FISHERY BULLETIN: VOL. 85, NO. 2 



with increasing size (regression coefficient = 1.64, 

 r- = 0.52). 



Based on the assumption that the relative growth 

 of the appendix mascuhna correlates directly with 

 maturity, we chose the point where this appendage 

 is 90% of its asymptotic value to define the length 

 at maturity for males. The 90% level was used in- 

 stead of 50% of the asymptotic value, as used for 

 the females below, because the 50% point would be 

 where the males are 50% mature, not where 50% 

 of the males are mature. Using this definition, the 

 carapace length at maturity (L^j) for males of H. 

 ensifer is 20.2 mm and that ofH. laevigatus is 35.7 

 mm. For H. longirostris, the length at maturity is 

 estimated at 31 mm. 



The length at maturity of females is perhaps more 

 important in assessment work since the females are 

 directly responsible for the production of recruits 

 to the population. For this study we used the pres- 

 ence of eggs (berried) as the indicator of maturity. 

 The length at maturity is defined as the size where 

 50% of the females are mature (Gunderson et al. 

 1980). When using the presence of eggs as the 

 measure of maturity, the sample must be restricted 

 to females collected during the time of year that egg 

 bearing can be expected. For H. laevigatus, the 

 breeding season is relatively discrete (November to 

 February), whereas for H. ensifer and H. longiros- 

 tris there are peaks in December and May (Fig. 4). 

 Data for each species were fitted to the same non- 

 linear regression model used for the males. Asymp- 

 totic values for percent berried by carapace length 

 are 66% for H. ensifer, 92% for H. laevigatus, and 

 55% fori/, longirostris. The carapace lengths asso- 

 ciated with values equal to one-half of the asymp- 

 totic values are the L^ for the various species. 

 These are 23.9 mm for H. ensifer, 42.7 mm for H. 

 laevigatus, and 37.4 mm for H. longirostris. 



YIELD ASSESSMENT 



The assumptions and methods of yield assessment 

 used in this study are presented in Polovina and 

 Ralston (1986) and Wetherall et al. (in press). Be- 

 cause H. laevigatus yielded the highest catch rates 

 and because it is generally regarded as having 

 superior market acceptability, most of our fishing 

 effort targeted this species. Hence estimates of total 

 biomass and sustainable yield for the pandalid re- 

 source are restricted to this species. 



GROWTH AND MORTALITY 



Estimates of asymptotic size (L^) and the ratio 

 344 



100 



80 



60  



40 • 



20 



H. longirostris 



I I 



Nov Dec Jan Feb Mar Apr May Jun Jul Aug 

 MONTH 



Figure 4.— Percentage of females bearing eggs by month for three 

 species of Heterocmyus with 95% confidence limits. 



of instantaneous total mortality to the instantaneous 

 growth constant {ZIK) were obtained by examining 

 the descending limb of the length-frequency distri- 

 bution using the regression method based on the 

 Beverton and Holt (1956) model (Wetherall et al. in 

 press). Table 4 lists the values of L^ and ZIK for 

 males, females, and pooled sexes for each of the 

 three major species. As anticipated, L^ values for 

 females are larger than those for males of the same 

 species. Estimates of length at recruitment to the 

 exploitable population (L^j) were obtained by apply- 

 ing the method of Gulland (1969) to the ascending 

 limb of the same length-frequency distributions. 

 Heterocarpus ensifer is recruited into the fishery at 

 a carapace length of 23 mm, H. laemgatus at 29 mm, 

 and H. longirostris at 34 mm. 



By fixing the L^ value at the estimates obtained 

 from the large length-frequency sample and using 



