90 MB. T. HICK OK A NEW FOSSIL PLANT 



only seen in transverse section, it is impossible to say whether the 

 fibrous structure is present or not. 



Medullary Bays (PI. XVII. figs. 3, 4, mr).— The medullary 

 rays vary in breadth from 0-085 millim. to 02 12 millim., and are 

 made up of rather large-celled parenchyma. Iu the broadest and 

 best preserved the cells seem to be distinctly elongated radially, 

 but neither in shape nor arrangement is there much regularity, 

 and the walls show little or no thickening. As mentioned above, 

 they run uninterruptedly through the tissue regarded as phloem. 



Pericycle (?) (figs. 3, 4, ^.—Surrounding the whole of the 

 vascular-bundle ring is a zone of parenchyma whose designation 

 must be provisional owing to its somewhat uncertain homology. 

 Its elements are small, averaging 0034 millim. in diameter, but 

 somewhat uniform in size, thin-walled, and in their general 

 appearance not unlike the cells of meristem. In some instances 

 the primordial utricle appears to be preserved, suggesting a con- 

 dition of functional activity. The line of separation between 

 this zone and the tissues outside is clearly marked in most of the 

 sections, but there are no indications of an endodermis. Even 

 where the line is less sharply marked there is a very obvious dif- 

 ference in the size, appearance, and arrangement of the cells on 

 the two sides of it (fig. 4, pc). 



^ Two views at least are possible as to the nature of this layer of 

 tissue. It may be regarded either as a portion of the phloem, 



or as a many-layered pericycle * lying between the phloem and the 

 outer tissues. 



If it were a portion of the phloem, the medullary ravs might 

 be expected to run through it and merge in the tissue "beyond. 

 This, however, does not occur in the majority of instances, as 

 most of the rays stop short at its inner margin. On the other 

 hand, there are rays which are prolonged into this layer, but 

 this only occurs where a vascular bundle is leaving the ring, as 

 already described (fig. 4). Where this happens, the medullary 

 ray follows the bundle, as it were, for a short distance, while 

 the layer m question is still visible on the outside (fig. 4 pc). It 

 would seem, then, that there is no real interruption of this layer, 

 which is continuous at all points, and cuts short the medullary rays 

 at its inner marg.n. On these grounds I am inclined to regard 

 it, at any rate provisionally, as a many-layered pericycle. 



* Van Ticghern, < Traits de Botanique,' 2nd ed. pp. 674, 739. 



