THE BEE-KEEPERS' REVIEW 253 



To investigate by experiment we will raise a lot of queens 

 from Italian stock such as is called pure, that is to say, one produc- 

 ing" uniformly banded workers but variable drones. 2^1 ate some of 

 these queens to drones from the same stock and some to blacks. If 

 the experiment is on a suflficiently large scale the first lot will be 

 found to contain some passing as purely mated, as judged by their 

 worker offspring' and some as mismated, even though we are sure 

 that no other drones were in reach. Of the so-called purely mated 

 ones a few may be found to produce uniformly colored and marked 

 drones and the others drones of varied marking. 



Of the queens mated to the black drones there will be found a 

 similar conditions as to drones, but there will also be found this: 

 the queens producing uniform drones will also be producing uniform 

 though dark workers, while the queens producing the miscellaneous 

 types of drones will also be producing a miscellaneous lot of workers. 



In the foregoing experiment absolute uniformity in any case 

 can scarcely be expected, as the queen possessed variable potentials. 



In the Italian by Italian mating we cannot tell whether the 

 drone in each case was from a queen producing uniform drones or 

 not, and so we cannot be sure whether the queen we classed as im- 

 purely mated was mated to an "irregular" drone, or she herself 

 possessed the irregularity. Were it possible to treat queens as we 

 do higher animals we could tell by the methods used with them. 

 But with the Italian by black mating we know the drones were pure, 

 so that all we have to consider is the relation of variation in the 

 worker offspring to variation in the drone off"spring. It will be seen 

 that where the drones were uniform in color and markings the 

 workers were also uniform, but with queens producing non-uniform 

 drones it is not so. \Miy is this so? Because the queen of sup- 

 posedly pure stock carried impure blood, to use current language. 



To make the matter a little clearer, let us take a queen of a 

 golden strain of known purity, a real thoroughbred, and mate her to 

 a black drone also a thoroughbred. The resulting female offspring 

 will be the most beautiful banded Italians one could ask to see, 

 while the drones will be uniform Goldens. I know of but one or 

 two bee men who would fail to pass that queen as a purely mated 

 Italian of high quality. But we must go a step further. We must 

 raise some queens from her, and we will mate them to golden 

 drones of the parent stock, not her drones, lest someone sa}^ they 

 were contaminated by her mating. Now what are the offspring? 

 Some queens giving fairly uniform Golden workers, some giving 

 banded Italians more or less uniform, and some giving a mixture. 

 The queens of the first and second class will be found to produce 

 drones of a fair degree of uniformity — not absolutely uniform, 



