28 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



The evidence, advanced in favor of this xiew, may be briefly 

 summarized as follows: Diffuse^ wood-parenchyma is primitive, 

 since it occurs in many Gymnosperms and in the Casuarinaceae, 

 Fagaceae, Betulaceae, Juglandaceae, and Rosaceae. Vasicentric- 

 wood-parenchyma is a derived and more recent type, since it 

 occurs in the highest Angiosperms, the Oleaceae, Ulmaceae, 

 Leguminosae, Compositae, etc. In the Salicaceae and Mag- 

 noliaceae the parenchyma is usually terminal,^ but may be vasi- 

 centric in roots, young stems, and traumatic tissue. Interpreted 

 in the light of the "laws of recapitulation, reversion, and retention," 

 these facts show terminal wood-parenchyma to have been derived 

 from vasicentric. Therefore, the Salicaceae and IMagnoliaceae 

 are specialized families that have originated by "reduction" from 

 advanced types of Angiosperms. 



There are, however, important objections to this type of reason- 

 ing. Even if it be admitted that laws of recapitulation, reversion, 

 and retention can be formulated, it is extremely difficult to apply 

 them logically in phylogenetic discussions, since frequently no 

 reliable evidence is available to show whether a given character 

 in a given region is cenogenetic or truly palingenetic. This would 

 be very likely to be the case in dealing with parenchyma, cells 

 that function in storage and other physiological processes, and 

 are particularly sensitive to the effects of environment. Further- 

 more, it is by no means certain that, because a selected character 

 is "progressive" or "regressive," a group of plants, the sums of 

 all characters, are moving in a similar direction. Thus, even if 

 it be granted that the simple flowers of the Amentiferae are 

 primitive (not reduced, as is held to be the case by many botan- 

 ists), and that the Magnoliaceae formerly possessed vasicentric 

 wood-parenchyma, it does not necessarily follow that all charac- 

 ters in the former group are primitive and that Tetnicentron and 

 Drimys once possessed vessels and ha\e lost them. 



In view of these facts, the writers decided to study the Mag- 

 nfjhaceae and allied families in the endeavor to secure evidence 

 that might indicate whether these families have become specialized 

 through reduction or have retained a nuni])cr of (rul\- primitive 

 characters. In the following pages are summarized the results 



1 Scattered iunoiiy the tracheids or fiber tracheids. 



* Clustered about the vessels. 



' Confined to the end of the year's growth. 



