Population and reproductive biology of the crab Uca burgersi (Crustacea: Ocypodidae) in three subtropical mangrove forests

: Population and reproductive biology were studied in three populations of the crab Uca burgersi Holthuis, 1967 , in the Indaiá, Cavalo and Ubatumirim mangrove forests (Ubatuba, São Paulo State, Brazil). Crabs were collected during low tide (August 2001 through July 2002), by digging the sediment, with a standard capture effort (two persons for 30 min.). Carapace width was measured, and gonad developmental stage was recorded from all specimens. U. burgersi was most abundant in the Cavalo mangrove, where the largest male was found. Juvenile crabs were found year-round at all three sites. However, the ratio of ovigerous females was very low, even null in the Cavalo mangrove. The gonad development rate indicated that U. burgersi was reproducing continuously, but more intensively during spring and summer, with recruitment occurring in winter. The synchrony between the populational and reproductive biology in the three areas showed that local features were not the limiting factors. It is suggested that this species is a habitat generalist. Rev. Biol. Trop. 55 (Suppl. 1): 55-70.

Ocypodid crabs of the genus Uca Leach, 1814 are the most characteristic group of invertebrates associated with estuaries of tropical and subtropical coasts, mainly mangroves.These crabs live in muddy areas and feed on organic matter deposited on the sediment surface (Jones andSimons 1983, Costa andNegreiros-Fransozo 2003).They have the habit of burrowing in the sediment, building complex galleries.These refuges protect the crabs from extreme temperatures, salinity variations and predators (Macintosh 1988), and they are also used for reproduction and molting (Crane 1975, Hyatt andSalmon 1978).
Crab populations have several features or properties, such as density, biotic potential, recruitment and growth, which are not attributes of an isolated organism, but when grouped they coordinate the dynamics of a natural population (Hutchinson 1981, Jones andSimons 1983).The individuals which are part of these populations interact, competing for food, refuge and reproductive pairing, among other resources, and the intra-and interspecific relationships maintain the equilibrium of the community (Fontelles-Filho 1989).
The population biology of ocypodid crabs has been analyzed mainly based on the distribution of individuals in size classes, seasonal abundance, population density, sex ratio, juvenile recruitment and reproductive intensity (Spivak et al. 1991, Trott 1996, Negreiros-Fransozo et al. 1999, Costa and Negreiros-Fransozo 2003, Litulo 2005).Information about a population provides knowledge of the ecological stability of the species in a certain habitat, as well as extending understanding of the biology of a species (Pillay and Nair 1971, Hutchinson 1981, Jones and Simons 1983, Santos et al. 1995).
These populations usually pass through long reproductive periods, when the females spawn several egg masses.Accordingly, the modal class varies as a result of reproduction and juvenile recruitment (Thurman II 1985, Macintosh 1988, Castiglioni and Negreiros-Fransozo 2005).
The study of the recruitment biology of crabs comprises two distinct aspects: events related to the cohort, and processes which precede and affect the mating behavior and the reproductive cycle of a species, i.e., maturity, egg development, and the relationship between molt and reproduction.Evaluation of these characteristics requires periodic sampling of the natural population (González-Gurriarán 1985).
The reproductive period for most brachyurans can be estimated by means of gonad development, and also by the ratio of ovigerous females throughout the year (Pilay and Nair 1971, Murai et al. 1987, Henmi 1992, Mouton and Felder 1995, Rodríguez et al. 1997, Chacur and Negreiros-Fransozo 2001, Flores et al. 2002, Negreiros-Fransozo et al. 2002, Colpo and Negreiros-Fransozo 2003).The size at onset of sexual maturity, based on external morphological features, can sometimes be misestimated when the curves for immature and mature specimens overlap (Somerton 1980a, b).Several researchers (Conan and Comeau 1986, Choy 1988, Fontelles-Filho 1989) mentioned that in some species, morphological sexual maturity does not coincide with physiological sexual maturity.The latter is related to the presence of gonads that are mature, i.e., producing gametes.Thus, when one intends to estimate the size at first maturation of a brachyuran, the degree of gonad development beyond the external morphological features should be considered (Watson 1970, Brown andPowell 1972).
Among the few published reports concerning the biology of Uca burgersi Holthuis, 1967 are those of Gibbs and Bryan (1972) who studied the account of strontium, magnesium and calcium in the environment and exoskeleton of decapods.Gibbs (1974) performed ecological studies in Barbuda, andvon Hagen (1982, 1984) investigated behavioral aspects.In Brazilian mangroves, Rieger (1998) worked on the larval development, Benetti and Negreiros-Fransozo (2003) studied the morphological sexual maturity, and Benetti and Negreiros-Fransozo (2004 a, b) studied the relative growth at two different sites and the symmetry of the chelipeds, respectively.
In this investigation we analyzed the population structure (size-frequency distribution, sex ratio) and the reproductive cycle of U. burgersi based on gonad development and molt stage.
The canopy of the Indaiá and Ubatumirim rivers is composed of Laguncularia racemosa (L.) and Avicennia schaueriana Stapf and Leecheman (Colpo 2001).The Cavalo River has only L. racemosa; because of its location about 300 m from the river, the canopy is not directly influenced by the mechanical changes of the tides.
