226 THE SCOTTISH BOTANICAL REVIEW 



that the longitudinal division of the embryo takes place before ^ the 

 formation of transverse walls in dicotyledons, and afterwards ^ in the 

 monocotyledons. To this rule, however, there are numerous 

 exceptions." 



Commenting on the further progress of this division, he says, "It 

 is at least an interesting coincidence that in the young embryo the 

 undivided foliar cells give rise to the single cotyledon, and the epi- 

 basal cell to the pair of cotyledons." 



The following remarks are so a propos that they are quoted at 

 length : — 



"Are there any indications which may help us to answer the 

 question of the origin of these two groups? Have dicotyledons been 

 derived from monocotyledons, monocotyledons from dicotyledons, 

 or both from a common ancestor? It must be admitted that on 

 theoretical grounds it is no more difficult to pass from two cotyledons 

 to one, than from one to two. Indeed, there have been not a few 

 botanists who have suggested the derivation of the monocotyledons 

 from the dicotyledons. When, however, one compares the two embryos 

 there is a slight preponderance in favour of the view that the struc- 

 ture is a little higher in dicotyledons than in monocotyledons. The 

 row of undivided cells in the embryo of the monocotyledon after the 

 third or fourth segmentation is certainly a lower structure than the 

 compact mass of cells constituting the ' octant stage ' of the dicoty- 

 ledonous embryo. The cotyledons themselves afford a slight sugges- 

 tion as to the relationship of the two groups. It is a well-established 

 principle in embryology that embryonic stages of higher organisms 

 resemble the adult stages of the organisms which are lower in the 

 same genetic hne. Applying this principle to the cotyledons, we 

 observe that while they bear some similarity to the leaves of both 

 monocotyledons and dicotyledons, the similarity is a little more 

 marked in the case of the monocotyledons. Compare the mostly 

 sessile, often clasping, usually elongated leaves of monocotyledons 

 with the cotyledons of either class, and contrast these with the 

 mostly petiolated, generally not clasping, and usually broad-bladed 

 leaves of the dicotyledons." 



Bessey also refers to the fact which Sir J. Lubbock (now Lord 

 Avebury) points out, that the earlier leaves of a plant are generally 

 quite different from the later ones. Young plants are the most 

 alike, and they begin to get more unlike as they mature. So that it 

 is often only possible with seedlings to distinguish the subclass, 

 later the family, and then the genus, whilst until the flower is pro- 

 duced the specific characters are hidden. 



The deductions he makes from ontogeny are : — 



1. All Angiosperms are essentially alike. 



2. The two subclasses (monocotyledons and dicotyledons) appear 



to be modifications of a common type which diverged from 

 one another at an early period. 



^ Present writer's italics. 



