Ill 



of the corresponding nerves. The difficulty is to determine for certain 

 how many muscles answer to each nerve ; or, in other words, whether 

 the motor supply of adjacent segments overlaps. For this purpose a 

 better method than that described above may be used. 



A vertical cut is made along the fin, passing at right-angles to 

 the radial muscles. The cut ends of the muscles can then be seen 

 in section as a regular dorsal and ventral series of squarish blocks 

 on either side of the median skeletal radials. On stimulating a given 

 nerve the corresponding muscles are easily observed to draw in and 

 out on contraction and relaxation. By this method there is less dan- 

 ger of mistaking mere drag for active contraction. The results are 

 as follows: At least two dorsal and two ventral muscles respond to 

 each nerve. It is the corresponding dorsal and ventral muscles above 

 and below which contract on the stimulation of a nerve. 3 dorsal and 

 3 ventral radial muscles usually respond to each nerve; this appears 

 to be the normal number of muscles supplied by each spinal nerve ^), 

 It sometimes happens in the case of the dorsal series, and more rarely 

 in the case of the ventral series, that only two muscles can be made 

 to contract on the stimulation of a nerve. This conclusion is based 

 on a very large number of trials. Only three times have four muscles, 

 above or below, appeared to contract with the electric stimulus; and 

 in these cases the result was not confirmed on the application of a 

 mechanical stimulus. Never with the mechanical stimulus have I ob- 

 tained certain evidence of the response of more than three pairs of 

 muscles to one nerve. On the other hand, by the stimulation, both elec- 

 trical and mechanical, of 8 consecutive nerves I have obtained the per- 

 fectly regular contraction of the muscles, three at a time, all along 

 the series of 16 2). 



The final conclusion is, therefore, that Mollier and Beaus were 

 right in thinking that the adult radial muscles are to some extent 

 mixed, and that the bridges connecting the base of the muscle buds 

 at a certain stage of their development really represent a migration 

 of muscle forming substance from one segment to another. But the 

 mixing in Raja appears to be very slight, so that the nerves meet 



1) Very often two of these muscles respond much more readily 

 than the third. 



2) In these experiments the stimulus was applied at the points 2 

 and 3 of Braus (1. c, p. 538, Fig. 3), and intei'mediate points, with the 

 the same result. 



