J. J. Simpson 
219 
3. Conception of Ghorisis. 
Cliorisis or reduplication is generally looked upon by botanists as a means of 
multiplication of the parts of a flower. It consists in the division or splitting 
of an organ in the course of its development by which tw^o or more organs are 
produced in place of one. Chorisis may take place in two ways : 
(1) transversely — when the increased parts are placed one before the other, 
that is, the resulting components are on the same radius ; this is known as vertical, 
parallel or transverse clioi'isis ; 
(2) collaterally— whan the increased parts stand side by side, that is, on the 
same circumference. 
Transverse chorisis is supposed to be of frequent occurrence ; thus the pistils 
of Lychnis and many other caryophyllaceous plants exhibit a small scale on the 
inner surface at the point where the limb of the petal is united to the claw. The 
formation of these scales is supposed by many to be due to the chorisis or unlining 
of an inner portion of the petal from the outer. 
Collateral chorisis is seen in different natural orders. In Strephantkus, in place 
of two stamens there is sometimes a single filament forked at the top and each 
division bears an anther. This is usually supposed to be due to collateral chorisis 
arrested in its progress. 
The flowers of the Fumitory are also generally considered to afford another 
example of this type of chorisis. In these we have two sepals, four petals in two 
rows and six stamens, two of which are perfect and four more or less imperfect. 
The latter are said to arise by collateral chorisis, one stamen being divided into 
three parts. 
Collateral chorisis may be compared, according to Bentley, to a compound leaf 
which is composed of two or more distinct and similar parts. 
Let us now consider chorisis in its bearing to the flower under consideration. 
In the description of the morphology of the flower we noted that in the inner 
whorl of the androecium there were four stamens arranged in pairs while in the 
outer whorl there were only two stamens situated singly. Various opinions have 
from time to time been advanced to explain this anomalous structure so that it 
might be well to briefly review these. Of the androecium of the Cruciferje Oliver 
says : " The two pairs of long stamens are generally thought to be due to chorisis 
or the division in the course of development of single antero-posterior stamens. 
Others have thought tliat the six glands represent abortive stamens and that these 
with the six stamens make up a normal series of twelve in three whorls." 
De Candolle held the view that the stamens formed a single, originally 
tetramerous whorl alternating with the petals in which the median members, 
i.e. the anterior and posterior, were cleft (chorised) in two. Since however the 
lateral stamens are inserted lower down than the median stamens and are also, 
