54 QUEEN-REARING IN ENGLAND. 



impossibility of knowing which drone out of hundreds a queen 

 may meet, breeding by selection on the male side cannot be 

 properly accomplished. But if we look at the work of 

 plant breeders, we see that a similar uncertainty often exists 

 as to which flower will supply the pollen by which the seed 

 is fertilised. Further, in cases where characters that do not 

 show themselves in the flowers, such as the size, shape, or 

 other qualities of seeds or fruits, are bred for, the selection 

 of flowers is useless. It is only essential that no flower 

 shall be employed that is not of the selected parentage or 

 strain. I think that even were it possible to select single drones 

 for breeding from, it would not be advisable, for we may 

 suppose that the honey-bee depends partly upon natural 

 selection of drones for its vigour and usefulness. It is not 

 asserted that one drone transmits exactly the same characters 

 as another, but this uncontrolled element of variation is 

 reduced to a negligible quantity by the employment of pure 

 stock and continuous careful selection. 



Mendelism. — Our knowledge of the laws of heredity 

 has latelv been much advanced by Mendelism, so named 

 after its discoverer, Gregor Mendel, monk, and afterwards 

 abbot, of Briinn in Austria, who announced its principles in 

 1865. In any attempt to improve livestock or plants by 

 breeding, a knowledge of Mendelism is of great value. 



Mendel made his classic experiments with the common 

 garden pea. In one series of experiments he crossed a tall 

 pea with a dwarf pea, and found that all the resulting hybrids 

 were tall, like their tall parent. He named the character 

 that pre\-ailed in this first generation of hybrids — in this case 

 of tallness — the dominant, and the character which disap- 

 peared, namely, shortness, the recessive. He sowed the 

 seeds from these hybrids, and found they produced tails 

 and shorts in the proportion of three to one. He was able 

 to prove that the tails in the second generation of the hybrids 

 were of two kinds, those that produced tails only, and those 

 that produced tails and shorts in the proportion of three to 

 one, like the tails in the first generation of the hybrids, and 

 also that the latter kind of tails were twice as numerous as 

 the former kind ; and that the shorts produced shorts only, 

 ^lendel showed that this remarkable train of inheritance 

 was true for six other pairs of characters in the pea, namely, ■ 



