THE TUEMATODE PARASITES OF FISHES. 489 



111 the prosostomate distonies we are accustomed to regard the relative 

 position of the genital glands as constant within narrow limits for the 

 same species, and any such difference as exists between P. squamatus and 

 P. acideatus would be sufficient to warrant generic separation. In the 

 Gasterostomata, however, it is evident that one cannot regard this feature 

 as a satisfactory basis of classification, and recourse must be had to others 

 of a more constant nature. Odhner has already (1905) denoted the 

 chief of these, namely, the structure of the copulatory organs, the struc- 

 ture of the head and the disposition of the yolk glands. 



Apart from the situation of the genital glands and the configuration 

 of the uterus, the species included in the genus Prosorhynchus appear to 

 form a homogeneous group. The same, however, cannot be said with 

 regard to the remaining species of Gasterostomes, included by Odhner 

 under the genus Gasterostomum { = Bucephalus). 



They all agree in having the yolk glands arranged in two distinct 

 groups, which are usually marginal in position, and, so far as is known, 

 the structure of the copulatory organs does not vary very much from the 

 type found in Bucephalus polymorphus {= Gasterostomum -fimhriatum). 

 It is in the structure of the anterior end that we meet with the most 

 pronounced features of difTerence. Three main types may be recognised, 

 (i) The anterior end may be provided with a simple sucker as in Gastero- 

 stomum gracilescens and G. tergestinum. This sucker closely resembles 

 the ventral sucker of the prosostomate distomes, and is regarded by 

 Odhner as the primitive type of head structure in the Gasterostomes. 



(2) From the sucker muscular prolongations may grow out in the form of 

 tentacles or fimbriae as in G. fimhriatum and G. trmiimmn Stossich. 



(3) The sucker may degenerate in musculature, become very shallow and 

 be surmounted by a contractile fan-shaped hood as in G. triglae and G. 

 viperae. It is apparent that some generic separation of these three 

 groups is desirable, and it is only on the structure of the anterior end that 

 this is practicable. Each of these three groups has already, in earlier 

 literature, been regarded as of generic or at least subgeneric importance. 

 The synonymy is slightly complicated. 



The monotypical genus Gasterostomum was founded on G. fifubriatum, 

 and as this has been shown to be the adult of the earlier known Bucephalus 

 polymorphus the name Gasterostomum fimhriatum must be regarded as 

 nomen nudum and the genus Gasterostomum as a synonym of Bucephalus. 



In 1855 Diesing erected the sub-genus Bucephalopsis for the larval 

 form B. haimeanus, and this has been shown to be the larva of Gastero- 

 stomum gracilescens (Rud.). This species is undoubtedly the type of a 



