Population and reproductive features of the western Atlantic hermit crab Pagurus criniticornis ( Anomura , Paguridae ) from Anchieta Island , southeastern Brazil

The population of the hermit crab Pagurus criniticornis (Dana, 1852) was studied based on seasonal abundance, size frequency distribution, sex ratio, reproductive period, fecundity and shell relationship. Specimens were collected monthly by SCUBA diving in the infralittoral area of Anchieta Island, Ubatuba. A total of 1,017 individuals was analyzed. Animal size (minimum and maximum shield length, respectively) was 0.7 and 2.9 mm for males, 0.6 and 2.8 mm for non-ovigerous females, and 1.0 and 2.5 mm for ovigerous females. The sex ratio was 1:1.29. Sexual dimorphism was recorded by the presence of males in the largest size classes. Ovigerous females were captured during all months along the year, with percentages varying from 8% (July) to 84.3% (February) in relation to the total females collected. Mean ± SD fecundity was 168 ± 125 eggs and tended to increase with increasing hermit size. Shells of four gastropod species [Cerithium atratum (Born, 1778), Morula nodulosa (Adams, 1845), Anachis lyrata (Sowerby, 1832) and Modulus modulus (Linnaeus, 1758)] were occupied by ovigerous females of P. criniticornis but fecundity was not significantly different in relation to the different shell types. The profile showed continuous and intense reproduction of P. criniticornis probably related to strategies developed to compensate for interspecific competition in the studied insular area.

Hermit crabs represent an important portion of many intertidal and moderately deep tropical benthic communities (FRANSOZO & MANTELATTO, 1998).Consequently, this group of decapods provides promising material for study because the establishment of these animals in such environments derives from the evolution of adaptive population strategies.Nevertheless, there is relatively little information on the life history or reproductive ecology of hermit crabs if compared to other decapods (F.L. Mantelatto, pers. comm.).
Aspects of reproduction, such as egg production and the estimate of the fecundity, are important facets that have been used to assess the reproductive biology and are essential for the understanding and to assure the natural renewal of the decapod populations, as well as to develop an adequate handling of the species (CADDY, 1989).Furthermore, the highly diverse and plastic reproductive characteristics of crustaceans may provide a good opportunity to study the fitness consequences of different reproductive strategies (GOSHIMA et al., 1996).The intriguing hermit crabs in particular, distinctive for sheltering their fragile and uncalcified abdomen in gastropod shell, afford an opportunity to study the interaction between resource (GARCIA & MANTELATTO, 2000) and reproductive strategy (MANTELATTO et al., 2002;MIRANDA et al., 2006).
The genus Pagurus Fabricius, 1775 is a heterogeneous pagurid group consisting of more than 170 species worldwide (LEMAITRE & CRUZ CASTAÑO, 2004).It is one of the most species-rich genera in Anomura and is distributed in shallow waters such as littoral, subtidal and intertidal areas.However, ecological information such as reproduction and life history is still scarce for the genus (GOSHIMA et al., 1996).
Pagurus criniticornis (Dana, 1852) is a tropical small species with geographic distribution restricted to the Western Atlantic (MELO, 1999).It constitutes an abundant population in the hermit crab community of São Paulo coast (MANTELATTO & GARCIA, 2002) representing promising material for study on comparative population biology and ecology with specimens from different areas.As far as we know, the specific knowledge on the population biology of P. criniticornis is retricted to the data available for one sample collection carried out at continental intertidal area of São Sebastião region (NEGREIROS-FRANSOZO et al., 1991).The present investigation analyzed the population biology of P. criniticornis from systematic samplings performed in the infralittoral area of Anchieta Island, southeastern Brazil, in terms of sex ratio, size frequency distribution, breeding season, fecundity and shell relationship.

MATERIAL AND METHODS
Specimens of P. criniticornis were obtained monthly from January to December 1999 from the infralittoral area of East Beach in Anchieta Island (23 o 33'S, 45 o 05'W), northern coast of São Paulo.The hermit crabs were captured during daylight by two SCUBA divers over a period of 30 min over an area of about 800 m 2 .After collection, the animals were frozen and transported to the laboratory where they were carefully removed from their shells by hand.If necessary the shells were broken.The animals were counted, measured for shield length (SL) and weighed (WW).Sex was checked from the gonopore position.Measurements were taken under a compound light microscope with a camera lucida.The voucher crabs were deposited in the Crustacean Collection of the Biology Department of FFCLRP (CCDB), University of São Paulo (Catalogue number: 1,476 to 1,484 and 1,741 to 1,745).
