FISHERY BULLETIN: VOL. 76, NO. 2 



val when feeding is only delayed for 1 or 2 days and 

 to ascertain whether tissue degeneration noted 

 during the period of early starvation is reversible. 



DISCUSSION AND CONCLUSION 



The effects of starvation could be seen through- 

 out the body of larval jack mackerel. There was 

 atrophy of all tissues, with the digestive tract and 

 its associated glands the first tissues affected by 

 starvation. The extent of cellular deterioration in- 

 creased with time of starvation. Many histological 

 changes, which were associated with starvation in 

 jack mackerel, were similar to changes described 

 in other starving fish larvae: 1) atrophy and cellu- 

 lar and nuclear degeneration of the liver (yellow- 

 tail, Umeda and Ochiai 1975; northern anchovy, 

 O'Connell 1976; plaice, Karl Ehrlich pers. com- 

 mun.), 2) separation of muscle fibers (northern 

 anchovy, O'Connell 1976; herring and plaice, 

 Ehrlich pers. commun.), 3) cellular and nuclear 

 degeneration in the pancreas (yellowtail, Umeda 

 and Ochiai 1975; northern anchovy, O'Connell 

 1976), and 4) decrease in the size of the epithelial 

 cells in the digestive tract (pike and carp, Kos- 

 tomarova 1962; yellowtail, Umeda and Ochiai 

 1975; herring and plaice, Ehrlich et al. 1976; 

 northern anchovy, O'Connell 1976). Some his- 

 tological changes present in starved jack mackerel 

 larvae, enlargement of the gallbladder and de- 

 terioration of the primitive brain cells, have not 

 been reported for other starved fish larvae. 



The onset of starvation in jack mackerel larvae 

 was manifested as 1) a change in acinar arrange- 

 ment of the pancreatic cells and 2) a sloughing of 

 mucosal cells from the midgut into the lumen. 

 These two criteria were shown to be critical vari- 

 ables and were histologically graded to assess the 

 nutritional condition of jack mackerel larvae. 

 0'Connell(1976) reported that the condition of the 

 pancreas was of primary importance for classify- 

 ing starving northern anchovy larvae. His criteria 

 for good pancreatic condition depended on the 

 abundance of zymogen as well as continuity of 

 cellular structure. The presence of zymogen was 

 not used as a criterion to classify jack mackerel 

 larvae; absence of zymogen was usually associated 

 with feeding, however, the correlation was not 

 consistent. O'Connell also noted mucosal cells in 

 the midgut of larval anchovy but he did not grade 

 this feature. Love (1970) suggested that sloughed 

 midgut cells may be used as an energy source. 



During starvation, growth was retarded, larvae 

 shrank, and the soft tissues collapsed causing 

 larvae to look abnormal; the body became bent and 

 thin and the head disproportionately large. An 

 array of larval fish morphometric characteristics 

 have been associated with nutritional condition. A 

 decrease in depth of the larval body has been con- 

 sidered to be an important indicator of starvation 

 although Blaxter (1971) did not find it to be a 

 sensitive character for elongate clupeoid larvae. 

 However, body depth divided by standard length 

 appears to reflect the condition of postlarval an- 

 chovy (Nakai et al. 1969) and older, larval north- 

 ern anchovy (Arthur 1976), but the ratio did not 

 identify feeding and starving post yolk-sac north- 

 ern anchovy (O'Connell 1976). Nakai et al. ( 1969) 

 also related the diameter of the posterior digestive 

 tract to condition in postlarval anchovy. The 

 height of the body in the posterior part of the trunk 

 was measured by Honjo et al. ( 1959) and was found 

 to differ between postlarval anchovy collected 

 from two fishing grounds. Honjo et al. (1959) 

 suggested that the anatomical difference was in- 

 dicative of a higher availability of food organisms 

 in one area than in the other. Another predictive 

 parameter for the condition of post yolk-sac larvae 

 was the relative condition factor reported by 

 Ehrlich et al. (1976) for herring and plaice. Its 

 estimation involves weighing each larva. All of 

 the above mentioned morphological criteria 

 reflect larval nutritional condition and may be 

 considered reference parameters, but no one 

 criterion accurately identifies larval condition on 

 an individual basis. The lack of sensitivity of a 

 single morphometric indicator may be due to 

 natural variability alone or more probably to the 

 rapid morphological changes which occur in larval 

 fish. For example, ZweifeF using the data of 

 Hunter ( 1976b) has shown that for feeding north- 

 ern anchovy larvae the condition factor weight/ 

 length^ was neither constant nor monotonic even 

 when food densities were carefully controlled. For 

 variation in food levels ( including no food) it would 

 be expected that there should be even more com- 

 plex relationships. 



Even though the histological and morphological 

 techniques have proven to be effective for predict- 

 ing condition of laboratory-reared animals, both 

 techniques must be field tested with wild-caught 



^Zweifel, J. 1977. A non-linear model for allometry in larval 

 fish. Unpubl. manuscr., 25 p. Southwest Fisheries Center La 

 Jolla Laboratory, National Marine Fisheries Service, NOAA, 

 P.O. Box 271, La Jolla, CA 92038. 



412 



