The distillate from the last distillation was examined spectroscopically by Professor Rowland and found to be free from all traces of impurity which would be detected by that method. The chemical test for arsenic was more delicate than the spectroscopic and this failed to reveal a trace. The preparation of pure nitric acid. The method of preparing the pure acid and of preserving and transferring it was the same as adopted by Morse and Burton in their work on the atomic weight of zinc. [Illustration: FIG. 2.] The simple form of apparatus is represented in fig. 2. A large platinum vessel containing fragments of ice was supported on a smaller platinum dish, from which it was separated by hooks of large platinum wire. The acid was distilled from a small flask as represented in the drawing. The purest nitric acid which could be obtained was diluted with about an equal volume of water. The vessel containing the acid was heated very gently that the distillation might take place without boiling. The dilute acid condensed on the cold surface of the larger dish and collected in the smaller, in which it was preserved until used. This acid gave no residue on evaporation. The arrangement of crucibles. [Illustration: FIG. 3.] The arrangement of the crucibles in which the determinations were made is represented in fig. 3. 1 is a small porcelain crucible, (00) from the exterior and lid of which the glaze had been removed by hydrofluoric acid. The lid was separated from the crucible by hooks made from thick platinum wire, to allow free communication between the contents of the crucible and the external air. This would facilitate the outward diffusion of the oxides of nitrogen when liberated from the nitrate. 2 is an uncovered porcelain crucible (no. II) in which 1 was placed. From the exterior the glaze had been removed to prevent the crucible from adhering to the unglazed porcelain scorifier on which it rested. The exterior was carefully brushed after treatment with hydrofluoric acid to remove all loose particles adhering to its surface. Crucibles 1 and 2 were not separated during a determination. 3 is a nickel crucible about two and a half inches in diameter. The porcelain crucibles were not allowed to touch the nickel at any point. The nickel crucible was covered by a lid of nickel. The mode of procedure. A piece of cadmium weighing from two to three grams was cut from the bar of the metal by means of a steel chisel. This was seized with steel forceps and filed with a hard steel file to about one half the original weight. Care was taken to remove the entire exterior portion of the metal which had come in contact with the chisel or had stood exposed to the air. The plug of metal was then carefully brushed and examined with a lens to insure the removal of all loose particles from the surface.