MATURITY AND SPAWNING OF ALBACORE 



291 



DISTRIBUTION OF MATURE EGGS WITHIN AN 

 OVARY 



Several tests were conducted to evaluate 

 sampling procedures. First, it was thought neces- 

 sary to determine whether advanced eggs were 

 evenly distributed within an ovary. For this 

 test, thin cross sections were taken from the 

 anterior, middle, and posterior parts of the right 

 member of one of the most advanced ovaries in the 

 collection (fig. 4a). The cross sections were 



Table 2. — Frequency distributions of diameters of the most 

 advanced group of eggs from the various parts of an ovary 

 (albacore No. 42 — Hawaii) 



CROSS SECTIONS 

 TAKEN AS SHOWN 



PIE- SECTION OBTAINED 

 FROM EACH OF 3 SECTIONS 



SUBSAMPLES TAKEN 

 FROM EACH PIE -SECTION 



Figure 4. — Schematic representation of method used to 

 obtain test samples for determining most efficient sam- 

 pling procedure. 



labeled A, B, and C, respectively. A triangular 

 section was then obtained from each of the three 

 cross sections (fig. 4b). Each triangular section 

 was further subdivided into three subsamples 

 representing, respectively, the center, mid-region, 

 and periphery of the ovary (fig. 4c). Thus, for 

 section A there were positions A u A, and A 3 ; for 

 section B, positions B u B 2 , and B 3 ; and for section 

 C, positions C\, C 2 , and C 3 . These samples were 

 weighed to the nearest 0.001 gram. All of the 

 eggs in the most mature group were then measured. 

 The weights and measurements are given in table 2. 



An analysis of variance (table 3) indicated a 

 significant difference (P<0.05) between positions 

 in the mean egg diameters but none between 

 sections. The diameter means. for sections A, B, 

 and C were 0.842 mm., 0.848 mm., and 0.842 mm., 

 respectively. The results showed that for sec- 

 tions A and B, the mean egg diameter was largest 

 at the periphery of the ovary. The situation was 

 reversed for section C where the mean diameter 

 was least at the periphery and greatest near the 

 central lumen (table 2). It has not been deter- 

 mined whether the differences between positions 

 are real or an artifact of preservation. 



The number of eggs in the most advanced group 

 to be expected from 0.1 gram of each sample was 



then calculated from table 2 and tested by 

 analysis of variance (table 4) . The results showed 

 that there was no significant difference (P>0.05) 

 in the relative numbers per unit weight of ovary 

 between sections A, B, and C. 



A comparison was made of the egg-diameter 

 frequencies of the right and left ovaries (albacore 

 No. 74, Hawaii). A random sample was obtained 

 from the central section of each ovary and all 

 eggs larger than 10-micrometer units (0.167 mm.) 

 were measured (table 5). The comparison re- 

 sulted in a chi-square of 19.77 (P=0.9, 30 d. f. 1 ) 

 indicating a general similarity in the distributions 

 between the two ovaries. 



The results of the various tests suggested that 

 representative samples of an ovary could be 

 obtained anywhere along its length, provided a 



Table 3. — Analysis of variance of diameters of the most 

 advanced eggs from different parts of the ovary (albacore 

 No. 42, Hawaii) 



•Indicate significant (p<0.05) mean square value. 



1 The last four classes in the frequency distributions were combined because 

 of the lew numbers. 



485320 O - 59 



