QUEEN-REARING IN ENGLAND. 67 



industrious worker in a hive vvliere most of the workers are 

 less industrious. Unfortunately, we have no means of dis- 

 covering such a queen, and characters of tliis kind that show 

 themselves in individual workers are almost sure to be lost 

 unless we can find them in a large proportion of the workers 

 in a colony. For such characters we are compelled to test 

 our bees in colonies, not as individuals. This is rather a 

 serious limitation, for while we can study the characters of 

 thousands of individual workers, comparatively few colonies 

 are available. On the other hand, the enormous number 

 of workers that a queen produces helps us very much in the 

 study of what gametes she is producing, and as they are 

 always surrounding her and continue to be produced for 

 years, we can study them at leisure. 



And here I may make a few remarks on estimating the 

 proportions of the different coloured offspring of a queen. 

 My practice is to do this, if possible, on the twenty-second 

 or twenty-third day after the queen begins to lay. Then 

 only a few hundred of her workers have hatched, and they 

 can be distinguished at once from the other workers in the 

 hive by their downy, soft, and immature appearance. At 

 this stage one can, with practice, estimate fairly accurately 

 almost at a glance the proportions, if only two types are 

 present. Often there is a risk of robbing, and then it is 

 not safe to keep the hive open more than a minute or two. 

 Later, when there are workers in all stages of immaturity, 

 it is less easy to estimate the proportions. It is not usually 

 possible to ascertain the proportions from mature bees until 

 the spring, and then the results may be rendered inaccurate 

 by the presence of bees from other hives. For accurate work 

 it is best to chloroform a few hundred bees, and then pick 

 out the young ones and count out the different varieties of 

 them before they have time to recover. 



I should like to say a few words on colour inheritance 

 in drones. Fig. 6 (in Fig. 34) shows a British golden 

 drone. The drone has seven dorsal segments — one more 

 than the worker or queen — but only five of these are seen 

 from above, for the first segment occupies the basal end of 

 the abdomen, and the last one is underneath. I should 

 state that British Golden drones vary a little in the extent 

 of the black on the fourth segment. It is well known that 



