110 MR. J. SCOTT ON THE FUNCTIONS AND STRUCTURE OF THE 
us the manner by which the at present normally characteristic 
* subdicecious hermaphrodism ” of the P. veris—and, of course, of 
all the other dimorphic species of Primalus—has been attained, 
and, furthermore, take us back to that period in its genealogy 
when non-dimorphism or perfect hermaphrodism was the genital 
characteristic of its line. Here, as I believe, we are afforded an 
instance of variability perfectly analogous to, though certainly 
less remarkable than, the Catasetum case, so ably elucidated by 
Mr. Darwin. And just as Catasetum and Monochanthus occasionally 
produce each other and likewise the hermaphrodite Myanthus, 
thus inducing us to regard the former sexual forms as the modified 
descendants of the latter, so in the case of the non-dimorphic 
plant of the Cowslip, we see first the original condition in those 
flowers whose stamens and pistils are of an equal length, and a 
mutual adaptation subsisting between their male and female 
sexual elements, and secondly, the earliest indications of a 
divergence from that condition and a tendency to the dimorphic 
in those flowers with stamens and pistils of different lengths. 
In the two forms of Primula Sikkimensis the stigmas differ 
little in shape or roughness, but greatly as respects the length of 
their styles, that of the long-styled being fully four times longer 
than that of the short-styled form. The stamens in the long- 
styled form rise little above the ovarium; in the short-styled 
form they are attached halfway up the corolla-tube; so that the 
relative differences in the length of these organs in the two forms 
are much less marked than those of their styles. There are also 
very marked differences in the pollen-grains of the two forms; 
those of the long-styled plants are sharply triquetrous, smaller, 
and more transparent than those of the short-styled, which are of 
a bluntly triangular shape. 
In respect to the relative fertility of the two forms, I may state 
that when carefully protected from insects they rarely produce a 
single seed. The long-styled form I have indeed found perfectly 
sterile when thus protected, while the short-styled form under 
similar treatment occasionally produces a few seeds. This slightly 
greater self-productiveness of the latter form is readily explained 
when we take into consideration the relations of its anthers and 
stigmas, these being relatively so disposed that it is scarcely 
possible for the dehiscence of the anthers to take place without & 
few of the pollen-grains falling upon the stigma. In the long- 
styled form, on the other hand, where the stigma rises high above 
the anthers, such a result can only be effected by insects or other 
