9438 - Laserfiche WebLink

Home Browse Search

CLARKSON AND CHILDS-TEMPERATURE EFFECTS ON FISHES 405 Replicate data for X. texanus and C. latiPinnis were pooled for analyses, whereas analyses of G. CYPha and P. lucius compared ranks of replicate means (individual data were lost after data sum- marization). Posthoc multiple comparisons of length and weight statistics across temperatures within each species and age at each sampling date were made using procedures of Zar (1984: 199-201). We computed specific growth rate (SGR; per- cent increase in body weight per day) for each species-age-temperature-replicate mean weight as 100 X [(loge final weight - loge initial weight) / days]. Kruskal-Wallis tests were used to compare SGRs across temperatures and ages within species. When null hypotheses were re- jected (corrected for multiple tests), nonpara- metric posthoc multiple comparisons were made using procedures of Zar (1984). RESULTS , Growth.-Plots of mean lengths (Fig. 1) and weights (Fig. 2) over time for each species-age group show enhanced growth with increasing temperature. The flat period of P. lucius growth between days 42 and 59 coincided with an out- break of costiasis. Comparisons of lengths and weights across temperatures for each X. texanus and C. latipin~ nis age group and sample date showed signifi- cant differences among mean ranks, with excep- tions of age 14-day X. texanus for length at day 2, age 41-day X. texanus for length and weight at days 10 and 24, and age 41-day C. latiPinnis for length and weight at day 35 (Tables 1-2). Tests of replicate mean length and weight comparisons across temperatures for G. CYPha and P. lucius were significant for each sampling date, with the exception of the 24-day compar- isons of length and weight for P. lucius. Correc- tions for multiple tests rendered all test proba- bilities nonsignificant for these two species. Kruskal-Wallis tests revealed significant differ- ences in SGR among sampling dates, tempera- tures, and ages for all species. A nonparametric multiple comparison test (Zar, 1984) showed that mean ranks of SGRs were significantly dif- ferent among all temperature comparisons and that SGR ranks of age-7 and age-14 groups dif- fered from those of age-41 fish but were not different from each other. Plots of mean SGR for age groups of each species over time are shown in Figure 3. Development.-AB determined by presence/ab- sence of the preanal finfold, smallest juvenile X. texanus in our experiments was 23.7 mm TL, and largest larva was 28.7 mm. Respective C. la- tipinnis lengths were 27.0 mm and 30.5 mm. To- tal lengths at transformation from larva to ju- venile are 27-30 mm for X. texanus and 28-29 mm (rarely to 31 mm) for C. latipinnis (D. E. Snyder and R. Muth, unpubl.). Length at time of finfold absorption for G. CYPha is reported as 26-27 mm, and 25 mm (rarely 24 mm) for P. lucius (D. E. Snyder, unpubl.). Transformation from larva to juvenile was first observed in age 7-day X. texanus 57 days following commencement of temperature ex- periments at 20 C (30% of sample), and all fish reared at 20 C transformed by day 126. In con- trast, no age 7-day X. texanus had transformed by day 126 at 14 C or 10 C (maximum TL = 22.0 mm and 14.2 mm, respectively). Ten per- cent of the sample of age 14-day X. texanus reared at 20 C were juveniles on day 51 (all transformed by day 120), but all fish remained in larval stage by day 120 for 14 C and 10 C groups (maximum TL = 24.1 mm and 20.0 mm, respectively). Juvenile X. texanus in the age 41-day group were first observed in the day 63 sample at 20 C (65% of sample) and by day 128 at 14 C (30% of sample), but none had trans- formed by day 128 at 10 C (maximum TL = 25.6 mm). Juveniles were first observed 59 days following commencement of temperature experiments at 20 C (70% of sample) in the age 7-day C. lati- pinnis group, and all had transformed by day 91. First observance of juveniles in samples of age 7-day C. latiPinnis reared at 14 C was day 91 (40% of sample), but no fish reared at 10 Chad transformed by that time (maximum TL = 21.3 mm). First detection of age 14-day C. latipinnis juveniles was in the day 63 sample at both 20 C (80% of sample) and 14 C (10% of sample) , but all 10 C fish remained larvae by day 92 (maxi- mum TL = 25.4 mm). All age 41-day C. latiPin- nis samples contained juveniles by day 35 at all experimental temperatures (63% at 20 C, 63% at 14 C, 50% at 10 C), but only at 20 C did all fish complete transformation by day 64. Based on replicate mean length data, "aver- age" age 41-day G. CYPha larvae transformed to juvenile stage by day 14 at 20 C (mean TL = 28.6 mm), by day 43 at 14 C (mean TL = 30.3 mm), and some by day 93 at 10 C (mean TL = 26.0). No G. CYPha larvae from any temperature group transformed to juvenile in the short-du- ration trials for age 7-day and age 14-day groups. "Average" time to transformation for age 14- day P. lucius occurred by day 42 at 20 C (mean TL = 26.6 mm) but not by 92 days at 14 C and 10 C (mean TL = 20.1 mm and 11.7 mm, re- spectively) .