THE VASCULAR CRYPTOGAMS 37 



homologies of the seed. One part of the seed, the testa, 

 is not represented in Selaginella, for the megasporangium 

 is without integuments. The megasporangium itself 

 corresponds to the nucellus of an ovule, as we have 

 already seen, but it never develops into anything of the 

 nature of a seed. This is because the megaspores are 

 set free from the megasporangium before fertilisation 

 takes place, and therefore long previous to the develop- 

 ment of an embryo. The megaspore, when it is filled 

 with prothallus and contains an embryo, bears a certain 

 resemblance to a seed, but there is no true homology ; 

 for, as we have already seen, the development shows that 

 the megaspore is homologous with the embryo-sac only. 



The seed, in fact, is a new development, peculiar 

 to Flowering Plants ; it corresponds to a persistent 

 megasporangium containing a single megaspore, which 

 has produced a prothallus and, after fertilisation, an 

 embryo, while still in situ', the shedding of the seed 

 would correspond to the detachment of the whole 

 megasporangium. Something of this kind takes place 

 in certain Cryptogams allied to the Ferns, though not in 

 Selaginella. 



TYPE V 



THE MALE FERN (Aspidium Mlix-Mas, L.) 



The Ferns are a vast group, enormously outnumbering 

 all the other Vascular Cryptogams put together. The 

 order in the widest sense includes at least sixty genera 

 and three thousand species. In our own native Flora 

 seventeen genera and about forty species are represented. 

 The limits of both genera and species are, however, very 



