Lycopodiales. 18 y 
conn ” even in the genus Lycopodium. These parenchymatous 
swellings may be looked upon as opportunist growths, rather than 
as persistent relics constant from a remote ancestry. This view is 
greatly strengthened by the occurrence of protocorm-like develop¬ 
ments in isolated cases among the Angiosperms. Phylloglossum 
with its large storage “ protocorm ” would then be the extreme type 
of a line of embryological specialization, not a form preserving the 
primitive embryological characters of the whole race ” (7). The 
opposite view, first elaborated by Dr. Treub and formerly held by 
Professor Bower himself (5), is held by Dr. Lang (15), by Professor 
Thomas (22), and by Miss Sykes (21). Dr. Lang brings forward 
several considerations in support of the primitiveness of the proto¬ 
corm. He points out that the type of prothallus found in Lyco¬ 
podium cernuum, L. inundatum and L. salakense is probably 
relatively primitive, since it is not modified in connection with 
epiphytism and with saprophytism. The habit of the gametophyte 
can legitimately be held to influence the extra-prothallial protocorm, 
for Professor Thomas has shown that, at least in Phylloglossum, 
the latter organ is formed directly the young sporophyte escapes 
from the embryo, while the prothallus often retains its vitality for 
some time afterwards. If, then, we admit that the prothallus of 
these species of Lycopodium is relatively primitive (and this seems 
to be the case as will be shown later), and that the structure and 
habit of the gametophyte may affect the young sporophyte, we 
should expect the young conditions of L. cernuum, L. inundatum, 
and L. salakense to be relatively primitive. Now these three 
species and Phylloglossum (whose prothallus corresponds in 
essentials to theirs (22) ) are those that develop a protocorm. 
Moreover, as Dr. Lang points out, the young sporophytes growing 
from the root tubercles of L. cernuum also produce protocorms (15). 
These can hardly be regarded as embryological specializations. It 
seems, too, that the occurrence of a protocorm in species of both 
genera supports the primitive nature of this organ, for it would be 
natural to suppose that their common ancestor possessed a proto¬ 
corm. But it is clear that Phylloglossum is very close to Lyco¬ 
podium, and it has been suggested (2), (21) that it arose by reduction 
from the section of the genus Lycopodium that includes L. cernuum, 
and that it should not be separated from Lycopodium. In support 
of the primitiveness of the protocorm Dr. Lang alludes to the 
presence of a rudimentary protocorm bearing rhizoids in an example 
of L. Phlegmaria (15). Professor Bower objects that the position 
