Differences in the Crab Fauna of Mangrove Areas
at a Southwest Florida and a Northeast Australia Location:
Implications for Leaf Litter Processing
CAROLE C. MCIVOR
T. J. SMITH

Abstract
Existtig paradigms suggest that mangrove leaf litter is processed primarily via the detrital pathway in forests in the Caribbean biogeographic realm whereas herbivorous crabs are relatively more important litter processors in the Indo-West Pacific. To test thii hypothesis, we used pitfall traps to collect intertidal crabs to characterize the crab fauna in a mangrove estuary in southwest Florida. We also tethered mangrove leaves to determine if herbivorous crabs are major leaf consumers there. We compared the results with previously published data collected in an analogous manner from forests in northeastern Australia. The crab fauna in Rookery Bay, Florida, is dominated by carnivorous xanthid and deposit-feeding ocypodid crabs whereas that of the Murray River in northeastern Australia is dominated by herbivorous grapsid crabs. No leaves tethered at five sites in the forests in Southwest Florida were taken by crabs. This contrasts greatly with reported values of leaf removal by crabs in Australian forests of 28-79% of the leaves reaching the forest floor. These differences in the faunal assemblages and in the fate of marked or tethered leaves provide preliminary support for the hypothesis that leaf litter is in fact processed in fundamentally different ways in the two biogeographic realms.

Discussion
Bias of pitfall traps for crabs, though uninvestigated, is likely a function of species-specific behavior. Eurytium limosum are highly active predators of invertebrates within the intertidal zone (Kneib and Weeks 1990). Wilson (personal communication) observed these crabs respond to the vibrations of clicking fiddler crab claws by running toward the source of sound. It is possible that the vibrations of crabs captive in the pitfall traps attracted Eurytium to the traps. Such a scenario could cause an overestimation of the relative abundance of Eurytium, and an underestimation of the abundance of any crabs that became prey for Eurytium in the traps. Whereas we cannot completely discount the overestimation of Eurytium, we do not believe predation was a major problem in the traps. If predation were very prevalent in traps with Eurytium or Panopeus, we would expect to find parts of crabs, particularly legs. This was not the case: detached legs were almost always matched with an individual crab for estimation of biomass. Furthermore, many crabs were moribund or debilitated when cleared from the traps, likely a result of drowning: many intertidal crabs have gills modified for aerial existence and drown when they cannot periodically climb out of the water. The pattern of increasing crab abundance with increasing tidal elevation observed in both southwest Florida and northeast Australia has also been observed in Malaysia (Sasekumar 1974). This pattern may be a function of a greater risk of predation from fish and swimming crabs (portunids) in the more frequently flooded low intertidal zone. The pattern could also result from longer feeding times afforded by shorter inundation periods in higher intertidal forests. In summer, Wilson (1989) found that predation risk for four species of crabs in fringing mangrove forests in south Florida was less in the canopy and on prop roots than on the forest floor, and less on low tide than on high tide, suggesting that the primary predators were nektonic. The greater diversity of the intertidal crab fauna found in a northeastern Australian site compared to a site in southwest Florida is in agreement with previously described patterns of high diversity of mangrove crabs in the Indo-West Pacific (MacNae 1968). Such a pattern is consistent with the species-area hypothesis (Abele and Walters 1979; Connor and McCoy 1979): mangroves occupy far greater area in northeastern Australia and southeast Asia than in Florida and the Caribbean (Saenger et al. 1983). The marked difference in average abundance of crabs in the two biogeographic areas was unanticipated, and begs explanation. A hypothesis, yet untested, is that greater predation from nekton in the Australian forests keeps crab numbers lower than in analogous habitats in Florida. Australian forests experience considerably greater depths of tidal inundation than do the forests of the Caribbean realm: mean tidal range in northeast Queensland is 2.6 m (Smith 1987) compared to 1.2 m in Rookery Bay. Deeper water permits access to the forest by large predaceous fishes (Morton 1990; Robertson and Duke 1990). It is reasonable to expect that shallower water depths in the forest, characteristic of the forests of Rookery Bay and other south Florida and Caribbean forests, preclude some larger predaceous fishes from foraging in these habitats. The markedly different composition of the two crab faunas in terms of trophic guilds (Table 1, relative abundance of family Grapsidae) is the most striking result of our research. At least in Rookery Bay, the sesarmid crab fauna appears to be poorly developed with the result that there is little if any consumption of freshly fallen leaf litter. (Melampus coffeus, the coffee bean snail, has been previously identified as a mangrove leaf detritivore [Smith et al. 1989; Proffitt et al. 19931, though it’s relative importance compared to herbivorous crabs is unknown.) Whereas the mangrove tree crab, Aratus pisonii, consumes some green leaf material in the canopy (Beever et al. 1979), this species is more omnivorous than herbivorous (McIvor unpublished stable isotope data; Wilson personal communication). Sesarma curacaoense, apparently more common near Flamingo in south Florida (Wilson 1989) than at our sites in Rookery Bay, almost certainly consumes mangrove leaf litter (Wilson personal communication) in addition to detritus (Abele 1973). Two congeners that occur in East Coast salt marshes, Sesarma reticulaturn and Sesarma cinereum, consume live and dead plant material (S’artina aZternu$ora) , and plant detritus respectively (Seiple and Salmon 1982). (These two species are uncommon at Rookery Bay: none were collected in our pitfall traps.) To what extent Rookery Bay is typical of south Florida and the Caribbean with regard to a poorly developed herbivorous crab fauna is not known. However, patterns of relative abundance of semiterrestrial crabs on mangrove-dominated isles off Belize (as determined in preliminary studies) are strikingly similar to those at Rookery Bay, that is, dominance by Eurytium limosum and several species of Ucu, lesser numbers of individuals of the leafeating genera Gonio@is and Ucides, and no sesarmids (Smith and Ruetzler unpublished data). Wilson (personal communication) reports that Sesarma curacuoense is a secretive, shy crab that spends much of its time in hiding. Abele (1973) reports that the same species is “sluggish.” Because of these behavioral traits, the species may be underrepresented in pitfall traps. However, the fact that none of our tethered leaves were taken by herbivorous crabs suggests to us that at least in Rookery Bay, the species, whatever its true abundance, is not common. In other parts of the Caribbean, herbivorous crabs may be more common.

Source :
Estuaries Vol. 18, No. 4, p. 591-597 December 1995
Next >>>