It was found, as has been stated elsewhere, that the lead compound of the poison could not be precipitated in 95 per cent. alcohol. Further experiments, however, showed that on extracting the poisonous gum with 50 per cent. alcohol, a portion of it dissolved, and this solution gave a precipitate with lead acetate which was a true lead compound. The remainder of the purified tar (about 10 gm.) was treated with 50 per cent. alcohol and filtered. Very little dissolved in alcohol of this strength, but on addition of lead acetate in 50 per cent. alcohol to the solution, a light colored precipitate was formed, which became dark on standing. It was filtered off, washed free from lead acetate, decomposed by hydrogen sulphide, and shaken out with ether. The ether left, on evaporation, a yellow resinous substance having a faint odor like garlic. By drying in a desiccator, a small quantity of a solid yellow resin was obtained which was completely soluble in alcohol. A very small drop of this solution applied to the skin on the end of a glass rod which had been drawn out to a point caused an eruption in about thirty-six hours. Following the nomenclature used by Maisch and Pfaff, this substance will be designated as _Toxicodendrin_, the ending "in" indicating its glucoside nature. The filtrate from the lead precipitate just described was freed from the excess of lead acetate by hydrogen sulphide, was tested for poison, and was found to be poisonous, showing that the precipitation by lead acetate was not complete even in 50 per cent. alcohol. On spontaneous evaporation of the solution, a yellow, sweet smelling resin was left. A portion of the alcoholic solution of the toxicodendrin gave a dark coloration with ferric chloride, did not reduce Fehling solution and was slightly acid to litmus. To see whether the toxicodendrin could be hydrolyzed, the remainder was dissolved in alcohol and dilute sulphuric acid was added. A fine, white precipitate was formed at once which rose to the surface on standing as a light flocculent substance. The mixture was heated for several days on a water bath, filtered from unhydrolyzed resin and the filtrate was neutralized and concentrated in the way already described. The solution obtained reduced Fehling solution. Not enough was obtained for further sugar tests, but all the hydrolysis experiments point to the conclusion that the poisonous substance is a rhamnoside, and is the source of the sugar in the plant. The reaction with ferric chloride observed whenever a lead compound of the poison is decomposed by hydrogen sulphide may be explained by the formation of traces of gallic acid or fisetin through the action of the weak acids present.