AUTOGAMY BY A CONTEACTION OF STYLE-BRANCHES. 359 



stances, the velvety stigmas cannot fail to receive the pollen that still adheres to 

 the anthers. 



It appears from what has been said concerning autogamy that in a large number 

 of plants the pollen developed in the anthers, especially if it be of the adhesive 

 variety, still occupies the recesses of the anthers, or sticks to the refiexed margins 

 of the lobes after their dehiscence, at the time when the flower fades. Even after 

 insects have brushed off a portion of the pollen and transported it to other flowers, 

 there is still invariably an abundant supply for the purpose of autogamy, and it is 

 only in rare cases that loculi, in which adhesive pollen has been matured, are com- 

 pletely emptied by the end of the flowering period. In some plants, however, the 

 adhesive pollen is swept out or removed in some other way from the anthers by 

 means of special contrivances as soon as it is mature, and is then deposited on some 

 particular spot in the flower where it is exposed for dispersal. In the case of the 

 pollen of Composites it is well-known that it is pushed out of the tube of connate 

 anthers by the style, owing to the growth of the latter organ which is sheathed 

 within the anther-tube, and that it appears at the top of the tube in the form of a 

 lump capping the extremity of the style. In Bell-flowers (Cavipanula), the entire 

 contents of the anthers are stored upon the surface of the style, and the same thing 

 happens in the various species of the Rampion genus (Phyteuma) and in some small- 

 flowered Gentians. The shrivelling of the anthers is in many plants the cause of 

 their shedding a portion of their pollen, and it may then collect on capillary append- 

 ages of the ovary, in cup-shaped petals, or on some other part of the flower where it 

 is stored up for future use. It must often happen, too, that when insects are in the 

 act of taking the honey they push against the stamens, and that the pollen shaken 

 out of the anthers by their impact adheres to particular parts of the corolla, calyx, 

 or perianth. This pollen is just as available for fertilizing purposes as that which 

 remains sticking to the anthers, and we meet with cases where the stigmas fetch the 

 pollen developed in the same flower from its temporary resting-place, and so bring 

 about autogamy. Contrivances for this purpose are not numerous, but the number 

 of species in which this form of autogamy prevails is extremely large. The abstrac- 

 tion of pollen deposited on the outer surface of the stylar column or its arms by 

 stigmatic tissue situated on the edges or the inner surface of these style-branches 

 occurs in hundreds of Campanulas and thousands of Composites, and shall therefore 

 be chosen as our first example of this type of process. 



Two modes of operation may be distinguished : first, a crossing; and, secondly, a 

 spiral revolution of the style-branches. The former process is observed particularly 

 in the Asteroidese (Aster, Bellidiastrum, Erigeron, Solidago), especially in the 

 tubular flowers in the middle of the capitula of these plants ; but it is also seen in 

 many Composites possessing ligulate flowers only. In Aster alpinus, the species 

 selected for illustration (see figs. 3021'^-^), the stylar arms are short and thickish; 

 their inner surfaces are smooth and flat, whilst their outer surfaces are a little 

 arched, and towards the free extremities are furnished with papilla-like sweeping- 

 hairs. The receptive stigmatic tissue is situated on the margins of the style- 



