standing several weeks while other work was in progress. The ester solution was then put in a vacuum desiccator over sulphuric acid and the alcohol evaporated. A black, tarry, solid mass was left having the ester odor. It was extracted with warm water and filtered from insoluble tar. The filtrate had a green color and the ethereal odor. It was shaken out with ether; the ether layer had a blood-red color while the water layer was deep green. The extraction with ether was continued until the water layer was no longer green. The combined ether extracts were evaporated in a desiccator without heat. A black tar-like solid was left very much like the original material, but it had the ester odor. It was partly soluble in water and readily soluble in alcohol. The alcoholic solution was tested on the skin and found to be not poisonous. The ester, or mixture of esters, was not investigated further in this connection, but was later shown to give the reactions for gallic acid and methyl furfurol. These reactions will be referred to in connection with other experiments.

After a few other preliminary experiments, it became evident that the original material was a complex mixture of substances and that it would have to be fractionated by some means and the fractions studied separately.

A portion of the original substance was treated with 50 per cent. alcohol and was found to be partly soluble in this medium. The solution was filtered from insoluble tar. A portion of the yellow filtrate gave a reddish yellow precipitate with lead acetate. The alcoholic solution was distilled in an Anschütze flask under diminished pressure; a yellow liquid condensed in the arm of the flask while most of the alcohol was collected in a bottle connected with the arm. The yellow liquid was acid to litmus. Water was added, the solution was shaken out with ether and the ether was evaporated. When the small residue was completely dry, it was a yellow solid soluble in dilute alcohol and acid to litmus. The substance was not volatile enough to justify the use of this method for getting it.

Chlorophyll could not be removed from the original substance because the solvents for chlorophyll such as alcohol, ether, fats, petroleum, and carbon bisulphide dissolve large quantities of the mixture.

A precipitate obtained by adding lead acetate to a filtered solution of the original substance in 50 per cent. alcohol was suspended in water, decomposed by hydrogen sulphide, shaken out with ether and the ether evaporated. The residue appeared at first to be a yellow oil, but on complete evaporation of the ether in a desiccator, a yellow solid was obtained—apparently the same as that obtained by vacuum distillation.

A solution of the original material in 50 per cent. alcohol was filtered through bone-black and the filtrate was colorless. Examination showed that everything had been removed by the bone-black and the filtrate was apparently pure alcohol and water.

In precipitating an alcoholic solution of the crude material with a solution of lead acetate, it was noticed that at least two kinds of precipitates were formed. The part that went down first was darker in color than that thrown down later. Pfaff used the last fractions in obtaining his oil and stated that these precipitates consisted of the lead compound of the oil in a pure state. It was found by experiment that the darker part was soluble in ether while the lighter part was not. This indicated that the darker part consisted of tarry matter which was brought down mechanically or separated out when the alcoholic solution was diluted by the lead acetate solution, or was perhaps a lead compound soluble in ether. To test this point an experiment was made as follows: Some of the crude material was thoroughly extracted with 50 per cent. alcohol. The tar insoluble in 50 per cent. alcohol was then treated with 95 per cent. alcohol; most of it dissolved; the solution was filtered and lead acetate in 50 per cent. alcohol was added. A greenish colored precipitate was formed which was filtered off and found to be completely soluble in ether and soluble to a considerable extent in strong alcohol. These experiments suggested that the light colored lead compound which was thought to contain the poison could be purified by extraction with ether in a Soxhlet apparatus more conveniently than by the tedious process of fractional precipitation. Further preliminary experiments showed that 50 per cent. alcohol extracted from the original material all of the substance or substances which gave the light colored precipitate and dissolved only a small amount of the tar.

Two hundred and eighty-eight grams of the crude material were then extracted several times with 50 per cent. alcohol and filtered; the insoluble tar was washed and saved for examination. To the filtrate was added an excess of a solution of lead acetate in 50 per cent. alcohol. The large precipitate, which will be designated as "precipitate A," was filtered and drained by suction in a Büchner funnel. The alcoholic "filtrate A" was saved. Precipitate A was extracted with ether in Soxhlet extractors until the ether came over practically colorless, the operation being interrupted from time to time to stir up the precipitate in the thimble. The green colored ether solution was saved for future examination. The lead precipitate, after extraction with ether and drying, weighed about 116 grams. A portion of this lead compound was decomposed by hydrogen sulphide in a mixture of water and ether which was well shaken during the operation. The ether was separated, filtered, and evaporated under diminished pressure without heat, and there remained a yellow oily looking residue having a pleasant odor. When the ether and water were completely removed in a vacuum desiccator, a hard, brittle, yellow resin weighing about 16 grams was obtained. It was soluble in alcohol, had a strong acid reaction and was free from nitrogen[16] and sulphur. The nitrogen tests were made by the Lassaign and soda lime methods,[17] and the sulphur test was made with sodium nitroprusside after fusing the residue with sodium. The main portion of the lead compound was decomposed under alcohol by hydrogen sulphide, filtered, and the alcoholic filtrate evaporated in vacuo. The same yellow acid resin was obtained. Experiments continuing through several weeks were made in applying solutions of this resin to rats, rabbits and guinea pigs. Finding it to be without effect upon these animals it was tried on the writer and found to be not poisonous.[18] In the meantime the resin was being studied in the laboratory.

GALLIC ACID.

An alcoholic solution of the resin was just neutralized with potassium hydroxide. During the titration, the solution rapidly became dark brown. After neutralization it was shaken with ether; the water solution remained brown while the ether layer was nearly colorless and contained practically no dissolved substance. A portion of the water solution of the potassium salt on being acidified with sulphuric acid and standing over night, deposited a slight precipitate. The solution of the potassium salt gave a heavy precipitate with lead acetate somewhat similar to the original lead precipitate A, and also slight precipitates with salts of zinc, mercury, copper, and silver (with reduction). It gave a