OF LOWER GTIERXSAXD PLANTS. Ill 



thick-walled. As is well known, tyloses in canals with thick- 

 walled epithelium are much rarer than in the species with a 

 thin- walled lining to the resin-canal. 



In transverse section in the first two or three annual rings 

 the medullary ray-cells are numerous and conspicuous, the 

 individual ray-cells often exceeding the tangential diameter of 

 the adjacent tracheids, the radial extension of the cells averag- 

 ing in the outer zones about 4-7 tracheids. The elements are all 

 thick-walled and pitted, the majority being ray-tracheids (PI. V, 

 fig. 2, and text-fig. 27). The ray-tracheids have round bordered 

 pits and a comparatively regular outline. In a few cases cells 

 which appear to be parenchyma-cells have single large pits in 

 each tracheid-field (text-fig. 27, p.}. Whether these are normal 

 pits the preservation of the fossil does not allow one to determine 

 with absolute certainty. The end-walls of the ray-cells show 

 the typical " abietiuean pitting " (text-fig. 27, a.). 



Comparisons. Among previously described Cretaceous species 

 of Pilyoxylon none appear to approach this new form. Con- 

 wentz (1890) figures a portion of pith and primary wood much 

 resembling our fossil (pi. x, fig. 5) from the Tertiary amber, 

 which he includes in his Pinus sucdni.fera. He uses Pinus in the 

 broadest sense, and points out the great difficulty of separating 

 it from Picea on the evidence of portions of wood arid bark alone. 

 Judging by Gothan's descriptions, the parenchyma-cells of the 

 resin-canals in the new species are similar to that in Piceoxyloii 

 antiquius, Gothan (1910 A, p. 20), though, unfortunately, no illus- 

 trations of this point are given by Gothan. The two fossils, 

 however, differ essentially in the presence of specialised ray- 

 tracheids in the Lower Greensand form and their absence in 

 the Sjdtzbergen fossil. 



AFFINITIES. If among the described Cretaceous fossils there 

 are none with which to compare this new fossil, it has many 

 features which make comparison with living genera profitable. 

 The extremely thick walls of the epithelium of the resin-canals 

 and their pits (ef. text-fig. 26) are very suggestive of Larix, as 

 is the thickened and pitted nature of the medullary ray-cells. 

 The tyloses which are so prominent in the fossil may at first 

 sight seem to militate against the view that the plant is allied 

 to Lftriv, but tyloses in. the living genus have boon found not to 



