Genoni, 1985), this
value was presumed to indeed reflect a low food availability.
Average and maximum burrow depth in arenas were greater than
those found in the salt marsh, but this may simply reflect a more
successful casting due to a lower water content. In fact, depths
observed in the arenas are comparable to the ones found in the
salt marsh studies cited above. The initial depth of artificial
burrows (15 cm) was presumably adequate, since crabs did not
remodel to greater depths.
The significant association of burrows with structural
elements is probably related to the lack of a root mat, which
makes the need of structural support for burrows particularly


36
from filling them completely. Katz (1980) found burrows as deep
as 15 cm, and Bertness (1985) 20 cm, a value also attained in the
arenas in this study. Basan and Frey (1977) found burrows 25 cm
deep. Fiddler crabs dig even more deeply in winter (Crane,
1975). The significant association of burrows with structural
elements of the substrate is probably related to the relatively
low root mat density, which helps reduce the probability of
burrow collapse (Bertness and Miller, 1984) Burrow
configurations reflected little intersection and branching, and
most burrows were J-shaped, with a straight shaft, an enlarged
terminal chamber and irregular walls. Other burrows were
branched or interconnected. This low degree of branching and
interconnection may be related to the relatively high burrow
holding capacity of the substrate (Bertness and Miller, 1984),
and the irregularity of walls to the protrusion of plant roots.
Allen and Curran (1974), Basan and Frey (1977) and Katz (1980)
give similar descriptions of configurations. These include
nearly straight (J-shaped), highly sinuous shafts, Y- or U-shaped
systems and more complex systems with straight shafts or U-shaped
components; the complexity decreases from the marsh edge to the
higher marsh. Most burrows have an enlarged lower end, where the
resident crab presumably remains during the high tide or when
molting or mating.
Effects of Burrow Density and Food Availability
Uca rapax also dug excess burrows. In all treatments,
fiddler crabs dug new burrows despite the presence of preexisting


47,
Greenspan, 1980). Regressions of propodus length and dactylus
length against carapace breadth and length were calculated. The
slopes and elevations of regression lines were compared between
the two sexes.
The analysis includes 330 males and 308 female Uca rapax,
ranging from 4.50 to 21.50 mm in carapace breadth. Carapace
breadth (CB), carapace length (CL), propodus length (PL) and
dactylus length (DL) were measured with vernier calipers to the
nearest 0.05 mm. CB was measured across the long axis of the
carapace at its maximum breadth; CL was measured medially along
the minor axis of the carapace, from the front to the posterior
margin; PL was measured along its lower margin, from the
articulation with the carpus to the tip of the immovable digit;
and DL from the articulation point on the propodus to the tip
(Crane, 1975). Linear regressions are good approximations of
correlations between dimensions of similar magnitude (Hartnoll,
1982). Linear regressions of the form y=ax+b, where y is the
dependent variable, x the reference variable, and a and b are
constants (Sokal and Rohlf, 1969) were done for CB vs. CL and CL
vs. CB for females and males. Because the growth of the major
cheliped in males is likely to follow a power function of the
form y=ax^ (Hartnoll, 1982), regressions of log(PL) and of
corresponding power function was derived. The adjusted
coefficient of determination (r ) and the standard error on the
estimate of the slope were calculated. Sexual differences in
slope and elevation between regression lines were compared by
t-tests (Zar, 1974)-


BIOGRAPHICAL SKETCH
Giulio Piero Genoni was born in Milano, Italy, in 1957. He
received the Licence es Sciences Naturelies (equivalent to the
degree of Bachelor of Science in biology) from the University of
Lausanne, Switzerland, in 1979. He worked at the Institute of
Biochemistry of the Zuerich Institute of Technology during 1980.
He took a course in oceanography at the University of
Aix-Marseille, France, during the 1980-1981 year. In the fall of
1981 he enrolled in the Department of Zoology of the University
of Florida, and he obtained the Master of Science degree in 1984*
His doctoral research was supported by research assistanships
from the Department of Fisheries and Aquaculture, teaching
assistantships from the Department of Zoology, the University of
Florida Marine Laboratory, and a grant-in-aid of research of the
Sigma Xi Research Society.
61


