FISHERY BULLETIN: VOL. 85, NO. 1 



frequent spawning is not consonant with the ap- 

 parent growth of oocytes in the 24 h after spawn- 

 ing and the near absence of fish with LMX <0.60 

 mm in the late afternoon samples. Some individuals 

 may, however, spawn more or less frequently than 

 every other day. In the early night samples, some 

 spawners carried larger oocytes than most non- 

 spawners and some of the latter carried smaller 

 oocytes than most of the former (Fig. 2). Thus a few 

 spawners appeared to be capable of ripening the 

 next batch in 24 h rather than 48 h, and the largest 

 oocytes of some nonspawners appeared unlikely to 

 be ready for spawning within 24 h. 



DISCUSSION 



Results of the present study indicate that the rate 

 of oocyte development in nehu is much faster than 

 in the northern anchovy, Engraulis mordax, or the 

 Peruvian anchovy, E. ringens, the only other species 

 for which comparable data are available. Hunter and 

 Goldberg (1980) showed that oocytes of £". mordax 

 which had spawned within 24 h averaged 0.46 mm 

 long and, during the peak spawning season, grew 

 to the size at which hydration begins in about 7 days. 

 Alheit et al. (1984) indicated that about 6 d are re- 

 quired in E. ringens. In nehu, oocytes in largest 

 mode just after spawning averaged 0.52 mm (mean 

 LMD of 54 spawners taken within 3 h after spawn- 

 ing); these appear to advance to hydration stage in 

 <48 h. Hunter and Goldberg's results also indicated 

 that about 7% of the oocytes in the largest mode 

 are not hydrated and spawned; whereas, in nehu it 

 appears that once a batch of ooc}i:es is separated 

 from smaller oocytes, oocytes in that batch are rare- 

 ly left behind and not spawned with majority of the 

 batch. 



Hydration, spawning, and degeneration of POF 

 after spawning are also more rapid in nehu than in 

 the Engraulis species. In E. mordax hydration 

 begins in the morning about 12 h before spawning 

 begins (Hunter and Macewicz 1980); Alheit et al.'s 

 (1984) data indicated that E. ringens is similar. Both 

 studies indicated that the Engraulis species spawn 

 over a broad period after sunset with peak spawn- 

 ing just before or near midnight. Nehu ova to be 

 spawned on a given night begin hydrating only a few 

 hours before spawning, and spawning occurs over 

 a rather brief period shortly after sunset. Whereas 

 POF are reliably identifiable up to 24 h after spawn- 

 ing in E. mordax and even longer in E. ringens 

 (Hunter and Goldberg 1980; Alheit et al. 1984), they 

 appear to degenerate to a similar point in about 16 

 h in nehu. 



My estimates of spawning timing and duration 

 conflict with those of Yamashita (1951) upon which 

 Tester (1955) apparently based his statements that 

 nehu spawn around midnight. As mentioned earlier, 

 studies in progress on appearance of newly spawned 

 eggs confirm the pattern indicated by presence of 

 females with hydrated ova in purse seine samples 

 after sunset. These studies further indicate that 

 Yamashita was probably not sampling deep enough 

 in the water column to collect newly spawned eggs 

 and that his "freshest" eggs were actually one or 

 more hours old. 



One of the broader implications of this study is 

 that, when dealing with tropical species, the time 

 scale of sampling must be on the order of hours 

 rather than weeks or days. The latter may be appro- 

 priate for investigation of species from higher lat- 

 itudes, but would miss many events or stages in the 

 reproductive cycle of nehu. Leary et al.'s (1975) con- 

 clusion that nehu spawn only once per lifetime was 

 in part based on the rarity of females with hydrated 

 ova in their samples. This was almost certainly due 

 to their not sampling during the short period be- 

 tween late afternoon and shortly after sunset when 

 hydrated ova are found in the current night's 

 spawners. Leary et al. stated that all females with 

 hydrated ova were captured between 2100 and 2300 

 h, i.e., well after the peak of spawning even in 

 summer. 



Both of the above studies of Engraulis species in- 

 dicate some degree of segregation of spawning 

 females at or near spawning time; spawners tended 

 to be overrepresented in such samples. Segregation 

 appears more extreme in nehu; the purse seine 

 samples taken just before and after spawning time 

 were almost all spawners. The greater percentage 

 of nonspawners in some purse seine samples taken 

 later after spawning and the nearly 1:1 ratio of 

 spawners to nonspawners in most day samples in- 

 dicate that spawning fish remix with others later 

 during the night and that segregation of the next 

 night's spawners does not occur until the mixed 

 schools leave shallow day areas at or near sunset. 

 The distribution of nonspawners early in the night 

 is not known. 



The winter-summer differences in nehu fecundity 

 were evident from both the comparison of relative 

 fecundity and the regressions of fecundity on either 

 length or weight. The G/S data for fish with ova 

 >0.75 mm also showed a higher mean and broader 

 range in summer. Other data (Clarke unpubl. data) 

 indicate that spawned nehu eggs are about 20% 

 heavier in winter, but this difference is insufficient 

 to compensate for higher fecundity in summer fish. 



134 



