570 R. A. SPAETH 



« August 8, 1912 



3.00 P.M. Scales from a dark, 8 cm. female were removed to seawater. 



3.06 P.M. Scales were stimulated with a ''Harvard" indu'ctorium, 

 with closed secondary and one dry cell. A stimulation 

 of one and one-half minutes had no .appreciable effect. 



4.00 P.M. Another scale stimulated for one minute showed no re- 

 sponse. 



From these observations it is clear that the conditions in F. 

 heteroclitus are essentially the same as in other teleost fishes as 

 regards the response to the induction current. NaCl tends to 

 antagonize the specific contracting effect of the current. This 

 is shown both by experiments with strong (0.5 M) NaCl solu- 

 tions and by the failure of the melanophores to respond to the 

 electrical stimulus when immersed in sea water. 



8. CONCLUSION AND SUMMARY 



From these experiments with the chromatophores of F. heter- 

 oclitus it seems certain that, in addition to the normal pigment 

 migrations which occur in the living fish under the influence of 

 the nervous sj^stem, the chromatophores are sensitive to a variety 

 of physiological stimuli which act directly upon the cells. Even 

 the possibility of an intercalated sympathetic reflex-arc seems to 

 be eliminated when the experiments with the various salt solu- 

 tions and ultra-violet light are considered. In the former case 

 the melanophores invariably showed a reaction which began at 

 the periphery of the scale and proceeded centripedally. Ob- 

 viously the torn ends of the sympathetic nerve-fibers were ex- 

 posed to the action of the solutions. Had such broken nerve- 

 endings been stimulated, a synchronous reaction should have 

 followed in the chromatophores. This never occurred. Again in 

 the case of the exposures to ultra-violet light the reaction always 

 began centrally, where the epithelium is thinnest, and proceeded 

 peripherally. The light rays stimulated the melanophores most 

 readil}'' at the centre of the scale. This response, as in the former 

 case, must consequently be considered a direct one (compare 

 figs. 1-5, pi. 1 with figs. 1 and 4, pi. 4). 



