ences in the daily mortality rates of embryonic 

 stages, rather than egg production, could have 

 profound influence on larval production at hatch- 

 ing. In other words, larval production at hatch- 

 ing may not necessarily be accurately related to 

 egg production without knowing mortality rates 

 of embryonic stages. Although crude estimation 

 of spawning biomass by conventional plankton 

 tows might be possible (Smith et al. 1970), a 

 radical departure from the egg sampling method 

 using information on vertical and horizontal dis- 

 tribution patterns of the eggs is required to de- 

 vise an egg production method for the saury. 



Lo (1985) calculated the time series of egg 

 production at fertilization of the northern an- 

 chovy in 1951-82. Year-to-year difference in the 

 egg production was more than 1,000-fold, which 

 is much larger than the difference in saury larval 

 production at hatching in our study, about 33- 

 fold. She also calculated mean yolk-sac larval 

 abundance for the years that showed a difference 

 of more than 1,000-fold. Annual fluctuation of 

 reproductive level in the saury seemed to be 

 smaller than the northern anchovy. This might 

 be related to the differences in spawning ecology 

 of these two species. The saury is reported to be 

 a multiple spawner that spawns in 2 mo intervals 

 in the spawning season (Hatanaka 1955) with a 

 batch fecundity of 500-3,000 egg/female (Hata- 

 naka 1953). The spawning frequency is 3-5 

 times/yr and the annual fecundity is estimated to 

 be 1,500-15,000. In contrast, time between 

 spawning incidence of the northern anchovy off 

 southern California is 6-8 days and the batch 

 fecundity is 389 eggs/g of ovary free female body 

 weight (Hunter and Goldberg 1980). The saury is 

 less fecund than the northern anchovy and seems 

 to be less variable in its annual fecundity owing 

 to a long maturation period of ovarian eggs. 

 Small year-to-year differences in the reproduc- 

 tive level in the saury might be the result of less 

 variable spawning effort. 



The correlation of the larval production at 

 hatching and the daily IMR in the individual 

 years may indicate that the mortality of young 

 saury is density dependent. Watanabe (1987) 

 showed an inverse correlation between the egg 

 abundance and the overall survival rate of larval 

 and juvenile Japanese sardine up to 1 yr old. He 

 further examined correlations between the egg 

 abundance and the biomass of larvae of the 40 

 mm size class, and between the biomasses of the 

 40 mm size class and of the 1 yr recruit size class. 

 He found that egg abundance and survival rate 

 up to 40 mm size class are inversely correlated. 



FISHERY BULLETIN: VOL. 87, NO. 3, 1989 



whereas the biomass of the 40 mm and 1 yr size 

 class are positively correlated. His conclusion 

 was that the mortality rate of larval sardine is 

 density dependent up to 40 mm. Thus, in some 

 pelagic fish, mortality rates of early life stages 

 could be density dependent. 



ACKNOWLEDGMENT 



We thank the late R. Lasker and P. E. Smith 

 of the Southwest Fisheries Center of National 

 Marine Fisheries Service, NOAA, for reading 

 the manuscript. Y. Watanabe would like to thank 

 the Science and Technology Agency of Japan for 

 funding his stay at the Southwest Fisheries 

 Center. 



LITERATURE CITED 



Dixon, W. J., M. B. Brown, L. Engelman, J. W. Frane, M. 

 A. Hill, R. I. Jennrich, and J. D. Toporek. 



1985. BMDP Statistical software. Univ. Calif. Press, 

 Berkeley. 

 Fisheries Agency. 



1985. Annual report on Japan's fisheries, fiscal 1985 

 (summary). Minist. Agric, For. Fish., Gov. Jpn., 

 Tokyo, 38 p. 

 Hatanaka, M. 



1955. Biological studies on the population of the saury, 

 Cololabis saira (BREVOORT). Part 1 Reproduction 

 and growth. Tohoku J. Agric. Res. 4:227-269. 

 Hatanaka, M., T. Watanabe, K. Sekino, M. Kosaka, and K. 

 Kimura. 

 1953. Studies on the reproduction of the saury, Colo- 

 labis saira (BREVOORT), of the Pacific coast of 

 Japan. Tohoku J. Agi-ic. Res. 2:293-302. 

 Hunter, J. R., and S. R. Goldberg. 



1980. Spawning incidence and batch fecundity in 

 northern anchovy, Engraulis nwrdax. Fish. Bull., 

 U.S. 77:641-652. 

 Japan Saury Fishery Association. 



1985. Sanma bouke-ami gyogyo kankei shiryo (Sta- 

 tistics of the saury stick-held dip net fishery). (In 

 Jpn.) Jpn. Saury Fish. Assoc. 23:l-.56. 

 Lasker, R. 



1985. An egg production method for estimating spawn- 

 ing biomass of pelagic fish: Application to the northern 

 anchovy, Engraulis nwrdax. U.S. Dep. Commer. , 

 NOAA Tech. Rep. NMFS 36, 99 p. 

 Lo. N. C. H. 



1985. Egg production of the central stock of northern 

 anchovy, Engraxlis mordax, 1951-82. Fish. Bull., 

 U.S. 83:137-150. 



1986. Modeling life-stage-specific instantaneous mor- 

 tality rates, an application to northern anchovy, 

 Engraulis niordax, eggs and larvae. Fish. Bull., 

 U.S. 84:395-408. 



Matsumiya, Y., and S. Tanaka. 



1978. Dynamics of the saury population in the Pacific 

 Ocean off northern Japan - III. Reproductive rela- 

 tions of large and medium sized fish. Bull. Jpn. Soc. 

 Sci. Fish. 44:451-455. 



612 