METHODS
Burrow/crab Ratios in the Salt Marsh
A sampling of Uca rapax population density, sex ratio, size
class distribution, and burrow density was done during Summer
1985 in an intertidal salt marsh at Matanzas Inlet, St. Johns
County, Florida (2945'50" N, 8117'20" W). The vertical tidal
range is ~1.9 m, and the distribution of vegetative zones is
fairly distinct. The site is dominated by a medium growth form
(50 cm tall) of cordgrass, Spartina alterniflora Loisel.
(140 shoots/m ). The sediment is fine mud with a small sandy
fraction.
To identify burrow/crab ratios, I used sampling techniques
as described by Frey et al. (1973), Wolf et al. (1975) and
Ringold (1979). An area of 12 by 12 m, which was apparently
homogeneous in Spartina height and density, Uca density, and
physical factors, was chosen. It was divided into 0.5 by 0.5 m
quadrats and marked permanently with dowels. Between 9 June and
6 September, 17 quadrats were chosen by the simple random method
(Snedecor and Cochran, 1980) for measurement of Uca and burrow
density. Each quadrat was enclosed with corrugated fiberglass
panels while water still covered the substrate and most crabs
were in their burrows (Wolf et al., 1975; Genoni, 1985). At low
tide, grass shoots within the quadrat were clipped near the
4


Table IX. Response of burrowing activity by each sex of Uca rapax to initial burrow density
Variable
Low density
High density N
Test
df Significance
(a) Females
2
% unbranched burrows
64.62 %
65.85 %
653
X =0.11
1
NS
Branching ratio
1.40+ 0.57
1.37+ 0.54
653
F= 0.27
1
NS
(1)
Burrow depth (cm)
6.03+ 2.76
6.07+ 2.64
532
F= 0.21
1
NS
(1)
% time burrowing
12.72 %
5.79 %
26
F= 3.27
1
NS
(2)
(b) Males
2
% unbranched burrows
65.05 %
68.82 %
732
x =1.17
1
NS
Branching ratio
1.40+ 0.59
1.35+ 0.55
732
F= 1.63
1
NS
(1)
Burrow depth (cm)
6.44+ 2.62
6.82+ 3.06
610
F= 2.30
1
NS
(1)
% time burrowing
5.33 %
3.09 %
27
F= 2.02
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.


Table XII. Response of burrowing activity by Uca pugnax of different sizes to
initial burrow density.
Variable Low density High density N Test df Significance
(a) Size class 1
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
(b) Size class 2
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
(c) Size class 3
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
(d) Size class 4
% unbranched burrows
Branching ratio
Burrow depth (cm)
% time burrowing
64.00 %
74-68 %
304
1.39+ 0.54
1.40+ 0.55
304
4.72+ 1.92
5.39+ 2.05
226
11.48 %
13.72 %
27
60.28 %
62.98 %
422
1.46+ 0.61
1.40+ 0.55
422
5.59+ 2.32
5.36+ 1.96
345
9.63 %
2.74 %
27
66.07 %
65.61 %
413
1.40+ 0.60
1.38+ 0.56
413
6.71+ 2.56
7.10+ 3.00
350
3.12 %
4.60 %
27
70.59 %
66.64 %
254
1.33+ 0.55
1.37+ 0.60
254
7.93+ 2.88
8.26+ 3-47
225
7.21 %
3.55 %
27
X2=4.08
1
p<0.0043
F= 3.88
1
p<0.0497
(1)
F= 6.63
1
p<0.0106
(1)
F= 0.01
1
NS
(2)
X2=0.32
1
NS
F= 0.87
1
NS
(1)
F= 0.17
1
NS
(1)
F= 5.13
1
p<0.03
(2)
X2=0.01
1
NS
F= 0.04
1
NS
(1)
F= 1.00
1
NS
(1)
F= 0.02
1
NS
(2)
X2=0.11
1
NS
F= 0.28
1
NS
(1)
F= 0.18
1
NS
(1)
F= 1.46
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EB29ZNHQB_T5WQJP INGEST_TIME 2015-04-01T19:01:47Z PACKAGE AA00029849_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


Table III. Effect of initial burrow density on burrowing activity of Uca rapax (all crabs)
Variable
Low density
High density
Skewness
Kurtosis
Levene
N
Test
df
Significance
All new burrows
39.63+11.76
30.06+13.24
NS
NS
NS
17
F=11.13
1
p<0.0014
New unbranched burrows
24.74+ 9.41
14-73+ 9.19
NS
NS
NS
17
F=19.43
1
pCO.0001
Maintained burrows (3)
New branches to initial
8.62+ 1.23
9.11 j+ 1.31
p<0.001
NS
NS
16
F= 2.41
1
NS
(1)
burrows (3)
2.41+ 1.06
2.48+ 1.11
p<0.01
p<0.001
NS
16
F= 0.03
1
NS
(1)
Maintained branches (3)
11.09+ 1.57
11.58+ 1.79
p<0.001
p<0.001
NS
17
F= 1.50
1
NS
(1)
Branching ratio
1.27+ 0.58
1.22+ 0.58
p<0.0001
p<0.001
p<0.0001
3340
F= 9.21
1
p<0.0024
(1)
Burrow depth (cm)
5.57+ 2.66
5.68+ 2.76
p<0.0001
p<0.001
pCO.0001
2596
F= 1.25
1
NS
(1)
g excavated sediment (3)
70.91+19-.23
69.68+16.48
NS
NS
NS
17
F= 0.08
1
NS
% time burrowing
6.56 %
4-56 %
-
-
-
26
F= 1.72
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.
(3) Divided by the duration of each trial in days.


