swordfish with longline gear is more successful dur- 

 ing the night than day (Ueyanagi, 1974), the low 

 catches may only be reflecting the fact that the 

 Hawaiian fishery operates principally during day- 

 light hours. Day fishing is carried out to maximize 

 the catch of tunas and species of billfishes other 

 than swordfish. 



Figure 1 shows the monthly landings of swordfish 

 for the period 1961-66. Although catches are small, 

 there is a pronounced increase in landings during 

 the summer months with the peak occurring in July. 

 The increase is due to an increase in availability and 

 not to an increase in fishing effort, since Yoshida 

 (1974) showed that the catch rates (catch per trip) 

 for blue marlin, Makaira nigricans, and striped mar- 

 lin, Tetiapterus aiulax, in the Hawaiian longline 

 fishery parallel the monthly landings, thus suggest- 

 ing that the monthly catch data could be used as a 

 general measure of availability. 



The average size of swordfish also shows a peak 

 during the summer period. As it will be discussed 

 later, the increase in average size accompanied by 

 the appearance of females in late stages of matura- 

 tion may be related to a spawning migration. 



MATERIALS AND METHODS 



The 16 swordfish ovaries were collected at the 

 Honolulu fish markets between June 1964 and May 

 1967 (Table 1). Since longline-caught fish are kept 

 refrigerated with crushed ice, the ovaries were kept 

 in an unfrozen condition until collected. 



In the laboratory, excess connective tissue was 

 removed from the external surfaces of the ovaries. 

 The ovaries were weighed to the nearest gram and 

 preserved in 10% Formalin. Detailed microscopic 

 examination of the ovaries was undertaken only 

 after the ovarian material had been thoroughly pre- 

 served, and shrinkage had stabilized. Generally, ova 

 diameter measurements were taken after preserva- 

 tion had exceeded 6 mo. 



For the maturation study, a small sample was 

 extracted from the ovary with a cork borer and 100 

 randomly selected ova were measured to obtain 

 mean diameter values for the most developed ova 

 size group. Individual ova diameters obtained were 

 not necessarily the maximum diameters. We fol- 

 lowed the method developed by Yuen (1955) for 

 measuring bigeye tuna ova and used by Otsu and 



Table I. — Summary of swordfish data used in maturation and fecundity study. 



Key: IM - Immature 



ED - Early developing 



ER - Early ripe 



RP - Ripe 



RS - Residual eggs present 



^ Gonad Index Is percentage of fresh ovary 

 weight to fish size. 



3 Weight estimated from fresh-preserved con- 

 version given in Figure 2. 



^ Ova diameters of fresh (non-preserved) sam- 

 ples placed In sea water averaged 1.571 mm. 



143 



