140 BOTANICAL GAZETTE [avcust 4 
the center of each stem bundle of Tmesipteris there is a cluster of — 
initial elements, the breaking-down of which often results in the i 
formation of a lacuna. Although the cylinder of the stem of Selagi- — 
nella is characteristically exarch, Greson (5) has noted, in the case — 
of the protostelic $. spinosa, that in the trailing portion of the stem, — 
all the protoxylem is in the center of the cylinder, completely inclosed 
by metaxylem. The root in the whole genus is also distinctly 
mesarch. In Phylloglossum, it is well known that the tubular 
stele at the base of the peduncle, the ring of bundles into which this — 
divides higher up, and the traces of the sporophylls, have their 
first-formed wood elements surrounded on all sides by later q 
ones. Exarch development seems entirely lacking, both in this — 
genus and in Tmesipteris. 3 4 
In Lycopodium, however, precisely the reverse seems at first ; 
‘sight to be true, and I have been unable to find a recorded instance — 
of the occurrence of mesarch development in the genus. 
In an endeavor to determine the presence or absence of such a char- 
acter, the following species and varieties of Lycopodium were investi- 
gated: L. inundatum L., and var. Bigelovii Tuckerm.; L. lucidulum 
Michx.; L. annotinum L.; L. obscurum L.; L. tristachyum Pursh; — 
L. complanatum L. var. flabelliforme Fernald; and L. clavatum L. 
L. inundatum and its variety are somewhat delicate forms, and 
have by far the most poorly developed vascular system of the species 
looked at. In both, the xylem rays are few in number (3-6); and 
along the much broadened end of each extends a row of crush 
protoxylem elements. There is no indication of metaxylem outside 
of these. In the leaf-trace, however, especially at a little distance 
from the cylinder, the smaller elements, ringed or loosely spiral, ten¢ 
to become clustered at the center of the bundle. Just before the a 
trace enters the leaf, these elements break down, leaving a protoxylem 2 
lacuna, completely surrounded by later-formed, closely spiral ele- = 
ments. This is perfectly evident in both transverse and radial sec 
tions (figs. 1, 2). 
In the case of L. lucidulum, the central cylinder is also small, 1 
comparison to the diameter of the stem, and its xylem rays are few 
in number and broadened at the ends. As in L. inundatum, howeVv — 
there is no indication of centrifugal wood in the cylinder itself, but ea 
