MICHIGAN ACADEMY OF SCIENCE. 315 



had been previously disinfected by treating with 1:1000 HgCL, for 10 

 minutes, and wasliing in three changes of sterile water, were planted. 

 The soil was watered every day from a beaker. The seedlings which 

 appeared above were not infected even after they had reached a height 

 of 6 cm. The plants at this stage were then separated into two groups. 

 The first group was watered as before (from a beaker), while the second 

 group was watered from above witli a hose, the water simulating more 

 or less the falling of rain. The soil was thus splashed uj) onto the lower 

 leaves. One week later the i)lants so watered were infected, wliile those 

 watered from below were not. 



The correlation between the spread of the disease toward the upper 

 parts of the plants in the field above described, and the precipitation 

 during the month of June, is striking, and corroborates what was proven 

 by exioeriment. The total precipitation from June 1st to June 10th in- 

 clusive was 0.46 inches, or a daily average of 0.016 inches. The total 

 precipitation from June 11th to June 21st inclusive was 3.28 inches, or a 

 daily average of 0.298 inches. Field observations showed that the spread 

 upward was most rapid between the last mentioned dates. 



From this experiment it is evident that while the soil and trash from 

 a diseased crop harbor the causal organism, they are not the only factors 

 in the production of infection. Splashing by rain is regarded as a trans- 

 locating agent. The rain, of course, also furnishes the moisture necessary 

 for germination after the spores have reached the plant. 



Field observations made during the course of this study seem to cor- 

 roborate the result of the above experiment. 



Seed : 



The most common agency introducing plant pathogenes to the field, is 

 the seed. Reference has already been made to the statement of Milburn 

 (1915) as to the presence of the fungus on the seed. Seed samples from 

 the 1915 and 1914 crops were examined for the presence of the fungus. 

 The fungus was not found growing upon any of the seed, as a parasite. 

 However, by centrifuging for one hour at a high rate of speed, spores of 

 M. sarcinaeforme were found in four out of nine samples so treated. 

 Incidentally, a large number of rust (Uromyces) and other fungus spores 

 were found. Some of the seed from samples containing disease spores 

 were planted in sterile soil, the pot containing it being covered with a 

 bell jar. None of those plants which came up developed any signs of 

 the disease, even after six weeks. This negative result is not entirely 

 conclusive, in that the experiment was not conducted upon a large enough 

 scale. But it seems likely that the spore containing seed is merely a 

 means of transferring the spores to the soil from which point, the infec- 

 tion takes place by splashing as has already been demonstrated. 



