124 BACTERIA IN RELATION TO PLANT DISEASES. 



The results at Arlington agree substantially with those obtained from the trial-rows 

 on the other farm. In both Cosmopolitan and Crosby's Early they point unmistakably to 

 infection of the plants from the seed-corn. In the other three varieties the cases are so 

 few that they may have arisen from proximity to the two diseased varieties. It is not diffi- 

 cult to understand how this could have taken place in blocks of healthy plants sandwiched 

 in between diseased ones, provided some of the latter developed the disease early in the 

 growing season. If the disease was actually derived from the seed-corn there probably 

 would have been some cases during the seedling stage, and fragments of these soft plants full 

 of the bacteria would have been blown upon neighboring plots, or dragged by cultivators, or 

 carried on the feet of men and horses, or bitten into by insects, or washed about by rains 

 and dews. There are ways enough to account for the dissemination of the bacteria and the 

 infection of a few plants where the distance is only a matter of a few feet. When plants 

 are well past the seedling stage there is much less danger of infection, even though the 

 infectious material is then much more abundant. That a larger number of cases did not 

 occur in the Crosby's Early and the Cosmopolitan is attributable to the poor soil and slow 

 growth as compared with that on the Flats and probably also to the gradual dying out of 

 the bacteria on the seed-corn between spring, when the first plantings were made, and 

 midsummer, when the last plantings were made. 



The experiment is equally interesting, however, if we assume all of the seed-corn to 

 have been free from Bact. stewarti. In such event the organism causing the disease must 

 have been present already in the soil of the field. If for the sake of argument such a 

 supposition be granted (and it can be granted for no other reason), it certainly stretches 

 the limits of probability to assume also that the organism was prevalent only in those 

 narrow strips where Crosby's Early and Cosmopolitan corn happened to be planted. If 

 the organism was present in the soil it is likely to have been there a long time and to have 

 been pretty uniformly distributed, just as it is in the infected soils on I^ong Island. Granted 

 these premises, then we may assume that 3 of the 5 varieties tested are very resistant to the 

 disease, and are, therefore, desirable varieties to plant on infected land, and important 

 to use as one parent in originating resistant varieties by cross-breeding. The other two 

 varieties, on the contrary, should never be planted on infected land. 



Subsequent experiments showed that an abundance of moisture stimulating rapid 

 growth is very favorable both to primary infection during the seedling stage, and also to 

 the general distribution of the bacteria through the stem of the plant later on. On the 

 contrary, dry soil and dry weather following planting interferes with infection, and any 

 decided check in growth later in the season interferes with the movement of the organism 

 through the plant by hardening the basal nodes and thus rendering it difficult for the 

 organism to get past them. In infected plants, however, the disease shows sooner, accord- 

 ing to Stewart, in periods of drouth than in rainy weather. The writer has observed the 

 same thing in diseased Cucurbitaceae (see this monograph, Vol. II, pp. 216, 284). 



EXPERIMENTS OF 1908. 



In 1908, Golden Bantam sweet corn attacked by Bacterium stewarti was sent to me 

 from a garden in Falls Church, Virginia, with the statement that the seeds were obtained 

 from a Philadelphia seedsman, who was believed by the planter to have sold him bad seeds, 

 because nearly all the plants of this variety were badly diseased, while other varieties 

 obtained elsewhere showed no disease or only very little. I asked the dealer where he 

 procured the seed of this variety and he said it came from Mr. - , Wakeman, Ohio, 

 the man to whose farm I traced the disease in 1903. 



This interesting fact led me to procure a half -bushel of this particular lot of seed (all 

 the dealer had left) for experiment. The first endeavor was to isolate Bacterium stewarti 

 from the surface of the kernels. In this effort we failed. The organism may have been 



