Equipment 



A reinforced concrete tank, equipped with an electrically in- 

 s\ilating liner, was used for the laboratory experiments. The insulating 

 liner was installed to prevent distortion of the electrical field pattern 

 by the highly conductive materials of the tank (reinforcing steel). The 

 inside dimensions were 210 inches long, 36 inches wide, and 36 inches 

 deep. A continuous flow of water was pumped from Lake Huron into the 

 tank during the tests. The depth was held at 12 inches by means of a non- 

 conducting overflow tube. 



The electrodes used in the laboratory tests were two 36-inch- 

 square pieces of l/2-inch-mesh galvanized hardware cloth. These electrodes 

 were suspended vertically 2 inches fron either end of the tank and per- 

 pendicular to the long axis of the tank. The immersed area of each elec- 

 trode was then 36 inches wide and 12 inches high. A barrier of small-mesh 

 cotton netting mounted on a wooden frame was placed across either end of 

 the tank 6 inches from the electrode and parallel to it. These barriers 

 protected the test animals frcan direct contact with the electrodes. 



A direct current power supply in combination with a motor-driven 

 switch (or conriutator) provided the electrical impulses used in the fish 

 leading experiments. A variable transformer, connected to a 220 volt AC 

 domestic line supplied any desired voltage up to 260 volts AC to the pri- 

 mary of a power transformer. The power transformer, having a 2 : 1 step- 

 up ratio and a center-tapped secondary winding, supplied power to a pair 

 of selenium rectifiers which were controlled by a switch to form either 

 a half- or full-wave rectifier circuit. 



The full-wave circuit delivered up to 260 volts of DC at 12.5 

 amperes, with a ripple frequency of 120 cycles per second, while the 

 half-wave circuit supplied up to 500 volts DC at the same current and 

 half the ripple frequency. Inasmuch as the output of the rectifier was 

 filtered when square wave pulses were desired after commutation, the 

 difference in ripple frequency is of little importance. In addition to 

 square wave pulses, the unit could be used to supply continuous filtered 

 DC or unfiltered DC at ripple frequencies of 60 cycles (half-wave rectifi- 

 cation) or 120 cycles (full-wave rectification) in continuous or cornmutated 

 form. The power supply, originally intended for field experiments in which 

 large power requirements were anticipated, might easily have been replaced 

 by a smaller unit providing the same voltage range at the low current le- 

 vels encountered in the insulated test tank. The 12 inches of water in 

 the tank presented a load of approximately 1,270 ohms to the power source. 



