80 ILLINOIS BIOLOGICAL MONOGRAPHS [80 



February 3, 1924. A similar experiment. The average of ten counts of 

 50 vibrations at a temperature of 4.5°C shows 100 vibrations in 2 minutes 

 and 24 seconds. 



This must be taken as an indication of the reduced activity of the fish 

 due to a low temperature, as control conditions were the same in both 

 cases. Hence it can be assumed that a drop in temperature has an effect 

 upon the fish. 



In order to ascertain whether or not the fish react to a drop in tempera- 

 ture, ten Labidesthes sicculus were placed in a galvanized iron tank four 

 feet long by eighteen inches wide by twelve inches deep. Water taken 

 from their own environment was placed in it, stirred up to give it a uniform 

 temperature, and temperature readings were taken (as well as hydrogen ion 

 concentration readings). The fish were placed in the tank and their move- 

 ments noted and traced. At the beginning of the experiment the fish 

 traveled consistently the length of the tank, turning back when they hit 

 the ends. As soon as the normal behavior was noted, a piece of ice was 

 suspended in the upper three inches of water at one extreme end and a 

 subsequent tracing of the movement of each of the fish was taken (the 

 experiment being performed separately for each fish). It was found that as 

 the fish approached the ever increasing zone of cooler water they turned 

 back sharply toward the warmer water, and the temperature was taken 

 at the point at which the fish turned back. The results of one of these 

 experiments, which has been selected as typical, is presented (Fig. 15). It 

 will be noted that the fish reversed itself at the first 0.4°C drop in tempera- 

 ture, and that a drop of 0.7°C was sufficient to confine it to one end of 

 the tank. Since the fish used in the experiment were immature individuals 

 measuring 4.5 cm, the conclusion is warranted that the young at least are 

 very sensitive to slight temperature changes. Their sensitivity is not quite 

 as marked as the case of Shelford and Powers (1915) where the reaction 

 occurred at a change of 0.2°C. 



The writer can not agree with Hubbs in his statement that the light 

 intensity is the same over deep and shallow water. In fact, when the bot- 

 tom is clear sand or fine gravel of a light color, the writer believes that there 

 can be no question but that the intensity of the light is much greater over 

 the shallow water, due to the reflection from this bottom. Since the bars of 

 Wisconsin lakes are largely light sand, plainly visible because of the clear- 

 ness of the water to a depth of twenty feet or more, the writer would em- 

 phasize the fact that as the fish approach the shore over water becoming 

 increasingly shallower, the light intensity increases in direct proportion 

 to the closeness with which the shore is approached. Hence with the failing 

 light of evening, if the fish be positively phototropic, one would expect a 

 shoreward migration in correlation with the decreasing daylight. Does 

 such a correlation exist? I have already mentioned the time at which the 



