That it may act with a greater valence was shown by R. Haafs[12]. He found that when zinc hydroxide was treated with hydrogen dioxide certain compounds of zinc and oxygen were formed containing more oxygen than the normal oxide ZnO. The close resemblance between zinc and cadmium led him to try the same reaction with cadmium. Hydrogen dioxide was accordingly allowed to act on cadmium hydroxide and the resulting product analyzed. There were formed Cd₅O₈, Cd₃O₅ and Cd₄O₇. In no case was the compound CdO₂ obtained. These compounds are described as fairly stable even at a hundred degrees. [11] Jahrb. 1855, 390. [12] Ber. 1884, 2249. The Preparation of Cd₄Cl₇. When anhydrous cadmium chloride is heated with metallic cadmium in a vacuum, or in an atmosphere of nitrogen, to the fusing point of the chloride, the molten chloride quickly assumes a garnet red color. In order to investigate this phenomenon a quantity of the chloride was prepared by dissolving the redistilled metal in an excess of hydrochloric acid, evaporating the chloride to dryness on a water bath, and finally removing the water of crystallization by heating in a current of dry hydrochloric acid gas. The heating was effected by placing the chloride in a long platinum boat, which was shoved into a large glass tube, through which was passed a current of the acid gas. The tube was heated by means of a combustion furnace and the chloride kept in the molten condition for two or three hours. By this means a perfectly white crystalline chloride of the composition CdCl₂ was obtained, free from water or oxychloride. The chloride and an excess of metal were placed in a long-necked flask of hard glass and after the displacement of the air by nitrogen, heated to the melting point of the chloride. The liquid chloride attained its maximum depth of color in a few minutes, nevertheless the heating was continued for five hours. When the temperature was allowed to rise much above the melting point of the chloride the red substance underwent decomposition and globules of metal collected upon the walls of the flask. For this reason no more heat was applied than was just necessary to keep the contents of the flask in a liquid condition. During the very gradual cooling of the flask it was shaken gently in order to facilitate the sinking of any metal, which might be mechanically retained by the chloride. On cooling, the solidified mass possesses a slightly greenish tint which disappeared when cold, the substance having then a grayish white color and a cleavage resembling that of talc or brucite. When examined under the microscope it was found to be perfectly homogeneous and free from metal. It gave no metallic streak when rubbed between agate surfaces. An analysis of the first preparation showed the following composition; Amount of chloride used .33541 gr. ” ” cadmium found .21559 ” ” ” chlorine ” .11943 ”