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



219 



dorf vials in a — 80°C freezer for no more than 45 days 

 until RNA, DNA, and protein determinations were 

 performed. Owing to the extremely small size of the 

 6-day-old larvae, each determination was performed 

 on samples consisting of two larvae pooled in the 

 same vial. This was the only case where pooling was 

 necessary. 



Determinations of RNA and DNA were performed 

 according to the methodology described by Bentle et 

 al. (1981) as modified for individual larvae of small 

 size (Nacci et al., 1992). Total protein determination 

 was assessed by a dye-binding assay ( Bradford, 1976) 

 in which bovine serum albumin was the reference 

 standard. Volumes were adjusted to 96-well micro- 

 titration plates and, after completion of the colored 

 reaction, absorbances were read at 600 nm in an EL 

 312 Bio-Tek automated microplate reader. 



The fraction of the sample destined for histologi- 

 cal examination consisted of 6 to 10 specimens pre- 

 served in Dietrich's fixative, embedded in paraffin 

 blocks, and completely sectioned every 4—5 pm on a 

 rotary microtome. Light microscopy analysis was per- 

 formed after staining with Cason's trichromic (Cason, 

 1950). 



The qualitative histological examination concen- 

 trated on the liver, pancreas, musculature, and in- 

 testinal mucosae. For quantitative purposes, mea- 

 surements of the cell height of the anterior and pos- 

 terior intestinal mucosae were performed as de- 

 scribed by Theilacker and Watanabe (1989). These 

 measurements consisted of the distance from the 

 basal membrane to the tip of the brush border and 

 were obtained under a microscope equipped with an 

 ocular micrometer eyepiece accurate to 0.02 urn. In 

 the anterior intestine, the site for this measurement 

 was the ventral row of cells located just cranial of 

 the intestinal valve complex (see Fig. 6, A and B). A 

 similar measurement on the posterior intestine mu- 

 cosa was performed caudal of the intestinal valve (see 

 Fig. 6, A and B). 



Control and starved group parameters at each sam- 

 pling time were compared by using Student's Mests. 

 The overall level of significance (a) for each data set 

 was fixed at a nominal value of 0.05. The critical t- 

 value for k number of tests was adjusted through 

 Bonferroni's correction as oik (Sacks, 1978). All values 

 were plotted as arithmetic means and standard errors. 



Results 



Morphometry and biochemistry 



Sampling of 6-day-old larvae was conducted at 24, 

 48, and 60 hours after initiation of starvation (Fig. 



1 ). Mortality in starved larvae began about 60 hours 

 after food deprivation. The mean standard length of 

 starved larvae was lower than their fed counterparts 

 at all sampling times tt 18 =3.39, P=0.003, at 24 h, Fig. 

 1A). Mean dry weight (Fig. IB) and the mean eye to 

 head ratio (Fig. 1C) did not differ significantly (dry 

 weight, £ 18 =2.39, P=0.028, at 24 h; eye/head ratio, 

 * 18 =1.31, P=0.208, at 60 h). The mean pectoral angle 

 of starved larvae decreased relative to the fed larvae 

 after 24 hours (f 18 =3.53, P=0.002; Fig. ID). Mean 

 RNA:DNA ratios of starved larvae were lower than 

 those of fed larvae « 18 =2.68, P=0.015, at 60 h; Fig. 

 IE). After 60 hours, mean RNA:DNA ratios had de- 

 creased from an initial value of 3.75 to 2.74 and 2.31 

 in fed and starved larvae, respectively. Levels of pro- 

 tein remained fairly constant throughout the experi- 

 mental period (Fig. IF). 



Sampling of 16-day-old larvae was conducted at 

 24, 48, and 72 hours after initiation of starvation (Fig. 

 2). Starved 16-day-old larvae began to die after 72 

 hours. Mean standard length of both groups was not 

 statistically different at any time U 18 =2.25, P=0.037, 

 at 72 h; Fig. 2A). However, differences in mean dry 

 weight were significant at 72 hours U 18 =3.04, 

 P=0.007; Fig. 2B). A comparison of the mean dry 

 weight at the beginning of the experiment and that 

 for each group after 72 hours indicates that fed lar- 

 vae incorporated body mass at a daily specific rate 

 of 7.9%/day, whereas starved larvae lost weight at a 

 rate of 10.4%/day. Similarly, the ratio of eye diam- 

 eter to head height became significantly different 

 only after 72 hours of starvation (t 18 = 4.41, P<0.001; 

 Fig. 2C). Little difference was observed in the mean 

 pectoral angle between the groups until 48 h 

 (f 18 =4.59, P<0.001 ) and 72 hours tf 18 =8.25, P<0.001; 

 Fig. 2D). For three days, the mean RNA:DNA ratio 

 of fed animals (2.97-2.99) remained near the mean 

 value at time (2.81; Fig. 2E). During the same pe- 

 riod, starved fish showed a steady decline in 

 RNA:DNA ratio to a final value of 1.93, although dif- 

 ferences were only significant at 72 hours (< 18 =3.47, 

 P=0.003). Mean protein content initially decreased 

 in both groups but became fairly constant and indis- 

 tinguishable between groups thereafter (Fig. 2F). 



Sampling of 33-day-old larvae was conducted at 

 24, 72, 120, and 192 hours after initiation of the ex- 

 periment (Fig. 3). Larvae began to die after approxi- 

 mately 8 days of food deprivation. Starved larvae 

 were significantly shorter than fed ones after 72 

 hours (* 18 =3.32, P=0.004; Fig. 3A). Daily specific 

 growth in length of fed larvae progressed at a rate of 

 2.5%/day but remained almost constant in starved 

 fish. Dry weight of fed larvae also increased signifi- 

 cantly relative to starved larvae (Fig. 3B). At the end 

 of the experimental period, the fed larvae had in- 