Table VII. Comparison of burrowing activity by each sex of Uca rapax
Variable
Females
Males
N
Test
df
Significance
% unbranched burrows
64.94 %
66.53 %
1391
X2=0.13
1
NS
Branching ratio
1.38+ 0.58
1.37+ 0.59
1385
F= 0.12
1
NS
(1)
Burrow depth (cm)
6.04+ 2.71
6.60+ 2.83
1142
Fs12.46
1
0
0
0

0
V
a.
(1)
Activity at surface
0.69
1.40
4493
X =560.88 1
p<0.001
% time burrowing
9.78 %
4-33 %
53
F= 4.33
1
P<0.04
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.


25
Table VIII. Fraction of the total numbers of shelters
constructed by Uca rapax, for each sex
and each size class. The interaction between sex
and size was not significant.
%
N
Test
df Significance
64
X2=36.04
1
p<0.0001
Females
18.8
Males
81.3
62
X2=51.48
3
pCO.0001
Size class
1
12.9
2
9.7
3
29.0
4
43.4


45
supply may not be merely a by-product of their burrowing activity
(Montague, 1980a). The increased investment in burrowing under
low food supply may be cost-effective because of the indirect
benefit accruing to the crabs.
It would be interesting to compare responses to food
availability of crabs of different sizes, in the salt marsh. If
small crabs are more sensitive to food limitation, and respond to
low food availability by increasing their burrowing, then their
burrowing activity should be inversely proportional to their
organic content. Additionally, if smaller crabs are more
efficient at farming Spartina, selective removal of large or
small crabs should have different effects on Spartina production.
The sand fiddler crab, Uca pugilator (Bose) would provide a
useful comparison, as it often occurs sympatrically with
Uca rapax or Uca pugnax in salt marshes. Burrows of
Uca pugilator are excavated in sandy substrates, and have usually
one opening. They include short temporary burrows (that last one
tidal cycle) and deep breeding burrows, the latter resembling, in
their configuration, those of Uca rapax (Basan and Frey, 1977;
Christy, 1982). These characteristics (short life of one kind,
use for breeding of the other kind) suggest that a flexibility in
adjusting time allocation to burrowing, burrow density, branching
and depth may not be as high as in Uca rapax. Moreover, since
this species feeds mainly on benthic algae (Miller, 1961), it
presumably does not farm grass, and although it does dig excess
burrows (L.W. Powers, pers. comm.), it may be predicted not to do
so in response to food limitation.


48
The relationships between carapace breadth and length
approximate closely straight lines (Figures 2 and 3) while the
ratio of male chela measurements to carapace breadth and length
follows a logarithmic straight line (Figures 4 and 5)> as in some
other Ocypodidae (Barnes, 1968) (Table XIV). Similarly, the
carapace breadth is proportionally larger in larger individuals.
It can also be noted from comparisons of the standard errors on
the estimates of slopes that for any given carapace breadth, the
size of the male chela shows a greater variation than carapace
length.
Regressions for carapace measurements of females and males
were significantly different in slope and elevation (t=4947 and
5-945 respectively, p<0.001, df=635).
Comparison of the standard errors on the estimates of the
slopes suggest that the measurement for the major cheliped
exhibit more variation than carapace measurements. This may be
due entirely or in part to smaller, regenerating chelipeds in
some individuals, and has been observed in other Ocypodidae
(Barnes, 1968). However, the range and rate of change of these
morphometric proportions may have taxonomic or biogeographical
significance (Barnes, 1968).


