The entrances to the two channels on the downstream end were blocked off with 

 plastic screens and the electrodes de-energized before a total count of the fish in the 

 traps was made. Before subsequent tests the experimental area was entirely cleared 

 of fish by forcing them into the channels at the upstream end. 



The results wer? calculated on a percentage basis from the total number of 

 fingerlings. recovered in both traps and the total number recovered in the narrow 

 channel. The index of effectiveness is defined as the difference of the percentage of 

 control fish collected in the narrow channel and the percentage of fish collected in the ■ 

 narrow channel following an electrical test during which the electrodes were energized. 



Before a useful comparison between a pair of electrical tests could be made, three 

 preliminary conditions had to be satisfied: (1) The respective controls had to be uni- 

 form, (2) The mean percentages of fingerlings recovered in the narrow channel had to 

 be the same and (3) The electrical tests being compared had to be uniform. While 

 nonuniformity could result from a response distribution with a large variance,^ the 

 foregoing restrictions guarantee a uniform experimental technique. All tests of 

 uniformity and of equal mean percentages were chi-square tests of significance at 

 the 5-percent level. The results of these tests are shown in tables 1 to 12. 



EXPERIMENTS 



In exploring the effectiveness of a narrow d.c . field under laboratory conditions, 

 the electrical characteristics and energy levels were held constant while the following 

 factors were varied: (1) angle of electrical field in relation to the flowing water; (2) 

 width of electrical field (distance betv/een rows of electrodes); (3) spacing between 

 electrodes and (4) diameter of electrodes. These experiments were exploratory in 

 nature and since the results varied considerably, it was difficult to interpret the data 

 conclusively. However, the results have been summarized in tables 13 to 18 and 

 graphically presented in figures 5 to 15. 



A maximum effectiveness occurred with a 2 -foot width of electrical field 

 (fig. 5). This maximum effectiveness occurred at a 40° angle of electrical field 

 with 1/2 -inch electrodes spaced 12 -inches apart. At the 40° angle and with 1/2- 

 inch electrodes the 2 -foot width of field was significantly more effective than the 3- 

 foot width of field. The difference in effectiveness between 2- and 3 -foot fields was 

 greatest at the closer electrode spacings . As the electrode spacing increased. from 

 12 to 36 inches there v/as a decrease in the ifference of the effectiveness of the two 

 widths of field; the effectiveness of the 2-foot width of -field also decreaseci. 



At a 60° angle of electrical field with 1/2 -inch electrodes the 2 -foot width of 

 field appeared to be more effective than the 3 -foot width (fig. 6). The difference 

 in effectiveness was greatest at a 6- and 36 -inch electrode spacing. 



8 



