Perhaps the most striking thing about ]\Icnders work is tlie 

 singularly suggestive and luminous interpretation which he gave 

 of just why the pea characteristics were transmitted exactly 

 as they were; why, in general, the peculiar numerical ratio l^e- 

 tween dominant and recessive should be, and why it should 

 persist so uniformly. This interpretation or explanation is now 

 well known in biology as the theory of the "purity of tlie germ 

 :ells," or, as Cuenot has called it, the theory of '' gmnctes dis- 

 joints/' or "la disjonction des characteres dans les gametes des 

 hybrides'' (the separation of characters in the germ cell of 

 hybrids), the Spaltimgsgesetz of de Vries. 



This interpretation is simply that in the young of the first 

 generation after a cross-mating, although because of dominance 

 but one of the contrasting pair of parental characters will show 

 itself in the body make-up, yet when these young form their 

 germ cells the two parental characteristics will be represented, 

 but only one in any one germ cell; that is, in the case of 

 Mendel's peas that the pollen cells and ovule cells produced by 

 the cross-bred young would carry each one of the alternative or 

 mutually exclusive parental varietal characters. If this were 

 the case and if, on an average, the pollen cells and ovule cells 

 were evenly divided as to the two characteristics, then by 

 miscellaneous or random mating (mating according to the law 

 of probabilities) between these cells we should get in the de- 

 veloped young just such conditions with regard to the con- 

 trasting characteristics as Mendel actually did get in his peas. 

 For twenty-five per cent of the pollen grains representing the 

 dominant character would unite with twenty-five per cent of the 

 ovule cells representing the dominant character, twenty-five per 

 cent of the recessive pollen grains with twenty-five per cent 

 of the recessive ovule cells, and the remaining fifty per cent of 

 each kind with each other; that is, of every four pollen grains and 

 every four egg cells we should get by random })ollination 1 ])<)llen 

 dominant X 1 ovule dominant; 1 pollen recessive X 1 ovule re- 

 cessive; 1 pollen dominant X 1 ovule recessive; 1 ])ollen reces- 

 sive X 1 ovule dominant. This condition would bring it al)()ut 

 that the fully developed young would sliow the contrasting 

 characteristics (remembering the dominance of one of tlie char- 

 acteristics in those cases in which dominant and recessive are 

 united), in this condition: 31), 111. Which is exactly what 

 occurred in Mendel's peas, and has since been noted to occur 



