No. 421.] CERTAIN ANOMALOUS DICOTYLEDONS. II 
leaf? This view is much strengthened by the persistence of 
the bilobed form in the leaves of the young plant for several 
years, and the permanently binate leaves of the allied genera 
Jeffersonia and Diphylleia. The position of the plumule at 
the base of the cotyledonary tube suggests a possible lateral 
origin for this, as in typical monocotyledons and in Nelumbo. 
The character of the flowers and the tissues in Podophyl. 
lum and Diphylleia, as well as in Caulophyllum, are in accord 
with a possible monocotyledonous affinity for these forms. It 
is, at any rate, highly desirable that a careful embryological 
study should be made of these extremely peculiar plants. 
The formation of a cotyledonary tube apparently similar to 
that in Podophyllum has been described for a number of widely 
separated forms, e.g., Anemone, Delphinium, various Umbel- 
liferze, Megarrhiza californica, Dodecatheon meadia. Whether 
in all of these there is really a coherence of two cotyledons 
must be determined by a study of the embryogeny. 
It is evident that the last word has not yet been spoken as 
to the interrelationships of the angiosperms. The numerous 
studies upon the development of the embryo sac, which have 
appeared during the past few years, have shown that there is 
much more variation in the structures of the embryo sac than 
was supposed to be the case. The most marked departure from 
the angiospermous type is the genus Peperomia. The embryo 
sac of this genus has regularly sixteen nuclei instead of the 
eight usually present, and in this respect shows an approach to 
the condition obtaining among the gymnosperms and higher 
pteridophytes. 
Peperomia, which appears to be genuinely dicotyledanpus, 
nevertheless in the structure and distribution of the vascular 
bundles, as well as in the character and arrangement of the 
flowers, approaches some of the simpler monocotyledons, espe- 
cially the Aracez. It seems possible, as the writer has else- 
where suggested, that there may be two points where the two, 
great divisions of angiosperms come together. 
From the evidence at hand it would appear that the two 
"uM Se and dicotyledons — are of about equal 
1 Holm, doc cit., p. 422. 
