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



In the next experiment a 45 per cent alcoholic solution of virus was 

 made up with virus X 23 as follows: 825 c. c. of this virus, which had 

 been previously filtered through filter paper, were shaken with 675 c. c. 

 of absolute alcohol on January 14, 191 6. On January 15 the clear, 

 supernatant solution was siphoned off. A portion of this was filtered 

 through hard filter paper and a second portion was left unfiltered. Of 

 the unfiltered portion 15 c. c. were also diluted with 15 c. c. of distilled 

 water. On January 15, 1,000 c. c. of the supernatant solution which had 

 been filtered through paper were set aside in a large, shallow dish to 

 evaporate. On January 20 the dry residue was taken up with 400 c. c. 

 of distilled water, which gave a somewhat stronger concentration than 

 the original virus. (See Table IV.) 



After decanting off as much of the supernatant solution as possible, 

 the heavy, semiliquid precipitate, or sludge, was treated as follows: A 

 portion was left unfiltered; a second portion was diluted by adding 15 

 c. c. of distilled water to 15 c. c. of the sludge. In addition to this, 200 

 c. c. of the sludge were placed in a beaker to evaporate to dryness at 

 room temperatures on January 15. The dry residue was taken up with 

 150 c. c. of distilled water on January 17. Inoculation tests were now 

 made with the virus after undergoing the various treatments outlined 

 above in connection with precipitations with ethyl alcohol. 



From Table IV it will be seen that the infective principle of the virus 

 has been completely destroyed in the 75 per cent and 80 per cent alco- 

 holic solutions, although the precipitates continued to give intense reac- 

 tions for peroxidase. In these strengths precipitation of the peroxidase 

 was complete, as the supernatant solutions gave no reaction for this 

 enzym. 



In the 45 per cent and 50 per cent alcoholic solutions, the infective 

 principle was not appreciably injured. The infective agent, however, 

 appears to have been carried down with the heavy, flocculent precipi- 

 tates, leaving the supernatant solutions free from infective properties. 

 Owing to the fact that the peroxidase remained in solution, the super- 

 natant solutions continued to give strong peroxidase reactions. Accord- 

 ing to Chodat and Bach (7), the oxygenase in the sap of a species of 

 Lactarius could be largely precipitated by 40 per cent alcohol, while the 

 peroxidase remained in solution. 



The writer's experiments indicate that concentrated solutions of per- 

 oxidase precipitated by strong alcohol from the sap of mosaic plants will 

 not produce infection in healthy plants. Furthermore, the writer has 

 carried out successive re-solutions in water and re-precipitations with 

 alcohol in order to obtain purer solutions of peroxidase. Such solutions, 

 however, have never produced infection, although giving intense reac- 

 tions for peroxidase and in some instances for catalase. 