Three samples of substrate (10 g) from each of the three study sites were collected for analyses of organic matter content in each season of the year.The samples were dried at 60 °C to constant weight, then incinerated for three hours at 500 °C and reweighed.The organic matter content was determined from the ash-free dry weight, and the means were compared by variance analysis (ANOVA; α=0.05) complemented by a posteriori Tukey's test among the mangroves and by year season; the significance level adopted was 5 %.
Crabs were sampled monthly (August 2001 through July 2002) during low tide by digging the burrows and removing the crabs during a period of 30 min. in a fenced area of approximately 0.01 ha, summing 12 samples.
Crabs were identified according to the key provided by Melo (1996).Cheliped size, abdominal morphology, and the number of pleopods assessed the sex of each crab.The ovigerous condition was also noted.
The carapace width (CW) of all crabs was measured to the nearest 0.1 mm using a caliper.For the population study, the values of carapace width were grouped into size classes, which were analyzed monthly for the three populations studied.The median size of each category, based on the carapace width, was compared among localities by Kruskall-Wallis, complemented by Dunn´s test (α=0.05).
The proportion of males was analyzed by means of a chi-square test for goodness of fit (α=0.05;Sokal and Rohlf 1987).
All individuals captured were classified as juvenile or adult, based on carapace width values obtained for sexual maturity, which have been assessed previously for the same populations by Benetti and Negreiros-Fransozo (2004a).These cutoff values were 7.8, 8.8 and 7.5 mm for males and 6.2, 6.2 and 6.8 mm for females from Indaiá, Cavalo and Ubatumirim mangroves, respectively.Thus, crabs with smaller carapace width than the above-mentioned values for each site were considered as juveniles, and larger ones as adults.Recruitment in each population was estimated monthly from the ratio of juveniles to the total number of crabs collected.The proportion of juveniles was compared among the seasons of the year for each population, using the multinomial proportions analysis (Curi and Moraes 1981).
To assess the degree of gonad development, all crabs were dissected and the gonads observed with the aid of an optical stereomicroscope.The stages of gonad development were determined according to the patterns described for other brachyurans (Haefner 1976, Abelló 1989a, b, Wenner 1989), as listed in Table 1.
Crabs were grouped as "reproductively inactive" (IM and RU; see Table 1) and "reproductively active" (ED, DE and AV; see Table 1).Subsequently, the relative frequency of these groups was determined for each sex and season.The proportions of the reproductively active group were compared by the multinomial proportions test (Curi and Moraes 1981).
Evaluation of the molt stage for each crab was based on the exoskeleton hardness and was confirmed by dissecting the crab.The moltstage features adopted were those proposed by Drach and Tchernigovtzeff (1967), Warner (1977) and Abelló (1989 a, b).Crabs were grouped into two categories according to molt stages: molt activity (stages A, B, D and E) and intermolt (stage C).The proportions of crabs in molt activity were compared by multinomial proportions analysis (Curi and Moraes 1981).
The numbers of crabs obtained from the three study sites as well as the descriptive measures of each demographic category of the populations are presented in Table 3.
In November 2001 in the Indaiá mangrove, a crab was found that possessed male abdominal morphology with gonopods, but it also had a pair of small chelipeds.Benetti and Negreiros-Fransozo (2003) described this intersex specimen.
U. burgersi was more abundant in the Cavalo mangrove (n=1 952 specimens) than in Ubatumirim and Indaiá with 925 and 933 specimens, respectively.Ovigerous females were very scarce during the study period, comprising only 0.42 % of the population at Indaiá during spring, and 2.76 % and 1.38 % at Ubatumirim during spring and summer, respectively (Table 3).The median size (CW) between the sexes differed from site to site.Adult males from the Cavalo mangrove were significantly larger than those from the other sites; adult females were largest at Ubatumirim (Kruskall-Wallis, complemented by Dunn´s test; α=0.05) (Fig. 1).
The histograms of frequency distribution of crabs captured at each site are presented in Fig. 2, 3 and 4. Juvenile crabs were not found in July (winter) in the Indaiá and Cavalo mangroves, or during March (summer) and April (autumn) in the Ubatumirim mangrove (Fig. 4).Ovigerous females were obtained only during November (Indaiá mangrove, Fig. 2) and November and January (Ubatumirim mangrove, Fig. 4).No ovigerous females were found in the Cavalo mangrove (Fig. 3).
The recruitment rate differed significantly among seasons in Ubatumirim (being higher during the winter) and was continuous throughout the year (multinomial proportions analysis; α=0.05) (Fig. 6).
The number of reproductively active crabs increased in early summer (multinomial proportions test; α = 0.05) (Fig. 7-8).For males, the higher proportions of active crabs were found in spring, and the inactive ones were recorded in autumn, presenting significant differences.The females showed a similar pattern, with exception of the Cavalo River population, where the proportions of active and inactive female did not differ statistically between autumn and summer.
The proportion of molting crabs was always lower than that of intermolting crabs.In the Indaiá mangrove, molting crabs were more frequent during the summer months.However, for molting crabs from the Cavalo mangrove during spring and summer and among those from the Ubatumirim mangrove, there were no significant differences (multinomial proportions test; α = 0.05) (Fig. 9).

