Morse, Behavior of Gojjionemus. 45 1 



the water and parallel to it. The medusas swam against the plate 

 exactly as in the case of the oil. The surface was then covered 

 with kerosene, which forms a very delicate film. In this case the 

 result was the same as in the case of pure water; the bells were 

 inverted. In order to show that it is not a matter of lack of 

 oxygen, a column of ammonium chlorid (NH4CI held in suspen- 

 sion in water vapor, formed by adding a few drops of HCl to a 

 similar amount of ammonia), was superposed on the surface of the 

 water with the result that the animals continued to invert the bells 

 and passively sink. 



Yerkes has shown that most probably the margin of the bell 

 which leaves the water receives a stimulus, by so doing, to contract, 

 inasmuch as that portion of the bell must contract earlier than the 

 other portions in order to cause the inversion. The writer has 

 found that the mechanism of varying the direction of swimming is 

 that of contracting one side of the bell more strongly or in advance 

 of the opposite side. The result is a more or less complete rota- 

 tion of the bell on a transverse diameter as an axis. Neither the 

 tentacles nor the velum exert any influence on the direction of 

 swimming, since when the tentacles are cut away from one-half of 

 the bell, the medusa swims either in a straight line or curved one as 

 before. Likewise, one-half of the velum was removed without 

 appreciable result and even cutting away completely both tentacles 

 and velum was without effect on the swimming movements as far 

 as direction is concerned. By paralyzing one side with a crystal 

 of silver nitrate which at the same time left a deposit of metallic 

 silver as a point of reference, the medusa swam in a circle with the 

 injured side on the inside of the circle. By cutting two windows in 

 the bell, near the apex and on one side, the pressure of the water 

 was removed from that side and the path in swimming was again 

 circular with the intact portion peripheral. 



Yerkes would account for the inversion of the bell by the 

 stimulus given to the exposed portion by light. It is difficult to 

 see why light affects only the exposed portion of the bell and 

 not the other parts. That it is not a matter of light striking one 

 side of the bell more strongly than the other is made evident by the 

 fact that if the light is admitted parallel to the surface of the water, 

 the invariable result obtains that the portion of the bell first striking 

 the water contracts first, although the opposite side may be toward 

 the light. The cause for reaction is not evident. The writer has 