The population structure was analyzed as a function of the size frequency distribution of the individuals and the sex ratio.The chi-square test (χ 2 ) was used to evaluate the sex ratio and to compare male and female percentage per month.The distribution frequency was tested by the Kolmogorov-Smirnov normality test (KS).The mean size of individuals of both sexes was compared by the Mann-Whitney test (ZAR, 1996).The reproductive period of the population was expressed as the percentage of females carrying eggs (ovigerous females) relative to the total number of females collected.
Shell species were identified according to RIOS (1994) and confirmed by Dr. Osmar Domaneschi from Instituto de Biociências da Universidade de São Paulo.The following measurements were taken: Shell Dry Weight (SDW), Shell Aperture Width (SAW), and Shell Aperture Length (SAL).Shell internal volume (SIV) was measured according to MANTELATTO & GARCIA (2000).
From a total of 306 ovigerous females captured, only 44 were used for fecundity analysis.We selected individuals with embryos in stages from late blastula to early gastrula (MANTELATTO & GARCIA, 1999modified from BOOLOOTIAN et al., 1959;Initial Stage: stages 1 to 4; Intermediate Stage: stages 5 to 8, and Final Stage: stages 9 and 10).This procedure was adopted in order to avoid confounding effects of embryonic diameter swelling during the late stages of development and because of egg loss from the brood during incubation.The eggs were carefully removed from the pleopods and counted under a light stereomicroscope.Frequency distributions of SL and fecundity were calculated for the females whose eggs were quantified.After frequency tabulation, mean egg number of brooded egg mass was determined for each size class.The average volume (mm 3 ) of eggs was calculated according to JONES & SIMONS (1983) from the formula 1/6 π I 3 (where I is the mean of two diameters, including the chorionic membrane tightly adhering to the embryonic surface) of a sub-sample of 15 eggs per female.The longest and shortest diameters of each egg (the eggs tended to be oval) were measured under a compound light microscope with a camera lucida.Data were analyzed statistically by regression of dimensions plots (Y = ax + b).
To determine correlations between fecundity and shell variables, regressions and correlation coefficients were computed.The ANOVA on Ranks test was used to compare the fecundity between the occupied shell species.The level of significance was p<0.05.

RESULTS
Population structure.A total of 1,017 individuals were collected, represented by 445 (43.76%) males, 266 (26.16%) non-ovigerous females and 306 (30.08%) ovigerous females.The highest number of P. criniticornis was recorded in February, and the lowest number was found in September (Tab.I).
The size frequency histograms presented non-normal distribution (KS = 0.076) of individuals and clear prevalence of specimens measuring 1.4 to 2.4 mm SL (Fig. 1).
Fecundity.The mean fecundity of the 44 ovigerous females analyzed was 168 ± 125 eggs.Individual fecundity ranged from 6 (SL = 2.1 mm) to 538 (SL = 2.3 mm) eggs.The general egg shape of P. criniticornis is slightly oval with mean diameter of 0.360 ± 0.046 mm and mean volume of 0.026 ± 0.011 mm 3 .
Ovigerous females with eggs in the initial stage of development were found at high frequency (more than 50%) along the months (Fig. 3).This pattern demonstrates an annual reproductive cycle with rapid embryonic development.
The egg number (FE) increased in accordance with female size (r = 0.44) (Fig. 4), with a variation in mean number of eggs for each size class studied, in agreement with the pattern of fecundity described for other decapods.
Fecundity analysis carried out in relation to shell dimensions demonstrated positive but not all significant correlations (Tab.II).Regression analysis in relation to the shell types were evaluated only to C. atratum and M. nodulosa shells because of the low number (< 1%) of occurrence of the others.It was verified that the fecundity of the ovigerous females in C. atratum shells was not significantly higher than that of the ovigerous females occupying M. nodulosa shells, indicating that egg number was not influenced by this factor.

