MODIFICATION OF THE BEHAVIOR OF FISHES 5 



introduced at one end of the experimental tank, while water differ- 

 ing as little as possible from that in which the fishes normally live 

 was introduced into the other end. Various kinds of treated water 

 were used as follows: (1) boiled water — oxygen, nitrogen, carbon 

 dioxide, and bicarbonates in part removed; (2) water with 

 varying amounts of carbon dioxide added ; (3) boiled water with 

 oxygen added (either against tap or boiled water at the other 

 end) ; (4) boiled water with either carbon dioxide, acetic acid, or 

 ammonia added. Various combinations of these factors were 

 also tried and nitrogen was added in a few experiments. 



When the difference between the solutes at the two ends of 

 the tank was not great, we found by chemical tests that the 

 central portion of the tank was a gradient between the character- 

 istic waters introduced at the two ends. Usually the end thirds 

 were essentially like the inflowing water. When the difference 

 in concentration was great the region of the gradient was pro- 

 portionally longer and the ends with the inflowing concentra- 

 tions were accordingly shorter. When the difference in con- 

 centration was very great the entire tank was gradient. 



During the experiments the two authors worked together. 

 Three fishes were placed in each of two dishes containing enough 

 water to barely cover them and set above the tanks. When all 

 was in readiness and at a time agreed upon the two lots of fishes 

 were emptied into the centers of the tanks. Marks on the sides 

 divided the tanks into sixths. The fishes nearly always swam 

 back and forth apparently exploring the tanks. The movements 

 of the fishes were recorded graphically as shown in figure 1 . For 

 this purpose sheets of ruled paper were used. Three vertical 

 double rulings correspond to the center and two ends of the tanks. 

 Two pairs of single rulings divided the space between two primary 

 rulings into three equal parts and the entire distance from right 

 to left into six parts. Distance from right to left was taken to 

 represent the length of the tanks; vertical distance to represent 

 time which was recorded in minutes and seconds at the center. 

 The width of the tanks was ignored. The graphs on the following 

 pages are copies of the originals with the time corrected to scale. 



Before or after the experiment, the headings of the sheets were 

 filled in with data regarding the kind, size, and number of fishes, 

 their previous history, the conditions in the tanks, concentra- 

 tions of the solutes and other significant data. Details of the 



