FISHERY BULLETIN: VOL. 85, NO. 3 



sediments at sites in this stratum since grab sam- 

 ples contained coral fragments and rubble. At 

 shoaler locations where golden crabs were abun- 

 dant, sediments were a mixture of soft silt-clay, 

 molluscan shell fragments, and foraminiferan 

 tests. Temperatures at sites where golden crabs 

 were collected ranged from 7.14° to 9.15°C. 



The number of golden crab per trap (11.4) and 

 the weight of golden crab per trap (9.37 kg) in the 

 Florida trap exceeded that in the Fathoms Plus 

 trap (7.0 individuals, 6.16 kg) for all combined 

 sets (Table 2). Statistical results by strata using 

 the two-sample ^-test or an approximate f-test 

 when variances were heterogeneous (Sokal and 

 Rohlf 1983), indicated significantly more crabs 

 were collected with the Florida trap than with the 

 Fathoms Plus trap from 367 to 457 m (stratum 2) 

 and from 458 to 549 m (stratum 3) (Fig. 1, 

 Table 2). Weight per trap was significantly differ- 

 ent for the 367-457 m stratum only. 



Size and Sex Composition 



Male G. fenneri were significantly more numer- 

 ous than females, outnumbering them by -18:1. 

 No ovigerous females were collected during the 

 sampling period. Dominance of males was statis- 

 tically significant for strata 1-3 (Table 3). In these 

 depth strata, males were 20 times as numerous as 

 females. In depths of 550-732 m, a male was the 



ranged from 85 to 193 mm in carapace width and 

 weighed from 100 to 2,109 g. Average weight of 

 male golden crab collected during the study was 

 927 g (s = 373.448, n = 1,640) while average 

 weight of females was 443 g (s = 289.385, 

 n =86). Carapace width-frequency distribution 

 for G. fenneri gave modes at 155 mm for males 

 and 100 mm for females (Fig. 2). The largest crab 

 collected measured 193 mm and weighed 2,091 g. 



Linear least-squares and functional regression 

 equations (Ricker 1973; Sokal and Rohlf 1983) 

 relating carapace length and live wet body weight 

 with width are in Table 4. Width-weight relation- 

 ships were calculated from data on individuals 

 that were not missing appendages. 



Of the 3,183 golden crabs examined for missing 

 appendages, 2.4% were missing one or both 

 chelae. Pereopods were missing from 307 individ- 

 uals (9.6%). 



Examination of carapace width and weight 

 statistics for each depth stratum showed that 

 mean size of male G. fenneri was greatest for the 

 shallowest (274-366 m) and deepest (733-823 m) 

 strata sampled (Table 5). For females, however, 

 mean carapace width and weight were greatest in 

 the deepest zone. At depths of peak abundance, 

 mean carapace width it^ = 4.70, P < 0.001) and 

 mean body weight it, = 2.70, P < 0.01) of male 

 crabs were significantly greater in the 367-457 m 

 than in the 458-549 m depth stratum. No signifi- 



Table 2 — Results of /-test (T^) comparisons of mean number and weight 

 (kg) per trap for two trap types (FM + and FLA) fished in each depth stratum 

 for Geryon fenneri. Standard deviation is noted in parentheses; ' indicates 

 significance at 0.05 level. 



only crab collected. In the deepest stratum sam- 

 pled (733-823 m), females significantly outnum- 

 bered males 2.9:1. Although the Florida trap 

 caught significantly more crabs than the Fath- 

 oms Plus trap overall, no significant difference 

 was noted in the number of female crabs between 

 those two trap types (x^ test, P > 0.5). 



The 3,217 golden crabs which were measured 



Table 3. — Frequency of male and female Geryon fenneri within 

 each depth stratum. Asterisks denote significant deviation 

 (P  0.05) from 1:1 by Chi-square analysis 



550 



