642 GRANATE. 
as seen on a transverse section carried through the upper 
portion have a marked correspondence, and in cases where 
one is deficient, one of the septa is much increased in size, 
indicating it may be fairly inferred an obliteration of the 
cavity, the number agreeing with the upper tier of cells. It 
seems to follow, owing to the greater permanence of the style 
or stigma than of the ovarial cavity itself, that no other car- 
'pella enter into the formation of the fruit, and hence the 
idea of an inner or lower series is untenable. Of the greater 
permanence of the styles or stigmata over the cavity of the 
carpellary leaf itself ample proofs exist in Composite Grami- 
nez Plumbaginez, and Cupuliferze ? 
It need not therefore be insisted on, it may be sufficient to 
state, that if Mr. Lindley’s view be correct, there must be 
the same number of stigmata as there are cells. This I do 
not find to be the case, my idea is this. The ovarium is formed 
of from 5 to 8 cells, which towards the base of the fruit be- 
come much distorted by various growths of the placentz, 
and it is possible that a sort of dislocation adds to the con- 
fusion, that the number of cells can only be ascertained by & 
transverse section carried through the upper portion of the 
ovarium, 
The placent: are various in the situation of the upper cells, 
they are always parietal, of the lower both axile, fundamen- 
tal and parietal. It is a general rule that towards its apex 
the pistilla when compound has a tendency to revert to its 
utmost possible simplicity of structure. 
The situation of the placentze of the upper cells is alone, a 
great objection to Mr. Lindley’s view, since the placental 
suture would, allowing it to be true, be turned away from, and 
not towards the axis, an anomaly quite unheard of. 
In my view the irregularity is due to cohesion, and is not 
opposed to any well known law. 
In very young ovaria before the appearance of the ovula, 
the cells are distinctly marked oat, and they are evidently 
