Table 2. — Mean effectiveness by classification with 

 respect to treatments and lot sizes. 



Meaui effectiveness for all release 

 group sizes was not significantly different, 

 and we are assured that our previous sauries 

 of 50 fish were adequate. 



Electrical Treatments 



44.32 20.96 35.83 



43.95 23.14 39.23 



37.91 23.05 37.01 



42.06 



22.38 



Table 3. — Preliminary analysis of variance. 



Sum of Degrees of Mean 

 squares freedom square 



19,611.74 



1/ Highly significant. 



Table 4. — Analysis of variance. 



The Jtnalysis of variance shown in 

 table 3 reveals no significant difference 

 resulting from any of the variables, with 

 one exception. This significant difference 

 occurred cunong treatments and is almost 

 entirely due to the difference between elec- 

 trical treatments and control (table 4). 

 This difference between electrical treat- 

 ments and control is emphasized by a histo- 

 grajti (fig. 5) which portrays relative effec- 

 tiveness by the ratio of the number of fish 

 recovered in the narrow channel to the total 

 number recovered downstream. 



Under the conditions of these experi- 

 ments it appears that polarity of the elec- 

 trical field of a sequentially energized 

 single-row electrode array is not an effec- 

 tive factor in guiding fish, as the analysis 

 shows that the only differences were between 

 electrical treatments and control. Because 

 polarity of field is not a contributing 

 factor, the response of the fish must have 

 been a simple avoidance of an unpleasant 

 stimulus, and thus electrotaxis was not 

 involved. 



This knowledge about field polarity 

 provides a greater degree of freedom in the 

 selection of methods of electrode energizing 

 thcui formerly existed. Future selection of 

 methods of electrode energizing will depend 

 upon power consumption, cost of control 

 equipment, and degree of simplicity in gear 

 desired rather than upon the electrotactic 

 effect of electrical fields. 



1/ Highly significant. 



but it was feared that any results might 

 have been obscured by the seemingly high 

 variance. This caused us to be concerned 

 about the size of our releases and, in the 

 last tests used for this report, numbers 

 of fish in each release group were added 

 as a variable. rhe mean effectiveness 

 with respect to lot sizes (table 2) is 

 summarized by over-all means: 



Release of 15 fish 

 Release of 50 fish 

 Release of 100 fish 



mean effectiveness 35.83 

 mean effectiveness 39,23 

 mean effectiveness 37.01 



SUMMARY 



The relation of field polarity to the 

 fish-guiding effectiveness of a single-row 

 electrode array was explored under controlled 

 laboratory conditions. Also explored was 

 the extent of variability due to sample size 

 in this type of test. 



Three manners of energizing the array 

 were compared: One condition maintained 

 positive field polarity toward the upstream 

 end, one toward the downstresun end, and a 

 third alternated the polarity in such a way 



