FISHERY BULLETIN: VOL. 69. NO. 



an increase in conversion efficiency. A similar 

 effect has been found in aduft fish (Ivlev, 1939; 

 Pandian. 1967) . There is some indication in the 

 results of the quantitative feeding experiments 

 that larvae convert food more efficiently after 

 7 days without food than after 7 days of feeding 

 (Table 5) , but the data are too meager to justify 

 any conclusion on this point. 



Omori (1970) showed that copepods from 

 areas poor in food tended to have lower C,'N 

 ratios than copepods from rich areas. In lar- 

 val fishes, condition factors have been used in 

 attempts to assess nutritional state (Hempel 

 and Blaxter, 1963; Blaxter, 1965). Both meas- 

 ures were compared in the present study (Fig- 

 ures 3 and 4). Although starved larvae had 

 lower C N ratios than fed larvae after day 4, 

 due presumably to catabolism of fat, the C/N 

 ratios of growing, fed lai'vae decreased with age 

 as a consequence of the rapid elaboration of 

 protein, while their condition factors increased. 

 Reflecting this same tendency, larvae 20 days 

 old had higher condition factors but lower C/N 

 ratios, the longer they had been feeding. Con- 

 dition factor seems to be somewhat more con- 

 sistent and reliable as an index of the nutritional 

 state of larval grunion than C/N ratio. 



In sum, larval grunion appear to be extremely 

 resistant to food deprivation. Under laboratoi\v 

 conditions it takes 3 weeks for all larvae to die 

 of starvation at 18° C (Figure 1). No matter 

 how long initial feeding is delayed, over 40'"r 

 of the larvae alive when food is offered will sur- 

 vive, and all larvae which survive 16 days with- 

 out food can commence feeding at this time and 

 survive (Table 1). Since grunion larvae hatch 

 from eggs deposited in the beaches of southern 

 California and northern Baja California and 

 must inhabit inshore waters almost exclusively, 

 and since the abundance of microplankton is ex- 

 tremely high in inshore as com|iared with off- 

 shore waters in this region ( Beers and Stewart, 

 1967), it seems unlikely that these larvae ever 

 experience high rates of mortality due to star- 

 vation. Major sources of mortality among 

 grunion larvae must be sought, rather, in pre- 

 dation and jihysical damage from waves. Tidal 

 variations may result in different incubation 



periods in grunion eggs from different spawn- 

 ings (Walker, 1952), but the effect of this on 

 larval viability has yet to be determined. 



These findings differ from results for clupeoid 

 larvae. In the northern anchovy a delay in in- 

 itial feeding of 2.5 days after yolk absorption 

 resulted in nearly complete mortality, even 

 though many larvae were alive when food was 

 administered (Lasker et al., 1970). This "point 

 of irreversible starvation" appears not to exist 

 for larval grunion, as starvation can in fact be 

 reversed at any point along the survival curve 

 of starved larvae (Figure 1). 



Larvae of the herring (Clupea harengus) 

 show a decrease in the percentage of larvae 

 which commence feeding as the jieriod of food 

 deprivation is lengthened, and the point at which 

 the percentage feeding is half that at the start 

 of the experiment has been termed the "point 

 of no return" (Blaxter and Hempel, 1963; Blax- 

 ter, 1965). Again, the grunion larvae show a 

 different pattern, with at least 80 ''r of the larvae 

 commencing feeding when food is offered after 

 periods of starvation ranging from 7 to 16 days 

 (Table 4). Some larvae which did commence 

 feeding after 7, 10, and 13 days without food 

 were nevertheless unable to survive and died 

 after gorging themselves with Artemia nauplii. 

 The interesting fact that all of the larvae alive 

 after 16 days of starvation commenced feeding 

 and survived, while the percentage feeding was 

 lower in larvae starved for shorter periods of 

 time, may be explained as a result of mortality 

 among the weakest larvae, so that by day 16 

 only the most hardy individuals were still alive. 



Thus, certain types of larvae would be more 

 likely than others to show a "critical period" 

 pattern of mortality at sea under conditions of 

 low food availability. If northern anchovy lar- 

 vae were not to encounter food within 2.5 days 

 after yolk absorption, there would ensue a ca- 

 tastrophic mortality concentrated in time (Las- 

 ker et al., 1970). In contrast, grunion larvae, 

 which hatch in a more well-developed and robust 

 state, would exhibit mortality extending over a 

 number of days if deprived of food and hence 

 would not show a "critical period" in the classi- 

 cal sense of Hjort. Obviously a sudden increase 



422 



