156 On the Species of the Gentis Primula in Essex.
veined, and so short that their tips scarcely protruded beyond
those of the anthers. The ovary containing the ovules
seemed at this period to be fairly well developed. The
stigma was very small, and supported on a short style at
about half the height of the anthers. This description
applies equally to both forms. Gradually, however, as the
flower developed the pistil of the long-styled form gained on,
and at last passed, the anthers; while in the short-styled
form the corolla-tube grew, raising the anthers far above the
pistil, until the normal state was reached.
There is one very important problem connected with hetero-
stylism on which scarcely any information is as yet obtain-
able. I refer to the transmission of its own form, or other-
wise, by the seed of any plant. Experiments on this point
are somewhat difficult to carry out owing to the time
required for the investigation, as Primulas do not often
flower the first year. The very meagre information which
Mr. Darwin gives on this point seems to have been collected
from various experiments on other matters which he had in
hand. He shows that an "illegitimate" union produces
offspring almost entirely of its own form, and adds (loc. cit.,
p. 208):—"The transmission of the two forms by hetero-
styled plants may perhaps be found hereafter to throw some
light on their manner of development." I am glad therefore
to be able to announce that our member, Mr. E. Rosling, of
Chelmesford, has undertaken to conduct experiments on the
point, in the hope of arriving at some definite results.13
13 Note on Primula, longiflora.—There seems to be some slight doubt
(vide 'Forms of Flowers,' p. 50) as to whether this species is heterostyled
or not. Having met with it this summer (1882) on the summit of the
Hill Marmore in the Engadine, I am in a position to say with certainty
that it is homostyled, although very variable in the length of its pistil.
In this extraordinary member of the genus the corolla-tube is generally
0.9 inch long, so that it is difficult to imagine any insect attempting to
reach the nectary. In withering, the corolla does not fall off, but bends
at the top of the calyx and hangs down. The edge of the entrance to the
tube is marked by a sharp ridge, immediately below which are the
anthers, the pistil projecting generally one-fourth of an inch at least
beyond them, although it is very variable in length, even in flowers on
the same umbel. Thus out of seventy-two umbels which I examined,
sixty-seven were truly homostyled. Of the rest, one seemed to be equal-
styled, and four seemed to be short-styled; this difference being almost
invariably brought about by non-development of the pistil, as in the
case with the other species. For instance, one umbel had one flower with
the stigma above the anthers, a second with it on the same level, a third
with it just below, and a fourth with it far below ; while the fifth flower
had its tube only half the usual length, and the stigma projecting far
beyond it. Another umbel bore three ordinary flowers, and two with the
stigma below the anthers. Yet another umbel had the stigmas at many