Upper Palaeozoic and Mesozolc Fossils. 321 



In tangential section the medullary rays are in single rows. 

 They are of a simple type and therefore do not carry horizontal 

 resin canals, as in the fossil wood referred by Kraus' to the type 

 Pityoxylon. 



The present specimen measures in its longest diameter about 

 9 cm. 



General Remarks on the Specimen and its Alliances. — The 

 stem now under description is clearly distinguishable from the 

 wood of Cordaites, on account of the definite annual rings seen 

 throughout the stem. Although these rings are somewhat 

 irregular in the earlier part of the wood, and show a tendency 

 To thin out at various places, causing an appearance of over- 

 lapping, they are very distinct later on, and throughout each 

 complete ring show the regular transition from the spring to the 

 autumn growth, and the sudden recurrence to the former with 

 less crowded cell-structure. This feature of differentiated wood 

 cells seems to be unknown in Cordaites, where the xylem is 

 practically uniform in sbructure." 



Tlie most important factor, however, in determining the 

 <lifference between the Coniferous wood and Cordaites is the 

 nature of the pith cylinder, which in the latter is unusually thick 

 and transversely ruptured, so as to leave numerous diaphragms 

 sepaiated by cavities. 



The pith is very small in our specimen, and the primary xyleni 

 is also of very limited extent. 



With regard to the bordered pits seen in the radial walls of 

 the tracheides, I notice that in our specimen of Araucarioxylon 

 they agree most closely with those seen in Danimara australis, 

 in that they are less crowded and only occasionally polygonal. 

 The bordered pits in the tracheides of the wood of Araucaria 

 cunninghami are perhaps more comparable with those seen in 

 Cordaites, although much larger. The central depression is 

 elliptical and oblique in all four types, so that this particular 

 feature cannot be relied on as a determinative character. 



There is, therefore, in view of the foregoing statements, no 

 <loubt as to the Coniferous affinities of our specimen. 



1 Tom. supra cit. See also Traitr de I'aluoiitolo^dc, pt. ii., 1801, l\ilaeoph.vtolO!,'ie, l».v 

 Schiinper and Schenk, p. 856, fig. 417. 



2 Scott -Studies in Fossil Botany, p. 419. 



