672 Journal of Agricultural Research Vol. vi, No. 17 



tive principle of the disease, leaving the clear, supernatant solution with- 

 out infectious properties, although showing good peroxidase reactions. 

 Similar treatment with nickel sulphate was not so satisfactory, as it gave 

 evidence of being more toxic to the infective principle than the aluminum 

 salt. 



(7) The virus is quickly killed at temperatures near the boiling point 

 of water. In some instances heating the virus for five minutes at 8o° C. 

 was sufficient to destroy its infectivity. In other tests, with a different 

 virus solution, heating for five minutes at oo° did not entirely destroy its 

 infectivity. In dried and ground mosaic material, rendered water-free 

 by drying over sulphuric acid in a desiccator, the infective principle 

 resisted much higher temperatures than it did in solutions. If virus 

 solutions are heated, the thermal death point of the infective principle 

 is lower than that of the peroxidase; or at least it is more quickly 

 destroyed than the peroxidase. 



(8) The virus is highly resistant to low temperatures. When frozen 

 to a temperature of — 1 8o° C. with liquid air, its infectious properties 

 were not weakened. 



(9) Unpreserved solutions of virus have sometimes lost their peroxi- 

 dase activities without losing their infectious properties. Dried and 

 ground mosaic material has also lost its peroxidase activities and still 

 remained highly infectious. Talc-treated material, while retaining its 

 infectious properties, has lost its peroxidase activities. 



(10) Solutions of virus sometimes lose their infectious properties and 

 continue to show intense peroxidase reactions, as when allowed to evap- 

 orate spontaneously in one instance. The feces of worms fed upon 

 mosaic plants have, in some instances, failed to produce infection, 

 although such material continued to give intense peroxidase reactions. 



(11) The writer's experiments show that peroxidase or catalase in the 

 sap of mosaic plants can not be responsible for the mosaic disease. The 

 same enzyms are normally present in healthy plants, but the sap of such 

 plants is without infectious properties. By evaporation the enzyms 

 present in healthy sap may be brought to a high concentration, and such 

 solutions never acquire infectious properties. By dilution, on the other 

 hand, the peroxidase content of mosaic sap may be diminished to such 

 an extent that peroxidase reactions are no longer discernible; yet such 

 solutions may remain highly infectious. 



Since it has been shown that the mosaic disease of tobacco does not 

 occur in the absence of infection, neither enzyms nor other normal con- 

 stituents in the sap of healthy plants can be considered responsible for the 

 disease. A specific, particulate substance not a normal constituent of 

 healthy plants is the cause of the disease. Since this pathogenic agent 

 is highly infectious and is capable of increasing indefinitely within sus- 

 ceptible plants, there is every reason to believe that it is an ultramicro- 

 scopic parasite of some kind. 