Table X. Response of burrowing activity by each sex of Uca rapax to food availability
Variable
Low food
High food
N
Test
df
Significance
(a) Females
% unbranched burrows
57.93 %
71.80 %
653
X =13.82
1
pCO.0001
Branching ratio
1.46+ 0.57
1.32+ 0.54
653
F=11.15
1
p<0.0009
(1)
Burrow depth (cm)
5.68+ 2.44
6.38+ 2.87
532
F= 7.34
1
p<0.007
(1)
% time burrowing
10.43 %
9.24 %
26
F= 0.15
1
NS
(2)
(b) Males
% unbranched burrows
60.72 %
72.65 %
732
X =11.74
1
p<0.001
Branching ratio
1.45+ 0.61
CM

0
o'

732
F=13.06
1
p<0.0003
(1)
Burrow depth (cm)
6.07+ 2.37
7.11+ 3.12
610
F=15.70
1
p<0.0001
(1)
% time burrowing
4.18 %
4.48 %
27
F= 0.30
1
NS
(2)
(1) Log-transformed data.
(2) Arcsine-transformed data.


APPENDIX
MORPHOMETRIC ANALYSIS OF CARAPACE AND CHELA PROPORTIONS
IN UCA RAPAX
Morphometric analyses in decapods have generated much
interest because of their relevance to many facets of life
history and their taxonomic diagnostic value (Teissier, 1960;
Barnes 1968; Hartnoll, 1982). Allometric constraints affect
growth and reproductive output as well as many aspects of basic
physiology (e.g. Miller, 1971; Savage and Sullivan, 1978; Hines,
1982). In burrowing animals such as Uca, they may affect
burrowing behavior and mating strategies: they allow predictions
of minimal burrow diameter (Bertness and Miller, 1984) or largest
mate (Uca rapax breed in the male's burrow; Greenspan, 1980).
This morphometric analysis correlates the basic
measurements, carapace breadth (or width) and carapace length (or
depth). Carapace breadth is often used as the reference
dimension for morphometric studies in brachyurans (see
e.g. Barnes, 1968; Hines, 1982; Davidson and Marsden, 1987).
However, in the case of Uca, carapace length is more satisfactory
(Miller, 1971; Crane, 1975). I have used both of these
dimensions. The development of the major cheliped in males and
its role in territorial defence, combat, display and courtship,
are of special interest in fiddler crabs (Crane, 1975). Indeed
it is a useful measurement for associating male size to mate size
and for predicting mating success of males (Miller, 1971;
46


6
then rehydrated, dried again, and weighed. The ash-free dry
weight was determined and the organic content was estimated as
the weight of the fraction lost in the ashing (Paine, 1971).
Total counts of burrow and crab density, and Spartina
biomass and density, were pooled for all 17 quadrats, whereas
data from subsamples were averaged over all quadrats.
Effects of Burrow Density and Food Availability
In a manipulative experiment in the laboratory, the effect
of food availability and burrow density on burrowing activity of
U. rapax was tested. Two sets of four circular arenas were
built, each consisting of a plastic pail on top of which a sheet
of metal was riveted to make a funnel with walls inclined at a
o
45 angle. The metal was lined with plastic and topped with a
10-cm strip of aluminium flashing that crabs could not climb.
The arenas were filled with sandy mud (collected from a creek in
the salt marsh) 45 cm deep, with an upper diameter of 52 cm and a
2
suriace area of 0.21 m Results from the field study were used
to determine the initial conditions used in the arenas. At the
beginning of each trial, complete randomization (Snedecor and
Cochran, 1980) was used in assigning each arena to one of four
treatments: (1) low food abundance and low burrow density, (2)
low food and high density, (3) high food and low density, and (4)
2
high food and high density. This 2 factorial experiment was
replicated in 17 trials, between April 7 and September 14, 1987.
The following responses were predicted: burrowing activity
should be highest in treatment (1), both to stimulate food and


35
should allow variations in crab burrowing activity in response to
food supply, because they help maintain a high enough Uca
population density to have a significant impact on Spartina.
Furthermore, burrows in this type of substrate may last long
enough to help maintain a burrow/crab ratio >1 and help enhance
Spartina growth.
Uca rapax density was reasonably high compared to reported
densities (Allen and Curran, 1974 Wolf et al., 1975; Aspey,
1978; Katz, 1980; Montague et al., 1981; Bertness and Miller,
1984; Cammen et al., 1984). The sex ratio (M/F) was 0.79. Sex
ratios higher than 1 are often observed (Wolf et al., 1975;
Bertness and Miller, 1984; Genoni, 1985), but this may be due
partly to sampling biases, as females often burrow more deeply
than males and move closer to the water edge in the days before
the release of larvae (Schwartz and Safir, 1915). There are also
discrete portions of the marsh where males display (display
areas; Greenspan, 1980); Bertness and Miller (1984) however,
noted that sex ratios may be lower in a medium Spartina zone than
in other areas of the same marsh, presumably because the
difference in burrowing ability between females and males is
broader in the substrate of this zone. As for size class
distribution, it was more skewed toward small sizes than in the
study by these authors. However, population structure can vary
remarkably between seasons and years in a salt marsh
(Shanholtzer, 1975; Cammen et al., 1984; Genoni, 1985).
_Maximum burrow depth was found to be 11 cm, which might be
an underestimate due to the limitations of the casting technique:
the presence of water in burrows sometimes prevented the resin


