339] PSEUDOPHYLLIDEA FROM FISHES— COOPER 51 



and the terminal excretory vesicle is also comprehensible in the light of this 

 method of development, for, since the nerve strands are situated close outside 

 the lateral excretory vessels at the constrictions, they simply turn in towards 

 the median line and unite immediately ahead of the junction of the latter with 

 the median vessel. 



As will be gathered from the foregoing description there is a most remark- 

 able resemblance between the scolex of H. globuliforme and that of the Try- 

 panorhyncha not only in the structure of the proboscides but also in the pre- 

 sence of the large mass of ganglionic cells associated with them posteriorly. 

 Each proboscis consists of three parts: (1) a hollow tentacle, capable of eva- 

 gination, (2) a short permanently protruded stump, armed with thickly set 

 minute, cuticular spines, and (3) a comparatively elongated bulb. Of these 

 parts (1) and (3) may be compared respectively with the proboscis and the 

 bulb of Tetrarhynchus or Rhynchobothrius. The proboscis, altho not pro- 

 vided with any kind of armature, is nevertheless supplied with a group of well 

 developed retractor muscles which are evidently analogous at least to the single 

 retractor muscle of the Trypanorhyncha. The bulb is not only provided with 

 a musculature arranged so as to diminish on contraction the volume of the 

 organ, but is also lined with an epithelium-like layer comparable to that of the 

 members of the latter group. But since the bulb extends to the point of exit 

 of the proboscis, there is no part corresponding strictly to the proboscis -sheath 

 of Tetrarh)mchus altho the stump would at first sight seem to be such. Fur- 

 thermore, the cells forming the large mass behind the bulbs in Haplobothrium 

 which are here interpreted as ganglionic cells, bear not a little resemblance to 

 those described by Braun (1896:1294) after Pintner (1880), Lang (1881) and 

 Niemiec (1888) as associated with the bulbs of Tetrarhynchus longicollis (v. 

 Ben.) { = Dibothriorhynchus ruficollis Monticelli) and considered by some to be 

 ganglionic cells and by others myoblasts. The distribution of the large nerve 

 trunks arising from the nerve conamissure is also somewhat suggestive of con- 

 ditions in a few of the tetrarhynchids (cf. Braun 1896:1293). 



While the writer is not prepared to go further into this comparison he would 

 like to emphasize the significance of the layers composing the walls of the 

 bulbs in H. globuliforme in connection with the possible origin of these most 

 aberrant structures. In discussing the homologies of the proboscides cf the 

 Trypanorhyncha Benham (1901) said: "It appears more probable (Pint- 

 ner) that each proboscis has been developed by the deepening and modification 

 of an ' accessory sucker' of some Tetraphyllidean as its relation to the bothridia 

 and its mode of development closely agrees with these structures. Func- 

 tionally too it is a perfection of the armature plus the accessory sucker of 

 three forms [Acanthocephala, Nemertini, and Taenioidea]; whilst there is 

 no doubt that the 'phyllidea' of the orders are identical. " The fact that here 

 the walls of the bulb, since they are composed of an outer layer of longitudinal 

 muscles, a middle layer of circular fibres and an inner cuticular layer are not 

 only comparable but directly continuous with the cuticula and cuticular mus- 

 cles of the body wall and in the reverse order would seem to lend support to 



