Reproductive Differences among Delmarva Grass Shrimp (Palaemonetes pugio and P. vulgaris) Populations

Populations of female grass shrimps (Palaemonetes pugio and P. vulgaris) were sampled from five coastal embayments in Delaware, Maryland, and Virginia (Delmarva) and compared with respect to reproductive and life history attributes. We observed interspecific differences in timing ofreproduction, carapace length, ratio of carapace length to total body length, body weight, clutch weight, clutch size, and egg volume. Onset of reproduction in P. vulgaris lagged behind P. pugio. Although there was no difference in the relationship between clutch size and carapace length for the two species, carapace length/total body length in P. pugio was greater than that in P. vulgaris. A multivariate analysis of variance indicated significant differences in carapace length, clutch weight, body weight, clutch size, and egg volume attributable to effects of species, population, and interactions between them. At all sites, P. pugio produced larger eggs than P. vulgaris. Although the two species did not differ in reproductive effort, both species exhibited increases in reproductive effort with latitude. Clutch size also tended to increase with latitude for both species. In populations where both species were abundant, adult females of P. pugio were longer and heavier and produced heavier egg masses comprised of fewer, larger eggs. INTRODUCTION In a classic paper, Hutchinson (1961) raised the issue of how so many similar species are able to coexist in the plankton given the prediction, from the principle of competitive exclusion (Gause, 1934; Hardin, 1960), that one species should outcompete the others. Coexistence of similar species is exemplified by the "grass shrimps" Palaemonetes pugio Holthuis and Palaemonetes vulgaris (Say) that abound in marshes and bays of the Atlantic and Gulf coasts of North America. These two closely related species often co-occur in estuarine sites (Williams, 1984). Although both species exhibit similar distribution patterns across their geographic ranges, they exhibit differences in within-habitat usage. In previous studies, it has been shown that habitat partitioning in these species is a consequence of interspecific differences in physiological tplerances toward salinity (Thorp and Hoss, 1975; Knowlton and Kirby, 1984; Present address: Department of Entomology & Center for Population Biology, One Shields Avenue, University of California, Davis, CA 95616. 2 Corresponding author. 36 VIRGINIA JOURNAL OF SCIENCE Knowlton and Schoen, 1984; Khan et al., 1995, 1997) and dissolved oxygen (Welsh, 1975), substrate and cover preferences (Arguin et al., 1989; Knowlton et al., 1994; Khan et al., 1995, 1997), and interference competition (Thorp, 1976). In the present study, we examine whether differences in reproductive strategies could also promote resource partitioning between P. pugio and P. vulgaris and how these differences are maintained across a range of environmental conditions. There is some evidence to suggest that reproductive strategies differ between the two species. Chambers (I 982) and Yan (1987) found that Massachusetts populations of P. pugio exhibit greater mean clutch size than P. vulgaris with the same reproductive effort (ratio of gonadal weight to body weight). Although seasonal breeding periods of P. pugio and P. vulgar is are thought to be similar (Knowlton, 1970), Hoffman ( 1980) observed that Delaware populations of P. pugio produced three or more clutches while P. vulgaris produced no more than two clutches within a single season. Within a species, reproductive characteristics might be expected to vary according to season and geographic location due to differences in temperature, photoperiod, and salinity. Latitudinal clines in egg number have been observed in birds, fishes, and mammals such that clutch size increases with latitude (reviewed by Fleming and Gross, 1990). Such variation in clutch size may reflect differences in the growing season. Salinity may also influence clutch size in estuarine organisms. For example, Alon and Stancyk (1982) found that P. pugio fecundity increased with prolonged exposure to lower salinities. The purposes of this study were to explore the extent to which P. pugio and differ reproductively and to examine these differences across a range of environmental conditions. Accordingly, we compared reproductive attributes of P. pugio and P. vulgaris from five marine and estuarine sites in Chesapeake and "outer" (Atlantic) bays of Virginia, Maryland, and Delaware, spanning a substantial salinity gradient and a wide range of latitude. We examined the effects of species and population level variation on reproductive characteristics. To determine broader geographical patterns in reproductive strategies, we compared results of this study of Delmarva populations with studies from populations in other regions. METHODS Palaemonetes pugio and P. vulgaris populations were sampled during the breeding period (May, July, and September 1987) at five locations (Figure 1; see Knowlton et al., 1994 for details). Collecting locations were selected so that two pairs of Chesapeake and Atlantic sites occurred at similar latitudes and spanned a broad range of salinities. Values at Chesapeake sites ranged from about 12 ppt at Station 5 to about 25 ppt at Station 3 while those at both Atlantic sites (Stations 1 and 2) were about the same, averaging about 30 ppt (Figure 1, Appendix A). Collections at each site were timed to occur at roughly the same time of day and stage of tide (about two hours prior to low tide, Appendix A). Samples, obtained using long-handled D-frame dip nets, were preserved on site in 95% ethanol. At time of collection, salinity was measured with a hand-held refractometer, air and water temperatures with a pocket thermometer. Dissolved oxygen (mg/L) was determined using a modified Winkler method (Hach Chemical Co., 1977). In the laboratory, we used a dissecting microscope to separate species per sample according to criteria in Holthuis ( 1952). Sex was determined by noting the form of the SHRIMP REP: ,-----:1a1t,. 1.., .:1.... ':b. FIGURE 1. Locations of collection sit, Bay, Chincoteague, VA; 3 = Kings Cre( near Saxis, VA; and 5 = Mezick Pond, ~ SHRIMP REPRODUCTIVE DIFFERENCES 37