ii8 



INTRODUCTION 



Hermann Miiller supposes that these different colours were evolved as a result 

 of the possession of and necessity for discriminative powers in bees, and it does not 

 therefore appear wonderful that not only white, yellow, red, violet, blue, brown, and 

 even blackish (Bartsia) are represented in the most varied degrees among Bee 

 Flowers, but several colours may even appear on the same flower, as in Polygala 

 Chamaebuxus, Viola tricolor, Cerinthe major, Galeopsis versicolor. Astragalus 

 depressus and alpinus, and so on. 



A large number of yellow-flowered Papilionaceae (Genista, Sarothamnus, Coro- 

 nilla vaginalis, Hippocrepis comosa) form the only exceptions. It would appear, 

 according to Miiller, that the yellow colour has here been so strongly transmitted 

 that ' variations, which of course constitute the necessary condition for the production 

 of different colours, may not have appeared at all.' 



Various flowers of this class are visited and pollinated only by a few species of 

 humble-bees, or even by a single species. 



Fig. 28. Humble-bee Flowers with Nectar concealed particularly deeply, (i) Aconitum Lycoc- 

 tonum L. (in outline). (2) Corydalis solida Sm. (the corolla-tube has been bitten through by Bombus 

 terrester). 



Aconitum Lycoctonum is of particular interest in this connection. The nectar 

 of this flower is hidden so deeply that only insects with an exceptionally long 

 proboscis are able to reach it. 



Especially interesting is the fact that in central Germany this flower is visited 

 exclusively by Bombus hortorum L., and in the Alps exclusively by B. opulentus 

 Gerst. ? . These two humble-bees possess a proboscis longer than that of any 

 other species of the same genus living in the districts named. That of B. hortorum 

 is 21 mm. long, and that of B. opulentus 22 mm. The latter species has hitherto 

 been observed on no other flower. We have here, therefore, vicarious species. 



In Jamtland (in central Sweden) C. Aurivillius (Bot. Centralb)., xxix, 1887, 

 pp. 125 and 126) frequently observed B. consobrinus Dahlb. acting as a pollinator, 

 in addition to B. hortorum Z. Both species are * particularly well adapted to reach 

 the nectar.' We have here, then, a secojtd vicarious species, characteristic of the 

 ' northern alpine regions,' or perhaps it is only a vicarious variety, for Schmiede- 

 knecht ('Apidae Europ.,' pp. 295, 297, 305) describes the former as a race of the latter. 



Lastly, MacLeod has again met with B. hortorum Z. as a visitor of Aconitum 



