BREEDING SNAPDRAGONS 3 



Snapdragon rust has been known for some sixty years; yet there are peculiari- 

 ties about the epidemiology of the disease which still are not clearly understood. 

 For example, climatic conditions appear to affect the prevalence and degree of 

 destructiveness of the disease. Maximum infection of snapdragon plants in the 

 field at Waltham, Massachusetts, generally occurs by early August, but some 

 years it may be mid-July or l?te August. Blodgett and Mehlquist (2) note that 

 in 1936 rust disease was extremely light on snapdragon seed field plantings at 

 Guadalupe, California; whereas at Lompoc, some th'rty miles south, the disease 

 situation was serious. In 1937, however, rust disease conditions in the plantings 

 were entirely reversed, being serious at Guadalupe but much less so at Lompoc. 



Materials and Methods 



Breeding work with snapdragons was initiated at Waltham, Massachusetts, to 

 develop strains which would be less susceptible to rust than commercial varieties. 

 The approach to the problem has been along two lines: interbreeding susceptible 

 varieties with rust-resistant strains developed by Dr. E. B. Mains (9); and inter- 

 crossing highly susceptible commercial varieties with one another to produce 

 resistant types of snapdragons different from those obtained by use of Mains' 

 material. The rust-resistant strains obtained from Mains were not suitable for 

 commercial use but were considered important because of their potential value for 

 breeding. Some eleven different resistant lines of these snapdragons were planted 

 in the field to determine their degree of rust resistance under climatic conditions 

 in Massachusetts. A number of these strains had been released some years earlier 

 by Dr. Mains to experiment station workers in Virginia, Arkansas, Michigan, 

 and California for testing and further use in breeding work. The following suscep- 

 tible garden varieties of snapdragons were used as comparative test material: 

 Rub}', Copper King, Old Gold, Canary Bird, Golden Dawn, Apple Blossom, 

 Cattleya, and Gotelind. 



Mains' rust-resistant strains all produced plants in which the flowers were 

 predominantly magenta colored, with the exception of strain GWl-Gl-Fl-GBl 

 which was white flowered. Individual florets on the plants were narrow and 

 pointed. The foliage as well as habit of growth was characteristic of the wild 

 species of Antirrhinum. These plant-growth habits in the hybrids are noted by 

 Mains (9) as being typical of A. gliilinosum, one of the parents from which the 

 rust resistant strains were derived. Strains 7-13-8-1-Gl-Fl-Fl, 7-13-8-1-Gl-FI- 

 Gl, and GWl-Gl-Fl-GBl were highly susceptible to rust at Waltham; whereas 

 2-1-1-1-Gl-Fl-Gl, 7-13-8-1-G6-F2-F1-G1, 292, 293, and 294 strains showed some 

 resistance. The number of progeny obtained from the different strains was not 

 sufficient to permit an analysis as to the hereditary ratio of segregation for rust 

 resistance. Garden varieties of snapdragons used in the field for comparison 

 with the rust-resistant strains were completely destroyed by the disease. Mains 

 (9), working with some of the same garden varieties, reported that the variety 

 Canary Bird showed considerable tolerance to rust in his tests; but at Waltham, 

 Canary Bird was severely rusted. 



Plants selected from the most resistant line, No. 294 (Table 1), were crossed 

 with the following susceptible commercial varieties: Ceylon Court, White Rock, 

 Cheviot Maid, Afterglow, W. W. Thompson, Jennie Schneider, Helen, Weld 

 Pink, Penn's Orange, Emily, Giant White, and Giant Yellow. Reciprocal crosses 

 were made between the No. 294 and Ceylon Court and White Rock. There were 

 7,000 to 8,000 plants of the Fi generation crosses tested for rust resistance in the 

 field. Check rows of susceptible varieties were interplanted with the hybrid 

 material to insure adequate inoculation with rust. 



