70 ZOOLOGY 



well-marked races gives rise to offspring whose nature can 

 best be accounted for by supposing that the hybrid produces 

 two kinds of germ-cell of each sex in equal numbers, and that 

 one type of germ-cell carries the differentiating character belong- 

 ing to the paternal race, and the other type the character 

 proper to the maternal race. When, however, two germ-cells 

 carrying different characters unite in sexual union, the resultant 

 offspring does not usually show a blend of the two characters, 

 but one of the characters alone appears ; this is termed the 

 dominant character; whereas the other, whilst equally im- 

 planted in the constitution of the hybrid, does not affect its 

 visible anatomy or appearance, and is therefore termed the 

 recessive character. Of course this law finds its most 

 obvious application in the character of the original hybrid 

 produced by the first cross, which usually favours father or 

 mother, but not both. The parent not represented in the 

 visible structure of the child makes, however, its presence felt 

 when that child comes to form germ-cells and produce offspring, 

 for in some of the grandchildren the features of the " recessive " 

 parent come forth again from obscurity. According to this 

 discovery of Mendel's it results that when two races cross, the 

 offspring of the hybrid gradually breaks up into the parental 

 races again and the hybrids become relatively less numerous 

 at each generation. Such a discovery, therefore, while it shows 

 how two races occupying different but contiguous territories may 

 keep distinct in spite of crossing, has no bearing whatever on 

 the formation of new species. But the question now arises, 

 what will happen when two races which differ from one 

 another in two differentiating characters are crossed? The 

 answer is that half the germ-cells of the first hybrid will con- 

 tain one of each pair of differentiating characters and half will 

 contain the other member of each pair, but that those germ-cells 

 which contain one character of one pair are not in every case 

 identical with those which contain one character of the other 

 pair. It is exactly as if one had a number of balls in a bag 

 and if one were to draw half of them out and paint them red, 

 leaving half of them uncoloured, and if one were then to put the 

 painted balls back again and shake the whole lot up together 

 and then draw out half the number again and paint black dots 

 on them. Half the number of balls would thus receive black 



