of Oligochoetous Annelids. 97 



tenth segment : the former are the testes, the latter the ova- 

 ries. There is only one pair of testes, not two or three as 

 supposed by Clapar^de, Avho, I imagine, did not examine the 

 youngest speciiuens. In the minutest details of structure the 

 ovary and testis are at this period identical, consisting of 

 nuclei scattered in a common protoplasm. The testicular 

 masses segment, forming groups of nucleated protojalasm, each 

 nucleus of which gradually developes around it a demarcated 

 area. The cells thus formed have the exact structure of the 

 young ova. At this point their development diverges ; for 

 whilst the ova increase in size individually without prolifera- 

 tion, the young sperm-cells exhibit most active multiplica- 

 tion by division of their nuclei into two, three, and four, 

 thus forming floating spherical or compressed aggregates of 

 young sperm-cells. The further development of these I have 

 carefully traced in several genera of Oligocha^ta. Several 

 phases appear subsequently in the development of each mass 

 of sperm-cells, which have not hitherto been described, and 

 require illustration. 'At one period -in the development of the 

 sperm-masses (the protoplasmic masses which give rise at 

 their periphery to S] )erm-cells) of Limnodrilus the whole mass 

 has a tendency to fibrillate into zoosperms ; and some of these 

 masses assume elongated forms far thicker than normal zoo- 

 sperms, and exhibit both protoplasmic contractile movements 

 and the flickering motion of a cilium. This fact has a 

 special interest in demonstrating the identity of ciliary and 

 amoeboid movement, of which Hiickel has lately written 

 (Biologische Studien, 1870). The innumerable spermato- 

 spheres which are thus developed from the original pair of 

 testes fill the segment in which they are formed, and also 

 dilate certain folds of the peritoneal membrane in con- 

 nexion with the septa which separate the ninth from the 

 adjacent segments ; and thus a sheath is formed for these 

 rapidly multiplying floating corpuscles. One thing is quite 

 certain, that this sheath is not part of the original testis, and 

 that at flrst the spermatospheres float freely in the perivisceral 

 cavity, as I have figured them in Chadogaster limnan. The 

 sheath is in all probability only a part of the dissepiment be- 

 tween the ninth and tenth fasciculate segments ; and it is 

 pushed down, as described by Clapar^de, through several suc- 

 ceeding segments as the spermatic elements increase in num- 

 l)er. This occurs equally in Nais. It also frequently happens 

 that a similar sheath extends forwards, distended with spermato- 

 spheres detached from the pair of testes. It will be observed 

 that this description difl:ers from that of Clapar^de chiefly as 

 to the position and character of the original testes. The large 



