DeMartini et al : Maturity, size, and sex composition of Xiphias gladius 



501 



equivalent to an estimated 124 cm EFL, considerably smaller 

 than that documented for females in any other fishery. This 

 estimate by de la Serna et al. ( 1996) was based on length- 

 based gonad indices that were histologically validated. 



For swordfish caught by the Hawaii-based longline fish- 

 ery, the disparate body lengths at median sexual maturity 

 of males and females are equivalent to a threefold differ- 

 ence between .55 kg RW (40 kg dressed-head off weight, 

 DW) for females and 19 kg RW ( 14 kg DW) for males. Cor- 

 reponding ratios of femaleimale body weights at L-,, for 

 swordfish in the tropical and western North Atlantic are 4.4 

 (Taylor and Murphy, 1992) and 2.7 (Arocha and Lee, 1996). 

 Details of computation notwithstanding, these disparities in 

 mass clearly represent biologically meaningful differences 

 between the sexes in swordfish. Sexual differences in rates 

 of biomass accrual — likely due to sexually dimorphic growth 

 rates — translate to important differences in the sustain- 

 able fisheries yield provided by each sex. With later matura- 

 tion, the period of rapid prereproductive somatic growth of 

 female swordfish extends several or more years longer than 



that of males. Taylor and Murphy ( 1992 ) for example esti- 

 mate ages-at-sexual maturity of swordfish from the Straits 

 of Florida, corresponding to median sizes-at-maturity, as 5.5 

 yr (for females) and 1.4 yr (for males). 



At present our characterization of maturity in North 

 Pacific swordfish is limited to body size. Age-at-maturity 

 estimates must await completion of size-at-age studies 

 being conducted at the Honolulu Laboratory of the National 

 Marine Fisheries Service. 



Sex and size composition 



Sex ratios of swordfish caught by the Hawaii-based long- 

 line fishery vary temporally and spatially. These variations 

 may partly reflect different availabilities or catchabilities 

 resulting from sexual differences in behavior that change 

 seasonally (e.g. during spawning periods). Variations in 

 the sex composition of catches must partly reflect differ- 

 ent natural distributions of the sexes. Sex ratios of catches 

 follow recurrent seasonal patterns among years, and spa- 

 tial differences in the sex composition of catches repeat 

 annually. Neither of these two observations would be likely 

 if, for example, the sex composition of catches represented 

 simple targeting of larger (female) fish by fishermen. 



Similarities in sex ratios-at-length between North Atlan- 

 tic (Stone and Porter, 1997) and central North Pacific 

 swordfish (our study) suggest that size-specific sex ratios 

 of swordfish catches can be predicted with adequate pre- 

 cision from data on the length composition of catches. If 

 true, this could obviate the need for a direct characteriza- 

 tion of the sex composition of swordfish catches — a task 

 complicated in fisheries like Hawaii's in which swordfish 

 are landed fully dressed. Sex ratios of swordfish caught 

 by the Hawaii-based fishery are identical for swordfish- 

 targeted and nonswordfish-targeted sets. Estimates of the 

 sex ratio of swordfish catches in the Hawaii-based fishery 

 thus seem robust to the specifics of capture, at least for 

 the period sampled. Further comparisons of sex ratio esti- 

 mates for catches with different gear types over a longer 

 sampling period would be required to thoroughly evalu- 

 ate possible sexual differences in catchability and expand 

 inferences on sex ratios beyond catch to stock. 



Seasonal patterns Temporal changes in the sex ratios 

 and related size composition of swordfish catches in the 

 Hawaii-based longline fishery resemble those observed 

 for swordfish in other known spawning areas. Taylor and 

 Murphy (1992) observed analogous temporal differences 

 in catches of male and female swordfish in and near the 

 Straits of Florida, where the fraction of males in catches 

 was significantly greater than that of females only during 

 peak (spring-summer) spawning periods. Temporal pat- 

 terns of changes in sex ratios thus are consistent with 

 sexual differences in migratory and spawning or other 

 reproductive behaviors such as the greater propensity 

 for males to aggregate in group-spawning pelagic species 

 (Hunter and Macewicz, 1986) and other fishes. 



Spatial patterns Arocha and Lee ( 1995 ) observed spatial 

 variations, complementary to Taylor and Murphy's ( 1992) 



