SANGER AND GIBSON. — DETERMINATION OF ANTliMONY. 727 



the same time. The generator which we have found most convenient 

 is one proposed by Richards, and is similar to the forms described by 

 him some years ago.^^ Any form of constant generator will answer 

 which provides for withdrawal of the spent acid without disconnection. 

 The zinc in the generator is conveniently sensitized, according to the 

 suggestion of Gooch,!^ by brief treatment with a solution of cupric 

 sulphate and subsequent washing. The acid is sulphuric, at a dilution 

 of one to eight. The hydrogen from our generator shows no arsenic 

 or antimony when run for hours at a time. 



As the hydrogen wnth which the antimony hydride is heated must 

 contain no hydrogen sulphide, as hereafter shown, the hydrogen from 

 the generator is passed through a ten per cent solution of cupric sul- 

 phate contained in an Allihn or other suitable washing bottle. From 

 this, the hydrogen, which needs no further purification, passes to a 

 Y-tube with two glass stopcocks. To these stopcocks are attached 

 the reduction flasks, which are wide-mouth bottles of 60 to 75 c.c. 

 capacity. These are fitted with a pure rubber stopper with three holes. 

 Through one hole passes, to the bottom of the bottle, a right-angle 

 tube connected with the stopcock ; through the second a tube pass- 

 ing nearly to the bottom of the bottle and extending 3 to 5 cm. 

 above the stopper. The upper end of this tube is open, the loAver 

 somewhat constricted. In the tube is placed a small funnel, blown 

 from narrow tubing, through which the solution to be reduced is added. 

 The third hole in the stopper carries the right-angle deliver)^ tube for 

 the hydrogen which passes just below the stopper. On its upper end is 

 a rubber stopper over which is placed a 15 cm. (total length) tube filled 

 with calcic chloride in fused sticks. We have found it convenient to 

 half fill the tube with calcic chloride, and then to introduce a second, 

 smaller tube filled with the chloride. By this arrangement the rear 

 portion of the chloride, which soon becomes moist, may be frequently 

 and conveniently renewed without disturbing the rest. 



To the calcic chloride tube is connected the hard glass reduction 

 tube drawn out to a straight capillary and ending in a capillary point. 

 This tube is supported throughout its length by three adjustable brass 

 hooks (No. 6 gauge, 4.1 mm.).' The middle hook, which has a shank 

 10 cm. long, is fastened to a stand by the ordinary double clamp. On 

 the top of the shank and about two thirds the distance from the 

 clamped end are soldered two ordinary screw connectors at right angles 

 to the shank. The other hooks, which are somewhat longer than the 

 first, are bent at right angles in the plane of the table and their shanks 



." Amer. Clicm. Jour., 20. 180 (1808). 

 " Amer. Jour. Sc-i., (3) 48, 202 (1894). 



