Table 4b. --Analysis of covariance and test of significance 

 of adjusted mean oxygen consumption values 

 for group 1 (20-45 fish) and group 2 (75-83 fish) 



♦♦Significant at 1 -percent level (P < 0.01). 



Table 4c. --Analysis of errors of estinnate from the average 

 regression within groups 



that the drop in oxygen consumption is probably a real one: and 2) the difference between the 

 slopes of the two curves is not significant (table 4c), indicating that the effect of flow rate on oxy- 

 gen consumption is similar regardless of the numbers of fish tested. 



The drop in metabolic rate with larger groups of fish deserves further discussion. 

 Schuett (1933) and Shlaifer (1938) both point out that 2 to 4 goldfish in a container consume signi- 

 ficantly less oxygen per unit weight thzin single fish. A similar "group effect" has been observed 

 by Geyer and Mann (1939a. b) in Perca fluviatilis . All these authors find that the effect disap- 

 pears when the size of the container is increased in proportion to the increase in numbers of fish. 

 The present data are inadequate to justify anything but tentative conclusions. The decrease in 

 metabolic activity is perhaps due to changing intensity of nnotion caused by the reduction in space 

 available per fish. No evidence is available, however, attesting to the possible influence on the 



12 



