of Lycopodium clavatum , L. 309 
Lastly the species of Lycopodium with saprophytic prothalli 
must be compared with the L. cernuum type in respect to 
their embryology,' before the attempt is made to estimate 
the phylogenetic significance of all the resemblances and 
differences that have been mentioned taken together. 
A well-developed protocorm is as yet known only in those 
Lycopods the prothalli of which belong to the L. cernuum 
type, though indications of it have been recognized by Treub 
in some of the young plants of L. Phlegmaria. It is thus 
confined to those species whose prothalli grow on the surface 
of the soil, with which the young plant soon comes into 
direct relation by means of the rhizoids of the protocorm, 
and thus early becomes independent of nutritive supplies 
from the gametophyte. In all the other cases known as yet, 
the prothalli being buried in the soil or in rotting wood, the 
young plants have to attain a more or less considerable size 
before they are capable of assimilation and become indepen- 
dent of the reserve materials in the prothallus. The larger 
size of all the saprophytic prothalli may be put in rela- 
tion with this need. The very young embryo in all the 
Lycopods of which the embryology has been followed in 
detail, exhibits a stage in which it consists of a suspensor and 
two tiers of four cells each. Slight differences in the succession 
of the divisions by which this result is brought about are 
known, the two tiers being sometimes separated by the 
second division in the cell from which the embryo will be 
formed, in other cases by the third ; little importance can, 
however, be attached to such a distinction when the similarity 
of the result is borne in mind. From the whole of the tier 
of cells adjoining the suspensor the foot is derived in all 
cases. In L. Phlegmaria and the other species with this 
type of prothallus, the foot remains comparatively small, 
though much larger than the same structure in L. cernuum. 
From the terminal tier there originate the stem-apex, the 
first leaf, and the first root ; in some examples, as mentioned 
above, the appearance of the root is delayed, and a rudimen- 
tary protocorm can be recognized. By the elongation of the 
