INSECTS AND FLOWERS. 
119 
anthers is provided with two long hairs, li, which, in this stage 
of the flower, project into the opening of the corolla, so that an 
insect thrusting its head into the flower necessarily strikes 
against them, and in so doing shakes the pollen out of all the 
anthers on to its proboscis, and then carries it to the next 
flower it enters. At a later stage the parts occupy the relative 
position shown in b ; and by this time the stigma is brown and 
withered, so that self-fertilisation is impossible. In the flower 
of the smaller form, (fig. 2), the anthers have completely 
discharged their pollen before the flower has fully opened (in 
the state shown in a); and the stigma, st , instead of over- 
topping the anthers, is slightly below and almost in contact 
with them. When the flower is fully open (b, c ) the stigma is 
already withered up, so that cross-fertilisation is almost 
impossible. Dr. Muller has observed three species of Diptera 
and four of Hymenoptera visiting the larger form of the 
Euphrasia, but never any on the inconspicuous ones. The 
same dimorphism is found, under similar conditions, in several 
other plants — Muller mentions especially Lysimachia vulgaris 
and Rhinanthus Crista-galli — accompanied by similar con- 
trivances for ensuring different modes of fertilisation. In the 
latter case the two forms have been described as distinct 
varieties, R. Crista-galli , a, and /3 Linn., and even as distinct 
species, R. major and minor Ehrh. In this last plant the 
small-flowered form appears to be frequently visited by insects, 
and to be self-fertilised only when by any chance it is left 
nnvisited. 
One of the most remarkable contrivances for ensuring cross- 
fertilisation is exhibited by the birth-wort ( Aristolochia 
Clematitis) a rare wild flower in our southern counties. Fig. 
3 a represents a flower cut through lengthwise when just 
opened. The stigmas, st , are in a receptive condition ; but 
the anthers, an, are still closed. A small insect, in, which 
has brought on its back a mass of pollen from an older flower, 
has passed through the tube of the calyx, t, and reached the 
globular cavity below ; the retrorse hairs which clothe the tube 
of the calyx effectually preventing its exit when once im- 
prisoned. During its imprisonment the insect is sure to 
deposit some of the pollen on the stigma, the lobes of which 
then curve up, as represented in 6. This curving up of the 
stigmatic lobes enables the anthers to open and discharge their 
pollen, some of which must adhere to the bodies of the still 
imprisoned insect. The hairs in the calyx-tube now dio 
away, so as to allow the insect to escape and carry the fresh 
load of pollen to other flowers. But the flower has now alto- 
gether altered its position. As long as the stigma was still in 
a receptive condition, the pedicel was erect, and the calyx open 