Table XIII. Response of burrowing by Uca rapax of different sizes to food availability
Variable
Low food
High food
N
Test
df
Significance
(a) Size class 1
% unbranched burrows
57.14 %
79.88 %
304
X =18.39
1
P<0.0001
Branching ratio
1.46+ 0.55
1.22+ 0.46
304
F=17.76
1
pCO.0001
(1)
Burrow depth (cm)
4.61+ 1.82
5.42+ 2.10
226
F= 8.60
1
p<0.0037
(1)
% time burrowing
5.01 %
20.69 %
27
F= 2.82
1
NS
(2)
(b) Size class 2
% unbranched burrows
54-71 %
69.35 %
422
X =9.53
1
p<0.002
Branching ratio
1.51+ 0.60
1.34+ 0.54
422
F= 9.16
1
p<0.0026
(1)
Burrow depth (cm)
5.27+ 2.05
5.75+ 2.24
346
F= 3.55
1
NS
(1)
% time burrowing
8.94 %
5.21 %
21
F= 0.04
1
NS
(2)
(c) Size class 3
% unbranched burrows
62.83 %
68.47 %
413
X =1.45
1
NS
Branching ratio
1.42+ 0.58
1.36+ 0.58
413
F= 1.04
1
NS
(1)
Burrow depth (cm)
6.47+ 2.47
7.24+ 2.96
350
F= 4.99
1
P<0.0261
(1)
% time burrowing
3.94 %
3.72 %
27
F= 0.04
1
NS
(2)
(d) Size class 4
% unbranched burrows
66.38 %
72.46 %
254
X =1.10
1
NS
Branching ratio
1.41+ 0.62
1.30+ 0.52
254
F= 1.79
1
NS
(1)
Burrow depth (cm)
7.38+ 2.40
8.66+ 3.58
226
F= 6.20
1
NS
(1)
% time burrowing
4.61 %
6.07 %
27
F= 0.10
1
NS
(2)
(1) Log-transorraed data.
(2) Arcsine-transformed data.


51
Log (CARAPACE BREADTH) ,
Figure 4* Regression of log (PL) (propodus length) and log (DL;
(dactylus length) against log (CB) (carapace breadth) for
Uca rapax.


% TIME BURROW BRANCHING EXCAVATED BURROWS
BURROWING DEPTH RATIO SEDIMENT (g) AND BRANCHES


I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Dr. Michele G. Wheatly
Assistant Professor of ZjQpiogy
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for
the degree of Doctor of Philosophy.
Dr. William J. ^¡indberg ^
Assistant Professor of Fisheries qnd
Aquaculture
This dissertation was submitted to the Graduate Faculty of the
Department of Zoology in the College of Liberal Arts and Sciences
and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of
Philosophy.
December 1987
Dean, Graduate School


43
Because small crabs are less involved in agonistic and courtship
activities, they may respond to a high population density by
digging more unbranched and branched burrows and remaining more
in their burrows, rather than by reducing their time allocation
to burrowing and constructing shelters as larger crabs do.
Response of crabs of different size to food availability
Smaller crabs burrowed more under low food availability, but
again dug more branches rather than increasing their time
allocated to burrowing. Small crabs may be particularly
sensitive to food availability because they may stand more chance
of deriving benefit from their investment during their lifetime;
because they can invest more time in burrowing as opposed to
displaying, etc.; because their better burrowing ability may
allow them to better adjust their burrowing to food availability;
and because their need for food may be greater for growth. Early
stages are particularly sensitive to food availability: megalops
larvae may select a settling substrate partly on the basis of its
organic content (Crane, 1975) and early crab stages may be more
limited by food supply (Genoni, 1985). Given these reasons,
smaller crabs may be more efficient at farming Spartina.
Furthermore, they constitute a large fraction of the population.
General Considerations
As with many complex biological interactions, the observed
results might be attributable to secondary interactions, or to
indirect cause-and-effect relationships. For example, burrowing


