134 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 54 



Still another method for fixing the atomic weight of sulphur was 

 adopted by Eichards and Jones.' Silver sulphate was converted into 

 chloride by heating in a current of pure, dry hydrochloric acid gas. The 

 data obtained, with vacuum weights, were as follows: 



0004 



In recent years attempts have been made to deduce the atomic weight 

 of sulphur from the density of sulphur dioxide, for which there are 

 several modem determinations. Leduc,^ in a series of measurements, 

 found the density to range between 2.2638 and 2.2641 ; in mean, 2.2639. 

 If we take these three values for the entire series the probable error 

 of the mean becomes ±.000067. For oxygen Leduc's density figures give 

 1.10514, ± .0000321. Hence the crude density ratio 0, : SO2 : : 32 : 65.553, 

 ±.0020. From these figures, with the aid of the compressibilities and 

 critical constants of the gases, Leduc' determines SO, = 64.056. From 

 the density of HoS he finds a molecular weight of 34.071. Hence S = 

 32.056. By the method of limiting densities, D. Berthelot,' from Leduc's 

 figures, finds S = 32.050. 



Jaquerod and Pintza' give for the weight of a normal litre of SO,, 

 2.92664 grammes. For the corresponding volume of oxygen their weight 

 is 1.4292 grammes. Hence the crude molecular ratio 32 : 65.528. Since 

 individual determinations are not given, the probable error of this ratio 

 cannot be calculated, and I shall assign it equal weight with Leduc's 

 determinations. Jaquerod and Pintza also measured the density of SOj 

 at pressures lower than the normal, namely, at 570 and 380 mm. Then 

 extrapolating to zero pressure they deduce SO, = 64.01, and S = 32.01. 



' Publ. Carnegie Inst., Washington, No. 69, p. 69. 1907. Richards and Jones give a thorough 

 criticism of the previous work on sulphur. 



^Compt. Rend., 117, 219. 1893. 



s Ann. Chim. Phys. (7), 15, 94. 1898. Leduc here puts the density of O = 1.1052. See also 

 ante, p. 33. 



*Journ. Ph3'siquc (3), 8, 26.3. 1899. 



' Compt. Rend., 139, 129. 1904. 



