18 Transactions. 



direct sunlight may have led to variation in this direction ; there 

 is also a considerable prabability that a blue or red colour is 

 expensive, requiring the expenditure of a certain amount of 

 energy. However colour change started, there is no doubt that 

 insects have exercised enormous influence on its further develop- 

 ment. Both birds and every group of insects have distinct 

 preferences for certain colours. This is least clearly obvious in 

 the case of beetles, flies, and perhaps the smaller bees, %vhich do 

 not seem to be more attracted by bright colours than they are by 

 yellow or white. Bees distinctly prefer red, and particularly 

 delight in the full purple colour of the bugle and Yicia Cracca. 

 Butterflies appear to like anything bright and vi\'id, but the 

 common cabbage white butterflies seem to me to prefer their 

 own colour (white) to anything else. Humming-birds have a 

 distinct and special love for a peculiar shade of red, which is not 

 found in any of our British plants. This shade, and a shape 

 corresponding to the long curved beak and head of these birds, 

 is however fairly common wherever they exist in large numbers. 

 Thus I have found members of such utterly different orders as 

 Leguminoscf, Bubiacerf, Scroplndariaceff, Lahiatff, Tridaceo'., and 

 the Indian Shot taking on this shape and colour, and in most 

 cases proved that they were visited by birds. 



The effect of the artistic preferences of bees is, however, clearly 

 traceable in our own wild flowers. Thus Gp.roninm pratevse is a 

 deep purple, and its large flowers are visited chiefl}' Ijy the larger 

 bumble-bees ; Geranhim silvnticnm has smaller purple flowers, 

 and is also visited by the large bumble-bees, though it is also 

 frequented hy numbers of small bees and the higher classes of 

 flies ■ G. sanguAnfium is not so distinctly purple, and has an even 

 more mixed clientele; our other forms — G. moUe, G. Rohertianum, 

 and G. dissectnm have pink flowers of a much smaller size, and 

 appear to be almost entirely dependent on very small bees anrl 

 flies of the upper and lower classes. Along with this difference 

 of colour there is in the vainous species a different arrangement 

 of stamens and peculiar methods of ripening, by which each form 

 is thoroughly suited to its main class of visitors. Now, let us 

 consider the Labiate family. We have Salvia pratensis and the 

 common bugle, which are purple, while the Woundwort, Stachys 

 silvaticus, is a strong red, and Wild Thyme is pink, not to speak 

 of Lamituv album, which is white. Why should there be this 



