LECTURES. 31 
the flower cannot be self-fertilised, while the pollen can be carried 
to older flowers. Afterwards the stigma ripens, and self-fertilisation is 
possible ; but still the chances are about even, for an insect is equally 
like to alight on pistil or petals ; in the former case cross-fertilisation 
is almost certain, and in the latter, though self-fertilisation is probable, 
it is not certain, for an insect may dust only one side of its head with 
pollen, and on approaching the stigmas, rub them with the other. 
In the Meadow Rue ( Thalictrum) the stamens play the part of 
petals. 7. minus, which is sometimes cultivated in gardens owing to 
the resemblance its leaves have to Maidenhair Fern, grows in exposed 
places, and it is fertilised by the wind; but 7: flavum frequents rich 
meadows which abound in insects, and its stamens are bright yellow, 
and attract pollen-feeding flies. In Anemone and Marsh Marigold 
(Caltha) the sepals play the part of petals Anemone has no honey, 
but probably provides succulent juices for its visitors, and we all know 
how honey shines on the carpels of Ca/tha. In all these cases self- 
fertilisation is not certain without insect aid, but it seems to be effec- 
tive. The Columbine ( Agui/egia) shows a great advance in adaptation 
for insects. ‘The five petals are hollowed out to form spurs 15 to 22 
m.m. long, which act as nectaries, while the sepals which encase them 
are bright blue, and make the flower conspicuous. The flowers are 
distinctly proterandrous (¢.¢., the anthers are ripe before the stigma), 
and so long are the spurs that only one humble bee (Sombus hortorum) 
can reach the honey in the proper way. The position of the stamens 
and stigmas is such as to ensure cross-fertilisation during the process ; 
but here we have an instance of incomplete adaptation, for the pirate 
humble bee (2. ¢errestris) bites through the spur and steals the honey 
without benefit to the plant. I have watched one thrust his proboscis 
into several flowers in vain, as his proboscis was too short, and then 
crawl round to the back, and finally bite through the spur. In default 
of cross-fertilisation, self-fertilisation can take place, as the anthers are 
still ripe when tbe stigmas open. The Larkspur (Delphintum 
elatum) is the first instance we have come across of a plant which 
is so habitually cross-fertilised that self-fertilisation has become im- 
possible. Its striking appearance and blue colour mark it as an object 
for the visits of many bees. ‘The crumpled texture of the sepal walls 
prevents the plunder which occurs in Agui/egia. The upper petals 
unite to form a spur containing the honey, and the lower ones make 
a platform for the bee, and also bear yellow hairs which serve as path 
finders, and leave the bee no choice but to thrust his proboscis into 
the right place. The stamens and carpels are endowed with a curious 
