270 HOVEY JORDAN 



of the reflector-box. This tissue constituted the so-called "meat 

 screen." With this apparatus were repeated experiments similar 

 to those on negative phototropism and regional sensitivity which I 

 have described above. The sensitivity of the spinal cord itself to 

 nearly direct stimulation i.e., without the meat screen was also 

 tested. The results obtained indicate that the spinal cord of Epineph- 

 elus striatus contains no photoreceptive elements, even though 

 light can reach it through the integument and muscles. 



It was always possible, contrary to expectation, to cause stimulation 

 with light which had traversed the screen of meat. The reaction 

 varied from the normal response only in its comparative slowness. In 

 one experiment, upon the posterior end of the body, the reaction-time 

 was 28.5 seconds longer than when no "meat screen" was used. This 

 merely shows that light is absorbed as it passes through the tissue, but 

 not sufficiently to render it ineffective. By local stimulation of different 

 areas of the integument with this apparatus, it was even possible to 

 obtain reactions which were consistent with the previously described 

 differences in regional sensitivity to unscreened light. The thickness 

 of the "screen" of tissue (about one-half inch) was the same as that 

 through which light would have to pass in order to reach the spinal 

 cord of an average sized fish. The possibility, then, of such stimulation, 

 provided that the cord contains the appropriate photoreceptors, is as- 

 sured. Consequently, the presence of half an inch of intervening mus- 

 cular body-wall cannot be used as a conclusive argument to prove the 

 absence of photoreception in the cord. 



The muscles of one side of the body were next removed from a 

 fairly large region anterior to the caudal fin, leaving the cord partially 

 exposed. Uninterrupted light was directed upon the column and cord. 

 A response occurred but it was not quickened, as one would expect if 

 the photoreceptors were located within the cord ; on the contrary it was 

 slower by 28.5 seconds than the response of normal fishes. It is signifi- 

 cant that this retardation is just the same as that which the "meat 

 screen" caused in experiments where the lateral body wall was not re- 

 moved. It indicates that in the present experiment the light passed 

 through the cord without affecting it; but that, after traversing the 

 body muscles of the intact side, it stimulated the integument of that 

 side from within. This suggests that the reactions of the blinded fishes 

 were due to photic stimulation of the integument only. 



In order to further check this result, the muscles and integument of 

 both sides of the body were removed and the cord and column were 



