KUNGL. SV. VET. AKADEMIENS HANDLINGAR. «BAND 51. N:O 7. 5 
ones occur on the one as well as on the other side. The tubercles would develope 
earliest on the upper side, as they can be seen distincetly here, while those on the 
lower side are hardly discernible yet. 
As for the degree of distinetness with which the tubercles appear, there is a 
great variation in different specimens. "They arte often sharply conspicuous, but at 
times they are only distinguished by slight folds on the macerated rachis. This 
variation sSeems to stand in intimate connection with the thickness of the cuticle, 
while, in a much lesser degree, it depends on the size of the specimens, for very 
broad stalks with a thin cutiecle can have often only very inconsiderable tubercles. 
The epidermal structure of the tubercles does not differ in any respect from 
that of other parts of the plant. The very biggest tubercles, however, possess ca- 
lottes of a more strongly cutinized tissue. 
It is quite a common thing for the calottes of the tubercles to be missing 
(pl. 2, fig. 8), so that the cuticle has more or fewer round or ovate holes. The for- 
mation of these holes may be due to different factors. As a thick fossil cuticle is 
often rather brittle, some of them probably arose during maceration, while others 
were obviously formed while the plant still lived. I will come back to this subject 
later on. 
Recent MSucculents sometimes possess similar tubercles on their laminae. But 
what mission they have to fulfil, I have not succeeded in ascertaining, and I am, 
of course, still less in the position to make a statement concerning those in Lepi- 
dopteris Ottonis. 
As to its structure the epidermis of the rachis (pl. 2, figs. 6—38) agrees per- 
fectly with that of the lamina, and is composed of rather thick-walled, isodiametric- 
ally polygonal, at times somewhat oblong cells, which sometimes possess, and some- 
times lack, distinct papillae. Stomata occur on both sides but are rather few in 
number. They (pl. 1, fig. 6) are surrounded by 4 to 6 regular subsidiary cells. The 
guardian cells are immersed. The subsidiary cells seem, probably on account of an 
oblique position of the walls, somewhat thicker than the rest of the cells, and each 
of them puts forth a rounded cuticular lobe from the wall directed towards the en- 
trance. These lobes meet in the centre of the entrance, which becomes reduced to 
a star-shaped passage. Hereby the efficacy of the stoma, no doubt, was greatly 
enhanced. 
SCHENK (1867, p. 54), who besides others examined the cuticle of this plant, 
mentions the regular arrangement of the subsidiary cells. But on the other hand 
he seems to have overlooked the occurrence of the lobes. "Though he has supplied 
drawings of the epidermis, these do not include any stomata. 
Stomata of quite the same construction have already been described by ZEILLER 
, (1882, p. 234; pl. 11 and 1900, p. 274; fig. 196) in Frenelopsis Hoheneggeri (ETT.) 
SCHENK, and by other authors in other species of the same genus. What I have 
spoken of as lobes here, ZBILLER, and others after him, have, however, interpreted 
as guard-cells in a stoma with 4 to 6 such cells, an opinion whose incorrectness 
THOMPSON (1912, p. 63) recently pointed out. 
