CONTRIBUTIONS TO THE ANATOMY OP THE HIRUDINEA. 495 



larity of structure throughout the group, and that structure is 

 of a most highly developed nature. As in the higher Annelids 

 and Arthropods there is a complete separation of ganglion cells 

 from nerve-cord, the number of ganglia is the same in every 

 genus, although this is not the primitive number, as is con- 

 clusively shown by the fact that certain ganglia fuse together 

 during the development. 



The posterior sucker exists and presents very minutely 

 similar structure throughout the group. 



The pharyngeal wall, whether so greatly modified as a whole 

 as to be capable of withdrawal into the anterior region of the 

 body, as in the Rhyncobdellidac, or remaining as the anterior 

 portion of the body, as in the Gnathobdellidae, presents, as I 

 have shown, a close agreement in its minute histological detail 

 throughout the group. 



The nephridial funnel, although differing entirely in position, 

 while the nephridium itself presents the two extremes of 

 variation in Pontobdella and Clepsine, shows an identity 

 of essential structure in these two genera. 



Although Clepsine differs as much as any leech from 

 Hirudo, Aulostoma and Nephelis and the nephridium of 

 Clepsine on the one hand and of these three genera on the 

 other presents great differences in complication, yet the main 

 duct has almost precisely the same course and re-enters (in itself 

 a most specialised condition) the cellular masses or lobes through 

 which it has already passed at precisely similar spots. At the 

 same time the nephridial funnel of Clepsine as stated above 

 resembles that of Pontobdella and differs widely from that of 

 Hirudoj there is, moreover a well-developed vesicle in 

 Hirudo and no trace of such in Clepsine. 



Passing, in the second place, to consider specially the varia- 

 bility of structure which may occur we may note the following 

 facts : 



The anterior sucker, present in Pontobdella, Piscicola, 

 and Branchellion is as little represented in their ally Clep- 

 sine as it is in the members of the Gnathobdellidae. 



The number of annuli which build up a somite, although 



VOL. XVIV NEW PER. L L 



