70 (^UEEN-REARING IN ENGLAND. 



The breeding true of an intermediate condition of colour 

 sucli as occurs in Italians has been found to be in accord 

 with Mendelian views in a case where only two factors were 

 concerned, when it was assumed that they interacted on one 

 another in a certain way and to different degrees. 



Fig. 8a shows a remarkable colouration found in a British 

 golden, bred in 191 2. Fig. 8b shows an unusual colour- 

 ing in several golden queens reared in August, 1912, from 

 a British golden queen bred the same year, that produced 

 all goldens. \Vhether these are fluctuations caused by cold 

 or some other agent, or whether they are new colourations, 

 it is not yet possible to say ; but I have ah\ays been on the 

 look-out for variations in goldens in the direction of darken- 

 ing, for it is probable that only by isolating such a varia- 

 tion shall we succeed in much improving the hardiness and 

 honey-gathering qualities of goldens. That a chill during 

 the later stages of dexelopment darkens the yellow colour 

 and causes the black to spread in queens, e\'ery breeder of 

 Italians or goldens knows. Fig. 9 shows a curious and 

 striking distribution of colour that I have obtained in inter- 

 mediate coloured queens chilled during development. 



Till now we have been considering only the ground colour 

 of the abdomen. But there is another character that greatly 

 affects its appearance in the worker, and that is the presence 

 or absence of a pronounced band of short white hair on 

 each segment, except the first and the last. This band is 

 well developed in the Italian bee, but feebly so in the 

 English black bee. The only outstanding difference between 

 the appearance of the English bee and that of the Carniolan 

 is the high degree to which these bands are developed in 

 the latter race. It would be interesting to ascertain if the 

 inheritance of these bands follows Mendelian rules. 



I am not without hope that swarming and even stinging 

 may be eliminated by breeding. 



In nature, the vast majority of queens are reared in 

 colonies that swarm, but by modern methods of queen-breed- 

 ing, colonies that never swarm may be bred from. British 

 goldens that had been bred artificially through at least ten 

 successi\c generations, regardless of swarming, were very un- 

 certain in the w,iy they swarmed. The swarms often issued 

 late in the season and many returned to the hive without 



