RACES AND RELATIONSHIPS OF HONEY-BEES. 215 



or mammals, since with these, under natural conditions, it is 

 almost invariably more or less protective. 



I do not mean to infer hereby that the colour is not protective 

 with insects ; — in fact, in many instances, and more particularly 

 during the stages of development, it is so to an extraordinary 

 extent; but this is not so universally the case as with the other 

 mentioned classes of animals. It is to a great extent com- 

 pensated by an extraordinary power of reproduction, and by 

 other protective qualities than that of coloration ; as, for in- 

 stance, offensive odours, the ejection of pungent fluids, or — as is 

 also the case with the bee — with organs of defence, which are 

 frequently venomous. A protective colouring is, therefore, not 

 essentially necessary to the bee, and its variability in this 

 direction is probably but little influenced by natural selection. 

 Consequently, I should not like to attribute the yellow colour of 

 the Egyptian bee to a selection in this direction, although it is 

 suggestive on account of the conformity of the coloration with the 

 general ground tint of the arid regions which this race inhabits. 



This seems to be borne out by the fact that the Cyprian bee 

 has not modified its colouring, although the climate and other 

 conditions of life in its habitat differ considerably from those of 

 Egypt and Syria, whence it was undoubtedly introduced, most 

 probably by the Phoenicians ; but, on the other hand, its pug- 

 nacity and greater strength of flight must be attributed to natural 

 selection, and are the result of an adaptation to altered sur- 

 roundings. If, therefore, colour variation with these bees cannot 

 be considered protective, and as resulting from natural selection, 

 it must be ascribed to a latent disposition which, under certain 

 not yet explainable external influences, has developed to a con- 

 siderable extent in certain varieties. This disposition towards 

 the coloration of the abdominal segments is found to pervade 

 nearly all varieties of the Brown bee, and may even be occa- 

 sionally observed amongst the most uniformly-coloured races, 

 e.g. the Heath bee. In the Nether- Austrian variety it is a 

 common occurrence, as also in the Attic bee. In this latter, in 

 fact, it is so pronounced that the Attic bee is, by some, con- 

 sidered identical with the bee found in Tessin, which is very 

 closely allied to the Ligurian. Such a moderate disposition 

 towards coloration in some varieties, and the more distinct one 

 in others, seems to point towards a variability in that direction 



