456 THE FERTILISATION OF FLOWERS. [PART III. 
accessible to all our humble-bees, the shortest proboscis among 
which, that of B. terrestris, is 7 to 9 mm. long. The mouth of the 
corolla is a slit 6 to 7 mm. long, which, however, is reduced to 4 
mm. by the position of the upper lip. The remaining posterior part 
of the slit is blocked by the stamens in the manner already described, 
so that the bee can only insert its proboscis beneath the anthers. 
The stigma curves down so far over the anthers (6, Fig. 156) 
that an insect’s proboscis must touch it before entering further and 
being dusted with pollen. In general, however, bees’ visits are 
scarce ; for the corolla is almost inclosed in the vesicular calyx, 
and is therefore inconspicuous. In absence of insects, self-fertilisa- 
tion always takes place; for the mouth of the corolla gradually 
opens more widely, the under lip bending downwards, and the sides 
of the upper lip spreading further apart (7, Fig. 156); at the 
same time the style elongates and curves downwards and inwards 
(7, 8), so that the stigma comes to stand under the anthers, or 
even between them if they have separated in withering. 
f, major, Ehrh.—In the large-flowered form, the tube is about 
2 mm. longer than in the small-flowered ; but this small difference 
suffices to exclude many of our humble-bees from the honey, or to 
cause them to bite a hole in order to gain access to it. Sprengel 
observed these holes, but they are not made, as he said, by “a 
large humble-bee, for whom the natural entrance is too small,” for 
all humble-bees insert only their proboscis into the flower ; but by 
those with the shortest proboscides, namely, as I have repentant 
observed, by B. terrestris, L. 9 and §, and Bb. pratorum, L. %, 
whose proboscides measure respectively 7 to 9 mm. and 8 mm. 
The size of the parts projecting beyond the calyx, viz. the 
upper and lower lips, differs more than the length of the tubes in 
the two varieties or sub-species. The upper lip is in &. minor 7 
to 8 mm., in &. major 10 to 11 mm. long. The flowers of 2. major 
are therefore more conspicuous, and their conspicuousness is in- 
creased by the pale colour of the bracts and by the violet colour 
of the tip of the upper lip, which is usually white in &. minor. 
As both species grow together in the same spots and flower at the 
same time, they give a good opportunity of proving that the large- 
flowered form is really much more plentifully visited than the 
other. In point of fact, insect-visits are so plentiful in the case of 
R. major, that it has been able to dispense with the power of self- 
fertilisation, and the style, instead of bending downwards, projects 
more and more from the mouth of the flower. 
We have thus in Rhinanthus major and minor two forms 
