426 CUPULIFERA. 
unilocular after dehiscence, the former being in every stage 
bilocular. 
Lastly the ovary offers the anomaly of a vascular fascicle 
in the same line or nearly so as the placente. 
The normal condition of an ovary is, to have one vascular 
fascicle in the axis of its dorsum, this may, or not be 
ramified in the ovary, but in almost all cases, it is single in 
the style; when not so, the number of carpella seems to 
be indicated by the grouping of the fascicles. This occurs 
in Cucurbitacez which seem to be binary carpelled, and the 
fascicles are divided into two groups, this point, however, 
requires fresh examination. 
In this plant the fascicles are 4, and one of these is 
larger than the others, and it would appear to be gene- 
rally separable into a fifth. As no fascicle has a right 
to exist in the situations these do, I conceive them to indi- 
cate the abortion of at least two, and perhaps three other 
carpella ; and moreover, as vascularity is a test of perfec- 
tion, that the apparently two styles, are nearly formed of 
2 each, or one of two, and one of three. This must be the 
case, if my views of the permanence of the test afforded 
by the number of vascular fascicles in the style as an indica- 
tion of the real number of: carpella, be correct ; because the 
styles, being more permanent than the ovary itself, must be 
developed considerably to enjoy such a supply of vessels. | 
. The true affinities of this plant are obscure, it is nearer Bur- 
seracez than Amyridez. 
. The nearest affinity is probably with Celastrinex. 
CUPULIFER.E. 
CARPINUS. 
Carpini sp. DLXV. Fig. I. 
Has. Bootan Tongsa, —— 
