C. GENERAL PHYSICS. IVo 



pounds, and float in the fused mass, from which they could 

 be removed by filtration, made experiments in this direction, 

 which were so far successfid that the expected definite com- 

 pounds were found upon tlie filter, though the metallic fil- 

 trate was still very impure. The filter was made of quartz 

 sand, slag, etc., which was not wet by the molten metal. 

 Subsequently Curter, in attempting to apply this method to 

 the purification of Bohemian tin, as a commercial operation, 

 sought to use for a filter a material which should be wet by 

 the material to be purified, and at the same time should not 

 be dissolved by it. For this purpose he chose iron, which, 

 while it has a comparatively high temperature of fusion, has 

 a strong adhesion to tin, as is evident in the tinned iron plate 

 of commerce. Five hundred strips of tinned iron, as thin as 

 paper, about 0.6 of an inch long, and one fourth as broad, 

 were packed tightly in a square iron frame by the aid of 

 wedges, and the frame was then luted into a suitable open- 

 ing in the bottom of a graphite crucible. The tin, melted 

 in a second crucible, was allowed to cool until the separation 

 of fine crystals on the surface was noticed, the thickening 

 metallic mass being then poured into the filtering crucible, 

 when the still fluid pure metal passed through and a pasty 

 magma was left, in which iron, arsenic, and copper, concen- 

 trated to a great degree, were found combined with tin, 

 while the filtered tin proved to be almost chemically pure. 

 Fifty hundred-weight were purified in the crucible described. 

 Other forms and other materials for filters are suggested, 

 and other possible applications of the method, as in the sep- 

 aration of silver from lead containing the former metal. 18 

 A, October 6, 1876, 89. 



RESISTANCE OF THE ELECTRIC AEC. 



Ayrton and Perry, of the Imperial College of Engineers, 

 Tokio, Japan, desiring to determine theoretically the best ar- 

 rangement of cells for the production of the electric light, 

 have measured, as a necessary preliminary, the resistance 

 of the electric arc. The first experiments were made with 

 sixty new Grove cells the immersed platinum plate be- 

 ing thirteen, and the zinc plate twenty -five square inches 

 used with a Duboscq regulator. The known resistance 

 introduced consisted of many meters of bare copper -wire 



