492 



Fishery Bulletin 98(3) 



45° 



35° 



25° 



15° 



January-March 



« 45° 



180° 170°W 160°W 150°W 170° E 



April-June 



35° 



25° 



15° 



_L 



_L 



180° 170° W 160° W 150° W 



October-December 



9 lA) 



O Females 

 A Males 

 Hawaiian Archipelago 



_L 



J_ 



_L 



J 



180° 



170°W 160°W 150°W 



180° 



170°W 160°W 150°W 140°W 



Longitude W 



Figure 2 



Spatial distributions of known-sex (female: hollow circles; male: hollow triangles) swordfish iXiphias gladius). 

 caught by the Hawaii-based pelagic longline fishery during January-March, April-June, July-September, and 

 October-December periods, between March 1994 and June 1997, illustrating seasonal geographic shifts in the 

 fishery. Areas of symbols are proportional to catch-per-unit-of-effort (CPUE, number of fish per set); CPUEs are 

 scaled continuously from a minimum of 1 to a maximum of 17 fish per set. 



that included a central nucleus and were spherical (least 

 deformed by processing), so that the coefficient of varia- 

 tion (CV) of the diameter estimate was kW-t . Diameters 

 measured for a subset of matched (formalin-fixed, histol- 

 ogy) specimens were used to convert between the two 

 types of measurements. 



The gonadal development of males was classified by 

 using the criteria of Grier (19811 as specified for billfish 

 by de Sylva and Breder (1997). Males were scored as 

 "immature" if spermatogenesis (meiosis) was absent or, 

 when occurring, if ripe (fully tailed) spermatozoa were 

 restricted to crypts and lobules (tubules) of testes. Males 

 were deemed mature only if ripe sperm were present in 

 spermatic ducts, as well as in crypts and lobules, or as 

 spent or resting if they had developed lobules but no sper- 

 matogenesis present (Table 1 ). 



Sex was histologically verified for 506 males and 830 

 females. The sex of only 23 swordfish (8 males: 70-153 cm 

 EFL, March-December; 15 females: 98-2 12 cm EFL, April- 

 December) was misclassified from gross visual inspection 



of gonads by observers aboard commercial longliners. Mis- 

 classification error was unrelated to true sex (x'=0.02: 

 df=l; P=0.9), and appeared unrelated to fish size or period 

 of capture, although the data were too few to formally 

 evaluate size and period effects. Because the misclassifi- 

 cation rate was so low (O.S'^r), we ignored it in further 

 analyses of the sex composition of swordfish catch where 

 field-sexed fish were used. 



For ovary specimens collected in spring 1997, frozen 

 specimens were thawed overnight in a refrigerator. Each 

 lobe of bilobed ovaries was weighed damp; accuracy was 1 

 g (if total ovary weight <500 g) or 20 g (if >500 g). Seven of 

 95 females caught during the spring 1997 spawning period 

 lacked weights for one ovary lobe; total ovary weights 

 of these fish were estimated by using the relationship 

 between both ovaries (total gonad weight, GW) and right 

 (RO) and left (LO) ovary weights (in g): GW = 1.807RO, 

 and GW = 2.170LO -h 133.6 (both r->0.98; «=88). Right 

 ovaries of the swordfish examined were 28'f heavier than 

 left ovaries (matched-pair /-test; f=6.9; n=88). 



