EFFECTS OF THERMAL EFFLUENT 559 



Statistical Evaluations 



Data on gi-owth and condition of both the Scripps and 

 generating-station control groups were evaluated by standard t-tests 

 for paired obsei-vations. In each case the initial and final size, weight, 

 volume, or W/L values for an individual test animal were treated as 

 paired observations to reduce effects of unwanted variability caused 

 by differences in size and other characteristics. Differences in growth 

 rates among the two control and the two treatment groups were 

 evaluated by comparing the slopes of regression lines for the growth 

 data by analysis of covariance and the Newman— Keuls multiple 

 range test. 



Evaluation of Control Conditions 



The control data in Table 5 suggest that conditions at the 

 Scripps laboratory were adequate for normal growth for most of the 

 size groups and species considered. The results of t tests for paired 

 observations indicated significant increases in linear size among all 

 size categories of S. purpumtus and for the smaller two size groups of 

 S. franciscanus in both the Scripps and generating-station controls 

 (P < 0.05). The largest size group of S. franciscanus, on the other 

 hand, showed no significant changes in size or W/L in either control 

 group (P > 0.05), possibly because of the slow growth of these 

 individuals and the short duration of the experiment. The results of t 

 tests for P. ochraceus and P. giganteus in both control groups 

 indicated significant increases in weight and volume among all size 

 categories (P < 0.05). On the basis of the limited information 

 available in the literature (Boolootian, 1966), the growth rates 

 exhibited by control animals of both species appear to be normal. 



The smallest size group of K. kelletii showed significant increases 

 in weight and W/L in both control groups; the largest size group 

 exhibited a significant increase in weight (P < 0.05). Weight and W/L 

 for all size categories of R. poulsoni showed no significant changes in 

 either control group (P > 0.05). Laboratory growth rates of 

 R. poulsoni controls were similar to those estimated for a natural 

 population by Fotheringham (1971). With one exception, the results 

 of t tests for A. undosa showed significant increases in body weight 

 and W/L ratios (P  0.05) for all size groups in both Scripps and 

 generating-station controls. The largest size category showed no 

 significant change in the W/L ratio in either control group; their 

 growth was comparable to that reported by Schwalm (1973) for a 

 natural population. These results suggest that the control conditions 

 at both the Scripps and generating-station laboratories were adequate 