23
(presumably in search of food), and agonistic encounters were
more frequent than in high-food arenas.
Comparison of burrowing by crabs of each sex
There was no overall difference between females and males in
the percent of unbranched burrows or branching ratio of burrows.
However, females were less active at the surface, allocated
different amounts of time to various activities, spent
significantly more time burrowing, and burrowed less deeply
(Tables VII and lib). Males constructed more shelters than did
females (Table VIII), and allocated significantly more time to
agonism (F=27.47, df=1, p<0.0001, N=53) and to display (F=14*56,
df=1, p<0.0004, N=53) (Table lib).
Response of crabs of each sex to initial burrow density
The three-level interaction between food, burrow density,
and sex, as well as the two-level interactions between pairs of
these factors, were not significant for all variables examined by
ANOVAs.
The effect of initial burrow density did not differ between
males and females with respect to burrow configuration, branching
ratio, depth, or time allocation to different activities, or to
burrowing in particular. This is suggested by a qualitative
comparison between the two sexes in Tables IX and IVb, and by the
non-significant interaction between sex and burrow density for
each variable examined. However, in arenas with a lower initial
burrow density, the trend for allocating a higher fraction of
time to burrowing was stronger for females.


40
layer, oxygenation by Spartina (metabolic oxidation and passive
0^ release), nutrient inputs, or burrows have a stronger effect
on these parameters (Howes et al., 1981; Bertness, 1985)* In
addition, digging branches near the surface may better enable
crabs to keep control of the original burrow, thus reducing the
risk of losing it to another crab, at the same time reducing the
risk of being without a burrow should the new burrow collapse or
encounter some construction obstacle. These constraints may be
particularly critical where food is limiting. In contrast, in
the low density situation, both unbranched and branched burrows
are excavated.
Comparison of burrowing by crabs of each sex
Females spent approximately half as much time outside their
burrows as did males. This observation may be an artifact of the
visual sampling technique, as females are actually more involved
in burrowing, removing them from view. Also they are less
involved in agonistic and courtship activities. Additionally,
gravid females need to protect their brood. A higher burrow
turnover rate may be expected for females, because (a) they have
an increased food requirement (see below); (b) they may be
dislodged by large males, which are less able to burrow due to
their large claw (Pearse, 1914; Hyatt and Salmon, 1977; Bertness
and Miller, 1984); (c) they allocate less time to agonistic and
courtship activities; and (d) non-receptive females dig more if
forced to interact with displaying males (L.W. Powers,
pers. comm.), an artifact of the arenas.


- 18
the number of maintained burrows, the number of branches added to
initial burrows and branches of new burrows, the amount of
excavated sediment, or burrow depth. However, in arenas with a
lower initial burrow density, there were significantly more newly
excavated burrows and the branching ratio was significantly
higher; there was a trend for an increase in the fraction of time
allocated to burrowing.
Differences between the reported fraction of time allocated
to burrowing in Tables III and IVa (and in comparisons presented
below) are the consequence of averaging across all treatments and
averaging by factorial univariate analysis, respectively.
Effect of food availability
Some of the variables describing burrowing activity were
affected by the food treatment. Results are summarized in Tables
V and Via and Figure 1. The data shown in Table V are collapsed
for the two levels of the food treatment, but the value of F is
for the main effect of food in the factorial ANOVA. There was no
difference between treatments in the number of newly excavated
burrows, the total number of maintained burrows, the amount of
excavated sediment or the fraction of time allocated to
burrowing. However, in arenas receiving no food additions, there
were significantly more branches added to initial burrows and
branches of new burrows, the branching ratio was higher, and
average burrow depth was lower. In addition, time allocation to
various activities was significantly affected by the treatment:
in low-food arenas, crabs spent less time eating, wandered more


DISCUSSION
Burrow/crab Ratios in the Salt Marsh
The burrow/crab ratio was found to be higher than 1; thus,
there were excess burrows. Burrow/crab ratios higher than one
have been found in previous studies. Both male and female crabs
dig new burrows with each tidal cycle. Burrows are relatively
stable in a muddy substrate with a root mat, particularly in the
medium Spartina zone. When abandoned, they persist through
several tidal cycles (Powers, 1973; Crane, 1975; Basan and Frey,
1977; Hyatt and Salmon, 1977; Katz, 1980; Bertness and Miller,
1984).
Spartina shoot density and biomass, and sediment organic
content were reasonably high though comparable to values for
other medium Spartina marshes (Kurz and Wagner, 1957; Basan and
Frey, 1977; Cammen et al., 1980; Gosselink et al., 1984) and root
mat density was relatively low (C.S. Hopkinson, pers. comm.).
Sediment water content was very similar to that found in a salt
marsh in Georgia, and indicative of a relatively firm substrate
and high burrow holding capacity (Teal and Kanwisner, 1961;
Kraeuter and Wolf, 1974; Bertness and Miller, 1984). All these
characteristics, together with a low degree of burrow
intersection (discussed below; see also Basan and Frey, 1977) are
"typical" of medium Spartina marshes. These characteristics
34


