886 



Fishery Bulletin 101(4) 



indicate that recruitment of P. formosa takes place in 

 deeper water. This conclusion is based on a decrease in 

 size with increasing depths, the higher proportion of fe- 

 males encountered in deeper water, and increasing crab 

 densities at depth. For P spinosissima this trend was not so 

 pronounced, although crabs of this species were also gener- 

 ally of a smaller size in deeper water However, contrary to 

 the case with P. formosa, females were more prevalent in 

 shallower water. Very few females of P. spinosissima were 

 encountered in deep water These rather unusual findings 

 might suggest ecological partitioning of the benthic habi- 

 tat, and warrant further investigation. 



Another unexpected result of our work was the discovery 

 of a third species of Paralomis, P. anamerae, at intermedi- 

 ate depths. This species was apparently not present in the 

 American Champion or Pro Surveyor catches, presumably 

 again because of the depth restriction in these cruises. 



Maturity 



For P. spinosissima our estimate of 67.3 mm carapace 

 length at 50% male maturity is similar to the 66.4 mm 

 carapace length found by Otto and Macintosh (1996) at 

 Shag Rocks for this species. Unfortunately, as can be seen 

 from Figure 8, relatively few small P. formosa males were 

 encountered and size at maturity for this species (64.0 mm 

 carapace length) is likely to have been poorly estimated in 

 our analysis. However, if it is assumed that male P. formosa 

 mature at the same size in relation to female P. formosa, 

 as in the case of P. spinosissima, the female maturity data 

 presented in Table 5 would suggest that male P. formosa 

 would mature at 60.4 mm carapace length (64.3 mm cara- 

 pace width) rather than the 64.0 mm shown in Figure 8. 



Watters and Hobday (1998) have also examined size at 

 maturity for P. spinosissima and P. formosa, although their 

 samples were taken from South Georgia rather than Shag 

 Rocks. Using a method based on finding the maximum 

 of the second derivative of smoothing spline fits to chela 

 height and carapace width data, they found that size at 

 morphometric maturity for P. spinosissima was 73 mm 

 carapace length. This size is similar to that which Otto 

 and Macintosh (1996) obtained for P. spinosissima at South 

 Georgia using the same technique as we did. Watters and 

 Hobday's (1998) results for P. formosa are, however, for a 

 higher size at maturity (80 mm carapace length) than that 

 for P. spinosissima, which would seem to be at variance 

 with our results and the apparent relative sizes of the two 

 species (see Fig. 6 and CCAMLR^). 



The minimum size limits for crabs at South Georgia were 

 set by CCAMLR in 1992 but, in common with many crab 

 stocks (Schmidt and Pengilly, 1993), these measures were 

 not accompanied by rigorous analysis of the effectiveness 

 of these measures in meeting management objectives. For 

 P. spinosissima. Otto and Macintosh's^ male maturity re- 

 sults for P. spinosissima were used, and allowing males 

 at least one opportunity to breed and an assumed growth 

 per moult of 15%, minimum size limits were calculated as 

 94 mm and 84 mm carapace width at South Georgia and 

 Shag Rocks, respectively (CCAMLR^). These results are 

 very similar to our own, but the CCAMLR limit of 102 mm 



width was based on the then-existing processing require- 

 ments rather than on these calculations. Our results sug- 

 gest, allowing at least one opportunity to breed, that the 

 limit should be 83 mm for P. spinosissima. For P. formosa, 

 taking our more conservative figure of 64.0 mm carapace 

 length at 50% maturity, the catch size limit should be set 

 at 78 mm carapace width (the less conservative figure, 60.4 

 mm carapace length, would suggest a size limit of 74 mm 

 carapace width). 



Hoggarth (1991) reviewed minimum size limits for a 

 number of stocks of lithodid crabs and found that mini- 

 mum legal sizes were about 70% of the maximum size for 

 males, which would suggest 85 mm and 84 mm carapace 

 width for P. spinosissima and P. formosa, respectively. It 

 should, however, also be taken into account that these 

 estimates were probably biased because of the greater 

 sampling effort made at Shag Rocks. Note that the length- 

 frequency distribution for P. formosa in Figure 6B appears 

 to indicate a lower maximum size for males of this species 

 than for males of P. spinosissima. However, the largest 

 P formosa actually encountered was 120 mm carapace 

 width. Furthermore, Figure 6B seems to be truncated at 

 the larger sizes, suggesting perhaps that a proportion of 

 the large adult population was not encountered during 

 fishing. 



Discard mortality 



Our results demonstrate that, although a high proportion of 

 crabs is likely to survive the physical strain of being hauled 

 to the surface from potentially great depths, some under- 

 size individuals and nontarget females can be expected 

 to die following discarding. The most significant factor 

 affecting discard survivorship was handling on board the 

 vessel. On the Argos Georgia, where crabs were unloaded 

 from pots and sorted on a conveyor belt, survivorships were 

 high, up to 88%, and P. spinosissima survived better than 

 P. formosa. By contrast, on the Argos Helena, where crabs 

 went down a chute prior to processing, survival rate was 

 between 38% and 58% and P formosa survived consider- 

 ably better than P spinosissima. In general, P. anamerae 

 was the most vulnerable species, followed by P spinosis- 

 sima, and the least vulnerable — P formosa. 



Studies of the discard mortality of lithodid crabs in 

 North Pacific fisheries have produced a variety of results. 

 Stevens ( 1990) found that crabs discarded from commercial 

 sole trawls suffered high mortalities (47.3%), but Byersdor- 

 fer and Watson'' and Zhou and Shirley ( 1995) both reported 

 relatively low mortalities (<2%) resulting from handling 

 when fishing with pots. Our results support these previous 

 studies and extend them to the Antarctic, clearly indicating 

 that where handling on a pot vessel is reduced, mortali- 

 ties are relatively low (<15% mortality). When crabs are 



' Byersdorfer, S., and L. J. Watson. 1992. A summary of bio- 

 logical data collected during the 1991 Bristol Bay red king crab 

 tagging study. Technical Fishery Report 92-14, 30 p. Alaska 

 Department of Fish and Game, Division of Commercial Fisheries, 

 P.O. Box 25526, Juneau, AK 99802-5526. 



