538 



THE AMERICAN NATURALIST. [Vol. XXXVIII. 



From such data it is clear that the distribution of the resin 

 cells bears an important relation to the recognition of sub-generic 

 groups and even of species. But viewing these structures from 

 the broader standpoint of the Coniferales as a whole, it is 

 obvious that they must be placed among the structural elements 

 which belong to the first rank for taxonomic purposes. 



We are now in a position to determine what relation, if any, 

 such resin bearing elements bear to questions of phylogeny, and 

 we may first of all consider the resinous tracheids. These 

 structures have been seen to be peculiar to Agathis, Araucaria, 

 and Abies, in which they occur only in certain species. In 

 answering this question, we cannot avail ourselves of evidence 

 derived from fossil plants, since it is in such cases of a nega- 

 tive character. Neither Cordaites nor Araucarioxylon affords 

 definite proof of the presence or absence of such structures, 

 since they do not appear in any of the published diagnoses, and 

 our own studies have not resulted in their recognition. If 

 originally present, they must have been obliterated in the course 

 of fossilization. We must therefore depend entirely upon such 

 evidence as is afforded by existing species. From this point of 

 view it is obvious that they furnish no evidence as to the origin 

 of either of the three genera in which they occur. It is, on the 

 other hand, possible to determine from other data, that both 

 Agathis and Araucaria are much inferior to Abies in point of 

 structural organization and development, and from this we may 

 be permitted to conclude that the resin tracheids of Abies are 

 vestigial forms of elements which were typically developed in 

 Agathis and Araucaria, and possibly characteristic also, of their 

 progenitors. If such inferences are to be regarded as justifiable, 

 they go far to support the idea of a common origin for all three 

 genera, and they thus lend force to conclusions which lead to 

 the same result, but upon the basis of independent data. 



trf)m a study of the distribution of the resin cells, it is 

 apparent that they fall into four categories in which the typically 

 scgrc-atcd cells may be held to represent the most primitive 

 orm of disposition. This view is greatly strengthened by the 

 observation that in all such cases, the resin cells are rarely if at 

 all accompanied by parenchyma tracheids, while 



