222 THE SCOTTISH BOTANICAL REVIEW 



escape competition. The arrest of the cotyledon is found to be 

 confined to one aquatic dicotyledon. The others are geophilous. 

 The sheathing leaf-base is common to both groups, and is correlated 

 with suppression of internodes. The absence of a cambium and 

 substitution of concentric rings of growth may be due to the number 

 of broad-based leaves inserted on a short axis in the case of under- 

 ground aquatics, e.g. NymphcEa. 



In Peperomoia Mr. A. W. Hill found one cotyledon, but there 

 were probably originally two, and one is hypogeal, acting as a 

 sucker j the other looks like and performs the function of a true 

 leaf. He extends this theory to monocotyledons generally. It is 

 geophilous, with stem-structure like Alisma. 



Then we have Miss Sargant's own theory in which fusion of 

 cotyledons is correlated with a geophilous habit. Suppose the 

 primitive Angiosperm were geophilous. The green parts of the 

 seedling would be reduced in early seasons and fusion would result. 



Podophyllian affords an example of a plant with united cotyledons 

 and scattered stem bundles which is a geophyte. 



The suggestion is thrown out that the glacial period may have 

 produced conditions most favourable to geophytes and afforded 

 scope for monocotyledonous development. 



In attempting to reconstruct the character of a race of primitive 

 Angiosperms^ Aliss Sargant discusses the question of monocotyledons 

 in a new light. The development from Pteridosperms, first as a 

 pro-anthostrobilus, then as a eu-anthostrobilus, up to the primitive 

 Magnolia and Liriodendron, as recently sketched out by Messrs. 

 Arber and Parkin,- is accepted, and she reconsiders the evidence for 

 regarding Angiosperms as monophyletic, and the reconstruction of 

 the pro- Angiosperm. As to the first, the community of descent of 

 monocotyledons and dicotyledons is proved by the many characters 

 they share in common. And it is pointed out that the angiospermic 

 flower is unique. The homologue of the flower, carpel, or endosperm 

 is by no means settled. 



The germination of the embryo-sac and formation of endosperm, 

 where the first nuclear division, origin of endosperm from three 

 nuclei, and the uniform nature of the two processes is so characteristic, 

 supports the monophyletic origin. Double fertilisation is by no 

 means uncommon. The history of the development of the endo- 

 sperm is not known. Commenting on Miss Berridge's suggestion 

 as to the resemblance between the endosperm of Ephedra and that 

 of Angiosperms, Miss Sargant says : " No tissue then in Gfietu??i, 

 nor, so far as we know, in IVeiwitschia, can be considered as the 

 direct representative of the angiospermic endosperm " ; '^ and she 



1 "The Reconstruction of a Race of Primitive Angiosperms," Ann. Bot. , 

 1908, vol. xxii. p. 121 et seqq. 



-"On the Origin of Angiosperms," Jour. Linn. Soc, 1907, vol. xxxviii. 

 p. 29. See further, Dr. D. H. Scott, "The Evolution of Plants," 191 1, Williams 

 & Norgate. 



^ See, however, Professor Pearson's recent researches on Welwitschia. 



