PHYSIOLOGY OF CHROMATOPHORES OF FISHES 563 



air at 20°C. In passing from 18 to 30°C. the change in the percentage 

 of dissociation is less than 0.4 per cent (Noyes and Coolidge '03, p. 

 374), This error is therefore neghgible. 



In the above experiment it follows that the contraction of the 

 melanophores could not have been due to an elimination of oxy- 

 gen from the solution since it occurred at 29°C at which tem- 

 perature a 0.1 M NaCl solution (under an oxygen pressure of 

 760 mm.) contains more than four times as much oxygen as a 

 0.1 M NaCl solution exposed to the air at 20°C. in which the 

 melanophores remain expanded for hours. The change of the 

 dissociation value is negligible. Consequently in this case also, 

 as in the first type of experiment the contraction of the melano- 

 phores must be interpreted as a response to the direct, specific 

 stimulation of heat. 



These experiments have thus proved (1) that at about 30°C., 

 heat acts as a specific contracting stimulus upon the melano- 

 phores; (2) that this contraction is independent of the oxygen 

 content of the fluid in which the melanophores are immersed; 

 (3) that at about 30°C. this specific heat effect antagonizes the 

 opposite effect of 0.1 NaCl. 



5. REACTIONS TO OXYGEN 



In the experiments with the effects of temperature upon the 

 melanophores of F. heteroclitus it was clearly shown that heat 

 causes a contraction. It remained to be proved (1) whether 

 the absence of oxygen would also bring about a contraction and 

 reciprocally, (2) whether the presence of oxygen could actually 

 bring about an expansion of the melanophores from the con- 

 tracted condition. 



April 16, 1912 



Scales from a 9.5 cm. female were immersed in a 0.1 M NaCl solution. 

 Thirty minutes later scales were removed to two identical specimen 

 bottles each containing about 10 cc. of a 0.1 M NaCl solution. A cur-, 

 rent of washed hydrogen was now run into one of these bottles. The 

 other bottle served as a control. All the melanophores under the hy- 

 drogen atmosphere were completely contracted at the end of thirty 

 minutes. In the control they were fully expanded. The stopper was 

 now removed and the contents were poured (solution and scales with 



