PLANT-BREEDING I763 



mitted in the same way as in a crossing between green plants and white 

 plants in which green is dominant. Thns when abnormal hereditary cha- 

 racters have appeared it has been possible to trace back the combination of 

 factors Cc as far as their association at the time of hybridisation, and even 

 to examine pure lines to seek for the eventual presence of the factor c. In 

 this investigation more than one half of the available seeds of two pure pa- 

 rent lines were sown (i ooo grains in each case) and it appeared that the 

 factor c did not occur in these lines, because all the plants raised were nor- 

 mally green, whereas the heterozygotes Cc shoul have produced white de- 

 scendants as well after self-fertilization. If one admits, with Nilsson-Ehle, 

 that the cause of the appearance of white plants is a mutation in which a 

 determinant chlorophyll factor has disappeared, this mutation, called a 

 " mutation of loss ", must have occurred at the time of hybridisation ; for, 

 on the one hand, the parents were pure, and on the other hand, the first 

 . appearance of the white plant:- in the generation F2 and their definite nien- 

 delism proves that the first association of C with c occurred in the genera- 

 tion P, i. c. in the hybridisation itself. 



As was shown at the beginning the crossing gave about 30 plants, of 

 which 26 were used for continuing the investigation. These all had the same 

 plant as mother and also the same male parent, as the pollen used for fertili- 

 zation was all taken from a single plant. Only one of these 26 plants had 

 any white descendants. The inevitable conclusion is that the supposed nmta- 

 tion had not affected the whole of the male or female parent, but only some 

 of the gametes, possibh' even one gamete only. It is obvious that of the 

 52 gametes (26 male, 26 female) which entered into combination in this cross 

 only one, either male or female, posessed the factor c, a predisposition to 

 the absence of chlorophyll. 



As in this case the factor c appeared at the time of crossing the idea 

 naturally suggests itself that the fact of crossing played a part in the forma- 

 tion of white descendants, and perhaps even caused the mutation. This 

 hypothesis is favoured by the fact that up to the present white individuals 

 have been chiefly noticed among cross pollinated plants. During the re- 

 searches on etiolated plants the author of this paper has found about i 000 

 white descendants in rye, i in 6-rowed barley, one in 2-rowed bent-eared 

 barley, i in oats, none in erect-eared barley and in wheat. 



With a view to solving these problems it is intended, during the next few 

 years, to grow both singly and in mass the pure lines from which the white 

 plants were descended. 



II. Rye. — In the autumn of 1915, 44 out of 104 lines (nearly 50 per 

 cent) of Saaleroggen (rye from the Halle district on Saale) divided up into 

 green plants and white plants. Nevertheless, in each line the proportion 

 of white ]ilants never exceeded 10 percent and on the average did not even 

 reach 5 per cent. The reason of this was that in rye after the crossir.g C 

 X c in the generation F^ no self fertilization occurs such as is necessary to 

 produce the mendelian numbers, but, on the contrary, the heterozygous 

 ovules Cc are fertilised by foreign honioz^-gous {CC) or heterozygous [Cc) 

 pollen ; it is only in the latter case that white plants can appear in the Fg 



