Elkins and Wieland — On Devonian Wood. 67 



Cordaites so closely resembles that of Araucaria that it is 

 " really indistinguishable."* 



Penhallow, in his classification of North American species 

 of Dadoxylon, places the species in two groups : in one group 

 the growth rings are distinct, in the other group the growth 

 rings are obscure or obsolete. In the first group the demarca- 

 tion between the fall and spring wood is quite abrupt and well 

 defined ; in the second group, the annular appearance is mainly 

 due to a line of tracheids with radially shortened diameters, 

 the transition into the elements of the surrounding tissue taking 

 place similarly and gradually on either side of the ring, so that 

 it would be impossible to determine either the inner or the 

 outer face if it were not for careful orientation. These two 

 groups strongly suggest Araucarian affinities and ancient 

 parallels to the Dammar a and Araucaria types of modern 

 Araucarise. 



Fig. 1. 



K ( I II 



I 53n\m. I 



Fig. 1. Showing the proportional distance of six growth rings from each 

 other, x 3. 



The species under consideration, which, for convenience, will 

 be referred to as the Indiana species, falls into the group with 

 somewhat obscure growth rings and is, therefore, an Araucarioid 

 type. 



In a transverse section of the Indiana species, the growth 

 rings, though not conspicuous, are quite apparent when viewed 

 either macroscopically or microscopically. To the unaided eye 

 they appear as fine parallel lines traversing a homogeneous 

 field. In the section studied, the dimensions of which have 

 already been given, there occur six growth rings ; these vary 

 in distance from each other in the proportion indicated in 

 fig. 1. With the microscope the rings can be identified only 



*Wood now referred to Cordaites was formerly called Araucarioxylon. 

 and Dadoxylon. According to some authors, Dadoxylon is a general name 

 for Paleozoic wood of Cordaitean type. Penhallow properly considers it a 

 noncommittal term harboring forms of doubtful affinity. Araucarioxylon 

 is of necessity often applied somewhat generally to post-carboniferous wood 

 resembling Araucaria, but which may actually include woods of diverse 

 families. Cordaites was first applied to woods known to accompany Cor- 

 daitean leaves and fruits. Now it is a group name including Dadoxylon and 

 Araucarioxylon. Other terms applied to Cordaitean wood of infrequent use 

 have here been avoided. Penhallow uses Dadoxylon as synonymous with 

 Cordaites. For further discussion of Cordaitean nomenclature see the intro- 

 duction to Penhallow's Notes on the North American Species of Dadoxylon ; 

 also see Coulter and Chamberlain's Morphology of Gymno sperms, Chapter IV, 

 and especially Knowlton in Proc. U. S. National Mus., Vol. XII, 1889. 



