were placed in another where their reactions 

 to each frequency could be observed closely. 

 Before the planned tests and controls were 

 started, it was decided to try the full range 

 of the low frequency equipment briefly to see 

 if one particular frequency response could be 

 singled out. Observations were made directly, 

 with Rounsefell standing motionless at section 

 No. I. Neal watched the large brood rainbow 

 in the live car near shore fronn a distance. 

 The following notes were made: 



1. 67 c.p.s.: Rounsefell noted that fish 

 appeared uneasy at low range especially when 

 sound was first turned on. Fish started and 

 faced away from the audio speaker but did 

 not swim off when sound continued. 



2. Siren effect 67 to 700 c.p.s. Rounsefell 

 observed whole school of fish face away from 

 speaker, but returned to normal in seconds. 



3. Intermittent operation all frequencies 

 from 67 to 3,000 c.p.s. No effect. 



4. Neal reported no response fronn brood 

 rainbow near shore. 



Following the brief tests described above, 

 those frequencies which elicited even the 

 slightest response were tested systematically 

 by octaves (i.e., 70 to 140 c.p.s., etc.). It was 

 evident from their reaction that the fish were 

 able to detect the source of sound at the 

 moment of starting. It is doubtful, but entirely 

 possible, that a visual stmriulus was received 

 in addition to the audio stimulus. The aluminum 

 piston of the audio speaker had a travel of 

 less than one-eighth of an inch. It was located 

 some distance from the fish and outside the 

 wire trough. 



Some sound emitted by the underwater 

 speaker escaped into the air and could be 

 heard plainly as a steady buzz at a distance 

 of 50 yards. At a distance of 1 yard the 

 escaping sound was likened to that of an 

 irritating door buzzer at arm's length. The 

 same sound intensity under water is multiplied 

 nearly 100 times--a fact familiar to the small 

 boy who strikes rocks together below the sur- 

 face with his head submerged. The fish, 

 therefore, were being subjected to an intensity 

 of sound much greater than is perceived by 

 the hunnan ear above the water surface. 



Figures 8-15 with histograms showing the 

 results of each test are presented to give the 

 reader an opportunity to compare the results 

 of the sound tests and the controls. For the 

 first several tests, fish were counted out 100 

 to a section and replaced 100 to a section 

 after each test. In each case, the darkened 

 portion of the histogram represents the nunn- 

 bers of fish found after each control or test 

 run. 



As there were obvious differences in the 

 sound test distributions and the control dis- 

 tributions it seemed desirable to apply con- 

 tingency tests to the results to determine 

 if the differences were significant. In all 

 cases (comparison of sound tests and controls 

 of the same duration of time) the results were 

 highly significant, thus indicating a marked 

 difference between the sound tests and com- 

 parable controls. 



This significant difference, however, cannot 

 and should not be interpreted as evidence that 

 sound waves either attracted or repelled the 

 fish. There is nothing to suggest that the 

 sound waves produced by this apparatus had 

 any influence on the distributions. 



For practical manipulation of sound waves 

 for leading and guiding fish into safe passages 

 around dams and other stream barriers, it is 

 necessary to have a stimulus which is very 

 close to 100 percent efficient. None of the 

 sounds produced by this first sound producing 

 equipment showed results which in any way 

 approach this efficiency. In no instance did 

 the "water hannmer" show a definite attracting 

 or repelling effect on the fish during any of the 

 tests. 



PIEZO-ELECTRIC CRYSTAL TRANSDUCER 

 HIGH FREQUENCY 



On December 15, D. W. Beecher of the Naval 

 Ordnance Laboratory assembled high fre- 

 quency equipment for our use at Leetown. 

 The oscilloscope and signal generator for 

 frequencies of 12 kc. to 70 kc. were connected 

 to the amplifiers used for the low frequency 

 tests. The transducer was a 4-inch brass 

 cylinder containing quartz crystals and castor 

 oil, covered with a rubber diaphragm. The 

 crystals changed dimensions when subjected 

 to a high frequency, alternating current. The 

 entire unit, a little over 8 inches in length, 

 weighed approximately 5 pounds. The speaker 

 could be beamed much like a flashlight and 

 had approximately a 60-degree cone of di- 

 vergence. The projector may be described 

 as having the following characteristics: At 

 frequencies of 12 kc. to 60 kc.--4,000 

 dynes/cm.^ or 72 db above I dyne/cm.^ at 3 

 feet, except for 10-db dips at each end and 

 at approximately 35 kc. 



For the first tests the transducer was 

 beamed directly upon the fish in section No. 1 

 in an attempt to frighten them out. Tests Nos. 

 47 and 48 (fig. 16) show that the fish moved 

 even faster when the sound was not on. Test 

 No. 49 appeared conclusive. The sound was 

 turned on and all fish moved away from the 

 transducer. The optimism was short-lived 

 when in test No. 50, a control, the fish moved 



