_>2J HKKKDIIS \\I) EVOLUTION l\ PLANTS 



certain species, in the formation of a cotyledonary ring, 

 or tube. Bucholz interprets the facts set forth by him- 

 self and other investigators as leading to the conclusion 

 that the more primitive gymnosperms had numerous coty- 

 ledons, that their number was reduced by the fusions of 

 their primordia and, in some species, a cotyledonary tube 

 or ring was formed. "Dicotyledony was attained either 

 by a general fusion of many cotyledons in two groups, or 



Fig. 105. Polycotyledonous seedlings of dicotyledonous species. A-C, 

 Silene odonlipctala, with hemi-tricotylous, tricotylous, and tetracotylous 

 seedlings; D-II, Pa paver Rhocas (semi-double cultivated form), dicoty- 

 lous, hemi-tricotylous, tricotylous, tetracotylous, and pentacotylous 

 seedlings; /, Acer Pseudo-Plnlnnnx, tetracotylous seedling. (All figures 

 re-drawn from de Vries.) 



by an extremely bilabiate development of a cotyledonary 

 tube" (Fig. 104). 



The final conclusion of Bucholx, based on the evidence 

 of comparative anatomy, supplemented by studies of 

 development, is that the polycotyledonous condition is 

 the more primitive, and the dicotyledonous one derived. 

 On the basis of this theory, the rather common abnormal 

 appearance of supernumerary cotyledons in dicotyledonous 



