CLARKE: FECUNDITY AND SPAWNING FREQUENCY OF NEHU 



The summer- winter ratio of mean relative fecundity 

 was 1.54; rouglily corrected for the egg weight dif- 

 ference, the ratio of mean effort per spawning would 

 be 1.28 (1.54/1.20), about the same as the ratio of 

 mean G/S, 1.31. There was no evidence that winter 

 fish compensated for lower effort per spawning with 

 higher frequency. 



The causes and adaptive value of the much greater 

 range and, on the average, higher effort by sum- 

 mer fish are not obvious. Similar differences have 

 been reported between different populations of 

 other species. For example, the northern population 

 of E. mordax appears to be more fecund than the 

 central population (Table 3). This difference is prob- 

 ably genetic and appears to reflect the shorter 

 spawning season (and lower number of batches) in 

 the northern population (Laroche and Richardson 

 1980). Since nehu live <6 mo (Struhsaker and Uchi- 

 yama 1976), it is difficult to postulate that the dif- 

 ferences between summer and winter fish are 

 genetic. It is, however, possible the winter fish may 

 spawn for longer periods and thus to some degree 

 compensate for lower effort per spawning. 



The winter-summer differences in nehu reproduc- 

 tive effort per batch may simply be physiological 

 consequences, perhaps with neutral or even nega- 

 tive adaptive value, which result from seasonal 

 differences in the environment. If output in nehu 

 is closely linked to recent feeding success (see 

 below), the output could be lower in winter fish if 



average daily ration were lower. There is, however, 

 no evidence of major seasonal differences in stand- 

 ing crop of the macrozooplankton upon which adult 

 nehu feed (Hirota and Szyper 1976). Also, nehu feed 

 almost exclusively at night (Clarke unpubl. obs.), and 

 actually have a longer feeding period per diel cycle 

 during the winter. Although the difference between 

 summer maxima and winter minima of temperature 

 in Kaneohe Bay is only about 5°C, it is possible that 

 metabolic processes overall, and consequently both 

 daily ration and reproductive output are slowed 

 enough in winter to account for the observed differ- 

 ence. 



Regardless of season, the relative fecundity data 

 combined with minimal estimates of spawner abun- 

 dance from purse seine catches predicts planktonic 

 egg densities 2 or 3 orders of magnitude higher than 

 those reported by egg surveys of Tester (1955) or 

 Watson and Leis (1974). Assuming all fish in a ca. 

 300 m- area were captured, catches of several 

 purse seine sets indicated 0.3-0.5 g dry weight of 

 spawning females/m^ and predicted egg densities 

 of 10^-10^/m". Studies in progress have shown that 

 such egg densities do in fact occur routinely, but that 

 most of the eggs are deeper than 5 m in the water 

 column. Thus the earlier egg surveys, which used 

 surface plankton tows, had missed over 90% of the 

 spawned eggs. 



Comparable fecundity data are available for only 

 a few other species of anchovies (Table 3), and most 



Table 3.— Fecundity-weight relationships for winter and summer nehu, Encrasicholina purpurea, and five other species of an- 

 chovies. Means and standard deviations of relative fecundity and power cur\'es for fecundity vs. weight were calculated from 

 available fecundity and weight data. Fish weight were ovary-free wet weights except for nehu, whose wet weights were estimated 

 from dry ovary-free weight data, and Engraulis ringens, for which the data were given as total fish wet weight. Power curves 

 are the antilog forms of equations based on Model II linear regressions of the natural logarithms; 95% confidence limits are 

 for the exponents. Relative fecundities of the smallest and largest female from each group were calculated from the extremes 

 of weight values and the appropriate power curve. 



135 



