squares for the unshocked and once-shocked 

 fish. The tabulated Q value for three 

 "treatments" and an error mean square based 

 on 108 d.f. is 3.37. Since each potential 

 mean is calculated from 72 observations. 



L.S.D. 



(14.09) (3.37) 

 V 72 



= 5.60 



Any difference of means exceeding 5.60 

 is significant at the 5 percent level. 



Since these three differences all exceed 

 5.60 the three potential means are all 

 significantly different at the 5 percent 

 level . 



Effect of pulse frequency . — The dif- 

 ferences between pulse frequencies were 

 highly significant for both unshocked and 

 once-shocked fish — F values were highest 

 in all three tables (tables 4, 5, and 6). 

 Examination of the data, as well as quali- 

 tative observations, leads to the conclu- 

 sion that optimum pulse frequency is below 

 5 pulses per second. For example, see 

 figure 8(b) and table 3 (page 10), 



Effect of pulse duration . — This vari- 

 able factor for the values tested (10, 20, 

 30 milliseconds) was not significant in the 

 three analyses (tables 4, 5, and 6), nor 

 was any significant interaction found. 



CONCLUSIONS 



A. Optimum electrical conditions : 

 Pulse frequency was the most critical vari- 

 able. Potential also was significant. 

 Substantiating the preliminary observations, 

 pulse duration was not significant, at least 

 in the range tested. The optimum values of 

 the electrical variables were as follows: 



1. Potential — Of the three potentials. 



50 r 



10 

 30 

 20 ■ 



L^ 



60 75 



(0 } Porentioi { voirs ) 



70 

 60 

 50 

 40 

 30 

 20 

 10 



Norlh Slock 

 South Block 



oMyv- 



2 5 



(D) Pulse Frequency (pulses per second) 



40 



30 

 20 

 10 



°T 



(c ) Poise Durolion ( milliseconds ) 



Figure 8. — Mean percentage of squawfish that 

 entered the "positive" end zone in north and 

 south blocks for each value of the variable 

 factors of potential, pulse frequency, and 

 pulse duration, linshocked and once-shocked 

 fish combined. Each mean percentage based 

 on 36 observations. 



60, 75, and 90 volts, 60 volts was 

 the optimum; 



2. Pulse frequency — Of the three pulse 

 frequencies, 2, 5, and 8 pulses per 

 second, 2 pulses per second was the 

 optimum; 



3. Pulse duration — With the three 

 pulse durations, 10, 20, and 30 

 milliseconds, no optimum was ob- 

 served. 



B. 



Direction of the electrical fields: 



The direction of the electrical fields was 

 a highly significant variable. The fish 

 showed a greater response when the electri- 

 cal fields moved toward the north end of the 

 laboratory tank than toward the south. The 

 reason for this reaction was not determined. 



14 



