— 553 — 
moir(1). A fragment from the roof of the Irapua coal, in Rio Grande do 
Sul, is cuneate, widening upward from avery narrow attachment. In 
both form and dimensions il agree well with that published by 
Zeiller (2) from Johannesburg in the Transvaal. 
Some interesting characters of the leaves of Noeggerathiopsis His- 
loptare brought out by the various conditions of preservation. On many 
of the impressions in the fine-grained cream colored shales from which 
the carbonaceous residue has for the most part been removed, leaving 
little else than a slight stain of brown in the mold, the surface is mar- 
ked by myriads of minute impressions, as of short hairs or down, the 
bases of which are marked by particles of coaly matter. The inference 
that the leaves of the species were villous is not borne oul by the fea* 
ture ofthe epidermis which have been portrayed in detail by Zeiller. (3). 
The latter is especially comparable in its details to the Cycadean cuti- 
cles. Someofour Brazilian specimens also exhibit zones of extremely 
minutely granular residue between the nerves, in characteres apparcn- 
tly agreeing exactly with the specimen (4) explained by Zeiller as due to 
punctations corresponding to stomata on the dorsal surface of the 
leaf. 
I have take care to point out the agreement between the South Ame- 
rican leaves and those from Indiaand other regions of the Gondwana 
flora for the reason that false nerves between the true nerves of Noegge- 
rathiopsis Hislopi are perceptible in many of the specimens before me. 
They are very faintly indicated even in portions of the large leaf shown 
in Fig. 5. In many cases they appear irregularly and only in small areas 
for short distances like the strands intermediate to the nerves in some of 
the Neuropteroid Cyeclopterids or in Dolerophillum. Certain of the brow- 
nish, carbonaceous shales associated with the Barro Branco coal near 
Candiota, Rio Grande do Sul, consist largely of the matted leaves of the 
Noeggerathiopsis Hislopi. A fragment of this laminated shale is shown 
in Fig. 6, in which the true nervation is distinctly seen. 
Under the lens, however, anumber, usually three or four, of inter- 
mediate or false nerves, as is photographically illustrated, in the four 
(1) Fl. Gondwana Syst., vol. IV, 1832, pt. 1. 
(2) Bull. Soc. Geol. Fr., (3) vol. xxiv, 1896, p. 372, pl. xviii, fig. 7. 
(3) Loc. cit., p. 372, figs. 16, 47. 
(4) Loc, cit., pl. xviii, fig. 8. 
5560 Bu 
