404 THE FERTILISATION OF FLOWERS. [parr m1. 
papillar inner surfaces are exposed to contact with the bee. So if 
bees visit the flower, cross-fertilisation seems to be insured; in 
absence of insects, self-fertilisation seems to be rendered impossiblal 
by the well-marked proterandry, and by the position of the anthers 
in regard to the ripe stigmas. I have only noticed humble-bees 
visiting the plant: viz. (1) Bombus agrorum, F. ¢ (12—15); (2) B. 
senilis, Sm. 9 (1415), both very frequent, sucking honey only. 
Gentiana ciliata, L., is proterandrous, and adapted for humble- 
bees. Honey is secreted by yellow fleshy areas of the base of the 
corolla between each pair of stamens (No. 6009, fig. 132). | 
Gentiana bavarica, l., and G. verna, L., are adapted for cross- 
fertilisation by Lepidoptera. The most important visitor seems to 
be Macroglossa stellatarum (No. 570, vol. xv. ; No. 609, fig. 131). 
G. nivalis, L.—The flower resembles the ine preceding species 
in structure, but is less conspicuous and less visited by insects ; 
case of need it fertilises itself (570, 609). 
297. GENTIANA AMARELLA, L.—Honey is secreted by five 
green fleshy spots at the base of the corolla, alternating with the 
stamens; it is sheltered from rain by the closing of the flower dur-_ 
ing dull weather, and is protected from flies by long, erect hai 
developed on the interior of the corolla at the junction of its tube 
and limb. The plant grows among grass, and accordingly the 
lobes of the erect corolla spread out to be mainly conspicuous from 
above. 
The tube is 16 to 18 mm. long, but as it is 6 mm, wide at th 
mouth a humble-bee can insert the whole of its head and reack 
the honey with a proboscis 10 to 12 mm. long. Unlike G. Pneu= 
monanthe, the flowers are homogamous. When the flower opens 
the anthers dehisce, after turning their dehiscent sides, which in the 
young bud were turned outwards, upwards so as to be touched the 
more certainly by the bee’s head. The two terminal lobes of the 
style are already expanded and provided with stigmatic papilla. 
In case of insect-visits, self-fertilisation is not impossible; but 
since the stigma stands above the anthers, the insect as a rule 
comes in contact with it first, and effects cross-fertilisation. After 
the pollen is shed the anthers again come to lie in a line with the 
filaments and place themselves close round the style. Ihave not 
observed whether self-fertilisation occurs in absence of insects. I 
have seen Bombus silvarum, L. ? (12—14), visiting the pam 
and sucking numerous flowers (Sept. 30, 1869). 
Gentiana tenella, Rotth. (@. glacialis, Thom. ), G. nana, 7 
