|
<br />200
<br />
<br />
<br />~
<br />,
<br />'tl
<br />., 0.35
<br />.c
<br />E 0.30
<br />
<br />0.25
<br />....,
<br />E- 0.20
<br />e
<br />E-
<br />.c
<br />~
<br />o
<br />...
<br />'"
<br />
<br />18
<br />
<br />lJ'
<br />'--
<br />'"
<br />r...
<br />'"
<br />
<br />G
<br />Q,
<br />f$
<br />~
<br />
<br />100 50
<br />Ration (nauplii fish-' dO' )
<br />
<br />o
<br />
<br />Fig, 1. Response surface of growth of Colorado squawfish larvae
<br />(mm TL d-1) in replicated treatments as a function of water tem-
<br />perature and food abundance. Mean (SE) growth for each treat-
<br />ment combination and the response surface model coefficients
<br />are in Tables 1 and 2; the star represents the optimal growth re-
<br />sponse as a function of the independent variables,
<br />
<br />Box- Wetz criterion, which states that if the regres-
<br />sion F-value exceeds 10 times the critial F-value, the
<br />function is adequately estimated. The critical F-val-
<br />ue with 5 and 114 degrees of freedom, and p = 0.01,
<br />was 3.18, which results in a Box- Wetz criterion of
<br />31.8. Deviance statistics were used to assess fit of the
<br />survival model and to determine if maximum likeli-
<br />hood estimates were appropriate. Significant inter-
<br />
<br />action terms were plotted and analyzed prior to in-
<br />terpreting main effects.
<br />
<br />Starvation resistance experiments
<br />
<br />Starvation resistance of Colorado squawfish was
<br />tested by delaying feeding of larvae after they were
<br />capable of feeding exogenously (6 d after hatching)
<br />and comparing survival and growth to a control
<br />group. Thirty fish were placed in each of eighteen 11
<br />beakers, and three replicate beakers were random-
<br />ly assigned to six treatments where first feeding was
<br />delayed 0 (control), 5, 10, 15, 17.5, or 20 days. Surviv-
<br />al in each beaker was monitored daily. Fish were fed
<br />brine shrimp nauplii ad libitum twice daily after
<br />feeding began. Growth was monitored by sampling
<br />and measuring larvae from additional cultures of
<br />fish that were first fed at the same treatment inter-
<br />vals used for delayed feeding trials. Beakers were
<br />cleaned as necessary and aerated to maintain dis-
<br />solved oxygen at about 6 mg 1-1. Survival and
<br />growth rates before and after first feeding were
<br />compared to a continuously fed control group.
<br />
<br />Table 1. Mean growth (G, mm TL d-1, (SE)) and survival (S, number of survivors/20 per replicate, (SE), number of experimental repli-
<br />cates) of Colorado squawfish larvae under various nominal temperature and ration (brine shrimp nauplii fish-1 d-1) treatments, Fluctu-
<br />ating and constant temperature treatments were pooled because no significant differences were detected.
<br />Ration Temperature (0 C)
<br /> 18 22 26 30 ..
<br />12.5 G 0,061 (0,0021) 0,071 (0,0029) 0,065 (0,0041) 0.043 (0,0058)
<br /> S 0.67 (0.0543) 0.45 (0.0565) 0,24 (0,0352) 0,13 (0,0167)
<br /> n=6 n=9 n=6 n=3
<br />28 G 0,093 (0.0025) 0,092 (0.0018) 0.094 (0.0021) 0,063 (0.0058)
<br /> S 0.86 (0.0455) 0,87 (0,0323) 0,73 (0,0279) 0.48 (0.0928)
<br /> n=6 n=9 n=6 n=3
<br />63 G 0.134 (0,0022) 0.138 (0.0012) 0.140 (0.0056) 0,129 (0,0061)
<br /> S 0.82 (0.0587) 0,92 (0,0264) 0,90 (0,0387) 0,63 (0.0167)
<br /> n=6 n=9 n=6 n=3
<br />142 G 0,170 (0,0039) 0.204 (0.0046) 0.217 (0.0042) 0,215 (0,0048)
<br /> S 0.91 (0,0239) 0.92 (0.0188) 0,93 (0,0357) 0,93 (0,0167)
<br /> n=6 n=9 n=6 n=3
<br />320 G 0.201 (0.0055) 0,273 (0.0055) 0.303 (0,0066) 0.291 (0.0013)
<br /> S 0.78 (0.1085) 0,94 (0.0176) 0.95 (0.0224) 0.92 (0.0167)
<br /> n=6 n=9 n=6 n=3
<br />
|