452 PART III. — THE CLASSIFICATION OF PLANTS. 



The degree of development attained by the endosperm in Angiosperms is 

 various. Whilst, as a rule, it completely fills the embryo-sac, leaving room, 

 however, for the embryo, in some cases it occupies but a portion of the 

 embryo-sac, as in the Coco-nut, where it forms a thick parietal layer ; or, as in 

 Nymphaea, Nuphar, Anthurium, Viscum, Lathraea, Thesium, Khinanthus, etc., 

 where the development of endosperm is confined to the upper half of the 

 embryo-sac. In some cases the endosperm is rudimentary, being represented 

 merely by a number of nuclei, as in Tropaeolum, Trapa, Naiadaceae, Alismaceae, 

 Orchidaceffi ; and in Canna even this rudimentary development is wanting. 



The antipodal cells do not, as a rule, undergo any further development, but 

 in some cases {e.g. some Graminaceae) they have been observed to divide and 

 give rise to a considerable mass of cells. 



The female organ is essentially an archegonium. In most 

 Gjmnosperms it is actually an archegonium, like that of the 

 Pteridophyta ; it is developed from a single superficial cell of the 

 prothallium at the micropylar end, and lias a neck, containing a 

 canal-cell, leading to a ventral cavity in which lies the female cell 

 or oosphere; but in Welwitschia the archegonium is reduced to a 

 single cell with a cell-wall ; and in Gnetum to a single cell with- 

 out a cell-wall, the oosphere. In the Angiosperms the female 

 organ resembles that of Gnetum, in that it is reduced to a single 

 naked cell : the three cells constituting the egg-apparatus repre- 

 sent each an archegonium reduced to a single cell ; but in one 

 only is this cell a true fertile oosphere, the other two (the syner- 

 gidiB) being sterile as a rale. 



In a few plants the egg-apparatus consists of but two cells (the rule in Santa- 

 linn album ; occasionally in Ornithogalum nutans, Sinningia, Gomphrena, some 

 Orchids such as OrcJds latifolia, Cypripedium Calceolu!<, Gymnadeaia Couopsea, 

 some Mimoseae) : it would aj^pear, that, in such a case, there has been but one 

 nuclear division in the micropylar end of Ihe embryo-sac. Both the cells are 

 susceptible of fertilisation, and are therefore both to be regarded as oospheres. 



In some plants (Crocus, Gladiolus, Torenia Fig. 285, Santalum) the syner- 

 gidae become elongated, and penetrate the wall of the embryo-sac ; their project- 

 ing ends are longitudinally striated and are covered with a coat of cellulose : 

 these structures are sometimes spoken of as the Jilifi)rni apparatus. 



In Casuarina, the cells of the egg-apparatus (when fertile) are surrounded by 

 cell-walls (like the oosphere of Welwitschia) : no antipodal cells are developed. 



Follination. In view of the fact that the female cell (oosphere), 

 and the prothallium bearing it, i-emain (as a rule) permanently 

 enclosed in the macrospore, and that the macrospore remains en- 

 closed in the indehiscent macrosporangium, it is clear that the 

 process of fertilisation can only be effected when the microspore 

 germinates in immediate proximity to the macrosporangium. 



