PHYSIOLOGY OF CHROMATOPHORES OF FISHES 565 



(2) that this tendency towards a contraction is greater than the 

 specific expanding effect of a 0.1 M NaCl solution and (3) that 

 oxygen differs from NaCl in that it probably acts not as a specific 

 expanding stimulus but passively, since it was found that an 

 expansion of contracted melanophores could not be induced 

 even after a long exposure to an atmosphere of pure oxygen. 



6. LIGHT EFFECTS 



V. Frisch calls attention to the present unsatisfactory condi- 

 tion of the evidence for the nature of the reaction of melano- 

 phores to light. In some cases (trout, v. Frisch '11 a) they 

 absolutely fail to respond; in others (eel, Steinach '91; triton, 

 Hertel '07) they are contracted, and again (chameleon, Briicke 

 '52; Phrynosoma, Parker '06; Crenilabrus, v. Frisch '12) they 

 may expand upon illumination. 



While attempting to photograph, by the light of an arc-lamp, 

 the widely-expanded, living melanophores of F. heteroclitus it 

 was found that before the exposure could be made the melano- 

 phores had almost completely contracted. The scales were im- 

 mersed in a NaCl solution at the time of the exposure and when 

 they were removed to the stage of a microscope illuminated by 

 an ordinary gas lamp, the melanophores showed a perfectly dis- 

 tinct reversal of the pigment migration. After a few minutes 

 (depending upon the length of exposure to the arc-light and the 

 strength of the. salt solution) all the melanophores were again 

 completely expanded. If these same scales were now exposed a 

 second time to the action of the arc-light, the same reactions 

 followed in the same sequence. 



Assuming this contraction in the NaCl solution to have been 

 due to a specific light stimulus, experiments were carried out 

 with NaCl solutions of varying strengths and varying exposures 

 to different intensities of light in the blue-violet region (439- 

 490 mm) of a spectrum in an apparatus similar to that described 

 by Day ('11). In all of these experiments the light was reflected 

 from the plane mirror of a compound microscope exactly as in 

 normal field-illumination. Control scales from the same region 

 of the body of the same fish were observed from time to time. 



