Sept. 1. 1920 Genetics of Rust Resistance 539 



These facts show that it is possible to transfer the rust resistance of 

 emmer wheats to common wheats by crossing and subsequent selection. 

 There is, however, an apparent partial linkage between rust resistance 

 and the emmer head type which makes it essential to grow large num- 

 bers in the F.^ and F3 generations. 



SUMMARY 



(i) Recent studies of the parasitism of the black stemrust of wheat 

 {16, 18, 21, 24) have shown that there are many biologic forms of Puc- 

 cinia graminis which can be differentiated only by their action on pure 

 line wheat hosts. This seriously complicates the breeding of wheat for 

 rust resistance. In the light of this knowledge differences of infection 

 of certain crosses in 1917, as compared with 1916 or 1918, shov/ that the 

 conflicting results may be explained logically by supposing that more 

 than one biologic form was present in the rust nursery in 191 7. 



(2) Sterility is a factor which must be considered in a study of crosses 

 between common wheats and durum or emmer varieties. Sterility was 

 shown in three ways: (a) pollen abortion; (b) the fact that F^ florets of 

 durum-common crosses set nearly 50 per cent less kernels than the 

 parent sorts, while Fj emmer-common crosses produced about 25 per 

 cent of barren florets; and (c) the large number of natural crosses which 

 occurred in some F2 plants as shown by the F3 results. 



(3) Crosses between durum and common wheats produced many dif- 

 ferent forms in the F2 generation, such as compact keelless commons 

 resembling club wheats, lax sharply keeled durums, both emmer and 

 spelt, as well as types \vhich resembled the poulard group. Lax and 

 compact durum, common, and emmerlike forms were obtained which 

 bred true in the F3 generation. The segregation in the Fj generation of 

 emmer-common crosses was not so wide as in the durum-common cross, 

 although both lax and compact keelless wheats which bore naked kernels, 

 as well as lax and compact wheats with adherent-glumed kernels, were 

 obtained. 



(4) The study of inheritance of rust resistance was made in a specially 

 prepared disease plot. Because of the conflicting results of 191 6 and 

 1 91 7 all barberry bushes were removed early in the spring of 1918 from 

 the immediate vicinity of the rust plot and the epidemic was induced 

 with a known form of rust. The data on rust infection are based on 

 these 1 91 8 results. 



(5) The following species and varieties were used in the study: 

 Triticum vulgare, varieties Preston, Marquis, and Pioneer; Triticum 

 dummy varieties Acme, D-4, Kubanka (C I 2094), and lumillo (C I 1736); 

 Triticum dicoccum, White Spring emmer (Minnesota 1165 and C 1 1524). 



The three common wheats were susceptible, the durums were com- 

 merically resistant, Kubanka (C I 2094) being somehwat less resistant 

 417°— 20 2 



