66 QUEEN-REARING IN ENGLAND. 



comparing their colonies for hardiness and honey gathering in 

 May and June, and breeding queens and drones from the 

 best only, I was able to get some improvement in these 

 directions, and a better bee for crossing with blacks. 



A valuable lesson that Mendelism teaches to breeders of 

 all kinds of plants and animals, and one that the l>ee-breeder 

 must not lose sight of, is that, in the words of Punnett, 

 "every possible variety arising from a cross appears in the 

 second generation if only a sufficient number is raised, and 

 of all these different varieties a certain proportion of each 

 is already fixed." More can be accomplished by a careful 

 analysis and isolation of the individuals of this generation 

 than by years of breeding by selection on the old lines. 

 The fear of losing the original variety by crossing it is 

 unfounded, for it can be recovered in a fixed state with all 

 the superadded vigour that follows from a cross. 



We must remembt-r that, underlying Mendelism, is the 

 doctrine elaborated by Weismann : that characters acquired 

 during the lifetime of the individual as the result of changes 

 in treatment or environment cannot be transmitted to the 

 offspring, though it is true environment may and does exert 

 selection in individuals carrying different kinds of gametes. 

 It is, therefore, useless to try to fix acquired characters — 

 fluctuations, as they are called — by breeding, unless the 

 treatment that causes them is always applied, as in the case 

 of worker characters as distinguished from queen characters. 



Robbing is generally an acquired character caused by 

 carelessness on the part of the bee-keeper. Bad temper 

 induced bv improper handling is another case of an acquired 

 character. Such characters, acquired by accident, need 

 cause us little concern. 



Most of the characters in the bee that we wish to improve 

 are in the workers. How, it may be asked, can these be 

 propagated, seeing the workers are sterile? We must do it 

 through their reproductive sisters — the queens. It is com- 

 parati\'ely easy to propagate the character if it is recognis- 

 able in the queens, even though much changed, as, for 

 instance, the colour character. But we may wish to pro- 

 pagate a character that is not recognisable in the queen, 

 such as industry. We may wish, for instance, to get the 

 queen containing the factors carried by a particularly 



