No. 464.] STUDY OF THE SALICACEE. 535 
rily leads to the greater simplification of the pits which rarely 
retain the bordered form but become simple perforations and 
commonly disappear altogether. Concurrently with these 
changes, the progressive increase in the thickness of the wall 
eventually leads to a more or less complete obliteration of the 
original cell cavity; and while all these varied changes have a 
profound bearing upon the ultimate strength of the general 
structure, they at the same time operate to eliminate completely 
even such slight capacity for conduction of fluids as the elements 
. may have had in their earlier stages of development. 
The hypothesis as thus presented, is suggested at the present 
time in view of the many facts which have come under observa- 
tion during the progress of studies bearing more particularly 
upon the anatomy of the Coniferales as elsewhere recorded (Pen- 
hallow, :04c), and because of the need of some working basis 
from which we may view the anatomy of the Salicacez, and 
upon which we may erect a possible point of departure for an 
interpretation of their phylogeny. The first real test of the 
validity of this hypothesis will be made on the basis of data to 
be derived from a study of the Salicacez as recorded in subse- 
quent pages. If it proves to have the support of observed facts 
it will then be necessary to supplement it with the further 
hypothesis that the angiosperms as a whole had their origin 
in some generalized type, possibly identical with or at least 
nearly related to the Cycadofilices; and in any event we can 
hardly agree with the suggestion of De Vries (:05) that the 
“Monocotyledons “are obviously a reduced branch of the primi- 
tive Dicotyledons," since our own investigations not only fail to 
lend support to such an idea, but they tend to establish evidence 
to the contrary, and to the effect that while the two may possi- 
bly have had their starting point in a common, generalized type, 
they nevertheless represent two distinct lines of descent. On 
the basis of such a hypothesis, we should reasonably expect to 
find in the early Mesozoic and Permian, and possibly even in 
the later Carboniferous, transitional forms which would carry 
the angiosperms back to some of the later but less highly spe- 
cialized types of the Cycadofilices. 
(79 be continued.) 