53
Table XIV. Summary of allometrie relationships for carapace breadth
(CB), carapace length (CL), propodus length (PL) and dactylus length
(DL) of Uca rapax. s.e.=standard error of estimate on the slope.
Type of
p
Sex
Variables
regression
Growth equation
s.e.
r
N
F
CL
vs.
CB
linear
y=0.6504x+0.0245
0.007
0.9691
308
M
y=0.6253x+0.0418
0.005
0.9783
330
All
y=0.6330x+0.0378
0.004
0.9739
638
F
CB
vs.
CL
linear
y=1.4901x-0.0030
0.015
0.9690
308
M
y=1.5645x-0.0394
0.015
0.9783
330
All
y=1.5386x-0.0282
0.010
0.9739
638
M
PL
vs.
CB
power
y:!-?x-o.2607
y=2!25S6x0-1172
0.055
0.9339
79
M
PL
vs.
CL
power
0.063
0.9236
79
M
DL
vs.
CB
power
0.064
0.9360
79
M
DL
vs.
CL
power
0.073
0.9247
79


56-
Go sse link, J.G. and C.J. Kirby, 1974 Decomposition of salt
marsh grass, Spartina alterniflora Loisel. Limnol
Qceanogr., Vol. 19, pp. 825-832.
Greenspan, B.N., 1980. Male size and reproductive success in the
communal courtship system of the fiddler crab Uca rapax.
Anim. Behav., Vol. 28, pp. 387-392.
Haines, E.B., 1976. Relation betwen the stable carbon isotope
composition of fiddler crabs, plants and soils in a salt
marsh. Est. Coast. Mar. Sci., Vol. 4 PP 609-616.
Haines, E.B. and R.B. Hanson, 1979. Experimental degradation of
detritus made from the salt marsh plants Spartina
alterniflora Loisel., Spartina virginiana L. and Juncus
roemerianus Scheele. J. Exp. Mar. Biol. Ecol., Vol. 40,
pp. 27-40
Hartnoll, R.G., 1982. Growth. In, The biology of Crustacea,
Vol. 2, edited by L.G. Abele, Academic Press, New York,
pp. 111-196.
Hines, A.H., 1982. Allometric constraints and variables of
reproductive effort in brachyuran crabs. Mar. Biol.,
Vol. 69, pp. 309-320.
Hopkinson, C.S. and J.P. Schubauer, 1984. Static and dynamic
aspects of nitrogen cycling in the salt marsh graminoid
Spartina alterniflora. Ecol., Vol. 65, pp. 961-969.
Howes, B.L., R.W. Howarth, J.M. Teal and I. Valiela, 1981.
Oxidation-reduction potentials in a salt marsh: spatial
patterns and interaction with primary production.
Limnol. Qceanogr., Vol. 26, pp. 350-360.
Hyatt, G.W. and M. Salmon, 1977. Combat in the fiddler crabs Uca
pugilator and Uca pugnax: a quantitative analysis.
Behaviour, Vol. 65, pp. 182-211.
Hyman, O.W., 1922. Adventures in the life of a fiddler crab.
Ann. Rep. Smiths. Insn. for 1920, pp. 443-459.
Katz, L.C., 1980. Effects of burrowing by the fiddler crab, Uca
pugnax (Smith). Estuar. Coast. Mar. Sci., Vol. 11,
pp. 233-237.


39
then, abandoned more often in low density arenas than in high
density arenas, and burrow turnover was higher. This sensitivity
to the existing burrow density may reflect a tendency to dig
burrows wherever possible. New burrows in a substrate with a low
density may be less likely to collapse, thus making excavation of
new burrows beneficial to the crabs. A second reason may be a
stronger competition for territories under higher burrow density.
Uca are responsive to the presence of otner crabs, and are more
likely to colonize or burrow in areas with lower apparent crab
densities, such as areas with vegetation cover (Bertness and
Miller, 1984; Vliet, 1981). Some species, including Uca rapax,
construct half-domes or shelters. One function of these may be
to reduce their visual contact with other males, thereby reducing
aggression (Zucker, 1974 1981).
Effect of food availability
Fiddler crabs dug more burrow branches, but average burrow
depth was lower, in arenas receiving no food additions. Thus,
time investment in burrowing was similar, but it was allocated to
digging new burrow branches rather than digging new burrows or
making existing burrows deeper. Digging burrow branches near the
surface may better stimulate Spartina growth. The
physico-chemical and nutrient requirements in the upper substrate
layers may be most critical to Spartina growth. Indeed, in the
upper layer there is a higher density of roots and rhizomes
(Valiela et al., 1976; Howes et al., 1981; Hopkinson and
Schubauer, 1985), and nutrients may be more easily extracted from
sediment (DeLaune and Patrick, 1980). Furthermore, in the upper


