2 8 PLATYHELMINTHES TURBELLARIA chap. 



Nevertheless, in some Polyclads it is probable, and in Stylo- 

 chus neapolitanus it is certain, that normal copulation takes 

 place. The sperm-masses are transferred to a coecal diverti- 

 culum of the female genital canal, and then by a delicate 

 mechanism, of which we know only the effects, one sperma- 

 tozoon obtains entrance into one matured ovum, which differs 

 from the ova of most Turbellaria in that it contains in its own 

 protoplasm the yolk necessary for the nutrition of the embryo. 

 In other words, there are no special yolk-glands. After fertilisa- 

 tion, the ovum in all Polyclads is coated with a shell formed by 

 the shell-gland, which also secretes a substance uniting the eggs 

 together. They are deposited on stones and shells, either in 

 plate-like masses or in spirals (like those of Nudibranchs). 

 Cryptocelis alba lays masses of an annular shape, with two ova 

 in each shell, and buries them in sand. 



Development. 1 The first stages in the embryology of 

 Polyclads appear to be very uniform. They result, in all 

 Cotylea and in certain Planoceridae, in the formation of a 

 Miiller's larva (Fig. 12) about a couple of weeks after the eggs 

 are laid. This larva (1-1*8 mm. long), which is modified in the 

 Planoceridae, is distinguished by the presence of a ciliated band, 

 running somewhat transversely round the body, and usually 

 produced into a dorsal, a ventral, and three pairs of lateral 

 processes. When swimming the body is placed as in Fig. 12, 

 and twists round rapidly about its longitudinal axis by means 

 of the strong locomotor cilia placed in transverse rows upon 

 the processes. The cilia of each row vibrate synchronously, and 

 recall the action of the swimming plates of a Ctenophore. It 

 is noteworthy that whereas Stylochus pilidium passes through 

 a modified or, according to some authors, a primitive larval 

 stage, its near ally, S. neapolitanus, develops directly. Most 



1 A full account of Polyclad development is contained in Lang's " Polycladcn," 

 with references to the literature of the subject. Since the date of that work (1884) 

 the embryology of Ctenophora has become better known, but, though the segment- 

 ation of the egg and early stages of development are very similar in both cases, 

 the elaborate investigations of E. B. Wilson {Journ. Morphology, vol. vi. p. 361) 

 show that the segmentation of Polychaet worms is again similar. The question of the 

 affinities of the Polycladida is also discussed by Lang (" Polycladen," p. 642 et seq.). 

 The work of the last decade has neither proved nor disproved his suggestion that 

 the Ctenophores and Polyclads have been derived from common ancestors. On this 

 subject the remarks made by Hatschek (Lehrbuch d. Zoologie, p. 319) are some of 

 the weightiest that have appeared. 



