232 THE SCOTTISH BOTANICAL REVIEW 



Structures in his investigations, and there he obtains the full history. 

 One is specially grateful to both of these able investigators for thus, 

 in their several fields, enabling us, by their combined researches, to 

 study both early and late stages. However much we may differ in 

 theory from either one or the other, the data upon which they have 

 been based lose none of their value whatsoever. In some respects 

 the plants Miss Sargant studied were anomalous forms, e.g. Pseudo- 

 monocotyledons, and primitive Liliaceae, and the^i evidence is thus 

 inconclusive. The lateral appendage view only leads back again to an 

 unnecessary transposition, in time, of the mono- and di-cotylar types. 



The interpretation of certain features, e.g. the absence of a 

 cambium, as due to reduction, instead of being held as a primitive 

 character, is a necessary consequence. 



There may really be some truth in Professor Lyon's view that the 

 fission of the cotyledons has to do with the freeing of the cotyledon 

 from the testa. 



Professor G. Henslow's theory that one cotyledon has been 

 suppressed in monocotyledons is based upon the preponderance of 

 aquatic orders m monocotyledons — and thought to be due to their 

 adaptation to an aquatic habitat, a physiological cause, which is not 

 universally applicable. 



In so far as it touches the origin of monocotyledons Messrs. 

 Arber and Parkin's paper is favourable to the view that a primitive 

 Angiosperm gave rise very early to both mono- and di-cotylar types, 

 the divergence being close to the origin of the primitive ancestor. 



Bessey's views seem to us to be so reasonable that, taken in con- 

 junction with Dr. Worsdell's, they lie nearest the truth. 



It is a remarkable fact that the earliest forms of both monocoty- 

 ledons and dicotyledons exhibit a sort of parallelism of development. 

 The first monocotyledons are apocarpous, just as are Ranunculaceae, 

 amongst the latter (except Nymphseaccce, which are epigynous). 



Then there is a higher series which becomes hypogynous, from 

 which the Calyciflorse early diverge and are perigynous or epigynous. 



So in monocotyledons the apocarpous forms (rarely petaloid) 

 become syncarpous, and then spadiceous or glumaceous, or hypo- 

 gynous, and finally epigynous. 



The researches of Professor Pearson and Miss E. L. Stevens as to 

 the embryology of Welwitschia and Penaeaceae favour the derivation 

 of Angiosperms from Gymnospermic or Bennettitean stock, but, 

 as yet, throw no light upon the Angiospermic origin nearer the 

 present time, and the differentiation into mono- and di-cotyledons. 



Speaking generally, it seems to us that the monocotyledons and 

 dicotyledons have diverged from a common primitive Angiospermous 

 stock. The earlier types of the former appear to belong to 

 Alismaceae, leading on to Liliaceae ; and the earlier forms of the latter 

 the Ranales, represented by Magnolia and Liriodendron, from which 

 Ranunculus and its aUies are later derivatives. This would allow of 

 the retention of some primitive characters in both groups which are 

 apparently explained only by supposing dicotyledons derived from 



