elxiv Proceedings of the Asiatic Society of Bengal. [N.S., XV, 
various groups of the Liverworts let us see what relation they 
can have with the adjacent groups. 
We have seen that no connection can be demonstrated 
at 
accept the view elaborated above that the simplest terms are 
the last terms in.a descending series. If these simple forms are 
derived from the Algae it is strange that no trace of this 
relationship has been left anywhere. On the other hand we can 
say that there is no trace because there has never been any 
relationship. 
On the other side we find several organs in common in the 
Liverworts and the Pteridophytes. The natural conclusion is 
that they have had some connection. Whether the Liverworts 
are derived from any of the modern groups of Pteridophytes 
through some unknown forms or from an extinct related group it 
is reasonable to suppose that, since evolution in this group has 
been towards simplification, the best developed gametophyte in 
the Pteridophytes would be somewhat like the ancestral form 
which gave rise to the Liverworts. Sucha gametophyte is met 
with in some species of Lycopodium, e.g. L. cernuum. It is 
erect, radial, with a basal cylindrical portion and an upper 
leafy or lobe-bearing portion with a meristem all round. It is 
generally admitted that the genus Anthoceros approaches 
the higher types more nearly than any other Liverwort. on 
account of the highly differentiated capsule and the imbedded 
sex-organs. In this connection the radial and erect specimens 
of Anthoceros erectus referred to above acquire # peculiar signi- 
ficance. These specimens are very much like the prothallus of 
Lycopodium cernuum in general appearance except for the lobes 
of the latter 
There must have been at least three different lines of 
simplification from such a type, represented by the Marchan- 
tiales, the Jungermanniales and the Anthocerotales. The first 
step would be a change from the erect and radial position to a 
prostrate and dorsiventral habit. Such a stage is very clearly 
shown by the prothallus of Equisetum debile described by 
the writer a few years ago. (Annals of Botany, 1913.) It 
shows a greater ressemblance with the types of the Marchan- 
tiales than the prothallus of any other Pteridophyte. It 
but it has no basal erect region at all, in which respect it 
differs from the prothallus of Lycopodium cernuum. nder 
certain conditions the prothallus of Equisetum debile develops 
only a single growing point like the prothalli of other species of 
this genus, and is then very much like an unbranched Riccia, 
only very small. The fully developed prothallus however may 
be as much as an inch across. The erect branched lobes of this 
prothallus correspond to the lobes of the prothallus of Lycopo- 
