weighing the empty boat. At this stage the sulphide was always from one to two milligrammes lighter than at the end of the determination. It was reheated for periods of one hour until the weight remained constant. This generally required from three to five hours. All weighings were reduced to the vacuum standard on the basis of 4.5 for the Sp. Gr. of cadmium sulphide, 3.31 for the Sp. Gr. of cadmium oxalate, 8.4 for the Sp. Gr. of brass weights and 21 for the Sp. Gr. of platinum weights. The results are as follows: Cd C_{2}O_{4} CdS At. Wt. Cd. I. 2.56319 1.84716 112.25 II. 2.18364 1.57341 112.17 III. 2.11643 1.52462 112.05 IV. 3.13105 2.25582 112.12 The first three determinations were made exactly as above described, the heating in hydrogen sulphide being done in a Bohemian glass combustion tube. The hydrogen sulphide was dried with calcium chloride. The fourth determination was made under somewhat different conditions. The boat containing the weighed oxalate was placed in a combustion tube which passed through an asbestus covered air-bath. The air was displaced by a current of dry hydrogen sulphide and the bath slowly heated. When the temperature had risen to 210°C. it was maintained there for three hours, and then raised to 250°C. for three hours. The sulphide then weighed 2.27 grammes, being 14 milligrammes heavier than when the determination was finished. It was replaced in the tube and reheated in a current of hydrogen sulphide at a temperature of 300°C. for four hours. It was then transferred to a porcelain tube and heated to redness for one hour. It then weighed 2.25437 grammes, being 1.45 milligrammes lighter than at the end of the determination. The weight did not become constant until it had been heated six hours more to redness in a current of hydrogen sulphide. When this oxalate was slowly heated in H_{2}S, a small amount of oxalic acid sublimed to the colder part of the tube, but, in the other cases where the heating was more rapid, only carbon monoxide, carbon dioxide, and water were observed. Discussion of the Method. When hydrogen sulphide is passed through a red-hot tube, sulphur is deposited on the colder parts because at this temperature hydrogen sulphide dissociates and the elements do not recombine on cooling. In this work, a faint sublimate was noticed before coming to the zone of sulphur deposit. On exposure to air, it deliquesced in a few minutes forming small yellow drops which had a saline taste, and gave tests for potassium and sulphur. The sublimate had a yellow color and was evidently formed by the action of sulphur on glass. It seemed to do no harm, but in the fourth determination an effort was made to avoid it by using a porcelain tube instead of a glass combustion tube for heating to redness in a current of hydrogen sulphide.