318 THE CROSSING OF FLOWERS, 



and of geitonogamy (from yeiruv, a neighbour, and yd/xos, marriage) when the two 

 flowers are on the same plant. We cannot, however, draw a sharp line between 

 the two. The offshoots of a plant, which become ultimately isolated, forming inde- 

 pendent plants, are, in point of origin, identical with the branches of a plant which 

 remain attached. Accordingly, when a crossing occurs between flowers produced on 

 plants that have arisen from one another by offshoots, the process is not essentially 

 different from the crossing which takes place between flowers on adjacent branches. 

 It is nevertheless convenient to keep the two cases distinct, on account of certain 

 processes connected with the greater or lesser distances between the flowers. 



Both in xenogamy and geitonogamy the transport of the pollen is effected 

 partly by wdnd and partly by flower-visiting insects. How this is carried out, 

 and the endless variety which exists, has been dealt with in detail in previous 

 chapters. Geitonogamy is not infrequently, however, brought about in other ways, 

 as by the pressing of mature stigmas on the liberated pollen of neighbouring flowers, 

 or by the actual falling of pollen upon them. Since these methods of cross- 

 pollination have only been incidentally touched upon, they must be described here 

 somewhat more fully. 



The conditions for crossing between neighbouring flowers are simple when the 

 flowers are crowded in heads, umbels, bunches, spikes, and the like, standing so 

 close together that the stigmas of one flower can easily touch the pollen-covered 

 anthers of another. And since this kind of crossing is actually very widespread 

 and is repeated in certain species with great regularity, generation after generation, 

 we are justified in regarding these forms of inflorescence as being particularly 

 associated with geitonogamy, and in assuming that a not unimportant advantage 

 of a crowded inflorescence lies in the possibility of crossing between the neighbour- 

 ing flowers of a plant (see vol. i. p 740). 



As w^e might expect, this particular form of crossing occurs with great frequency 

 in Compositae, w^hose flowers are crowded so densely into capitula that the whole 

 inflorescence might be taken, at first sight, for a single flower; consequently this 

 extensive family of plants, which includes more than 10,000 species, will be the most 

 suitable in v/hich to describe the phenomenon of geitonogamy. We will commence 

 with those Composites whose heads only contain "ray" or ligulate florets. The 

 term ray or ligulate floret is applied to florets whose corolla is tubular only at the 

 base, the free end being flattened and projecting outwards like a tongue or strap, as 

 in the Dandelion. In Prenanthes each capitulum consists of only five such ray- 

 florets. In each floret the long, thin style is inclosed in a tube of anthers. The 

 style is covered with stiff outwardly-directed bristles which are called " sweeping 

 hairs". When the style elongates, immediately after the opening of the flower, 

 these hairs sweep out the pollen w^hich has been ah^eady shed into the interior of 

 the anther-tube. The long style, rendered quite yellow by the pollen it carries, now 

 projects from the empty tube of anthers. The two branches of the style which 

 bear the stigmatic surface are at first folded together, but they soon separate, and 

 the stigmas can then be fertilized by the aid of insects with pollen brought from 



