﻿154 Canadian Record of Science. 



and it is extremely difticult to deteniiiue precisely upon 

 what it depends, but it seems probable that it is 

 determined by a peculiar disposition of the cells in relation 

 to the medullary spaces. 



Large transverse sections also exhibit radial fissures due 

 to shrinkage, but there appears to be a total absence of 

 those radial bands simulating medullary rays, so con- 

 spicuous in K Logani. On the other hand, the medullary 

 spots, already described, are • connected radially and 

 tangentially by more continuous and open tracts as 

 medullary spaces, which thus form a sort of netted system 

 between the various sub-divisions of which the large cells 

 lie in distinct and often more or less rounded groups. 

 This distribution of the elements gives the transverse 

 section a very characteristic appearance. It had already 

 •been noted in the previously described specimens of N. 

 crassum, but owing to the very limited area of the Gaspe 

 sections, and the highly altered character of the specimens 

 from the Hamilton group, a proper description was not 

 possible, and this structural feature was, therefore, omitted 

 from the diagnosis. It is, nevertheless, an important 

 diagnostic element, under the present circumstances of 

 limited material, since it seems to definitely differentiate' 

 this species from all the others. 

 Longitudinal Section. 



In longitudinal section the cells of the medulla are 

 somewhat strongly interlacing, while groups of a dozen or 

 more often cross the general direction of growth more or 

 less abruptly, and sometimes turn off nearly at right 

 angles for a short distance. These features also appear in 

 previously described specimens, both from Gaspe and 

 from New York. The intercellular hyphie are freely 

 interlacing and cross the large cells in all directions, but 

 their structure is so altered by decay as to render it 

 impossible to determine if they are septate or not. 

 Nowhere have trumpet hyphic been found, thus con- 

 firming previous observations in this respect. 



