210 University of Calif ornia Pulilications iti Agricultural Sciences [Vol. 2 



DISOUSSION OF RESULTS 



One Avould expect that the germ-plasm of an old wild species 

 had been largely purified of the genes which cause the production of 

 abnormal and harmful characters by the elimination of weak forms 

 through natural selection, but our experience with Crepis demon- 

 strates that such is not the case in a race partially cross-pollinated. 

 The germinal material of Crepis capillans is maintained in a hetero- 

 zygous condition by natural cross-pollination, as is the case in the 

 cultivated species of maize. This is shown first by the marked re- 

 duction in the size of the plants and their slower rate of development, 

 and secondly by the fact that we have isolated, by inbreeding and 

 selection, constant breeding forms wiiich differ in the characters for 

 which selected. 



The maximum amount of the effects of inbreeding appears to 

 occur in the second and third generations. Forms have been observed 

 in inbred cultures which have not been observed in wild colonies ; 

 no doubt a more extended observation would show that they do occur, 

 though rarely. Pollen sterility is one of the results of inbreeding 

 and one plant has appeared in a third generation culture which pro- 

 duced almost no pollen. In the culture produced by growing seed 

 of wild plants which themselves grew in New Zealand we have also 

 found one plant (N. Z. P^, 1920) which produces no pollen at all; 

 other plants of this culture appear normal in pollen production. 

 This is evidence that this character may also appear in wild plants. 



Strains of fasciated plants have been isolated in Crepis tectorum 

 which are so weak that it is only by starting a large number that we 

 can get a few to live long enough to produce seed, yet the plants in 

 the heterozygous condition seem to be in no way affected. 



Most of the plants were grown in four-inch clay greenhouse pots. 

 In order to determine whether this limiting of the root space would 

 in any way accentuate the dwarfishness of the inbred plants, part of 

 inbred strain No. 128 and part of hybrid culture No. 129 were grown 

 in both four-inch and six-inch clay pots and placed in adjacent rows 

 on the bench. Plate 40 showing plants in six-inch pots and plate 39 

 showing similar cultures in four-inch pots answer this question in a 

 very definite manner. 



