364 
POPULAK SCIEJ^CE KEVIEW. 
Vauciier added other illustrations, but it was not till the 
publication and experimental researches of Darwin that the 
subject received the attention it merited. Darwin’s classical 
paper on Primula, in the “ Journal of the Linnean Society,” fol- 
lowed up by similar memoirs on Ly thrum, Linum, &c., stirred 
up a host of observers. Large numbers of new facts were 
recorded, all confirming the idea that while self-fertilisation is 
not impossible, and is indeed in some few cases inevitable, yet 
in the majority of instances some provision for cross-fertilisa- 
tion is afforded, if not constantly, at least occasionally. Hilde- 
brand, Delpino, Alfred Bennett, and others, have correlated 
these facts, and, making allowance for exceptional instances, 
they one and all confirm the views expressed by Darwin. The 
observations and experiments of these gentlemen are so well 
known, and the records of them are so accessible, that we 
do not propose now to occupy space by making further allu- 
sions to them. Suffice it merely to say, that these observa- 
tions have reference to the facilities offered by various 
structm'al modifications for favouring cross-fertilisation and 
preventing self impregnation or vice versa, and they all 
tend to show the advantage that accrues from an occasional 
cross. 
Our object in the present communication has reference to 
another department of the subject ; and it is one we think 
it desirable to call attention to, as it appears not to receive so 
much attention as its importance demands. We allude par- 
ticularly to the circumstances promoting the development of 
pollen-forming or ovule-bearing flowers respectively. 
Before proceeding further we ought to explain that in using 
the expressions male, female, or hermaphrodite flowers, we do 
so, unless otherwise stated, with reference to structural con- 
ditions rather than to physiological office. The term “ bisexual” 
is preferable to that of hermaphrodite, as not implying any 
physiological distinction. 
For our present purpose, then, a flower with stamens and 
pistils perfectly formed is bisexual or hermaphrodite, even 
though its pistil be not fertilised by its own pollen, but by that 
derived from some other source. 
For convenience sake, we take a bisexual flower as our 
starting-point ; and we propose to allude to various not infre- 
quent changes observed in flowers of that description, in conse- 
quence of which their sexual organisation becomes more or 
less materially modified. 
A plant usually producing flowers, bisexual or hermaphro- 
dite as to structure, may bear flowers of one sex only by the 
simple arrest of growth. Thus, if the stamens of any given 
flower be arrested in their development, the blossom becomes 
