130 PKOCEEDINGS OP THE GEOLOGICAL SOCIETY. [DcC. 20, 



leaves or fruits attached ; but, from tlie associations observed, I believe 

 that the former were long, narrow, rigid, and two- or three-nerved 

 ( Cyperites), and that the latter were Trigonocmya, borne in racemes 

 on the upper part of the stem. These trees attained to a great size. 

 I have seen one trunk four feet in diameter, and specimens of two 

 feet or more in diameter are common: some of these trunks have been 

 traced for thirty or forty feet without branching. The greater 

 number of the erect stumps preserved at the Joggins appear to belong 

 to this genus, which also seems to have contributed very largely to 

 the formation of coal. Judging from the paucity of their foliage, the 

 density of their tissues, and the strong structural resemblance of 

 their stems and roots to those of Cycads, I believe that their rate 

 of growth must have been very slow. 



The genus Uhytidolepis, in which the areoles are large, hexagonal, 

 and tripunctate, and the ribs narrow and often transversely striate, 

 ranks as a coal-producer next to Sigillaria proper, and is equally 

 abundant in the Coal-measures. These trees seem to have been of 

 smaller size and feebler structure than the last mentioned, and are 

 less frequently found in the erect position ; but they are very abun- 

 dant on the roofs of the coal-beds. Judging from such specimens 

 as I have seen, their roots were less distinctly Stigmarioid than in the 

 last genus, though this appearance may arise from difference of pre- 

 servation. Their leaves were of the same type as in the last genus ; 

 and their stems bear rings of irregular scars, which may mark stages 

 of growth, or the production of slender racemes of fruit in a verti- 

 ciUate manner. The woody axis of the stems of this genus was com- 

 posed of scalariform and coarsely porous tissues, much like those of 

 modern Cycads. I figure, as an illustration of the genus, a fragment 

 of S. scutellata showing one of the belts of abnormal scars. 



The genus Favularia is represented in IS'ova Scotia principally by 

 the typical species S. elegans of Brongniart. The admirable investiga- 

 tions of the structure of the stem of this species by Brongniart, with 

 the further illustrations given by Corda, Hooker, and Goldenberg, still 

 afford the best general views of the structure of Sigillariw which 

 we possess*. It is to be observed, however, that Brongniart's speci- 

 men was a young stem or a branch, and that it affords a very imper- 

 fect idea of the development of discigerous and bast tissues in the 

 full-grown stems of Sigillaria proper. The trees of this genus ap- 

 pear to have been of small growth; and they branched in the manner 

 of Lepidodendron, the smaller branches being quite destitute of ribs, 

 and with the areoles elliptical and spirally disposed. The stems show 

 joints or rings of peculiar scars at intervals, as in the last genus. 

 The leaves differ from those of the other genera, being broad and with 

 numerous slender parallel veins, almost in the manner of Cordaites. 

 (Figs. 26 and 27, Plate YII.) 



The genus Clathraria is evidently closely allied to the above, 

 and is possibly founded on branches of trees of the genus Favularia. 

 It is a rare form in Nova Scotia. 



* See also Binney " On some Fossil Plants showing Structure, from the Lower 

 Coal-measures of Lancashire," Quart. Journ. Geol. Soc. vol. xviii. p. 106. — Edit. 



