342 THE BEEKEEPERS' REVIEW 



is probably an advantage, for such weather caused great loss of 

 queens at Ripple. 



Kazabazua also differs from Ripple in usually having a fairly 

 heavy honey flow throughout August and up to about September 

 15. This comes from willow-herb (August), golden-rod (August 

 and September) and aster (September). At Ripple very little nec- 

 tar was collected in August and September. The longer continued 

 honey-flow at Kazabazua discourages robbing and in this and other 

 ways facilitates safe mating, but it encourages the local black col- 

 onies to keep some of their drones alive for a time and even rear a 

 few in August. 



A number of the daughters of a non-swarming Italian queen 

 were mated at Kazabazua in the presence of about 4,000 drones bred 

 from the same queen, during a period of clear sky and light south- 

 erly to westerly airs extending from October 1 to October 5, 1914, 

 this being about a month later than the latest date that can be re- 

 lied upon for mating at Ripple. During this period the temperature 

 rose to about 63°, 68°, 72°, 75° and 78° on each successive day, 

 though there were at first sharp frosts at night. The results of 

 these matings will be duly recorded. 



It should be mentioned that an apiary of sixteen black colon- 

 ies in box hives three miles from the Kazabazua mating station was 

 watched for five minutes at 11 :40 a. m. on October 4 when yellow 

 drones were flying very freely at the mating station. No drones 

 flew in or out of those hives. On the previous day an apiary of 

 three colonies, also three miles from the mating station, was 

 watched for five minutes at 1 :25 p. m. Four black drones were seen 

 to fly in and one out. 



For autumn mating the bees need not be brought to the mating 

 station until the end of August. This is a great advantage in that 

 the local black colonies are not likely to have virgins flying so late 

 in the season, so that they are not likely to get crossed with the 

 yeflow bees; consequently the same mating station may be used 

 in succeeding autumns without fear that the distinction between 

 pure matings and mismatings will become blurred. By maintain- 

 ing this clear distinction there is good reason to believe that three- 

 banded Italians as well as goldens can be bred by color selection. 

 Indeed some daughters of a pure imported leather colored Italian 

 queen mated at Kazabazua by local drones in 1913 produced work- 

 ers distinctly darker that the pure Italians, their sisters. In the 

 writer's mind there is no doubt that, for northern climates, the 

 three-banded Italian is superior to the pure golden. 



