144 SEVENTEENTH REPORT. 



origin was assumed for both classes in an ancestral plexus of plants 

 very closely allied to and scarcely different from the apocarpous 

 Ranales, Alismatales and Rosales, which were made, respectiyelj, 

 the first orders in three divergent evolutionary lines. (PI. XIII 

 fig 1). This system has been worked out in greater detail in sub- 

 sequent publications by the same author.' The fundamental point 

 of this classification is the assumption that the most primitive type 

 of flower is that type most nearly representing a strobilus; i. e. 

 witli numerous separate sporophylls on a more or less elongated 

 axis. This requirement is met in the apocarpous flower of the 

 Ranales (e. g. Magnolia (PI. XIII, figs. 2-4). Ranunculus (PI. XIII, 

 figs. o-G) etc.) and of the Rosales (e. g. Rubus. Potentilla). both 

 belonging to the Dicotyledoneae, and of the Alismatales in the 

 Monocotyledoneae (Sagittaria, Alisma (PI. XIII, figs. 7-11), etc.) 

 This scheme of classification thus places as lowest in tlie evolu- 

 tionary scale in both Monocotyledoneae and Dicotyledoneae fonns 

 whose similarity had long ago been recognized.^ 



Others have also proposed classifications with the same general 

 features, viz. Ranales and Alismatales (Helobiae) as the beginning 

 groups of the Dicotyledoneae and Monocotyledoneae, respectively, e. 

 g. Delpiuo-', Hallier^'^', and Lotsy^^ These three authors suggest that 

 actually the Monocotyledoneae are directly derived from the 

 Ranales, a view which C. E. Bessey soon accepted. This is sup- 

 ported on anatomical grounds by Miss Sargent^-. This author 

 also comes to the conclusion that the apocarpous (i. e. strobilar) 

 type of fiower is the most primitive. In 1904 Fritsch^^ expressed 

 his idea that the apocarpous Monocotyledoneae are the most primi- 

 tive and derived from the vicinity of the Ranales in the Dicotyle- 

 doneae. 



In the discussions as to ihe phylogeny of the Anthophyta most 

 writers seem to have had in mind the herbaceous families as the typi- 

 cal ones through which the evolution of the various groups has passed. 

 The woody forms have been looked upon as having originated in- 

 dependently from herbaceous ancestors by the increase in cambial 



7. See Bibliography. 



S. Biiclionau. Fr. Alisniaroap. in : Enjrlcr & Prantl. Die Natiirliclion Pflanzon- 

 faniilion. 2 Teil. Aht. 1. pp. 227-2.^2. 1S89. Particularly p. 220 whore tho author 

 remarks upon the similarity in appearance of many Alisiuataceae with Ranunculaceae ; 

 also 



Mussat. article Allsmacecs In : Baillon. A. Dlctionnaire de Botanique, Vol. I. , p. 

 lie. 1870. who writes — "Finally, let us artil that the Alismataceae show tho preatest 

 analogy in their floral structure, with the Ranunculaceae, with which they form a 

 parallel group among the Monorotylerlonoae." 



9. Delpino. T. Applicazionc di nuovi criterii per la classificazionc delle plantc. 

 Memoria VI. 1S9G. 



10. Ilallior, H. Ueber die Morphogenle, Phylogenle und den Generatlons-wecbsel 

 dor Achsonpflanzon, 1002. and other papers (see Bibliography). 



11. I>otsv, J. P. Vortriige iiber Botanische Stammesgeschichte. Band 3. 1911. 



12. Sargant. Ethel. The Origin of the Seed-leaf in Monocotyledons. 1902. (Sep 

 IMbllography for other papers.) 



13. Fritsch, K. Die Stellung der Monokotylen Im Pflanzensystem. 1905. 



