H YMENOPTERA BEES 



i6i 



The varying length of proboscis is the cause of this difference in habit. 

 B. Gerstackeri ? possesses a proboscis 18-21 mm. in length, with which it can 

 easily suck the nectar from A. Lycoctonum, but the workers, having a proboscis 

 only 11-12 mm. long, are unable to do this. 'There was, therefore, no alternative 

 for the workers but to resort to flowers with less deeply seated nectar, and as 

 A. Napellus corresponds, in regard to nectar, to A. Lycoctonum perhaps more than 

 to any other plant, while at the same time both species are very conspicuous at 

 a distance, and exhibit at the same spot racemes that rival one another in splendour, 

 the division of spoil between queens and workers is not difficult to explain ^.' 



V. Dalla Torre sees in this ' heterotrophy ' an advantage to humble-bees, because 

 it enables the comparatively short lives of these insects to be employed to the best 

 advantage. 



Co ^ 



Geograx>hical Distribtitton 



Bomhuj 



Aconitum 



Fig. 70. Map to illustraie the distribution of Bombus and Aconitum 



(after Kronfeld). 



With regard to the relation between the distribution of humble-bees and certain 

 flowers that are pollinated by them, Kronfeld (Bot. Jahr., Leipzig, xi, 1890, p. 19), 

 in an interesting paper, points out that the genus Aconitum is dependent upon 

 Bombus. He gives a map showing the areas of distribution of monkshoods and 

 humble-bees, and a glance at this shows that the distributional area of the former 

 is completely included in that of the latter (see Fig. 70). 



The adaptations of bees to flowers are intimately connected with the length 

 of the proboscis and the other bodily dimensions, for on the former depends, as 



' According to Hoflfer, the ' heterotrophy ' of Bombus Gerstackeri Mor. (in which v. Dalla Torre 

 states that old queens exclusively visit Aconitum Lycoctonum Z., while males and workers visit 

 A. Napellus Z. as exclusively) does not apply to localities where A. Napellus is abundant but 

 A. Lycoctonum rare. In this case all three castes of the Ijee in question suck A. Napellus, as Alfken 

 was able to prove. I have gone more fully into this in vol. IL 



DAVIS 



JI 



