EVOLUTIONARY STATUS OF POLYCOTYLEDONY 
107 
yledonary tubes are the direct result of the incomplete fusions, it is just as 
easy, on the basis of the anatomical facts presented by Hill and DeFraine, 
to pass from polycotyledony to dicotyledony as vice versa. In support of 
the view that polycotyledony is primitive, Sister Helen Angela (i) has 
prepared a series of diagrams based on vascular anatomy, which show all 
stages in a series of intergrading forms from the polycotyledonous to the 
dicotyledonous condition, in Coniferales as well as in Cycadales. But the 
evidence mustered by vascular anatomy or by the study of the occasional 
freaks in which the cotyledons seem to be partly divided, is merely proof 
that one condition has without doubt arisen from the other, and leaves us 
with no very positive clue as to which course evolution has been taking — - 
which condition has actually given rise to the other. 
The study of fossil material has frequently furnished us with a def- 
inite record of the past history of a group. The fact that the embryos that 
have thus far been found in the most primitive gymnosperm seeds, those of 
the Benettitales, are dicotyledonous has led to a rather widespread impres- 
sion that geological proof establishes dicotyledony as the more primitive 
condition. On the other hand, it is now well known that the cycad line, 
of which the Benettitales are the Mesozoic representatives, has been dis- 
tinct from the conifer line since Paleozoic time. The Cordaitales were 
probably the ancestors of the whole conifer line which includes Ginkgoales 
as well as Coniferales. The cycad line, on the other hand, has been derived 
from the Cycadofilicales which also existed in Carboniferous time along 
with the Cordaitales, and these two phylogenetic lines of seed plants have 
been distinct from each other since very early times. Thus the dicotyle- 
donous embryos of the Bennettitales do not represent the ancestral condition 
from which the conifers were derived, and we have no knowledge of the 
ancient conifer embryos from a study of fossil material. Coulter and Cham- 
berlain (5), who do not accept the conclusions of Hill and DeFraine as final 
but favor the opposite view, point out that "probably our oldest group of 
Coniferales, older even than the Cycadales and Benettitales with which 
we are acquainted, is the extreme illustration of polycotyledony, while the 
youngest of the Coniferales are dicotyledonous or nearly so." Until some 
paleobotanist describes embryo-bearing seeds of the most ancient conifers 
or Cordaitales, we shall need to look to our living material for our informa- 
tion, or content ourselves with a philosophical discussion of the question. 
Investigation 
This paper is the result of a study of the ontogeny of the cotyledons in 
various living species of conifers, in the hope that this evidence may reveal 
modern evolutionary tendencies and afford a safe criterion from which to 
determine in what direction this particular evolution of the cotyledons has 
been proceeding. When such evidence is not taken from a single isolated 
