346 



Fishery Bulletin 103(2) 



Males 



stage 1 immature ("virgin" in other studies); 



stage la immature, developing; 



stage 2 mature, resting; 



stage 3 reproductively developed, ripe; 



stage 4 spawning. 



The immature, developing stage identified females that 

 were immature and unlikely to spawn but had ovaries 

 containing cortical alveoli stage oocytes (which other- 

 wise identified mature, developing females). 



Division of the microscopic staging system for ovaries 

 into three spawning stages was based on the pres- 

 ence of migratory nucleus stage or hydrated oocytes 

 within the ovarian lamellae (stage 5a), the presence of 

 hydrated oocytes within the ovarian lumen (stage 5b), 

 and the presence of postovulatory follicles (POFs) in 

 the lamellae (stage 5c). In tropical fish species POFs 

 may remain up to 24 hours in the ovaries before being 

 resorbed (West, 1990), and there is evidence suggesting 

 this is the case for S. commerson in Queensland waters 

 (McPherson, 1993). In the present study POFs observed 

 in the ovaries of females were categorized as either 

 "new" or "old" based on their degree of degeneration 

 (Mackie and Lewis 2 ). 



Gonadosomatic indices (GSIs) were calculated by us- 

 ing ratios of gonad weight to whole body weight, head 

 weight, and head length. The latter two ratios were 

 used to assess the usefulness of head and viscera sam- 

 ples in future monitoring of S. commerson. 



Scomberomorus commerson is a serial-spawning spe- 

 cies (Munro, 1942). Estimates of batch fecundity were 

 made for preserved prespawning (stage 5a) ovaries from 

 counts of hydrated oocytes within three samples taken 

 from the anterior, middle, and posterior region of one 

 lobe (each 130-200 mg). A section of each ovary was 

 also processed by using histological methods to confirm 

 suitability for estimation of fecundity. Some ovaries 

 were subsequently rejected for fecundity estimates be- 

 cause the most mature batch of oocytes had not fully 

 hydrated and were less easy to distinguish from earlier 

 stage oocytes. These ovaries tended to provide an over- 

 estimate of batch fecundity (Mackie et al. 3 ). 



The daily timing and frequency of spawning were 

 determined for females captured in the Kimberley re- 

 gion during September 1999 when 94% of ovaries were 

 retained for histological analysis (« = 344). Spawning 

 frequency was determined as the inverse of the spawn- 

 ing fraction (the number of ovaries with hydrated or 

 migratory nucleus stage (MNS) oocytes divided by the 

 total number of mature ovaries in the catch). These 

 data were compared with estimates made by using the 

 number of ovaries macroscopically identified as having 



3 Mackie, M. C, D. J., Gaughan, and R. C. Buckworth. 

 2003. Stock assessment of narrow-barred Spanish mackerel 

 iScomberomorus commerson) in Western Australia. Final 

 report to the Fisheries Research and Development Corpora- 

 tion (FRDC) on project no. 1999/151, 242 p. Department 

 of Fisheries, Perth, Western Australia, 6020. 



hydrated oocytes. Analyses of sex ratios were based on 

 data where the whole catch or a known random sample 

 of the catch was processed. 



Results 



The gonads of 5128 male, female, and juvenile S. com- 

 merson were macroscopically staged during this study. 

 Of these, 1624 were also processed with histological 

 techniques for more detailed analyses. 



Biological analyses 



Body lengths ranged from 58 to 1720 mm FL (62 to 1840 

 mm TL), and whole weights ranged from 0.0015 to 40.6 

 kg. Regression analyses incorporating step-wise reduc- 

 tion (using analysis of variance) of a fully parameterized 

 model indicated that differences in length and weight 

 relationships between regions and sex were minor com- 

 pared to measurement error. Thus, the simplest models 

 which adequately explain the pooled data were 



Whole weight (kg) = 3.40e - 9 x FL (mm) 3 12 (re=2842) 

 (SE of constants: a=2.78e-10, 6 = 0.01) 



TL 'mm) = 42.74 + (1.06 x FL (mm)) 



(« = 1679, r 2 =0.996). 



Overall sex ratios were biased towards females, with 

 the M:F ratio varying between 1:1.2 and 1:1.6 in the 

 three regions. However, there was considerable varia- 

 tion between samples, from a peak M:F ratio of 1:2.6 

 for samples obtained in the nonreproductive period, to 

 a male bias of 1.1:1 in pooled samples obtained during 

 the peak spawning period. This slight male bias during 

 the spawning period occurred in successive years; the 

 sex bias, however, was variable between daily samples. 

 Sex ratios also changed over from a male to female bias 

 with increasing size class, with a 1:1 ratio occurring at 

 about 1000-1050 mm FL. 



Ovarian weight ranged from 2.00 to 1908.30 g and 

 testes from 0.84 to 840.10 g. Gonads of juvenile S. com- 

 merson were small and contained no recognizable germ 

 tissue. The smallest fish with differentiated gonads was 

 a 301-mm-FL male. The smallest female was 396 mm 

 FL. Two abnormally large juveniles (1170 and 1251 

 mm FL) were captured whose gonads had remained 

 unusually small and undifferentiated. Body lengths of 

 immature females (largest=1195 mm FL [13.8 kg WW]) 

 overlapped substantially with those of mature females 

 (smallest=641 mm FL [2.3 kg WW]). 



Estimates of the size at which 50% of females were 

 mature were calculated by using all available data as 

 well as data taken only during the reproductive season 

 (October to April). Data for each area were pooled to 

 provide sufficient samples (virtually all samples of im- 

 mature fish were obtained from the Pilbara region). 

 Both data sets provided similar estimates; 809 mm 

 FL, ±9.8 SE (898 mm TL) for all data, and 788 mm 