DISCUSSION
In carcinology, the populational information has been assessed in several ways.Size frequency distribution is the most used parameter to characterize the population structure since it changes throughout the year as a result of reproduction, recruitment from larvae and death (THURMAN, 1985).In the present study, the population presented an unimodal pattern of distribution which usually characterizes the occurrence of slight monthly variations.It may usually result from the balance between continuous recruitment without class disruption and mortality rates.This is a common pattern observed in other decapod crustaceans as the brachyuran fiddler crabs Uca longisignalis (Salmon & Atsaides, 1968) and Uca thayeri (Rathbun, 1900) [studied respectively by MOUTON & FELDER (1995)  Similar patterns of size distribution were found for males and females.Both presented non-normal distribution with a peak of occurrence in intermediate classes whereas the distribution of males skewed toward larger size classes.This and the unequal sex-ratio probably can be attributed to factors such as differential mortality and growth rates between sexes and migrations (WENNER, 1972;HARTNOLL, 1982), with males reaching larger sizes in less time than females but being influenced by shell limitation.On the other hand, natural selection may favour large males if they have an advantage over small males either in competition for receptive females and/or more frequent successful copulation (ABRAMS, 1988).The larger size of the males reflecting sexual size dimorphism has been a common pattern among hermit crabs and especially for Pagurus and Clibanarius (see MCLAY, 1985 andMANTELATTO &GARCIA, 2000 for a review), a fact that may represent an advantage in competition for resources.
Crustacean populations, in almost all cases, have sex ratios differing from 1:1 (WENNER, 1972).In the present study the proportion obtained (1:1.3)differed statistically from the expected value (1:1), which is not in agreement with the Fisher Theory (FISHER, 1930).According to that author, natural selection favours an equal proportion between the sexes.However, differences in the sex ratio can be related to differences in life cycle, growth rate and behavior of males and females (WENNER, 1972).
The sex-ratio analysis by size classes revealed an anomalous pattern (see WENNER, 1972 for review) which is characterized by a decrease in the males/females ratio in the intermediary classes and subsequent increase in the largest size classes.Probably, the deviation of proportion males/females found in the present study (females more abundant in the middle-size classes) may be attributed to a faster growth of males, to sexual selection and to smaller longevity of females (BERTNESS, 1981;ABRAMS, 1988;BENVENUTO & GHERARDI, 2001).This pattern has been the most common one reported in the literature (WENNER, 1972;MANJÓN-CABEZA & GARCÍA RASO, 1995;GHERARDI & NARDONE, 1997;FRANSOZO & MANTELATTO, 1998;GARCIA & MANTELATTO, 2001) for hermit crab populations inhabiting both the intertidal and infralittoral areas.Unfortunately, we are not able to compare the present results with those of NEGREIROS-FRANSOZO et al. (1991) which were obtained from one punctual collection.
Ovigerous females with eggs in the early developmental stages were collected round all year   indicating that reproduction is not interrupted at any time of the reproductive cycle.Considering the occurrence of small sized ovigerous females (from 1.0 -| 1.2 mm range), the low number of juveniles (n=16) and the continuous reproductive pattern observed in the population studied, we may infer that sexual maturity occurs early in the life cycle of P. criniticornis in Anchieta Island.Based on the small size of ovigerous females captured (1.0 mm of SL), we may also infer that sexual maturity may occur at approximately 2 and 3.5 months in the life cycle of this species in this area.
According to LANCASTER (1990), early sexual maturity seems to be a tendency among hermit crabs, which may be determined by interespecific competition and shell availability.Similar pattern of sexual maturity was observed to Pagurus annulipes (Stimpson, 1862) (CARLON & EBERSOLE, 1995), Diogenes pugilator (Roux, 1829) (MANJÓN-CABEZA & GARCÍA-RASO, 1999), Paguristes tortugae Schmitt, 1933(MANTELATTO & SOUSA, 2000) and Pagurus brevidactylus (Stimpson, 1859) (MANTELATTO et al., 2005).Furthermore, high adult mortality leads to small size at maturation, whereas high juvenile mortality leads to large size at maturation (STEARNS, 1989).This implies that future reproduction prospects determine the timing for reproduction of each individual: individuals with higher chances of survival would invest more energy into growth instead of reproduction, whereas individuals with poorer prospects for survival would invest more energy into reproduction (YOSHINO et al., 2002).
