MOSAIC DISEASE OF TOBACCO. 87 



parts of a diseased plant will cause infection, it would be natural to sup- 

 pose that if bacteria were the causal agent, it should be possible to demon- 

 strate their presence in the different parts of a diseased plant. This has 

 never been done, and in the writer's study of the anatomy of diseased 

 plants it has never been possible to demonstrate the presence of bacteria 

 in the different tissues. The writer has many times attempted to obtain 

 cultures of bacteria from diseased tissue, and in some cases cultures of 

 organisms were obtained on various media, but they proved in every case 

 to be secondary in character, and were not capable of reproducing the 

 disease. In the Ught of all later investigations the evidence points over- 

 whelmingly to the absence of bacteria, in the present-day sense of the 

 term, as the causal agent of the disease. 



Dissemination Agents. 

 Insects. 



The fact that many fungous and bacterial diseases are often transmitted 

 by insects, as well as other agents, has been long known and thoroughly 

 estabUshed, but until Allard {loc. cit.) called attention to the fact that 

 the mosaic disease could be carried by aphids, and one in particular 

 {Macrosiphum tobaci Perg.), nothing had been published on this phase of 

 the matter. Allard in well-controlled experiments demonstrated beyond 

 a reasonable doubt that the disease was so communicated. Clinton 

 (loc. cit.) made a few observations on the infection of healthy plants by 

 the tobacco horn worms which had been feeding on diseased leaves, but 

 was unable to demonstrate that the disease could be so transmitted either 

 by the excreta ejected by the worm or by its biting and feeding on the 

 healthy plants. His results were negative in the few experiments made. 

 Observations made in the field during the progress of the writer's work 

 have not shown conclusively that the disease is communicated by biting 

 insects, such as the tobacco horn worm, grasshoppers and a small black 

 flea beetle of more or less common occurrence in our fields. 



Occasionally aphids have been found infesting the leaves of tobacco in 

 our fields, but so far as could be judged were present in too small numbers 

 to be active agents in transmitting the trouble. As a rule, comparatively 

 few aphid infestations are found in our tobacco fields. 



In the greenhouse during several winters tobacco plants grown in benches 

 were infested with white fly, and it was at first feared that they might 

 carry the infection from diseased to healthy plants in the same benches. 

 This, however, was not the case, and it has never been possible to demon- 

 strate positively that the white fly is an active agent in the spread of the 

 disease. This insect is, of course, of rare occurrence in our fields, but 

 may possibly do damage in the south. It apparently feeds and breeds 

 freely under greenhouse conditions on the underside of the leaves. 



In order to ascertain more definitely the possibility of infection by these 

 insects, adult white flies from badly mosaicked leaves were carefully re- 



