FOSSIL PLANTS. 497 



or out of it, like some of our actual water plants which have for their suste- 

 nance in water peculiar organs, sometimes a mere swelling of their petioles, as 

 in Trapa natans for example, which disappear on the emerged parts. A like 

 lubulose, bladdery form characterizes the leaves of the species of Stigmaria, 

 whose long stems were sustained by these floating organs, and we could there- 

 fore easily admit a dimorphism of the leaves of plants which, like Annularia, 

 evidently lived partially in water. But the cylindrical form of the leaves of a 

 Lepidodendron, like those of L. rigens, cannot be explained in the same manner. 

 These leaves are evidently aerial organs, and by their form expose a new charac- 

 teristic not yet surmised in species of this genus, though it was already ob- 

 scurely marked by the position of the vascular lines seen in a different relation, 

 according to the plane in which their leaves have been flattened in shale. 



The shales of Morris and Colchester have remains of small branches of a 

 Lepidodendron, referable to L, elegans or L. gracile, Brgt? one to two inches 

 thick, mostly dividing perpendicularly to their axis, bearing short flat leaves 

 and so abundant that they fill the shale to the thickness of one foot or more, 

 extending and covering a large space. These remains, scarcely varying in 

 thickness, do not look as though pertaining to erect stems, but rather appear 

 like creeping branches, extending all around, like those of some of our species 

 of jLycopodiacece. The roots of Lepidodendra are unknown as yet, and I be- 

 lieve that some species of this genus, heretofore considered as branches of trees, 

 are mere creeping stems, which, in some circumstances, only bear flowering 

 stems, or true Lepidodendra, It is the same with the genus Sigillaria, the 

 species of which I consider as fruiting stems of Stigmaria. It is easily con- 

 ceivable that large trees, like those of Lepidodendron and Sigillaria, could 

 not be sustained upon the soft surface of the swamps of the Coal period, 

 without a peculiar kind of support ; and this solidifying process of the surface 

 could only be afforded by a vegetation like that of floating or creeping stems 

 of the same kind of plants. Some Lycopodes of our time, when growing in 

 swamps, Lycopodium inundatum and L. clavatum for example, cover the soft 

 ground with their interlaced creeping branches, bearing their rare flowering 

 stems here and there, out of the reach and influence of water. Many aquatic 

 plants of our time also multiply their stems, extending them in every direction 

 by constant division, and fill large basins, even small lakes, never bearing any 

 flowering stem until they have formed, by compact netting, a kind of support 

 strong enough to sustain them out of water for fertilization. This is the case 

 especially with some species of our mosses, Hypnum Lycopodioides, H. fluctans, 

 Sphagna, etc. Some of our species of Utricularia have two very distinct 

 modes of vegetation. U. intermedia, for example, has, in water, its stems infi- 

 nitely expanded and divided, sustained as they are by their utricles, while in 

 sand the same species has a simple stem dividing at the base into three 

 63 



