EFFECT OF ELECTRICAL PARAMETERS ON YOUNG SALMON 



233 



Table 10. — Analysis of variance of the delayed effect (2d to 

 SOth day) of water resistivity (rows), wiring pattern and 

 pulse shape (columns), and voltage and frequency (treat- 

 ments), upon the mortality of fingerling silver salmon 



Table 12. — Analysis of variance of the delayed effect (3d to 

 SOth day) of the low (1,000 and 5,000 ohm cml) and high 

 (10,000 and 15,000 ohm cm.) levels of water resistivity on 

 mortality of fingerling silver salmon 



after 30 days as they did after 1 day. These 

 differences, however, were not significant because 

 of an increase in residual error. 



Pulse frequency. — The difference in mortaUty 

 which resulted from the difference between levels 

 of frequency' actually reversed in sign. This is 

 presumablj* another indication of the increase in 

 residual error. 



Water resi.'^iiivity. — Mortalities which resulted 

 from each of the four levels of water resistivity 

 were combined in a low and a higli group as they 

 were for the anah'sis of the immediate effect. 

 Analysis of variance (table 12) reveals no signifi- 

 cant difference in mortality between the two 

 groups after 30 days. 



Table 11. — Cumulative mean percentage mortality (2d to 

 SOth day), resulting from each of the test conditions 



Chi-square analysis showed that after the 30-day 

 holding period there was no longer a significant 

 difference in mortality between tlie control fish 

 and the test fish. The chi-square value of this 

 comparison was 3.476, with 1 degree of freedom. 

 In this analjsis the total percentage mortality of 



1 See table 10. 



the control fish, measured from the second through 

 the thirtietli day, was compared with the cor- 

 responding mortality of tlie test fish. A prelimi- 

 nary chi-square analysis showed no significant 

 differences in mortality between the control 

 groups (A'^= 6.8737, with 3 degrees of freedom) 

 and analysis of variance (table 10) showed that no 

 differences in mortalitj' had resulted from differ- 

 ences in the test variations. 



CONCLUSIONS 



Although analysis of the immediate effect (effect 

 m the first 24 hours) of the conditions tested 

 shows that a significant number of mortalities 

 occurred when the fish were subjected to combi- 

 nations of two or more of the deleterious variables 

 (low water resistivity, high frequency, and square- 

 wave pulses), it should be remembered, before 

 reading the conclusions of this manuscript, that 

 the fish tested in this experiment were forced to 

 swim completely through the electrical fields 

 created by the arraj'. This situation is not likely 

 to occur in our more successful field experiments, 

 since our method of diverting fish with electricity 

 is based on the principle of an avoidance reaction. 



Immediate effect 



1. Of the two pulse shapes tested, the square- 

 wave pulses resulted in a higher mortality than the 

 half sine-wave pulses. 



2. High frequency (30 p.p.s.) caused a signifi- 

 cantly higher mortality than low frequency (15 

 p.p.s.). 



3. Low levels of water resistivity (1,000 and 

 5,000 ohm cm.) resulted in a higher percentage 

 mortaUty than the high levels (10,000 and 15,000 

 ohm cm.). 



4. Of the two levels of applied voltage tested, the 

 liigh voltage (250 v.) resulted in a higher percentage 

 mortality than the low voltage (165 v.). This dif- 



