364 THE CANADIAN NATURALIST. [June 



of the whole. Growing as these trees did in swampy flats close 

 together, and the bark of which they were chiefly composed being 

 less susceptible of rapid decay than most kinds of wood, and too 

 impervious to fluids to be readily penetrated by mineral matter, 

 they were admirably fitted for the production of the raw material 

 of coal. (Fig. 161.) 



* # * * 



I include under Sisfillariae the remarkable fossils known as 



o 



Stigmaria, being fully convinced that all the varieties of these 

 plants known to me are merely roots of Sigillaria ; I have verified 

 this fact in a great many instances, in addition to those so well 

 described by Mr. Binney and Mr. Brown. The different varieties 

 or species of Stigmaria are no doubt characteristic of different 

 species of Sigillaria, though in very few cases has it proved 

 possible to ascertain the varieties proper to the particular .species 

 of stem. The old view, that the Stigmari^e were independent 

 aquatic plants, still apparently maintained by Goldenberg and 

 some other palaeobotanists, evidently proceeds from imperfect 

 information. Independently of their ascertained connexion with 

 Sigillaria, the organs attached to the branches are not leaves, but 

 rootlets. This was made evident long ago by the microscopic 

 sections published by Goeppert, and I have ascertained that the 

 structure is quite similar to that of the thick fleshy rootlets of 

 Cycas. The lumps or tubercles on these roots have been mistaken 

 for fructification ; and the rounded tops of stumps, truncated by 

 the falling in of the bark or the compression of the empty shell 

 left by the decay of the wood, have been mistaken for the natural 

 termination of the stem.* The only question remaining in regard 

 to these organs is that of their precise morphological place. 

 Their large pith and regular areoles render them unlike true 

 roots ; and hence Lesquereux has proposed to regard them as 

 rhizomes. But they certainly radiate from a central stem, and 

 are not known to produce any true buds or secondary stems. In 

 short, while their function is that of roots, they may be regarded, 

 in a morphological point of view, as a peculiar sort of underground 

 branches. They all ramify very regularly in a dichotomous 



* For examples of the manner in which a natural termination may be 

 simulated by the collapse of bark or by constriction owing to lateral 

 pressure, see my papers, Quart. Jour. Geol. Soc, vol. x. p. 35, and vol. 

 vh. p. 194. 



