THE VASCULAR CRYPTOGAMS 33 



almost exactly like that of a Gymnosperm, such as Picea, 

 and which develops in just the same way. The arche- 

 gonia of Selaginella are formed from superficial cells of 

 the prothallus, exactly as those of the Fir are formed 

 from superficial cells of the endosperm. The develop- 

 ment, moreover, of the prothallus itself is just like that 

 of the endosperm. In both cases free nuclear division 

 first takes place, then cell- formation begins, and the tissue 

 thus formed goes on growing until it has filled all the 

 available space. We can have no doubt, then, that 

 the female prothallus of Selaginella corresponds to the 

 endosperm of the Fir, which we may, if we like, call 

 by the same name. The only difference is that, in 

 Selaginella, the prothallus grows a little way out from 

 the megaspore, becomes partly green, and forms a few 

 root-hairs. It is, in fact, a more independent structure 

 in the case of the Cryptogam, developing freely on 

 the ground, instead of within the closed tissues of the 

 ovule. 



Now the cell in which the prothallus of Selaginella 

 develops is the megaspore, while that in which the 

 endosperm of Picea arises is the embryo - sac. We 

 therefore arrive at a new term in the comparison ; 

 the megaspore is the equivalent of the embryo-sac. 



There are some differences, however, to be dealt with 

 at this point : in normal cases the megaspore is set free 

 and completes its development on the ground, while the 

 embryo-sac remains always enclosed in the ovule or seed. 

 We must remember, however, that the megaspore itself 

 begins its germination while still in the rnegasporangium, 

 and, in exceptional cases, even the embryo-plant may 

 be developed in this position. The thick cuticular- 

 ised wall of the megaspore is obviously a necessity 

 for its protection when it becomes freely exposed. It 

 3 



