parti.) | THE MECHANISMS OF FLOWERS. 217 
Darwin had also shown that cross-fertilisation by insect-agency 
¢ takes place to a large extent in Phaseolus. Mr. Coe planted four 
rows of Negro Dwarf Kidney Beans between some rows of white 
and brown Kidney Beans; near by were some Scarlet Runners. 
He let the black Kidney Beans run to seed, and over # of the 
beans produced showed all gradations from light brown to black, 
and some were mottled with white. Of the plants reared from 
these seeds every one differed from the rest in stature, leaves, 
colour and size of flower, time of flowering and of ripening fruit, 
size, form and colour of the pods ; and the beans produced by them 
were of all shades between black and light-brown, some dark- 
purple, some’ slightly mottled, and of various shapes and sizes 
(151, 152). 
In P. multiflorus the carina, with the inclosed style, are so bent 
that when the carina is depressed the style emerges pointing 
downwards and towards the left, so that a bee can only accomplish 
cross-fertilisation if it enters the flower to the left of the coil. 
_ Francis Darwin has pointed out that the tenth, free, stamen bears 
an appendage which prevents the bee from taking any other way 
towards the honey (169). 
at eee es NUS 
if a : 
In regard to Treviranus’ opinion that self-fertilisation is the general rule in 
 Papilionacez, it is needless to discuss his arguments, since the only objection 
which he made to Darwin’s experiments, viz. that the nets sheltered the plant 
_ from movements of the air (742) was experimentally refuted by Darwin. For 
the flowers in which Darwin imitated the action of the bees, though they grew 
beneath the net, were completely fertile. 
RETROSPECT OF PAPILIONACE. 
The Papilionaceze which we have studied are all fertilised by 
bees, and in spite of their manifold peculiarities of detail they all 
agree in the following points regarding the arrangement and 
function of the parts of the flower: 
__ The flowers stand more or less horizontal; except Sarothamnus 
they expose the stigma and pollen to contact only with the 
| ventral surface of the bee, since the reproductive organs occupy the 
inferior side of the flower and are only curved upwards at the 
_ extremity. In the bud the reproductive organs are inclosed by the 
_ two inferior petals, these by the two lateral, and these again by the 
superior petal. 
_ The two inferior petals cohere to form a “carina,” which 
