July 6, 1900.] 



SCIENCE. 



27 



tetrachlorid, and as the chlorin seems to 

 have been generally estimated by loss, 

 analysis would reveal no discrepancy, but 

 in one case at least, the chlorin was directly 

 determined, and these figures can be ac- 

 counted for only on the supposition that in 

 this case it was a tetrachlorid which was 

 analyzed ; so that it would seem possible 

 that Claus actually formed the tetrachlorid, 

 although he did not distinguish it from the 

 nitrosochlorid. Even now the conditions 

 of formation of the tetrachlorid are obscure, 

 and not less so is the cause of a phenomenon, 

 noticed first by Claus, and since his day 

 used as a test for the detection of ruthen- 

 ium, and which is familiar to all of you 

 who have experimented at all with this 

 metal. I refer to the beautiful indigo-blue 

 color assumed by the solutions of ruthen- 

 ium trichlorid when hydrogen sulfid is led 

 into them. Since, at the same time, sulphur 

 is precipitated, and since the trichlorid 

 also assumes [a blue color on treatment 

 with metallic zinc,* it was assumed by 

 Claus that reduction takes place and hence 

 that the solution contains ruthenium bi- 

 chlorid, RuClj. When ruthenium is heated 

 in a current of mixed chlorin and carbon 

 monoxid, it increases many times in vol- 

 ume and there is formed an anhydrous tri- 

 chlorid. This is insoluble in water and in 

 strong alcohol, but dissolves with consider- 

 able readiness in dilute alcohol to a similar 

 deep blue solution. Joly succeeded in dis- 

 tilling olf the alcohol and water from this 

 solution, in a vacuum, and obtained a blue 

 deliquescent substance which he considered 

 to be an oxychlorid, EuOHClj. I have also 

 formed this blue solution by electrolytic 

 action, and while it seems to be formed by 

 a reducing action, this is not perfectly clear. 

 Considerable work upon this solution, how- 

 ever, leads me to agree with Claus that it is 



* It has already been mentioned that Vauquelin 

 had noticed this "blue color, but not knowing of ru- 

 thenium had attributed it to osmium. 



probably a lower chlorid of ruthenium, but 

 it has not been proved. I have dwelt per- 

 haps unduly upon these compounds for the 

 purpose of showing the obscurity in which 

 even such seemingly simple points are en- 

 veloped, for it well illustrates how much 

 work must yet be done before we acquire 

 any adequate knowledge of the nature of 

 even the commoner compounds and reac- 

 tions of these elements. 



Of the simple salts of oxy-acids few are 

 known of any metals of this group except 

 the lower series, iron, cobalt and nickel; 

 a single sulfate of rhodium, one of palla- 

 dium, and perhaps a double sulfate of plati- 

 num, a chromate of iridium, a basic car- 

 bonate of palladium, two or three nitrates, a 

 phosphate of rhodium, and a hypophosphite 

 of platinum; such is practically the whole 

 list. The platinum metals have little ten- 

 dency to form crystalline salts with oxy- 

 acids, and many such salts are unquestion- 

 ably incapable of existence, but in many 

 cases at least the difficulty is our ignorance 

 of the condition of formation of such salts. 

 And herein, I may say, is one of the most 

 marked differences between investigation 

 in organic and in inorganic chemistry. In 

 the former the field has been so thoroughly 

 studied that the conditions of reaction are 

 often well known and the course of a reac- 

 tion can be foretold with considerable cer- 

 tainty; in inorganic chemistry the work is 

 like exploration in an almost wholly un- 

 known land. "We know neither the possi- 

 bility of existence of conjectured com- 

 pounds, nor the conditions under which 

 alone such formation or existence is possi- 

 ble. For this reason inorganic research is 

 slower and far more apt to be fruitless. No 

 better example of this can be cited than the 

 fact already referred to that Professor Joly, 

 as well as myself, exhausted every method 

 which occurred to us for the formation of 

 the tetrachlorid of ruthenium, and failed in 

 our efforts by missing just the pi-oper con- 



