RACES. 
619 
it. Such is not the case : the following summer, all 
the drones produced in the hybrid stocks will be 
Carniolan, and, unless the supply of Cyprian blood be 
kept up, the stock will, collectively, gradually revert 
towards the Carniolan type (see page 321). This fact 
seems to have escaped the attention of those who 
wrongly assert that, unless we can absolutely control 
fertilisation, no advance can be made in the produc- 
tion of a special strain. It is clear that the queen 
exerts three times the influence over her posterity 
that the drone does, since her blood is one-half of 
that of the fecundated daughters, and the whole of 
that of the sons. 
The circumstances just enumerated greatly favour 
any attempt at establishing one race in lieu of 
another; e.g., in what is called “ Ligurianising,” one 
or two, or any number of Ligurian queens, are 
introduced into an apiary. They are encouraged in 
drone-breeding, and other drones are, as far as 
practicable, banished. Some queens are purely mated, 
some crossed. The crossed ones the second year 
raise the drones, which are still pure, and the purely- 
mated furnish the queen-cells ; and in this way, theo- 
retically, Ligurians may be made to supplant blacks, 
or any one race any other. Practically, unless a 
position of great isolation can be secured, it is not 
so, for bees seem to prefer crossing. Many years 
ago, when bees were few about me, I placed twenty 
pure queen-cells in as many stocks. I had abundance 
of Ligurian drones, and none of any other race — 
nineteen of the queens crossed ; this has been cited 
as evidence of the difficulty of securing pure im- 
