THE PAST HISTORY OF MONOCOTYLEDONS 223 



thinks "the germination of the embryo-sac and the history of the 

 endosperm isolate monocotyledons and dicotyledons from all other 

 plants." The flower also is distinct from other "flowers," and 

 characteristic. 



In the reconstruction of the primitive Angiosperm Miss Sargant 

 relies on the evidence adduced by Arber and Parkin as to the form 

 and character of the primitive flower, with angiospermic foliage, 

 flowers, seeds, and ovules enclosed in an ovary, three nuclei in the 

 pollen-tube, etc. The flower was elaborate in structure, with sepals, 

 stamens, carpels, spirally arranged in acropetal succession, as in 

 MagtioUa. 



The early types were of the form of a bisexual strobilus as in 

 Magnolia, with o sporophylls below, 5 above, protected by a series 

 of sterile bracts forming a. basal perianth. Such a type is found in 

 the Jurassic Bennettites. The immediate ancestor had an antho- 

 strobiloid flower, with perianth leaves, stamens, carpels, arranged 

 acropetally. The strobilus was the primitive form, and Bentiettites 

 stands midway between it and Magnolia. 



Monocotyledons and dicotyledons differ in stem-anatomy and 

 number of cotyledons. How did the primitive Angiosperm differ 

 from or resemble either ? As to stem-anatomy, monocotyledons do 

 not possess a cambium, or rarely. Gloriosa, however, exhibits 

 collateral traces and a cambium in second-year tubers. Gymno- 

 sperms exhibit cambial activity, and so did many fossil vascular 

 cryptogams. Dicotylar seedlings early exhibit a cambium. Some 

 monocotyledons, as Yucca, etc., do. Thus evidence points to the 

 primitive Angiosperm possessing a cambium. As to the number of 

 cotyledons, monocotyledons possess one, dicotyledons two. No 

 monocotyledon possesses two, but several dicotyledons possess one. 

 Many groups of Ciymnosperms are dicotylar, and the most primitive 

 Araucarieae are polycotyledonous. In what respect did the primitive 

 Angiosperm differ from or resemble either recent group? Analogy, 

 as above, favours the primitive nature of the dicotylar type. What 

 do we learn from embryology within the embryo sac? The whole 

 of the evidence in this connection centres around the position of the 

 cotyledon. The terminal position of the monocotyledonous form 

 with the derivation therefrom by fission of the two is not favoured 

 by Miss Sargant. Can a terminal cotyledon be a true leaf? 

 Opinions differ as to this. If extensions of the axis, they are 

 terminal and not true leaves. The difficulty is thought to be 

 avoided by considering the cotyledon as only apparently terminal, 

 and so lateral. 



In Alisma (PI. IV. figs. 1-6) the cotyledon is terminal at first, 

 then lateral as the plumule displaces it. In Pistia and Sparganiuvi 

 early leaves are terminal. Are they cauline? Or are they due to 

 suppression of the axis? They are primitive forms and aquatic. 

 The stem exists in the mature Pistia, but not in the embryo. The 

 reduction in stem is due to the aquatic habit, and causes terminal 

 leaves. Can it cause the cotyledon to be terminal ? The form of 



