The tests demonstrated that it was not only possible to stimu- 

 late the desired movement of rainbow trout in the field, but also that 

 the fish could be made to enter an enclosure. Each rainbovj trout was 

 tested individually. The first 21 experiments were conducted with the 

 device installed as shown in Figure 3. In 11 experiments the fish were 

 allowed to swim unmolested into the area between the electrodes before 

 the power was applied. Of the 11 rainbow trout subjected to this proce- 

 dure, 8 entered the trap and 3 were electronarcotized. One of the 3 

 turned over enroute to the trap and the other 2 displayed a galvanotaxic 

 response similar to that classified during the laboratory tests as "over- 

 control". The location of the fish in relation to the electrodes made 

 no detectable difference in results. It was found also that as long as 

 the power remained on, the fish did not, or could not, leave the trap. 

 The polarity was reversed in a few of the tests after the trout had 

 entered the trap. On each occasion, the test animals immediately left 

 the trap and went to the opposite electrode. This procedure could be 

 repeated three or four times with an active fish before exhaustion re- 

 sulted in its being carried out of the electrical field by the stream 

 flow. 



The remaining 10 rainbow trout were released individually well 

 below the electrode array with the power applied constantly in an attempt 

 to determine the reactions of the fish upon contact with the extremity 

 of the electrical field. Of the 10 trout, U were successfully drawn to 

 the anode, but only after an appreciable increase in the electrode volt- 

 age. The rest invariably turned back downstream upon encountering the 

 weak fringe of the electrical field. 



Thirteen of the 3U rainbow trout were subjected to tests with 

 a different electrode array. The original electrodes were replaced by 

 two pieces of 2-inch diameter aluminum tubing, suspended vertically, 

 spaced 11 feet apart, and immersed to a depth of 6 inches. Results with 

 this arrangement were poor even with an electrode voltage as high as 230 

 volts. At this voltage level, power consumption was less than that of 

 the original array with an electrode voltage of 90 (ll5 watts as compared 

 to 135 watts). The above tests with the extremely small electrodes were 

 conducted to determine the influence of immersed electrode area on the 

 voltage gradient as portrayed by the reactions of the fish. Other studies 

 aimed at development of an electrical sea lamprey control structure in- 

 cluded tests to determine the relative efficiency of electrodes of various 

 sizes, shapes, and materials. 2./ These studies demonstrated that elec- 

 trode size does effect voltage gradient. McMillan (1928) also conducted 

 an extensive investigation of the same nattire. 



6/ See footnote 2. 



20 



