No. 449-] NORTH AMERICAN CONIFERALES. 357 



the sporadic recurrence of thin walls with local thickenings 

 represents the persistence of a primitive character. 



Ray tracheids probably constitute one of the most valuable of 

 the structural elements as an indication of development. This 

 has its foundation (i) in the fact, previously shown, that they 

 arise as secondary structures from the parenchyma elements, 

 with which they exhibit interchangeable relations, in dired 

 response to the requirements of a higher degree of or-^anization, 

 and (2) their general relation to progressive development. The 

 complete absence of ray tracheids from the Cordaitales and 

 Ginkgoales, as also from the Taxaceae and more primitive Con- 

 iferae, while they are invariable features of the higher Coniferae 

 in which they attain their most complete development, admits of 

 only one interpretation. The fact that they are exclusive fea- 

 tures of the Coniferae emphasizes their inferior value for deter- 

 mining the derivation of that group, while it points to their 

 superior importance as a factor in the sequence of the various 

 coniferous genera. They occur sporadically in Thuya (i), Cu- 

 pressus (2), Juniperus (i), and Abies (i). They are prominent 

 features of Tsuga, Pseudotsuga, Larix, Picea and Pinus. Their 

 invariable absence from Sequoia would appear to suggest that 

 this genus is more primitive than Thuya, but there are other 

 reasons which serve to suggest the opposite relation. Apart 

 from this exception it will be seen that in accordance with the 

 relations exhibited in the table of anatomical data, the genera 

 enumerated form a continuous series, commencing with those 

 showing sporadic tracheids, and ending with those in which such 

 structures attain their highest expression. From this we are 

 justified in the conclusion that the rare occurrence of tracheids in 

 Thuya, etc., is to be interpreted as the first evidence of a tend- 

 ency in development which is only fully realized at a later period, 

 and this appears to be justified by a closer examination of the 

 last five genera in this respect, since it is found that in them 

 the tracheids not only show a progressive numerical develop- 

 ment, but their structure likewise becomes more complicated in 

 direct relation to the evolution of higher types of genera and 

 species. We must therefore look upon the tracheids, with their 

 thin, simple walls, as the primitive form, while those with the 



