ii8 
JOHN T. BUCHHOLZ 
Dicotyledony may have been attained in more than one way. One of 
these would be by checking the growth of the cotyledonary tube in two 
places during its development, much as a strongly bilabiate corolla develops. 
Another method is by a fusion of the cotyledons in two groups. This is 
strongly suggested by the bilateral grouping of the cotyledons in Pinus 
Laricio, and doubtless the tendency for the cotyledons to fuse, acting in 
concert with the cause which brings about this bilateral symmetry, could 
produce dicotyledony from polycotyledony. 
That cotyledonary tubes are frequently found among gymnosperms was 
brought out by the work of Hill and DeFraine. The recent work of Hutch- 
inson (7) on Keteleeria also shows that this embryo has a well developed 
cotyledonary tube and at the same time only four cotyledons, a rather 
reduced number. The fact that cotyledonary tubes have been found in 
angiosperms connects these with the polycotyledonous gymnosperms. 
Coulter and Land (4) have shown in a recent investigation how mono- 
cotyledony has been derived from dicotyledony by a zygomorphic develop- 
ment of the cotyledonary zone of an embryo which has a cotyledonary" 
tube in an early stage of its development. 
The writer has expressed the opinion that the cotyledonary tube had 
its origin in cotyledonary fusions. This is further suggested by the fact 
that Agapanthus has two primordia on a cotyledonary tube while Cyr- 
tanthus has four primordia (4), and no doubt these are still further examples 
of embryonic recapitulation of an ancestral character, therefore distinctly 
pointing to polycotyledony as the more primitive condition. 
The account which Coulter (3) gives for grass embryos as well as the one 
for Cyrtanthus completes our analogy between embryo and corolla develop- 
ment, for these monocotyledonous embryos come to correspond to the uni- 
labiate or ligulate corolla by becoming extremely zygomorphic. 
Summary and Conclusions 
The results of this investigation show that in a number of conifers fu- 
sions of the cotyledons occur during their embryonic development. It is 
significant that no evidences of splitting cotyledons were found in any spe- 
cies. 
The larger number of primordia found in the species exhibiting fusion is 
a recapitulation of a more primitive condition in which a larger number of 
cotyledons existed. 
The fusion of cotyledons has given rise to a reduced number of cotyledons 
and also to cotyledonary tubes in some species. The occurrence of coty- 
ledonary tubes in gymnosperms and the retention of this feature in some an- 
giosperm embryos points to polycotyledony as the primitive condition. 
All the definitely known facts brought out by others, as well as the in- 
vestigations of the writer, may be used to support the following conclusions; 
