ygA LECTURE XLV. 



lobe of the corolla (5, Fig. 451) closes over the mouth of the throat, stopping 

 the entrance to the flies, which have now nothing more to do here. 



(2) Flowers with simultaneous dehiscence of stigmas and anthers, but in 

 which self-pollination is rendered impossible or difficult by the position of the 

 organs, and by mechanical obstructions. The transference of the pollen on to 

 the stigma is here also usually dependent upon insects, mostly so that a stigma 

 can only.be pollinated by the pollen of another flower; occasionally however, 

 (as in the Asclepiadese) pollination by the pollen of the same flower, in addition 

 to that froih a distance, is not excluded entirely. The adaptations here' are 

 extraordinarily manifold, and sometimes so complicated that their explanation can 

 only be detected by means of close research. To this sub-division, for instance, 

 belong the species of Iris, Crocus, Pedicularis, many Labiates ; also Melastomaceae, 

 Passiflorese, and Papilionaceae. Among the most inter- 

 ^ esting are the Asclepiadese, where these relations, how- 



ever, can only be explained with the aid of numerous 

 figures and prolix descriptions. 



The mechanism for the avoidance of self-fertilisation 

 and for the ensuring of crossing between different flowers 

 of the same species is unusually elegant and easily com- 

 prehended in the case of Salvia pratensis (the Meadow 

 Sage) and some other species of this genus. Fig. 452 A 

 shows a iflower of this species seen from the side, n 

 being the bilobed receptive stigma, and the position of 

 one of the two stamens being indicated by means of a 

 dotted line within the upper hp of the corolla. If a 

 needle is pushed in the direction of the arrow into the 

 throat of the flower, both the stamens spring forward 

 as at A a. If a Bee does the same with its proboscis, 

 striving to obtain honey thence, the dehiscing anthers 

 strike it on the back and there deposit their pollen at 

 a definite spot ; on the insect coming into the same 

 position on another flower, it rubs its pollen-laden back 

 against the stigma and pollinates it. The cause of the 

 springing forward of the anther is sufficiently clear from 

 Fig. 452 ^; here are shown the short true filaments y_/" fixed by their bases on 

 to the sides of the throat of the flower, and supporting the long connectives c x, 

 ■which can be swung to and fro about the points of attachment. Only the upper long 

 thin arm of each connective c bears a half-anther a, the lower short arm at x 

 being devoid of any anther, and so connected with that of the other stamen, that the 

 two together form a sort of lounging-chair. If then the honey-seeking proboscis 

 of a Bee strikes this apparatus in the direction of the arrow, the short arm of 

 the connective is driven down and the upper arm c is moved forwards. 



The impossibility of self-fertilisation in Viola tricolor depends upon mechanical 

 arrangements of quite a different kind. Fig. 453 A and B illustrate the position and 

 arrangement of the floral parts in this case. The base of the flower is invested by the 

 floral leaves, and completely filled up by the anthers and the ovary which they surround, 



Fig. 452. 



