SPECIFICITY OF THE PRECIPITIN REACTION IN TOBACCO MOSAIC DISEASE.

1. Leaf extracts of Sudan grass, Hippeastrum equestre Herb., lily, and Abutilon striatum Dicks. (A. Thompsoni hort.), each affected with its respective mosaic disease, and peach affected with yellows disease, were tested for their ability to precipitate antiserum for virus extract of tobacco mosaic disease. No precipitate occurred. 2. Nicotiana glutinosa L., N. rustica L., and Martynia louisiana Mill. were added to the list of hosts of tobacco mosaic virus which have been tested with antiserum for the same virus in N. tabacum L. var. Turkish. The object was to determine the presence or absence of material reacting with the specific precipitins such as that already demonstrated in extracts of tomato, pepper, and petunia affected with the same virus. The presence of specific substances was demonstrated in every case. 3. The viruses of ringspot and cucumber mosaic diseases were multiplied in Turkish tobacco and leaf extracts of the affected plants were used in turn as antigens in precipitin tests with antiserum for tobacco mosaic virus extract of Turkish tobacco. A slight precipitation resulted in the tubes containing undiluted antiserum and virus extract such as occurs when juice from normal tobacco is used with undiluted antiserum. No precipitate was demonstrable that was specific for virus extracts of tobacco affected with either ringspot or cucumber mosaic disease. 4. The results favor the interpretation that the specific antigenic substance in virus extract of tobacco mosaic disease is foreign antigenic material, possibly virus itself, not altered host protein.