364 AUTOGAMY. 



discussed (see p. 361), but the moment the anthers are retracted some pollen is 

 invariably caught by the hairs of the corolla, and when the flower opens a portion 

 of the pollen is always to be seen adhering to them. The humble and hive bees 

 which visit the flower may, of course, bring about cross-fertilization, just as they 

 do in the other Bell-iiowers. The style-branches in Campanula Trachelium do 

 not, at the close of the flowering period, roll back so far as the central column; a 

 less degree of bending is here adequate to bring the receptive tissue on the tips of 

 the style into contact with the pollen sticking to the hairs. 



In Dianthus neglectus, a species of Pink indigenous to the Southern Alps, and 

 in the Glacier Pink (Dianthus glacialis), the laminas of the petals are beset with 

 hairs. The pollen is fii'st exfiosed to the chance of dispersal by insects, but after- 

 wards the stamens curve outwards, and some of the pollen becomes aiBxed to the 

 hairs of the petals, which usually receive in addition a small deposit as a result of 

 the scattering action of insect-visitors. The flowers are protandrous. The stigmas, 

 which are situated in front of the entrance to the floral interior, wait till the period 

 of the flower's bloom is nearly at an end, for the chance of being touched by insects 

 bearing extraneous pollen. But sometimes no insects come, and in that case the 

 pollen stored upon the hairy laminae of the petals is made use of at the last 

 moment. The trausj^arent papillose stigmas wind themselves into the shape of the 

 letter S, and, sweeping like a brush over the petals, collect the pollen from them. 

 This operation is materially assisted in both the Pinks under discussion b3^ the fact 

 that the lamince of the petals grow some millimetres longer during the flowering 

 period, the result of which is to bring the hairs besmeared with pollen a little nearer 

 to the stigmas. In Dianthus neglectus there is besides an involution and uprising 

 of the laminae in the evening, which, likewise, assist the process of sweeping up the 

 pollen by the stigmas. 



In Ballota nigra, a Labiate with protandrous flowers which grows commonly in 

 hedges on cultivated land, some of the pollen falls at the very commencement of the 

 flower's bloom upon the hairs clothing the borders of the upper lip. If no pollen is 

 brought by insects to the stigma of a flower of this plant, the lower stjde-arm bends 

 down at the end of the flowering period and takes up the pollen from the hairy 

 mantle above referred to. The same kind of thing happens in a few other Labiates, 

 as, for instance, in Salvia viridis, of the Mediterranean flora, whose style bends 

 down in the event of a failure of insects, and brings the stigma into contact with a 

 store of pollen resting upon the under-lip, where it fell at the very commencement 

 of the flower's bloom. I have hitherto observed only in the cases of Tozzia aljnna 

 and Pyrola media the phenomenon of a style curving down to take pollen from 

 cup-shaped hollows in the corolla in which it has lain stored, but it probably occurs 

 in many other plants besides. 



The curious case of the pollen being taken from the hairs of the so-called pappus 

 by the stigma is illustrated in figs. 303^' ^' "■ ^^ In the Stinking Groundsel (Senecio 

 viscosus), which will serve for an example, the style-branches are furnished at the 

 tips only with bunches of collecting-hairs. As the stjde elongates these hairs sweep 



