670 Journal of Agricultural Research vol. vi, No. 17 



On the other hand, by evaporation the enzyms present in the sap of 

 healthy plants may be brought to the highest possible concentration, but 

 such solutions never acquire infectious properties. 



That oxidase (producing the blue color with guaiac alone) can not be 

 responsible for the mosaic disease may be shown by heating the solution 

 to 70 C. for several minutes. This temperature destroys the oxidase, 

 according to Loew (15,' p. 31), but does not affect the peroxidase or the 

 principle of infection. As a matter of fact, the oxidase of the tobacco sap 

 appears to be an unstable enzym and very soon disappears entirely from 

 untreated solutions on standing. 



Although the enzym termed "catalase" by Loew is very often a normal 

 constituent of healthy and mosaic plants, it can be shown that the presence 

 of this enzym has nothing to do with infection. As shown by Loew (16, 

 p. 19), catalase is destroyed by heating the solution for a minute or two at 

 8o° C. Such solutions, although no longer showing reactions for catalase, 

 may yet retain their infectious properties. 



Although it is known that other enzyms than oxidase, peroxidase, and 

 catalase occur normally in the sap of healthy tobacco plants (18), such 

 enzyms can not be considered in a causal relationship to the mosaic disease 

 if it has been established that this disease is not of so-called physiological or 

 spontaneous origin — that is, it can not occur in the absence of infection. 

 Furthermore, the writer sees no reason to believe that any specific enzym 

 occurs in a mosaic tobacco plant which would not be found in healthy plants. 



Although it has been shown by various workers that the enzymic rela- 

 tions and reactions in plants become disturbed as a result of disease and 

 unfavorable conditions of growth, there is no reason to believe that these 

 disturbances, when associated with the mosaic disease of tobacco, hold a 

 causal relation to the disease.. It is now well known that various factors, 

 aside from pathological conditions caused by an unknown infective prin- 

 ciple, may change the quantitative relations of enzyms in plants, as 

 Bunzel (5) has shown in studying the curly-top of sugar beets. It yet 

 remains to be shown that an increase in the amount or activity of 

 enzyms in diseased plants is anything more than a symptom or an indi- 

 cation of disturbed metabolism as a result of the disease. 



In the writer's experience all evidence at hand indicates that the 

 mosaic disease of tobacco is dependent upon a specific pathogenic agent 

 which must be introduced into healthy plants from without before the 

 disease can arise. That this pathogenic entity is highly infectious and 

 is in some manner reproduced within the plant are established facts. If 

 these facts are interpreted according to those fundamental principles 

 upon which all our scientific conceptions in pathology and biology are 

 based, that infectious diseases are associated with parasitism and that 

 self-reproduction is a characteristic of living things alone, it must be 

 admitted that the pathogenic agents responsible for the mosaic disease of 

 tobacco must be parasites. If from the facts stated above it is held 

 that nonliving chemical substances such as enzyms or toxins engender 

 the disease, our fundamental biological conceptions no longer hold true. 



