LIGHT REACTIONS OF THE MEDUSA GONIONEMUS. 361 



favorable light areas ? Although Yerkes does not hold it so, I 

 believe he has indicated one way in his observation : 17 " When 

 an individual in swimming about chances to cross from the sunlit 

 region into the shadow, it very quickly ceases swimming and 

 sinks to the bottom." In another connection but bearing out 

 the same point, Morse ls has this to say - - " being in motion 

 almost incessantly, and swimming in all directions, it is obvious 

 that sooner or later they will enter the dark area. Once having 

 entered this area, the stimulating effect of sunlight being cut off, 

 they remain as in a trap." Even if we do not hold that only 

 light stimulates, or that optimum light acts like a trap the ex- 

 planation may hold. 



More recently Morse 19 has described a way in which Gonio- 

 nenius may get from a location that is unsuitable with respect to 

 light to a more suitable one. A medusa was placed in the end 

 of an aquarium through which the sunlight was reflected hori- 

 zontally. The medusa swam to the surface in characteristic 

 fashion, each time bending its upward course a little farther from 

 the vertical, and therefore away from the source of light. Thus 

 ultimately it got to the farther side of the aquarium, into weaker 

 light. The promising feature of this explanation is that it is 

 based on the peculiar habit of the animal - - swimming to the 

 surface when disturbed. The other case that Morse records of 

 a " strong swimmer " moving directly toward the less illuminated 

 end of the aquarium I should consider an exception, as in repeat- 

 ing the experiment I have observed that some medusas move 

 almost directly to the lighter end after proceeding, by stages, to 

 the darker end. In my trials I have found that about 25 per cent, 

 get to the farther end of the aquarium by the method indicated, but 

 instead of its being a regular method of progression it seemed to 

 me to be characterized by irregularity. This may be due to the 

 fact that Morse seems to have worked with few individuals, 

 whereas I placed a number in the aquarium at the beginning of 

 the experiment and added to these as the experiment progressed, 

 so as to get results representative of more than individual be- 



17 Amer. Jour. Physiol., 1903, Vol. IX., p. 282. 



l *Jour. of Comp. Neural. Psychol., 1906, Vol. XVI., p. 454. 



19 Am. Nat., 1907, Vol. XLI., p. 684. 



