634 



Fishery Bulletin 92(3). 1994 



SOUTH CAROLINA 





Figure 1 



Map of Breach Inlet, South Carolina, site of collection of postlarval 

 Penaeus. 



Ringo and Zamora (1968). Postlarvae with overlap- 

 ping characters were identified to genus. 



Counts of postlarvae were converted to density 

 (numbers per 1,000 m J ). Preliminary analyses re- 

 vealed these data to be nonnormal and with large 

 variation; therefore, potential effects of time of day, 

 tidal stage, and location of nets (surface versus bot- 

 tom) on ingress of postlarvae collected by consecu- 

 tive sampling were tested with the nonparametric 

 Mann-Whitney test on densities of postlarvae (flood 

 tide only for time of day and location of nets; Siegel, 

 1956). The effect of lunar phase on catches made 

 during regular sampling (daytime bottom collections 

 made on flood tide during season of peak abundance) 

 was tested with the nonparametric Kruskal-Wallis 



test on densities grouped by four lunar periods ( new, 

 full, first, or last quarter) pooled from the entire 

 year's data (Siegel, 1956). Differences among groups 

 were considered statistically significant at the P<0.05 

 level. The above data were log-transformed to facili- 

 tate graphic representation of means and standard 

 deviations (Figs. 2 and 3). An annual index of abun- 

 dance (Xlog((number/l,000m 3 ) + l)/number of 

 samples); Elliot, 1977) was calculated for each spe- 

 cies for use in regression analysis. The index was 

 computed from February, once densities reached 20/ 

 1,000 m 3 , through April for P. aztecus, and from May 

 through August for P. setiferus. Annual indices of 

 postlarval abundance were regressed against annual 

 estimates of harvest to ascertain possible relation- 



