128 Gibson.— Contributions towards a Knowledge of 
dorsal leaves, as in the case of the ventral leaves, bear stomata 
on both sides. 
Erikson’s monograph 1 deals in considerable detail with the 
leaf-structure in the Lycopodiaceae. The author has examined 
thirty species of Selaginella , apparently chiefly from herbarium 
material ; of these I have examined twenty. His work, to 
a large extent, covers the same ground as my own ; still, as 
I have had opportunities of studying thirty-two species 
which Erikson has not had access to, and as I am able to 
supplement to some extent many of his observations, and to 
add several points which he has not touched, I may venture 
to consider his memoir as not rendering the publication of 
the present paper altogether unnecessary. Into the details 
of his work I need not at present go, but merely give, for 
convenience of comparison, his scheme of classification of the 
species according to their leaf-structure. Erikson distinguishes 
three types of leaf : the first division includes such as have 
leaves with the upper epidermal cells conical and with, at the 
same time, a homogeneous spongy mesophyll, e.g. 5. Martensii , 
&c. ; the second group includes those which have leaves 
provided with a distinct palisade-layer, some of these, like 
S. Kraussiana , with upper and under epidermal layers similar, 
others, like 5. Wildenowii , with upper epidermis different 
from the under. The third group have leaves without 
palisade-layers, again subdivided into those with a similar 
epidermis above and below, e.g. S. spinosa, and those with 
epidermis above and below dissimilar, e. g. S. serpens. In 
comparing Erikson’s scheme with that of Dangeard, it will 
be seen that the former takes the presence or absence of a 
palisade-layer as the basis of primary division, and subdivides 
the groups according to whether the epidermis of the upper 
side is like that of the lower side or unlike ; whilst Dangeard 
selects the character of the epidermis as his primary basis of 
classification, and subdivides according to the nature of the 
mesophyll. It seems to me quite immaterial which plan is 
1 Bidrag till Kannedomen om Lycopodinebladens Anatomi : Arbet f. Lunds 
Bot. Instit., 1892. 
