104 SCIENCE PROGRESS. 



bell. After a short course in this direction these elongated 

 cells bend at right angles and form together with similar 

 cells proceeding from other sense-organs a circular band of 

 fibres, which completes the circuit of the bell. In connec- 

 tion with these circular and radial nerve tracts there are 

 bundles of muscular fibres arranged in a regular and rather 

 complicated manner, and there seems to be little doubt that 

 the rhythmic contraction of these fibres causes the rhythmic 

 contraction of the bell as a whole. It cannot be said that 

 Hesse's observations on these histological points are alto- 

 gether novel. Claus, Eimer, Schafer and others have 

 partially described them before, but with the help of more 

 modern methods of research Hesse has been able to give 

 us a more complete and satisfactory description than any of 

 his predecessors. 



Without referring to further details of the distribution of 

 this important nerve plexus we may now, in view of the 

 physiological experiments to be described, consider it to be 

 proved that the so-called nerve sheath of the inner sensory 

 pit of the tentaculocysts is continuous with bundles of fibres 

 — or elongated bipolar ganglion cells ; that these fibres run 

 radially from the margin and then circularly to form a broad 

 tract, and that finally these tracts are confined to the sub- 

 umbrellar aspect of the bell. 



It can be shown that if semicircular incisions be made 

 through the subumbrellar epithelium round each of the 

 tentaculocysts, so as to cut through radial tracts, the effect 

 is the same as if the tentaculocysts were entirely removed, 

 that is to say, the rhythmic contractions cease. Moreover 

 if the epithelium of the base of the tentaculocyst itself be 

 severed the effect is the same. On the other hand if the 

 epithelium on the upper side of the ball be cut in a similar 

 manner and with it the subjacent mesogloea the contrac- 

 tions continue, so long as the subumbrellar epithelium is 

 intact. 



These experiments prove then conclusively that the 

 nervous impulses which stimulate and regulate the move- 

 ments of a medusa traverse the subunibrellar epithelium 

 only in definite directions, which correspond with tracts of 



