LIMITS OF LEPIDODENDRON AKD SIGILLAKIA. 535 



structure ; but though some of the Erian or Devonian species are probably of 

 this type, in the lower Carboniferous, where the lepidodendra first became ira« 

 portant, the species with leaf-bases separated by wrinkled bark or by expan- 

 sion of the cortical tissues between the leaf-bases are apparently predominant, 

 though others also exist, and the type which we are now considering perhaps 

 culminates in the coal formation. 



We may first refer to L. costatum of Lesquereux, with vertical rows of 

 corrugated leaf-bases, but separated by distinct longitudinal spaces of 

 wrinkled bark. This is a lower Carboniferous species, and is compared by 

 Lesquereux with his L. hrittzi and with L. volkmanniannm, Sternberg, of 

 the European Carboniferous, both of which have strong points of resem- 

 blance in the characters of the leaf-bases, though diflTering in the scars and 

 in the leaves, so far as known. The L. wortheni of Lesquereux is based on 

 fragments closely allied in general form to our species. So also is S. dip- 

 lotegioides, a species found in the lower coals as far west as Arkansas. None 

 of these species are, I think, sufficiently near to be identified with our New- 

 foundland and Nova Scotia species, though as most of them are known only 

 by the bark of old stems, this may admit of doubt. In any case, lepido- 

 dendra of this general type and aspect were widely distributed, both in 

 Europe and America, in the Carboniferous, aud especially in the lower 

 portions of the coal formation, to which in all probability the Newfound- 

 land specimens belong. 



I may add here that Zeiller* figures a species as L. veltheimianuni which 

 can scarcely be that species and may be a branch of L. murrayanum, with 

 which it agrees very closely. The same plant is figured by Kenault.f The 

 leaf-bases of the Newfoundland species have also some resemblance to those 

 of L. aculeatum, Sternberg, but differ in detail. 



Another interesting question rises here as to the limits of Lepidodendron fxud 

 Sigillaria, as determined by their surface markings. The markings of the 

 latter have usually been considered as characterized by the leaf-scars being 

 placed in vertical rows and often on continuous prominent ribs, and also by 

 the fact that the lateral vascular scars are much larger than the central one ; 

 but in such a case as Lesquereux's species, L. costatum, the confluent leaf- 

 bases in vertical rows have the effect of ribs, and in a less degree the same 

 remark applies to L. murrayanum. I may add that when one happens to 

 find young stems of Sigillaria not compressed, the leaf-bases are seen to pro- 

 ject in the manner of those of Lepidodendroii, and that in some non-ribbed 

 Sigillarids, as in L. elegans, the very young branches have the scars arranged 

 spirally.^ In connection with this I may observe that Sauveur§ has described 



* Vegetaux fossiles du Terrain Huillier, 1880, pi. xxii. 



t Coins de Botaniqiie Fossile, 1881, pi. v, fig. 2. 



j Acndian Geology, 1878, p. 435. 



§ Fossil Flora of Belgium, 1848, pi. lvi and lviii. 