DISCUSSION
The rate of production and export of organic matter characterize the mangroves as the most productive ecosystems on Earth.Because of this, mangroves play an important role in the maintenance of ecological equilibrium in estuaries where they are present (Odum andHeald 1972, Koch andWolff 2002).Maintenance and retention of organic matter and nutrients in the mangroves are also influenced by the local hydrology and the sediment texture.At sites with less tidal action, the production of organic matter is increased  while export of litter is diminished and the decomposition rate is higher (Twilley 1985).
Fiddler crabs present a behavioral pattern associated with the sediment features (both grain size and organic matter content).Such behavior can be exemplified by both the burrows and the feeding activities.Several species of the genus Uca present a spatial distribution according to the grain size (Thurman II 1987, Macintosh 1988, Mounton and Felder 1996) and some morphologic features adapted to the organic matter extraction of the sediment during feeding (Costa and Negreiros-Fransozo 2001) Because U. burgersi showed no preference for a particular kind of substratum or organic matter content, we suggest that it is a generalist species with respect to its habitat.According to Costa and Negreiros-Fransozo (2001), this species is capable of extracting organic matter from several fractions of sediment, aided by several types of setae on the mouth appendages, which allow them to filter the ingested sediment.
Many environmental factors seem to act on the population growth of U. burgersi in the sites studied.Scarcity or low quantities of food, latitudinal variations, and adverse environmental factors (salinity, temperature, air exposure and pollution) have been pointed out as the main factors influencing the growth of decapods (Conde and Díaz 1989, 1992, Hines 1989, O'Connor 1990, Costa and Negreiros-Fransozo 1998).
Considering that fiddler crabs feed on particulate organic matter, the organic content of the sediment can promote better nutritional conditions for them.In this sense, we expected to found larger crabs in sites with   Comparisons were made among seasons, for each site.Bars with at least one letter in common did not differ significantly (multinomial proportions analysis; α=0.05).
higher organic matter content.Thus, the larger males of U. burgersi were found in the Cavalo River mangrove, as it was expected because this is the most productive of the sites studied.Nevertheless, females grew larger in the Ubatumirim River mangrove than at the Cavalo River; but this case suggest a sampling artifact, as the area in which crabs were found is very small compared to other sites.If one considers the predation risk, males being more conspicuous to predators than females because of their major chelipeds, they therefore might suffer higher mortality rates and do not reach larger sizes.Nevertheless, another hypothesis may explain the observed size differences: females grow slower than males because they invest more energy in the production of gametes (with yolk) than in somatic tissue.
The sex ratio of fiddler crabs is usually biased toward one sex or the other (Genoni 1985).In the present study, U. burgersi showed a higher proportion of males in the intermediate size class in the Indaiá and Ubatumirim mangroves.This difference is common among marine crustaceans, and can be attributed to several factors, as droving for males, which can favor one sex over the other.The tendency toward biased sex ratios is notable in species of Uca as mentioned by Wenner (1972) and Johnson (2003).
Some studies (Frith and Brunenmeister 1983, Emmerson 1994, Croll and McClintock 2000), using sampling techniques with sediment excavation and wide temporal and spatial variability, have shown a tendency in fiddler crabs toward male-biased sex ratios.Johnson (2003) mentioned that 14 species of fiddler crabs tend toward male-biased sex ratios, suggesting that this tendency has an ecologic meaning and does not represent a sampling artifact.Rieger (1998), studying the larval development of U. burgersi, collected ovigerous females during November and February, i.e., spring and summer.The same pattern occurred in the present investigation, with seven ovigerous females recorded in spring and summer.
All the egg masses found were in the final embryonic developmental stage.This indicates that these females may be ready to disperse the larvae in the estuary.According to Salmon (1987), females of broad-fronted fiddler crabs as U. burgersi, can incubate their eggs underground to protect them from extreme environmental conditions and to provide a uniform environment, thus promoting synchrony in embryonic development and larval hatching.Christy and Salmon (1984), Murai et al. (1987) and Henmi (2003) also observed such behavior in Uca pugilator (Bosc, 1802), Uca lactea (DeHaan, 1875) and Uca perplexa (H.Milne Edwards, 1837) respectively.Moreover, ovigerous females with large broods remained in their burrows during the entire incubation period and did not feed during this phase.It can be inferred that ovigerous females of U. burgersi have the same kind of behavior in the mangroves studied by us.
