378 AUTOGAMY. 



the most convenient manner possible for nocturnal moths to visit it. In flowers 

 adjusted in anticipation of such visits, the stigma takes up a position which pre- 

 cludes the possibility of its being dusted with pollen from the anthers in the same 

 flower. In the act of introducing their long probosces into the honey secreted in 

 the interior of the flower, Sphingidse come into contact first with the stigma and 

 then with the anthers, and as they travel from flower to flower they are the means 

 of efiecting cross-pollination in this as in so many other cases. But should no 

 moths come upon the scene, autogamy invariably takes place through the inflection 

 of the corolla-tube already referred to. The stamens are adnate to the corolla-tube, 

 and undergo inflection with it, thus bringing the anthers, still covered with pollen, 

 into direct contact with the stigma, which, in the horizontal position of the flower, 

 was stationed a little lower than, and in front of, the anthers. 



In respect of the manner of their autogamy the last-mentioned plants exhibit 

 a transition to a large group in which self-fertilization is prevented during the 

 early stages of flowering by the relative positions of anthers and stigmas, but is 

 eflected towards the end of the period of bloom, when certain changes in the 

 position and direction of the flower -stalks have taken place and brought the 

 pollen and stigmas into conjunction. These alterations of position are usually 

 associated with one of the many other contrivances already described. Thus,- for 

 instance, the styles or the filaments may undergo elongation and inflection, or the 

 corolla may grow up and carry with it pollen affixed to its petals, or the stamens 

 themselves, and so forth; but these processes would not of themselves be sufficient 

 to induce autogamy if it were not for the part played by the flower-stalks. To put 

 it briefly, the stigmas and the anthers become, in the absence of cross-fertiUzation, 

 so situated by the growth and inflection of the flower-stalk as to render autogamy 

 inevitable. When we consider that the changes in the position and direction of 

 pedicels, and the consequent drooping or straightening up of flowers, serve other 

 purposes of great importance in the life of plants, and that, in particular, to these 

 inconspicuous movements are often due the protection of pollen from moisture and 

 the placing of the entrance to a flower in the position most convenient to insects 

 whose visits are profitable to the plant, we cannot be surprised to find that this form 

 of adaptation is one of the commonest of all. A combination of advantages, either 

 simultaneous or in rapid succession, is secured, and contrivances of this kind which 

 best contribute to the economy of plant-life are found by experience to be invariably 

 the most widely distributed. 



We will first consider flowers in which the stigma begins by being situated 

 outside the line of descent of the pollen as it falls from the anthers — a circumstance 

 which is advantageous inasmuch as it favours cross-fertilization — ^but where subse- 

 quently the entire flower assumes a different position in consequence of a growth 

 or an inflection of the flower-stalk, whilst the direction and situation of stamens, 

 style, and stigmas remain the same as before. In several species of Narcissus, e.g. 

 the graceful Narcissus juncifolius, and in some Boragineee, such as the common 

 Wood Forget-me-not {Myosotis sylvatica), the flowers at first have their mouths set 



