FISHERY BULLETIN, VOL. 69, NO. 2 



or determined its significance for year-class 

 strength. 



The response of larvae to food deprivation in 

 the laboratory may provide badly needed evi- 

 dence of how susceptible they are to starvation 

 at sea. There have, however, been few attempts 

 to determine experimentally the effects of de- 

 layed initial feeding on the larvae of marine 

 fishes. Fabre-Domergue and Bietrix, the two 

 pioneers of marine fish culture who coined the 

 term "critical period," which Hjort later adopt- 

 ed, believed that feeding before the yolk supply 

 was exhausted was essential to assure larval sur- 

 vival in laboratory rearing attempts. They 

 stated that among larvae which received food 

 prior to yolk absorption, ". . .la phase que noiis 

 avians nornmee periode critique post-larvaire 

 n'existe jms" (Fabre-Domergue and Bietrix, 

 1898: 468) . These authors went on to say that 

 larvae which did not receive food early would 

 subsequently become weak and unable to capture 

 food and would exhibit considerable, if not total, 

 mortality (Fabre-Domergue and Bietrix, 1898). 

 The importance of early feeding for marine fish 

 larvae was not further investigated in the lab- 

 oratory until Blaxter and Hempel (1963) stud- 

 ied the effects of delayed initial feeding on the 

 behavior of larval herring, Chipea hareiu/us L. 

 By feeding larvae after successively longer times 

 without food, Blaxter and Hempel determined 

 the time beyond which the larvae failed to ex- 

 hibit feeding movements when supplied with 

 food, a time they called the "point of no return." 

 This point came 5 to 9 days after complete yolk 

 absorption, at temperatures of 12° to 8° C, much 

 later than the statements of Fabre-Domergue 

 and Bietrix would have led one to expect. Re- 

 cently, Lasker et al. (1970) observed the mor- 

 tality of larvae of the northern anchovy, En- 

 graulis mordax Girard. which had been fed at 

 progressively later times after hatching. At 

 temperatures of 1.5° to 22° C, larvae for which 

 initial feeding had been delayed until 2.5 days 

 after complete yolk absorption, showed the same 

 pattern of mortality as grouiis of starved con- 

 trols, while larvae receiving food 1.5 days after 

 yolk absorption exhibited good survival — a phe- 

 nomenon which these authors termed "irrever- 

 sible starvation." 



The purpose of the present study was to in- 

 vestigate in detail the changes which take place 

 in starving larvae and in larvae whose initial 

 feeding is delayed for various lengths of time, 

 and thus to bring more evidence to bear upon 

 the perennial questions of how susceptible lar- 

 val fishes are to food deprivation and whether 

 they do pass through a "critical period" at the 

 time of yolk absorption. This study also sought 

 to broaden the range of our knowledge of larval 

 fish ecology by utilizing a species belonging to 

 a group other than the Clupeiformes or Pleuro- 

 nectiformes, or which nearly all of our infor- 

 mation has hitherto been based. The fish chosen 

 for study was the grunion, Leuresthes tenuis 

 (Ayres), a member of the family Atherinidae. 

 Atherinids produce rather large demersal eggs 

 (Breder and Rosen, 1966), and the well-devel- 

 oped larvae which hatch provide an interesting 

 contrast with flatfish and clupeoid larvae. Spe- 

 cifically, these experiments were designed to de- 

 termine the effects of delayed initial feeding on 

 mortality, on growth, and on the ability of 

 grunion larvae to begin feeding and to utilize 

 ingested food, and to ascertain what changes 

 in the morphology and chemical composition of 

 the larval body occur during starvation. 



MATERIALS AND METHODS 



SOURCE OF EGGS; HATCHING 



The grunion is best known for its unusual 

 habit of spawning on the sandy beaches of south- 

 ern California and northern Baja California 

 (Thompson and Thompson, 1919; Walker, 1952; 

 Breder and Rosen, 1966). The eggs are depos- 

 ited in the sand at night at certain times in the 

 tidal cycle and are washed free some days later 

 by a succeeding high tide, at which time the lar- 

 vae hatch if the develojimental period has been 

 sufficiently long. The spawning season extends 

 from late February or early March to late Au- 

 gust or early September, with spawning inten- 

 sity reaching a peak in April and May (Walker, 

 1952). During this time it is relatively easy 

 to collect grunion eggs, which therefore provide 



412 