Usually female fiddler crabs move to the entrance of their burrow during the highest high-tide (spring-tide) to disperse their offspring, and larvae will be transported to the sea.Thus leaving their burrow would not improve larval dispersal, but would increase predation risk.
The similarity of juvenile recruitment among the three populations shows that larval development must be occurring under the same near-shore conditions.Recruitment was observed year-round, which may indicate that reproduction is continuous, as in other intertidal brachyurans (Leme 2002).
The definition of the reproductive period as the time interval when ovigerous females are present in the population (Sastry 1983) cannot be adopted for this fiddler crab species, considering that they incubate their eggs inside the burrows (Salmon 1987).Other reports have emphasized that U. pugilator (see Christy and Salmon 1984) and U. lactea (see Koga et al. 2000) incubate their eggs underground and do not feed during this period.
The many factors influencing the reproductive cycle, such as sexual maturity, can be endogenous or exogenous, or a result of the interaction between them.Most ocypodid crabs from temperate regions show seasonal reproduction, e.g., U. lactea, U. pugilator, U. pugnax, Macrophthalmus japonicus, Scopimera globosa and Ilyoplax stevensi (see Greenspan 1982, Christy and Salmon 1984, Henmi 1989, Snowden et al. 1991, Koga et al. 2000).Even some subtropical crab species can show seasonal reproduction: Pachygrapsus transversus Gibbes, 1850 (see Flores and Negreiros-Fransozo 1998) and Uca rapax (Smith, 1870) (see Castiglioni et al. 2007).The reproduction of U. burgersi was continuous in the area studied, with pronounced peaks during spring and summer.Cardoso (2003), studying U. leptodactyla, andCosta andNegreiros-Fransozo (2003), studying U. thayeri, both in the same region as the present investigation, found that reproduction occurs throughout the year.This reinforces the supposition that favorable conditions for intertidal fiddler crabs exists continuously in the area.
The ratio of molting individuals of U. burgersi was very low (<20 %) throughout the year in all three mangroves, which is expected for terrestrial or semi terrestrial crabs.Because the molt is a critical period, crabs remain inside the burrows to prevent attacks by predators (Hyatt and Salmon 1978, Christy and Salmon 1984, Salmon 1987, Atkinson and Taylor 1988, Koga et al. 2000).
Terrestrial and semi terrestrial crabs (males and females) are known to copulate even in intermolt periods, as long as the female's genital opening is not calcified.In contrast, most aquatic crabs copulate just after the female has molted (Bliss 1968, Greenspan 1982).U. burgersi showed a high proportion of females in the intermolt stage with developed ovaries.This may indicate that mating occurred during the intermolt.
The antagonism between the processes of reproduction and growth is widely known among crabs, because of the competition for energy resources for each process.Thus, there is a wide diversity of growth and reproduction patterns, which allow each species to maximize its reproductive potential within the limits of its plasticity (Hartnoll 1985).Certainly this argument can be admitted for U. burgersi as a result of the interaction between reproduction and growth.The findings presented here clearly indicate a growth period prior to the spawning season in spring and summer, and a peak in the establishment of juveniles in the subsequent autumn and winter.Molting preceding the reproductive period has also been reported in other semi terrestrial crabs (Pillay and Ono 1978, Fukui 1990, Flores and Negreiros-Fransozo 1999).
According to the present data, U. burgersi presents similar patterns of both population structure and reproductive cycle among the three studied sites, despite of the different environmental features encountered in each site.Benetti (2003) studied the Cavalo River mangrove and found its sediment is mainly compounded by coarse sand, and organic matter content is higher than in the other two mangroves.She also found monotypic vegetation: only L. racemosa inhabits this area.In the other study areas (Indaiá and Ubatumirim rivers), which are exposed to mechanic effects of wave action, very fine sand were predominant (Benetti 2003), and the vegetation was composed by L. racemosa and A. schaueriana.
The synchrony between the population and reproductive biology of U. burgersi in the three studied areas shows that the specific features of each site were not the limiting factors in its biology and, probably other factors (biotics and or abiotics), or also the interaction among them, are the cause of the stability and synchrony of its life cycle.Moreover, it is suggested that U. burgersi is a generalist with respect to the habitat.

