140 Elementary Biology. 



they might well be, on the same thallus inside the same 

 spore-wall. 



Again, how must we account for the disappearance of 

 the thallus which ought to have been formed by the micro- 

 spore, and the retaining of the thallus formed by the 

 macrospore ? Obviously when the cells which will produce 

 sperms have been formed there can be no further use for a 

 thallus bearing them only ; whilst not only has the thallus 

 producing ovaria to give origin to sexual cells, but it has also 

 to nourish for a time the embryo that results from the ferti- 

 lisation of the ovum. The thallus, in short, has to act as 

 nurse to the fertilised ovum it has itself given birth to. 

 Hence its persistence for a variable period dependent on the 

 time taken by the embryo to develop. Moreover, we saw 

 that the tissue in the interior of the macrospore is divisible 

 into what has been termed thallus and endosperm. This 

 latter would apparently correspond to the purely vegetative 

 portion of the thallus of a fern, whilst the thallus proper is 

 that portion which gives rise to the ovaria. It is possible, 

 however, that the endosperm is an altogether new formation 

 which has arisen during the gradual assumption of para- 

 sitism by the sexual generation, and for the special purpose 

 of affording nourishment to the embryo during its develop- 

 ment whilst inside the spore-wall. 1 



1 Although speculations without detailed proofs are in many cases 

 looked on with suspicion (and rightly so), yet they do so much towards 

 relieving detailed description from the charge of dulness that no further 

 apology is given for their introduction here. In the present instance, 

 when one considers the fact that the unisexual thallus, whether formed 

 from a microspore or a macrospore, must have, at one time in phylo- 

 genetic history, been a hermaphrodite structure, and that the spermaria 

 are developed on a different part of the thallus from the ovaria ; that, 

 moreover, the presence of root-hairs on the exposed ovaria-bearing 

 portion of the enclosed thallus of Seiiginella clearly points to that being 

 the morphologically ventral surface of the thallus, one cannot but sus- 

 pect that the endosperm really corresponds to the spermaria-bearing 

 portion of the hermaphrodite thallus. The division of the nucleus of 



