528 



Fishery Bulletin 9I|3). 1993 



nia if CN oocytes were not present. Perinucleolar oo- 

 cytes were not included because this stage was not 

 present in all immature females. 



To develop the criterion, histological sections of ova- 

 ries from randomly-selected resting (n=21; 319- 

 449mm) and immature (n=32; 252-410 mm) females 

 were examined and the area of one lobe was measured 

 with an image analysis system. Five specimens were 

 selected, if available, per 25-mm size interval for each 

 of the two reproductive states. Atresia, which is evi- 

 dence of previous oocyte development, was present in 

 each resting female, but it was not present in each 

 immature female. The oocytes and/or gonad area in 

 resting females were noticeably larger than those in 

 immature females; additionally, body length (<ca. 

 290 mm) indicated that some specimens were imma- 

 ture. For each specimen, a histological section and one 

 of two axis orientations on the section were also ran- 

 domly selected. The minimum and maximum dimen- 

 sions of each of the five largest CN oocytes along the 

 selected axis were measured and an average diameter 

 was calculated for each oocyte. The largest CN oocytes 

 were selected because previous studies have shown 

 that the size of the largest oocyte can be used to as- 

 sess ovarian development (see MacGregor, 1957; 

 Greeley et al., 1987). We measured five oocytes in- 

 stead of one oocyte to more accurately assess the size 

 of the largest oocytes. 



The accuracy of the criterion was assessed by com- 

 paring the length-frequency distributions of resting fe- 

 males and females with evidence of certain maturity 

 (e.g., developing, ripe, or spent state). Because oocyte 

 diameter and gonad area data were not normally dis- 

 tributed, the Mann-Whitney test was used in statisti- 

 cal analyses. 



Results 



Otoliths from 1,039 Spanish mackerel were examined, 

 of which 132 were unreadable (i.e., considered aber- 

 rant by one or both readers). Readers agreed on ages 

 for 96.5% of readable otoliths. Except for the small, 

 secondary peak caused by the November sample of 

 only seven individuals, the distribution of monthly per- 

 centages of otoliths with opaque margins was unimodal. 

 Annual ring formation peaking in May-July is indi- 

 cated I Fig. 1). Linear regressions of FL on otolith ra- 

 dius (OR) were male FL = 113.8 + 8.3 OR, n=228, 

 r 2 =0.55; female FL = 36.2 + 11.6 OR, n=502, r 2 =0.71; 

 combined FL = 40.1 + 11.2 OR, rc=748, r 2 =0.66. Fe- 

 males grew more rapidly and lived longer than males 

 (Table 2). The oldest males were age 6 (n=3) and the 

 oldest females, age 11 (n=2). The von Bertalanffy 

 growth equation parameters (L„, k, and t lh respectively) 



1 3 3 



Figure 1 



Monthly percentages of Spanish mackerel with marginal in- 

 crements of zero (= opaque margins!, with number of zero 

 marginal increments over number in sample for each month. 



were 538 mm, 0.31, and -2.31 for males; 723 mm, 0.24, 

 and -1.80 for females; and, 760 mm, 0.18, and -2.44 

 for sexes combined. The von Bertalanffy estimates of 

 L„ were reasonable compared with maximum observed 

 lengths. One male (600 mm) and two females (760 and 

 730 mm) exceeded the asymptotic lengths. 



All female specimens were used in the assessment 

 of maturity because resting and immature females 

 could be distinguished based on the size of the largest 

 CN oocytes or oogonia if atresia was not present. The 

 largest CN oocytes or oogonia in resting ovaries (with 

 atresia) were significantly larger than those in imma- 

 ture ovaries (Fig. 2; [7=633, P<0.001, df=l). The aver- 

 age diameter of these oocytes was usually >70 urn in 

 resting ovaries. It was not necessary to use the CN 

 oocyte/oogonium criterion often because >90% of the 

 resting ovaries had beta or delta stage, and with less 

 frequency gamma stage, atretic follicles. Gonad area 

 was also larger in resting ovaries compared to imma- 

 ture ovaries ([7=43, P<0.001, df=l; immature= 3.78+ 

 3.24mm 2 per lobe; resting=9.93+3.33 mm 2 per lobe); 

 however, oocyte characteristics should be used to as- 

 sess reproductive state. 



Use of the CN oocyte/oogonium criterion to assess 

 maturity was not free of error, as the CN oocyte/oogo- 

 nium size distributions for resting and immature fe- 

 males overlapped at 60-80 urn (Fig. 2). This error was 

 negligible because the length-frequency distributions 

 of resting females and females with evidence of cer- 

 tain maturity (e.g., developing, ripe, or spent state) 

 overlapped, and the smallest individuals of both groups 

 were in the same size interval (Fig. 3). 



Female Spanish mackerel become sexually mature 

 later and at a larger size than males. Mature gonads 



