354 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



upon these structures as parasites, but a closer examina- 

 tion of their development has led me to suggest that two 

 kinds of spermatozoa are present. On this understanding 

 the same terminology will be adopted in the following 

 description as was used above in describing the develop- 

 ment of the normal spermatozoa. 



The central blastophore from which the spermato- 

 cytes are constricted off is larger, and more irregular in 

 outline, than that on which the normal spermatozoa are 

 formed. The spermatocytes themselves, too, are less 

 regularly arranged, often tending to be formed in groups 

 of four or five together (PL Y, fig. 36). The nuclei of the 

 spermatocytes stain very deeply. The changes taking 

 place during metamorphosis can be easily followed. The 

 spermatocytes, which are at first rounded, become oval in 

 shape, the outer free end being considerably more pointed 

 than the opposite end, which is embedded in the blasto- 

 phore (PL V, fig. 37). The spermatids formed by 

 division of the spermatocytes are much larger and less 

 numerous than those developed from a normal sperm 

 polyplast, and their arrangement on the blastophore is as 

 irregular as that of the spermatocytes from which they 

 are derived (PL VI, fig. 38). In the earlier stage of 

 metamorphosis there is no semblance of a true tail, but 

 when the head has lost its oval form and becomes much 

 elongated, the tail is gradually differentiated. At first it 

 is quite short, but it increases in length as the head of the 

 spermatozoon becomes fully developed, and in the final 

 stages the tail is as long as, or a little longer than, the 

 head (PL VI, fig. 39). The mature spermatozoa are quite 

 irregularly arranged, but they tend to lie together on the 

 blastophore in bundles of five or six, just as the 

 spermatocytes have been described as doing, and from it 

 they finally separate. In some cases I have been able to 