38
critical. Burrow configurations were similar to those observed
in the marsh, and could be assigned to the same categories
described above. However, burrow walls were smoother, due to the
lack of a root mat. The lack of a root mat would presumably
reduce burrow longevity. However, longevity in bare sediment was
found by Bertness and Miller (1984) to be in excess of 1 wk.
Since each trial was ended at or before 7.5 d, results should not
be significantly biased by the collapse of abandoned burrows.
Fiddler crabs allocated 5*9 % of their activity to
burrowing, which is probably an underestimate due to the visual
sampling technique, since burrowing activity will continue below
the surface. The display activity seemed low, perhaps due to the
lack of display areas in the confinement of the arenas; to
insufficient time to establish display activity; to the reduced
light intensity in the laboratory; or to the phase of the tidal
cycle (Greenspan, 1980). Although season affects burrowing
activity (Crane, 1975) this did not complicate the present study
since field and laboratory experiments were conducted during the
same season.
Effect of initial burrow density
Fiddler crabs dug new burrows and branches even in arenas
with a high initial burrow density. A lower initial burrow
density, however, resulted in a higher number of branched and
unbranched burrows. There was a trend for an increase in time
spent burrowing and for less deep burrows, suggesting that time
for burrowing was allocated to digging new burrows and branches
rather than to making existing burrows deeper. Burrows were,


ACKNOWLEDGEMENTS
I wish to express my gratitude to my graduate committee,
Drs. F.J.S. Maturo, Jr., C.L. Montague, F.G. Nordlie,
W.J. Lindberg, and M.G. Wheatly, for their generous assistance
and guidance. I am especially indebted to Dr. Maturo, who
provided much support with facilities, supplies and advice, to
Dr. Montague, who stimulated my interest in this study, and to
Dr. Lindberg, for helpful discussion.
I also thank Drs. M.D. Bertness, W.F. Hermkind,
C.S. Hopkinson and L.W. Powers for their helpful comments and
G. Nemes, V. Orcel and A.M. Fiore for help in the lab and in the
field. D. Harrison prepared the illustrations.
ii


5
substrate. Shoots were counted and later dried 48 h at 70 C
and weighed to the nearest 0.1 g with an Ohaus trip balance. To
determine the characteristics of each burrow and identify
resident crabs, burrow openings were marked with numbered sticks,
and burrow casts were taken with polyester resin (Dembowski,
1926; Shinn, 1968). The resin caused resident crabs to escape.
Their sex was noted and their carapace breadth was measured to
the nearest 0.05 mm with Vernier calipers. Because some burrows
are kept plugged by resident crabs at any time (Crane, 1975),
this procedure was repeated on the next two days with newly
opened burrows. Association of burrows with structural elements
in the substrate (grass stems or mussel shells,
Geukensia demissa) was estimated by scoring burrows as either
occurring on bare substrate (>2 cm from a stem or mussel) or
structure-associated (<2 cm) (Bertness and Miller, 1984).
On the third day, burrow casts were removed and their depth,
angle from the substrate surface, diameter and configuration were
noted. To sample root mat density, a 20-cm deep sediment core
was taken with a PVC tube (5*2 cm in diameter). To sample water
content and organic content, two 7-cm deep cores were taken with
a PVC tube (3*0 cm in diameter). In the laboratory, the larger
sediment core was sieved through a 0.5 mm mesh, and the roots
were dried 48 h at 70 C and weighed to the nearest 0.01 g with
a Mettler H8 analytic balance. One smaller core was weighed wet,
dried 48 h at 110 C and weighed again to determine the water
content. The upper 0.5 cm from the other core was weighed wet,
dried 48 h at 70 C, then weighed dry and left for 4 h in a
Sybron Corp. Thermolyne 10500 muffle furnace at 500 C. It was


CARAPACE BREADTH (cm)
50
Figure 3. Regression of carapa.e breadth against carapac
for Uca rapax. F=females, M=males, ALL=all crabs.
length