538 Brebner . — On the Anatomy of 
description leaves some doubt about the matter, and it is 
possible that the absence of a stipular commissure may be 
a generic difference. Archangiopteris , Angiopteris and 
Danaea have distinct, not to say well-marked, commissures. 
Kaulfussia was not examined for this feature. In Marattia 
alata only a very few, three or four, meristeles pass into the 
leaf-base from the stem, but they then branch very freely, so 
that they become very numerous. They are not very 
definitely arranged in the leaf-base, but, by the time they 
have reached the pulvinus, they have taken up a definite 
and characteristic position. In the case of a frond of 
Marattia alata , about three and a half feet long, three or four 
meristeles passed into the leaf-base ; there they branched 
freely, giving rise to 33-34 meristeles, as seen in transverse 
section of its middle region. In the middle region of the 
pulvinus there were likewise 34 meristeles. Their arrange- 
ment in the pulvinus and petiole is illustrated by Figs. 11 and 
12 (PI. XXII) from M. fraxinea , which is a much smaller 
plant than M. alata , and therefore there are correspondingly 
fewer meristeles. The arrangement is essentially the same in 
all the species of Marattia examined. In this genus, as also 
in Angiopteris and Danaea , the meristeles (ms. in the figures) 
form a ring, somewhat flattened towards the upper surface, 
and besides there is a sort of indentation formed by a single, 
or a few, meristeles (ms/). In all these vascular strands the 
protoxylem is directed towards the centre. Then further 
there is a small number of meristeles (ms!') nearer the 
middle of the pulvinus, or petiole, in which the protoxylem 
faces upwards, or outwards towards the ring. As already said, 
there is the same general arrangement in all the Marattiaceae, 
but in Angiopteris there are several almost concentric sets 
of meristeles. In a petiole of Angiopteris , § in. thick, there 
were three such sets. The possession of more than one set 
of concentrically arranged meristeles in the petiole seems to 
be generically distinctive of Angiopteris. It may be that, 
in the full-sized fronds of the larger species of Marattia, 
the same thing may be found, but in the absence of the 
