534 SIR J. W. DAWSON — FOSSILS FROM NEWFOUNDLAND. 



dendra took place in three different ways : In some, as in L. Sternbergi, the 

 bark retains its vitality in such a manner that the leaf-bases increase in size 

 and do not become separated from each other. In others, as in L. velthei- 

 mianum a,ud L. pidoense, the leaf-bases remain small and the intervening 

 bark becomes torn in strips, leaving wide gashes without any scars. An in- 

 termediate type is that which we have in L. rimosum and L. corrugatum, in 

 which the scars increase only slightly in size and then become separated by 

 rims of slightly wrinkled bark. It would appear, from the observations of 

 Williamson and others, that the first condition appertains to those Lepido- 

 dendra that possess only a very slight development of the woody axis, while 

 the second occurs in those species in which the woody zone becomes thick 

 and strong. 



The two species above referred to evidently belong to the first category ; 

 and, as the stems found are not large, still older stems would probably show 

 larger leaf-bases. Such species of lepidodendra approach nearer than others 

 to the genus Lepidophloios in the expansion of the old leaf-bases and the 

 small development of the woody axis ; and it is interesting to notice that they 

 also resemble them in the great length of the leaves and the thickness of the 

 branches. The lepidodendra whose branches end in slender sprays are usually, 

 if not always, those in which the woody axis is large and the bark of the old 

 stems torn and wrinkled. 



I may add that these differences are most important in the discrimination 

 of species of the genus Lepidodendra )i by the markings on the stems, though 

 they have been too often overlooked. 



Another noteworthy point is the manner in which the fruit of L. clifton- 

 ense is borne on slender branchlets with few and short leaves, extending from 

 the thick branches. Such branchlets might, if alone, be r-eadily mistaken 

 for branches of other species. They also help to explain the scars of fructi- 

 fication often found on lepidodendra, as well as on the so-called ulodendra, 

 some of Avhich, however, are not geuerically distinct from the lepidodendra, 

 and on Lepidophloios. In some species, especially of the latter genus, these 

 scars are seen from their form to represent sessile cones, usually of large 

 size; but in other cases they are merely round marks, as if indicating the 

 insertion of branches or buds. The little fertile branchlets of L. cliftonense, 

 which would probably die after the maturity of the fruit, would leave such 

 scars, and may probably account for some of the less intelligible of them. 



If now we compare our two species above described with others found in 

 America and Europe, and most of which are characterized merely by the 

 forms of the leaf-bases and scars, we may exclude from consideration all 

 those in which the leaf-bases do not expand in growth, and confine ourselves 

 to those having living and expanding leaf-bases. At first sight we might 

 imagine that these would be the oldest, as being simpler than the others in 



