160 ANATOMY OF THE GYMNOSPERMS 



not answer the object in view, since its most prominent feature 

 would suggest the idea of a continuous series of conterminous 

 members, from which lateral members would arise at intervals. 

 There is no evidence that any phylum represents such a series ; 

 on the contrary, there is every reason to believe that such 

 relations do not exist among the various groups of plants. 



In those branches of the Norway maple which exhibit slow 

 growth various forms of arrested development are manifested. 

 These take the form of atrophied buds, or of branches in all 

 stages of development, and there thus arises a modified monopo- 

 dium which eventually becomes, in many cases at least, a true 

 sympodium. In comparing this with the monopodial branch of 

 vigorous growth, it appears that the alterations involve more 

 than mere suppression. In the monopodium the average angle 

 of divergence for the lateral members is 45.3 degrees, while for the 

 derived form it is 34. i degrees. The latter will be seen to com- 

 pletely fulfil all conditions with respect to the development of 

 a phylum, even to indicating the position of missing members. 

 In the construction of this figure an attempt has been made to 

 show all normally developed buds (O) and their relative dimen- 

 sions ; atrophied buds (o), the position of which is recognizable ; and 

 atrophied branches (Y) which are still visible, but it is obvious 

 that the figure does not show many members all evidence of the 

 former existence of which has completely disappeared. Selecting 

 from this we obtain the accompanying figure, which embodies 

 our final conclusions as to the general succession of the different 

 gymnosperms, and from it we may gather that the highest repre- 

 sentative, Pinus, is the terminal member in the main line of 

 descent from the Cycadofilices through Poroxylon, while from 

 such a central line both the Cordaitales and Gingkoales have 

 been given off as side lines. 



The general results of these investigations serve to confirm in 

 a very striking manner the probable monophyletic origin of the 

 gymnosperms, as already expressed by Coulter (11), while they 

 also show that the real transition ground, at least for all but the 

 Cycadaceae, was probably represented by Poroxylon, as indicated 

 by Scott (81). 



