304 QUESTIONS IN RELATION TO THE 
with four or even, under pressure, more especially towards the interior 
of the flower, and rarely in the outer circles, three and two. In cruci- 
form flowers the calyx and corolla have four parts each, the stamens 
are six, unequal, and there are seemingly but two carpels though with 
an anomalous connectiou between their opposite edges, which demands 
explanation. The late eminent Robert Brown, than whom a higher 
authority cannot be appealed to, considered the fruit as really consist-. 
ing of two carpels, whose placental edges are at the part where they 
first touch each other, but the exterior covering of each of which 
extends until the parts meet in a median line, thus forming a spurious 
partition. There is nothing impossible in this explanation since the 
separation of the principal portion of the carpel leaving the placenta 
in its position, occurs in other instances, and there are probably 
examples sufficient to justify the notion of the spurious partition 
though it is something extraordinary. Considering the cases in which 
a line is observable down the middle of the partition, and others in 
which there is a partial or even entire separation into two parts, it. 
must I think be agreed that the partition is due to the meeting im the 
middle of two parts projecting from the placental hnes; but I confess 
I greatly prefer another theory which had occurred to myself many 
years ago, and which I have since ascertained to have been proposed 
by Lindley and defended by Kunth. This is the supposition that the 
fruit is really formed of four carpels, two of which are abortive, their 
remains forming the partition, whilst the remarkable circumstance of 
the stigmas being in the line of the placente is accounted for by the 
fact that each stigma is double, formed by the union of one from 
each carpel, the tip of the carpel dividing into two portions as in some 
other instances. This explanation is greatly confirmed by the manner 
in which the alternate circles of cruciferous flowers exhibit increased 
development in opposite directions, the largest pair im one circle 
being opposed to the lesser pair in the next. On the whole, though 
Dr. Gray adheres to Brown’s theory, 1 cannot but consider the other 
as better explaining all the facts of the case, and it is especially 
confirmed to my mind, by the consideration of the deviative structure 
of Parolinia. In vain it is contended by Moquin Tandon and Webb, 
in their ingenious article on the subject, that the prolongation of the 
valves into extremities with two horns is an unmeaning and unimport- 
ant aceident. I cannot look at the figures which are said faithfully to 
represent the fruit of Parolinia, and which I have copied for your 
