FISHERY BULLETIN: VOL. 78, NO. 2 



population abundance of larval anchovies by 

 length and region for the 1967 through 1975 

 California Cooperative Oceanic Fisheries Investi- 

 gations net tow data, and length dependent mor- 

 tality rates were calculated from the abundance 

 estimates. For larvae of 7.5 mm SL, which approx- 

 imates the median length of those showing 

 symptoms of starvation in the present study, the 

 estimated average daily mortality rate in the San 

 Pedro Channel area was 21%. If it is assumed that 

 all larvae showing symptoms will die directly or 

 indirectly from starvation, the observed 8% with 

 symptoms in the March 1977 survey could indicate 

 a net daily mortality from starvation of 8%, which 

 is 409c of the average total daily mortality for this 

 length group. If starvation tends to contribute this 

 substantially to total mortality, variations in the 

 proportion of larvae observed to be starving may 

 relate reasonably well to the magnitude of ongoing 

 total mortality and consequently to recruitment 

 from the year class. 



How well the proportion of starving larvae from 

 a given sampling in 1 yr will predict the eventual 

 recruitment of that year class will only be evident 

 from correlation of the two variables for at least a 

 few years. As for 1977, with a northern anchovy 

 "starvation ratio" of 8%, there were indications 

 that recruitment would be good. The winter and 

 early spring were relatively mild, a condition con- 

 ducive to development of high density patches of 

 larval food organisms, particularly from di- 

 noflagellate blooms (Lasker in press). Growth rate 

 of northern anchovy larvae was also shown to be 

 above average in the San Pedro Channel area for 

 March 1977 (Methot and Kramer 1979). The above 

 average growth rate may have been valid for much 

 of the population developing in the region without 

 applying to the patches of emaciated larvae, which 

 were taken at different locations than the grovd;h 

 samples. Estimates from recent catch data indi- 

 cate that the 1977 year class is of moderate size, as 

 compared with the large 1976 and 1978 year clas- 

 ses and the small 1974 and 1975 year classes (J. S. 

 Sunada^). Thus, to the extent that 8% starving 

 larvae is a reliable estimate of that parameter for 

 1977 and to the extent that the parameter is as- 

 sociated with recruitment, both higher and lower 

 occurrences of starving larvae are likely pos- 

 sibilities from future surveys. 



''J. S. Sunada, Assistant Biologist, Marine Resources Region, 

 California Department of Fish and Game, 350 Golden Shore, 

 Long Beach, CA 90802, pers. commun. May 1979. 



Reliability of the estimate of starving larvae is 

 probably reasonably good for 1977 in that tows 

 were most concentrated in a region of high abun- 

 dance, but reliability in future sampling efforts 

 could probably be improved by more diligent 

 stratification in respect to population distribution. 

 In addition to sampling at more than one point in 

 time, an effective strategy might be to expand 

 sampling in several areas that show high abun- 

 dance, particularly of larvae under 10 mm SL, 

 along a preplanned survey track. Expanding sam- 

 pling in this way would likely result in a quantity 

 of samples that would be formidable if analysis is 

 entirely dependent on histological or physiological 

 parameters. The distinctive appearance of the 

 aggregated larvae from those few tows of the pres- 

 ent survey that contained predominantly 

 emaciated larvae, however, suggests that histologi- 

 cal analysis can be greatly reduced. 



Assuming that starvation of any consequence 

 will occur in patches, a stereomicroscope scan of 

 the total aggregation of larvae from each tow 

 should suffice for the identification and enumera- 

 tion of such patches, with histological processing 

 reserved for selected verification. Undoubtedly 

 some of the starving larvae that occur as scattered 

 single cases would be missed under such a proce- 

 dure, but this should have little effect on the over- 

 all estimate. 



ACKNOWLEDGMENT 



I would like to acknowledge the very able assis- 

 tance of Pedro Paloma, who sorted and measured 

 all the samples and prepared all the slides. 



LITERATURE CITED 



Blaxter, J. H. S. 



1971. Feeding and condition of Clyde herring lar- 

 vae. Rapp. P.-V. Reun. Cons. Int. Explor. Mer 160:128- 

 136. 



Blaxter, J. H. S., and G. Hemple. 



1963. Influence of egg size on herring larvae (Clupea 

 harengus L.). J. Cons. 28:211-240. 

 CIULLO, R. H. 



1975. Intestinal histology of Fundulus heteroclitus with 

 observations on the effects of starvation. In W. E. Ribe- 

 lin and G. Migaki (editors). The pathology of fishes, p. 

 733-767. Univ. Wis. Press, Madison. 



EHRLICH, K. F. 



1974. Chemical changes during growth and starvation of 

 herring larvae. In J. H. S. Blaxter (editor). The early life 

 history offish, p. 301-323. Springer- Verlag, Berl. 



EHRLICH, K. F., J. H. S. BLAXTER, AND R. PEMBERTON. 



1976. Morphological and histological changes during the 



488 