The fecundity recorded for P. criniticornis from Anchieta Island was lower than that reported for Pagurus pollicaris Say, 1817(FOTHERINGHAM, 1976) and Pagurus longicarpus Say, 1817(FOTHERINGHAM, 1976), and similar to P. brevidactylus (IOSSI et al., 2005).Variations in fecundity for species of the same genus can be related to environmental adaptations, variations in animal size and geographic location (FOTHERINGHAM, 1976;SASTRY, 1983;CARLON & EBERSOLE, 1995;MANTELATTO & FRANSOZO, 1997).In this case we agree with IOSSI et al. (2005) who proposed that not only these factors can influence the fecundity of a species but also the interspecific competition (especially in terms of competition for resources).
The low fecundity observed in some individuals may occur as a function of the presence of very young ovigerous females assumed to be spawning for the first time.According to SOMERTON & MEYERS (1983), these primiparous females produce a lower number of eggs than multiparous ones.A variation in number of eggs was observed within the same size class, probably indicating a process of multiple spawning within the same reproductive season (AMEYAW-AKUMFI, 1975;MANTELATTO & GARCIA, 1999;MANTELATTO et al., 2002;IOSSI et al., 2005).This variation in fecundity by size class is influenced either by the number of spawns in the same reproductive cycle or by inadequate shell occupancy.
According to HINES (1982), the egg size is the best reproductive variable to determinate if the spawn has numerous and small eggs or few and large eggs, resulting in large variation in fecundity between species.Larger eggs produce an offspring better adapted to the search for food and with greater competitive ability (BERTNESS, 1981).The eggs of P. criniticornis were smaller than eggs of the other species of Paguridae except P. brevidactylus (see IOSSI et al., 2005 for review).This fact may indicate that these species compensate the smaller eggs produced by releasing numerous, but less competitive planktonic larvae during their continuous reproductive cycle.
On Anchieta Island, ovigerous females of P. criniticornis presented a low plasticity of shell occupation, being C. atratum significantly more occupied than other species.This pattern may be associated with the preference of this hermit species for C. atratum shells in the natural habitat as well as to its availability in the area.MANTELATTO & MEIRELES (2004) found 32 species of gastropod shells available in this area where C. atratum was the most available (38.63% of the total).However, we may also speculate about the adaptation of the ovigerous females to a given shell species as a function of its availability in the enviromment and of the fact that these females do not change shells frequently as the others (males and non ovigerous females).
According to FOTHERINGHAM (1976), shell weight directly affects the amount of energy used in reproduction, shifting it to activities such as locomotion and the search of food, among others.REDDY & BISESWAR (1993) observed in Clibanarius virescens (Krauss, 1843) and Calcinus laevimanus (Randall, 1839) and ELWOOD et al. (1995) in Pagurus bernhardus (Linnaeus, 1758) that the number of eggs produced was larger among ovigerous females occupying lighter shells.In the present study we detected no significant difference in the mean weight of the shell species occupied.
Differences on the fecundity in relation to the shell species occupied were not registered and thus did not reflect a real influence of shell type on the number of eggs produced.However, the influence of the shell type and size on egg production of hermit crabs was reported: 1) P. bernhardus ovigerous females inhabitants of Littorina obtusata (Linnaeus, 1758) shells produced more eggs than those in Gibbula spp.shells (ELWOOD et al., 1995); 2) shell type influenced the fecundity of P. tortugae depending on the size of the ovigerous females (MANTELATTO et al., 2002); 3) the fecundity of P. brevidactylus in M. nodulosa shells was significantly higher than that of the ovigerous females occupying C. atratum shells (IOSSI et al., 2005).This later also postulated that although fecundity is an important tool used to estimate the reproductive potential of a hermit species, it should be seen as part of the strategy developed to guarantee survivorship, which is influenced by a number of factors including habitat resource, intraand interspecific competition, and the intrinsic way of living of the species.
In this respect, the population and reproductive aspects of P. criniticornis were related to a strategy to compensate for interspecific competition, i.e., high and continuous reproductive effort with low fecundity and small eggs produced, continuous recruitment and early maturity.

Table I .
Total number of individuals of Pagurus criniticornis (Dana, 1852) collected monthly in 1999 in the infralittoral area of East Beach, Anchieta Island, Southeastern Brazil.