Bisbal and Bengtson: Starvation in early life stages of Paralichthys dentatus 



221 



length as an estimate of age is a coarse alternative 

 when age data are not available. If this is the case, 

 then the analysis should be restricted to a limited 

 size range (Martin and Wright, 1987). 



Among the biochemical criteria, protein data had 

 the largest associated variability. Similar variation 

 in the protein content of winter flounder, Pleuronectes 

 americanus, larvae has been obtained by Cetta and 

 Capuzzo ( 1982 ). Other studies have shown that pro- 

 tein breakdown is the major source of energy during 

 starvation of herring (Ehrlich, 1974a I and plaice 

 (Ehrlich, 1974b), at least during early larval stages, 

 when lipid reserves are negligible or nonexistent. 



The RNA:DNA ratio showed less individual vari- 

 ability and provided a more sensitive index to feed- 

 ing condition than did protein. The ratio of total RNA 

 to DNA in tissues has been extensively used as an 

 indicator of recent growth rate and changes in feed- 

 ing levels of various larval fish ( Buckley, 1984; Bulow, 

 1987). In recent years, the relative ease and sensi- 

 tivity of this analysis have stimulated the develop- 

 ment of several procedural variations of the tech- 

 nique. Thus, discretion should be exercised in directly 

 comparing RNA:DNA values obtained with different 

 methods and standards (Caldarone and Buckley, 

 1991). In addition, it has been demonstrated that 

 temperature can affect the RNA:DNA ratio in fish 

 larvae (Buckley, 1982, 1984; Buckley and Lough, 

 1987). In the 6-day-old larvae used in our study, the 

 RNA:DNA ratio declined by about 30% over the 60- 

 hour experiment, even in fed larvae. After that de- 

 cline, which was similar in magnitude to that ob- 

 served in fed winter flounder larvae 4 days after yolk 

 absorption (Buckley, 1980), the mean RNA:DNA ra- 

 tio of fed larvae remained within a narrow range (2.7 

 to 3.1) for the remainder of the larval period. There- 

 fore, it appears that a mean RNA:DNA ratio of less 

 than 2.7 strongly suggests food limitation in floun- 

 der. The equilibrium RNA:DNA ratio for P. dentatus 

 larvae reared at 14, 16, or 18°C has been reported to 

 be 2.4, 3.1, and 2.6, respectively (Buckley, 1984). Win- 

 ter flounder and striped bass, Morone saxatilis, lar- 

 vae also appear to establish narrow RNA:DNA equi- 

 librium ranges (Buckley, 1980; Wright and Martin, 

 1985). After metamorphosis, the RNA:DNA ratio of 

 summer flounder increased to between 8.2 and 8.9, 

 whereas that of starved fish was never above 6. A 

 similar increase in RNA:DNA ratio after metamor- 

 phosis has been observed in fed winter flounder 

 (Buckley, 1980). 



Although RNArDNA ratio and pectoral angle were 

 both able to discriminate fed from starved summer 

 flounder, pectoral angle was more sensitive to star- 

 vation than was the RNA:DNA ratio in larvae, 

 whereas the opposite was true for juveniles. The 



quick response of RNA:DNA ratio to food depriva- 

 tion noted by Buckley (1980), Wright and Martin 

 (1985), and Martin and Wright (1987) was not ap- 

 parent in summer flounder. An advantage of bio- 

 chemical methods for field use is that larvae dam- 

 aged by sampling gear can still be analyzed (Fraser 

 et al., 1987) and distortions due to chemical fixatives 

 are avoided. We conclude, therefore, that RNA:DNA 

 ratios may be useful as indicators of nutritional limi- 

 tation in summer flounder larvae and juveniles. 



Histological analyses indicated that food depriva- 

 tion of summer flounder larvae and early juveniles 

 had a marked effect on several internal structures. 

 Starvation was readily manifest in the intestine, fol- 

 lowed in time by changes in the pancreas, liver, and 

 skeletal musculature, as previously seen in other 

 teleost larvae (Umeda and Ochiai, 1975; Ehrlich et 

 al., 1976; O'Connell, 1976, 1980; Theilacker, 1978, 

 1986; Cousin et al., 1986; Margulies, 1993). The nu- 

 trient shortages that result from food deprivation 

 have an almost immediate manifestation in the in- 

 testinal epithelium. In starved summer flounder, 

 lipid and protein inclusions progressively disap- 

 peared from the intestinal epithelial cells until they 

 were no longer visible, similar to the previous obser- 

 vations of Ehrlich (1974a), Ciullo (1975), Watanabe 

 ( 1985), and Govoni et al. ( 1986). By contrast, Kjorsvik 

 et al. (1991) reported that pinocytic inclusions were 

 visible at all stages of starvation in cod larvae. 



Mucosal cell height in summer flounder was ex- 

 tremely sensitive to starvation when applied to the 

 posterior intestine, whereas the height of the ante- 

 rior intestinal mucosa varied with increasing size or 

 age, or both. The mean height of the posterior intes- 

 tinal mucosa showed a stable boundary for discrimi- 

 nation of fed and starved individuals (above 10 (im 

 for fed larvae, below for starved) regardless of indi- 

 vidual size or age. This criterion therefore provides 

 the best tool to assess starvation in summer floun- 

 der during the first 60 days of life. Previous investi- 

 gators have noted the utility of histological exami- 

 nation of intestinal mucosa, especially cell height, 

 for determination of starvation (Ehrlich et al., 1976; 

 Theilacker, 1978, 1980; Watanabe, 1985; Umeda et 

 al., 1986; Theilacker and Watanabe, 1989; Kj0rsvik 

 et al., 1991). The discriminating power of the mu- 

 cosal cell height criterion incorporates the well known 

 advantages of other traditional histological evalua- 

 tion procedures. As with the biochemical criteria, 

 specific equipment and some technical proficiency are 

 required to process the samples. One advantage to 

 this criterion is that samples can be preserved on a 

 ship and no subsequent calibration is necessary for 

 shrinkage due to capture or fixation, or for individual 

 size or age. 



