REPRODUCTIVE ORGANS IN THE PRIMULACES. 83 
are non-dimorphie so far as structure is concerned: in respect to 
their functional performances, I will here merely remark that, 
from my experience, two of the above species, viz. P. mollis and 
Scotica, seem to be perfectly self-fertile ; in respect to a third, 
P. verticillata, this is very doubtfully the case, as the experiments 
which I will give in a subsequent part of my paper will show. In 
the genus Primula, then, as in that of Linum already illustrated 
by Mr. Darwin, we have species structurally and functionally 
hermaphrodite, as well as species which, though hermaphrodite as 
to structure, have undergone such differentiations in their male 
and female sexual elements.as to render their mutual functional 
action so highly imperfect, that they have been aptly designated 
“ subdiecious hermaphrodites.’ How the former of these, the 
* non-dimorphie," should have escaped the notice of such excellent 
observers as Koch and Tausch, I fail to understand; but I feel 
convinced that a careful re-examination of the species will cer- 
tainly induce them to cancel the above statement, if indeed the 
evidence which I have already adduced and that which I have 
yet to lay before the Society is not sufficiently demonstrative. 
3. Gregoria presents at least one structurally “ dimorphic” 
species, Œ. Vitaliana : respecting the others I am in perfect igno- 
rance, never having seen specimens, In the “ stort-styled " form 
of G. Vitaliana the anthers reach the mouth of the corolla-tube, 
while the pistil is about half the length of the latter. In the 
*]ong-styled" plants the anthers are attached about one-third 
below the mouth of the corolla-tube, the pistil in this case being 
exserted beyond the latter and fully twice as long as in the 
“short-styled” form. The pollen-grains in both forms are very 
similar—of an ovoid or oblong shape: the stigmas likewise are 
nearly of the same shape in both forms, apparently rougher in 
the “long-styled”; but I cannot speak confidently as to this, 
having only had dried specimens of the latter form to judge from. 
In respect to the reciprocal functional relations of these forms, I 
have no knowledge ; but I think the following experiments on the 
“ short-styled " form fully justify us in predicating the existence 
of a functional dimorphism. Three plants on which I had an 
opportunity of experimenting produced in the course of the season 
twenty-one flowers, which I treated in the following manner: 
five were left to natural agencies; eight were fertilized with own 
pollen; the remainder (eight) were fertilized with pollen from one 
of the other plants. The results were the complete abortion of 
every capsule of the two former, while in the latter case two cap- 
