MATERIALS AND METHOD 

 COLLECTION OF OVARIES 



This study is based on 402 pairs of ovaries col- 

 lected on seven exploratory fishing cruises from 

 August 1956 through June 1958. The approxi- 

 mate locales of collection are shown in figure 1. 

 Most of the ovaries were collected from fish caught 

 at the surface by pole and line. This method of 

 fishing, as used in the Hawaiian skipjack fishery, 

 was described by June (1951). 



Plans for all of the pole-and-line fishing cruises 

 called for sampling 25 skipjack from each school 

 fished in order to obtain an estimate of their size 

 and sex composition. From the 25 fish thus 

 selected, the first 5 females picked at random were 

 cut open and their ovaries removed for examina- 

 tion. Skipjack were caught by pole and line 

 from 92 schools during this study, and ovary 

 samples were secured from fish from 81 of these 

 schools. Longline fishing and incidental trolling 

 provided the few remaining samples. A descrip- 



-Cj>ti 



; a 



• NUKIHIU 



o 



•MARQUESAS 



• \ jHVA FOU 



./ ISLANDS 





•d 



fill Itn, 



I hi re 1. — Locations where skipjack ovaries used in this 

 study were collected. 



tion of the longline fishing method was given by 

 Mann (1955). 



The ovaries were preserved in about 10 percent 

 formalin. At the time of collection a record was 

 made of the date, locality, method of fishing, and 

 the fork length of the fish. 



EXAMINATION OF OVARIES 



Several investigators made detailed studies of 

 the distribution of mature ova within ovaries: 

 June (1953) for yellowfin (Thunnus albacares); 

 Yuen (1955) for bigeye (Thunnus obesus); and 

 Otsu and Uchida (1959) for albacore (Thunnus 

 alalunga). All showed that a representative 

 sample of ova could be obtained anywhere along 

 the length of an ovary or from either member of a 

 pair. 



Similarly, an analysis of variance of the mean 

 size of mature ova within a skipjack ovary indi- 

 cated that they were homogeneously distributed. 

 Therefore, the following sampling method was 

 adopted for this study. The formalin-preserved 

 ovaries were weighed after excess moisture and 

 tissue had been removed. A cross-section about 

 K-inch thick was taken, usually from the middle 

 of the right ovary, from whicli a wedge-shaped 

 (triangular) sample was cut. In many instances 

 it was not possible to distinguish the right ovary 

 from the left. In these instances the smaller of 

 the pair was selected for examination; it appeared 

 that the right ovary was usually the smaller of the 

 two. I assumed that the distribution of ova 

 within the right and left ovaries is similar, as has 

 been reported for the tuna cited above. 



The ova from the triangular section were teased 

 apart in a shallow dish and measured in a Sedg- 

 wick-Rafter counting chamber. Measurements 

 were made with an ocular micrometer having a 

 magnification of 0.016 mm. per micrometer divi- 

 sion. Since the ova were not perfectly spherical, 

 the diameter which was measured was the random 

 diameter that fell parallel to the lines in the 

 counting chamber. 



At the outset of the study, 300 randomly 

 selected ova, 10 micrometer divisions (0.16 mm.) 

 or larger, were measured from each ovary in order 

 to characterize the size of ova at the different 

 developmental stages. Since this required a great 

 amount <>f time, and since the determinations of 

 the stage <>l development of the ovaries were 

 based on the most developed ova, the -ample size 



4SH 



U.S. FISH AND WILDLIFE SERVICE 



