68 ROYAL SOCIETY or CANADA 



The radial section shows the bordered pits to be hexagonal in 3-4, 

 chiefly 3 rows. The elements of the medullary mys are all of one kind, 

 and their lateral walls bear from 2-4, chiefly about 4 pits i)er trachcid. 

 These pits are commonly disposed in radial series. 



In tangential section the r'ays are seen to Ix^ 1-2 seriate. 



Transverse. — Tracheids about 47 x 56 mic. broad, the walls 9.i> mlc. thick. 

 Growth rings none. 



h'adiat. — Bordered pits numerous throughout the tracheids, In 2-4, chiefly 3 

 rows, about 12 mic. broad. Ray cells equal to about 3-4 tracheids; 

 the lateral walls with oval pits, 2-4, chiefly 4 per tracheid and dis- 

 posed in radial series. 



7'anj/fHn«/.— Rays of medium height, 1-2 seriate in part, the oval cells chiefly 

 about 25-37 mic. broad. 



CORDAITES ANNULATUM, Dn. 



Bib. :— Quart. Jn'l Geol. Soc, 1886. 146; Knowlton, Foss. Woods & Lig. of the 

 Pot. Form., 1889, 52, and Proc. U. S. Nat. Mus., XII., 1890, 610; Can. 

 Nat, VIII., 1863, 433i« 



Dist. :— Middle Carb. of the Joggins. N. S. 



The original description of this i>\)cdes in 18()3, wns derived from 

 a specimen six inches in diameter, whidi is described as having about 

 fourteen to sixteen growth rings ami a i)yritised pith about one incn in 

 diameter.* 



A critical examination of the original specimen shows that there is 

 nothing of the nature of growtli rings, but as the description states, 

 there are "layers of structureless coal rei)resenting cellular tissue or 

 very dense wood." These layers run diagonally to the rtidial lines in 

 which the tracheids fall, and therefore diagonally to the medullary rays, 

 buit sooner or later they cur\-e in such a manner as to often become more 

 or less parallel with them. It will thus be observed tliat it is quite im- 

 l>ossible for them to represent growth rings. The alternating layers of 

 badly preserved wood have been taken for the summer wood of the 

 gi-owth rings, but in following out one of these for a long distance 

 radially, no evidence of growth rings appears. It would thus seem that 

 the statement of the first diagnosis was based upon a misinterpretation 

 of the features resulting from decay localised along diagonal or radial 

 lines. The confusion resulting from this has led recent observers to 

 question if this plant has any real affinity with Cord'aites,* but although 



' Can. Nat., VIII., 1863, 433. 



' Proc. U. S. Nat. Mus., XII., 1890. 410. 



