TOBACCO WILDFIRE IN 1922. 5 



the seed-bed with chaff from infected pods of the previous year. It seems 

 improbable, however, that any considerable proportion of the spring 

 infection in the Connecticut Valley beds starts from the seed, because (1) 

 growers now know the disease well enough so that few of them would 

 save seed from infected plants; (2) many of the growers during the last 

 season used old seed (grown previous to 1920) and yet they did not escape 

 infection; (3) those who sterilized the seed were apparentlj^ no more 

 successful in eliminating the disease from the beds than those who did 

 not ; ^ (4) even those who advocate most strongly the sterilization of seed 

 do not present convincing data to prove that the disease organism is 

 carried on the seed. 



In the Soil. 



From the plant the bacteria may get into the soil in two ways: (1) they 

 may be washed from the plant by the rain during the growing season; 

 and (2) when the leaves or other infected parts are turned into the soil or 

 left to rot on the soil, the bacteria probably remain alive for a long time. 

 It is important that we should know how long they remain alive there and 

 capable of infection and whether they may survive the winter in this 

 habitat. 



Experiment 1. — In order to see whether the organisms could be carried from one 

 crop to the next through the medium of naturally infested soil, such soil was taken 

 from three beds of diseased plants at different times during the summer of 1921 

 and seeded with sterilized seed. The plants grown in this soil did not become in- 

 fected. On the other hand, in one of the greenhouse beds which had grown a num- 

 ber of diseased crops, sterile seed was planted in the spring of 1922 and the seedlings 

 became diseased before the plants were an inch high. 



Experiment 2. — In this experiment one pot of soil was inoculated by spraying a 

 suspension of bacteria over it, while another pot had an equal amount of water 

 sprayed on it. Both were seeded shortly after sprinkling, and wildfire developed 

 in the inoculated pot but not in the check. 



During some control experiments in Whately, it was observed that 

 even when all diseased leaves were removed from the plants, others became 

 infected after rains and almost always on the tips which were beaten down 

 into the soil. It appeared as though the bacteria had been washed from the 

 diseased leaves into the soil and then splashed from the soil to other leaves. 



In two fields in Hadley and North Hadley which were under constant 

 observation by one of the writers during 1922, the plants became so badly 

 diseased during June that all were pulled and carted from the fields. Both 

 fields were set later with healthj'^ plants, but in both cases there was a 

 very heavy reinfection before the new plants were half grown. The 

 second infection must have come by way of the soil. 



Clinton and McCormick (2 : 404) buried wildfire leaves under healthy 

 plants, and by this means the infection was increased to 63 per cent as 

 compared with 13 per cent on adjacent plants not so treated. 



1 Records were kept on the beds of 11 growers in Massachusetts who treated their seed with 

 mercuric chloride. Wildfire afterward appeared in 5, while the other 6 had no wildfire in the beds. 



