1880.] On the Essential Properties, Sfc, of Beryllium. 



m 



and colour resembled needles of polished steel. The metal was, as 

 might be expected, not absolutely pnre. The analysis was somewhat 

 difficult, the mean result of a number of accordant determinations 

 being : — 



Be (metallic) 86'94 per cent. 



Be 3 3 9-99 



Fe 2-08 



Si0 2 0-99 



100-00 



We next determined its specific heat by the method of Bunsen (ice- 

 calorimeter) . We here met with quite unexpected difficulties ; but, 

 having given up the original arrangement of the experiment described 

 by Bunsen* as impracticable, we, by means of an arrangement 

 similar to that recommended by Schulier and Wartha,f obtained the 

 following results : — J 



Specific heat of Be=0'4084 between 0° and 100°. 

 Atomic heat of Be = 5'64 [Be=13'8]. 



These results, not in accordance either with those recently obtained 

 by Mr. Reynolds or the periodic law of Mendeleeff, were not accepted 

 without hesitation by chemists. Notwithstanding that, the editor 

 of the new edition of Gmelin's " Handbuch d. Chemie," Professor 

 Kraut, altered the formulas of beryllia and its compounds in the part of 

 the great encyclopaedia of chemistry then passing through the press, 

 several other chemists publicly or privately commented on our work, 

 and urged us to pursue our inquiries further. Among the objections 

 thus made we deem the following most worthy of discussion. 



Mr. Lothar Meyer § hints that the equivalent of beryllium may be 

 wrong, and suggests a revision of that number. If it should be found 

 lower than 4*0, instead of higher, beryllium may still be considered a 

 three-atomic element, without upsetting the periodic law. In that 

 case it would only be necessary to interpose a new group of metals 

 between the trivalent and quadrivalent elements. We had no reason 

 to doubt the accuracy of the old number ( = 4'6), which we had found 

 and verified by many analyses, but we redeemed our promise to Mr. 

 Lothar Meyer, 1 1 and undertook the following determinations of the 

 atomic weight of beryllium. 



In order to determine this value with the utmost accuracy, we 

 thought it safest to choose the simplest possible method, viz., the 



* " Pogg. Ann.," cxli, p. 1. 



f " Pogg. Ann.," [2], ii, p. 359. 



% Obs. — Allowance is made for the impurities of the metal. The specific heat of 

 Be 2 3 was found = 02471 between 0—100° C. 



§ "Ber. d. Deutsch. Chem. Gresellsch.," xi, p. 576. 

 || Ibid., p. 906. 



E 2 



