302 
H. F. Wernham 
stage, and before ovule-economy had proceeded even so far as in 
the Geraniales and Sapindales (see diagram below). 
POLEMONIACEAE 
I 
RANALIAN ANCESTOR 
Diagram illustrating the origin suggested for the Polemoniaecze. 
In connection with the Transitional Group it is necessary to 
draw attention to a noticeable tendency in certain of its components, 
namely, to secondary or “ spurious ” septation of the ovary. The 
essence of this septation lies, presumably, in its initiation subsequently 
to the primordial development of the pistil; it results in the division 
of the ovary into a number of loculi exceeding the number of carpels. 
The problem in any particular case is, of course, intimately associated 
with development, and a wide and useful field of enquiry is open in 
this regard. For the number of loculi is by no means an infallible 
guide to the number of carpels, especially when this is an even 
number* In the case, say, of a three- or five-locular ovary it is 
