MODERN PLANT-BREEDING METHODS 719 



In spite of the severity of the attack, a fair quantity of grain 

 was obtained at harvest, though very few grains of Michigan 

 Bronze were produced. It is possible that the exceptional 

 vigour of the hybrids enabled them to withstand the attacks of 

 the parasite better than the parent, but on the other hand, it 

 may be that the actual severity of the attack was not so great 

 as in the case of Michigan Bronze, though, if amount of rust on 

 the foliage is the reliable indication we take it to be, this 

 was not the case. Whether we are dealing with the ordinary 

 phenomena of dominance, or whether the heterozygote is more 

 or less intermediate between the parents in this respect, is 

 a matter of no immediate importance ; the essential fact is that 

 the hybrid was excessively susceptible to the disease which 

 attacks one of its parents. 



All the available grain was sown in the following season. 

 The rust was late in appearing, and the epidemic was compara- 

 tively slight, but the plots taken as a whole were badly attacked. 

 About midsummer, when the epidemic was considered to be at 

 its height, the plots were examined plant by plant, with the 

 result that amongst this mass of rust-laden individuals numbers 

 were found to be free from any sign of disease. That this 

 freedom from disease was in any way due to lack of opportunity 

 for infection was disproved by the fact that rust-coated leaves 

 were continually rubbing against the plants from their diseased 

 neighbours, with the result that loose spores were found in 

 abundance on the foliage. These disease-resisting plants were 

 kept under observation until the time of harvest. They retained 

 their immunity till the end, their clean vigorous foliage and 

 stems standing out in vivid contrast to the diseased and 

 shrivelled plants around them. The proportion of susceptible 

 to immune plants was approximately as three to one, the 

 expected Mendelian ratio if immunity is recessive to suscepti- 

 bility. Experiments with slightly susceptible and extremely 

 susceptible parents showed in a similar manner that the ex- 

 cessively susceptible habit was again dominant, and crosses 

 between very susceptible varieties gave either all diseased 

 offspring in the generation raised from the hybrids or failed 

 outright owing to the severity of the disease. Wherever it 

 was possible to obtain statistics this has been done, and the 

 nett result is that, if figures are any guide, immunity and 

 susceptibility to the attacks of yellow rust form as sharply 



