152 W. Carrutkers — Coal Plants Jt^om Brazil. 



is really a specimen of Fle77iingites gracilis. In passing, I may say that 

 this supposed species was made the receptacle for all indistinct and 

 badly preserved specimens of such cones ; the variable appearance of 

 the specimens arising from their belonging to different species, and even, 

 as it now appears, to different genera. In all the specimens, however, 

 of Flemingites which I have examined I have seen no indication of the 

 branch or tree on which they were supported, and the characters of the 

 genus have been consequently confined to those of the cone. It is not 

 a little interesting that the materials for completing our acquaintance 

 with the fossil should be brought from South America, — a continent 

 which, as far as I know, has not hitherto yielded any palaeozoic fossil 

 plants. The examination of the plate and of the description given will 

 show, however, that the one organ wanting in the specimens from 

 Brazil is the cone on which to a considerable extent I founded the 

 genus. The sporangia abound, and though the co-relation of these 

 organs with the lepidodendroid stem depends not upon direct observa- 

 tion but upon induction, yet that induction is so conclusive that it 

 appears to me to place the matter beyond doubt. 



The small round sporangia belong, as far is known either to Sigil- 

 laria or Flemingites. They abound in these Brazilian shales, and the 

 only plants associated with them to which they could belong is that 

 which is described in the diagnosis of the species. The arrangement 

 of the leaf scars and the form of the leaf conclusively establish that this 

 is not a Sigillaria. Among Mr. Plant's small collection there are over 

 twelve different specimens of the stem, with numbers of sporangia scat- 

 tered over the surface of the fragments on which they are preserved, 

 and in one specimen several sporangia occur among the mass of true 

 leaves which remain attached to the end of the branch, though not re- 

 lated to them as in the described cone of Flemingites. 



The spiral arrangement of the leaves on these stems also agrees with 

 the arrangement of the fruit-bearing leaves on the cone, so that there 

 can be no doubt that the stems belong to Flemingites. They have the 

 ordinary aspect of the stems of Lepidodendron, the scars being arranged 

 in a spiral order. There are nevertheless points of considerable im- 

 portance by which the two forms can be distinguished. The sears are 

 small, approaching in size and form those of Lycopodiolites cordatus, 

 Sternb. (Flora d. Vorwelt, tab. Ivi. fig. 1), from Yarrow in Durham, 

 a fossil which has been overlooked by subsequent authors, except that 

 Professor Morris, who never overlooks anything, records it in his Cata- 

 logue of British Possils under the name of Lycopodites cordatus. In 

 the Brazilian fossil the scar is not cordate, but perfectly rounded on 

 its upper margin (see enlarged scars, PL Y. Pig. 11). But the most 

 important character in respect of the scar is that it presents no im- 

 pressions from an articulating surface like what is seen in Lepidoden- 

 dron. This arises from the fact that the bases of the petioles per- 

 manently invested the stem, the leaves disarticulating at a line about a 

 quarter of an inch along the petiole from the stem surface. A somewhat 

 similar structure is described by Corda in his genus Lomatophloyos 



^ Prof. Morris has shown me specimens of Corda' s L. crassicaule from English 

 strata, and I have a second species from the beds of Tolcanic ash in Arran to which I 

 have given the name of L. Wunschianus, after my friend Mr. E. A. "Wiinsch, of 