ACKNOWLEDGMENTS
To the Fundação de Amparo à Pesquisa do Estado de São Paulo-FAPESP (94/4878-8; 98/03134-6) and the Conselho Nacional de Desenvolvimento Científico e Tecnológico-CNPq for financial support.To the members of the Núcleo de Estudos em Biologia, Ecologia e Cultivo de Crustáceos-NEBECC for their help during fieldwork.To Janet W. Reid for English corrections.

Fig. 6 .
Fig. 6.Uca burgersi.Comparison of proportions of juveniles among seasons.Bars with at least one letter in common did not differ significantly (multinomial proportions analysis; α=0.05).

Fig. 7 .
Fig. 7. Uca burgersi.Proportions of actively reproducing and inactive males.Comparisons were made among seasons, for each site.Bars with at least one letter in common did not differ significantly (multinomial proportions analysis; α=0.05).

Fig. 8 .
Fig. 8. Uca burgersi.Proportions of actively reproducing and inactive females.Comparisons were made among seasons, for each site.Bars with at least one letter in common did not differ significantly (multinomial proportions analysis; α=0.05).

Fig. 9 .
Fig. 9. Uca burgersi.Proportions of molting individuals.Comparisons were made among seasons, for each site.Bars with at least one letter in common did not differ significantly (multinomial proportions analysis; α=0.05).

TABLE 1
Uca burgersi.Description of each gonadal stage for males and females and reproductive condition.Organic matter content (%; mean ± SD) in the sediment from studied sites.Descriptive measure for each demographic category, in the localities studied.