156 BLASTOGENIC VARIATIONS. 



parent forms, and to show little or no trace of the 

 other. The characters thus appearing were termed by 

 Mendel dominant, and the characters becoming latent 

 in the process, recessive. In the next generation, how- 

 ever, the seeds from these dominant hybrids betrayed 

 their mixed origin, for instead of maintaining the pure 

 dominant character, on an average one out of every 

 four of the plants or seeds obtained reverted to the re- 

 cessive parent form. The following are the actual 

 numbers of plants and seeds examined by Mendel in 

 respect of the various differentiating characters above 

 mentioned: 



Proportion of Dominant 

 to Recessive. 



(1) 253 hybrids (gave 7324 seeds) . . . 2.96:1 



(2) 258 " ( " 8023 " j I . . . 3.01:1 



(3) 925 " ... 3.15:1 



(4) 1187 " ... 2.95:1 



(5) 580 " ... 2.82:1 



(6) 858 " ... 3.14:1 



(7) 1064 " ... 2.84: 1 



Mean 2.98:1 



It will be seen that the proportion of 3:1 is fairly 

 evenly maintained in respect of all the characters ob- 

 served. 



The observations on the next and succeeding genera- 

 tions afforded an even more remarkable result than this, 

 for they proved that the recessive forms obtained in the 

 second generation were absolutely pure. Thus the 

 seeds obtained by crossing them amongst each other, 

 or by self-fertilisation, yielded offspring which never 

 showed any trace of the dominant grand-parental char- 

 acters. The dominant forms, on the other hand, which 

 of course were self -fertilised, underwent a further split- 

 ting up. A third of them yielded plants which in sub- 



