56 Journal of Agricultural Research voi. xvii. No. a 



owing, probably in large measure, to poor infestation; and the second 

 test was ruined by a heavy infection of nematodes. All the soils were 

 then sterilized by steam, and two of each series again inoculated, this 

 time by the application of equal quantities of chopped-up, air-dried, 

 heavily infected roots which had been taken in the fall from the tobacco 

 field. The pots were again set to young seedlings of White Burley from 

 steamed soils. The infestation now proved to be good, but complica- 

 tions arose in the limed end especially, owing either to the influence of 

 the lime itself or to the sterilization alone or more likely to the two 

 combined. It seems most probable that the lime was concerned in 

 rendering the phosphates unavailable to the plants, but, as was expected, 

 this injury was probably not a factor in the following experiment. How- 

 ever, this test, which has been designated as experiment I, in Table III, 

 produced some fairly striking results in the infested series. 



It was evident during the early growth of the plants that those in 

 the infested soil of highest acidity were making considerably better 

 growth than those at the alkaline end. It also appeared early that the 

 line of demarcation between heavy infection and reduced infection in 

 this series did not lie near the point of neutrality but well into the acid 

 end and so the soil requiring 4.6 tons of lime per acre was as productive 

 of disease as any at the alkaline end. Growth of all the plants was 

 slow as the soil was not very fertile and the light poor. The plants, 

 therefore, were harvested when still relatively far removed from the 

 blossoming stage. They were dried at about 80° C. for several days, 

 and then allowed to come to air-dry weight. The data given under ex- 

 periment I, Table III, sufficiently illustrates the results obtained. The 

 roots were carefully removed and examined for lesions of the disease, 

 and, as was expected, these were correlated with the growth of the plants. 

 In the soil requiring 9.38 tons lime per acre only a most careful search 

 revealed any T. hasicola at all. In the next lowest series (7.19 tons per 

 acre), although considerable disease was present, plainly its development 

 was markedly checked, whereas in all the series below this practically 

 no uninfected portions of roots existed. 



The crocks were now replanted with Connecticut Havana tobacco, a 

 semi- resistant type. These were allowed to grow for about 50 days be- 

 fore being harvested. The air-dry weights are given under experiment 

 II in Table III, It will be noted that the soil at the alkaline end prac- 

 tically recovered from the injurious properties previously described in 

 the control series. In the infested series it ma}' be noted that the point 

 of effectiveness of the acid reaction in reducing the disease shifted to a 

 somewhat lower degree of acidity, undoubtedly due to the greater re- 

 sistance of the variety (PI. 4, I-II). For some unexplainable reason 

 the plant in one of the pots of series 5 and also one in series 7 failed to 

 become as seriously infected as those in the neighboring pots. The in- 

 creased yields in series 5 and 7, however, are not considered as inter- 



