EAST: INTERCROSSES BETWEEN SELF-STERILE PLANTS 147 



as if each carried the composition of the mother plant from which it 

 came. In other words, as far as its action in fertilization is concerned, 

 a pollen grain partakes of the character of its mother plant and is 

 ITkeTtF sisters; as far as the hereditary characters carried on to the 

 next generation are concerned, sister pollen grains may differ both 

 from their mother and from each other. 



A part of our evidence on these points we shall present. For 

 further details the reader is referred to a forthcoming paper in Genetics.^ 



The first experiment to which attention is called is an inbreeding 

 experiment performed on a cross between Nicotiana forgetiana and 

 Nicotiana alata. If sister plants are mated in successive generations 

 after an original mating Aa X Aa, by Mendelian recombination there 

 results a gradual approach to 1/2 AA, 1/2 aa and o Aa. Expectation 

 of homozygosis in successive matings is 1/2, 5/8, 11/16, 24/32 ••• i 

 (Jennings, 1916). If, therefore, plants of like constitution as far as 

 effective factors are concerned are cross-sterile with each other, 

 cross-sterility should become more and more apparent in generations 

 succeeding Fo. To test this possibility, a comparatively small number 

 of cross-matings was made on the F2, F3, F4 and F5 generations. In 

 the F2 generation, out of 131 intercrosses on 20 plants only 4 were 

 unsuccessful. The percentage of unsuccessful matings increased from 

 this time on, until in the F5 generation about 21 percent of the cross- 

 matings tried on 20 plants were impossible to make. 



In this experiment as well as in all_others, results showed that 

 reciprocaj_ ^Grnsses were alike in their compatibility. If two plants 

 werg_iertile together, they were fertile reciprocally; if two plants 

 were incompatible, they were incompatible reciprocally. This is 

 proof of the sporophytic behavior of the factors affecting the behavior 

 of self-sterile plants. 



The two crosses to be described next are reciprocals made with the 

 same two individuals. Made with Nicotiana alata and Nicotiana 

 forgetiana as parents, they are in a sense repetitions of the cross just 

 described, but it is hardly probable that they duplicate it. Both of 

 these species must consist of plants which differ among themselves in 

 the factors which afifect self-sterility, hence any crosses in which 

 different individuals are used may show different results. 



All of the individuals resulting from this cross were grown in a 

 greenhouse as^glted-plantsr The Fi generation came into blossom 

 durmg the latter part of the winter. Conditions were extraordinarily 

 favorable for growth and the pollinations were all made while the 

 plants were vigorous, hence scarcely any trouble arose over classi- 

 fication of the results through end-season pseudo-fertility. 



'This paper has since appeared. See "Studies on Self-sterility I. The 

 Behavior of Self-sterile Plants." Genetics 2: 505-609. 1917. 



