SANGER AND GIBSON. — DETERMINATION OF ANTIMONY. 731 



through the upper and lower sides of the frame. As the antimony 

 mirrors are aflected by moist air, and in the course of time by lii^ht, 

 we take the precaution to seal the tubes and to keep the set of stand- 

 ards in the dark. The deterioration of the standards with time is not 

 a serious objection to the method, since a fresh set can readily be made 

 when desired. ^^ 



The amounts of antimony used by us in the set of standards shown 

 in the plate are, in milligrams of antimonious oxide (SbjOs) as follows : 

 ().(>(>5; 0.01; 0.015; 0.02; 0.025; 0.03; 0.035; 0.04; 0.045; 0.05; 

 O.OG ; 0.07. Above 0.07 mgr. it is of no advantage to make a mirror, 

 as the recognition of the dilferences in the higher mirrors is difficult. 

 These mirrors may be used as standards when viewed by reflected 

 light, as shown in the upper set of the plate, but it is much better to 

 use transmitted light. This is best arranged by mounting the frame 

 in a box, similar in form to the fluoroscope, which may be of wood or 

 metal. This box, about 25 cm. long and open at each end, is larger at 

 the bottom than at the top, and carries a rabbet at the bottom through 

 which the frame can be slipped. It is provided at the top with means 

 of shading the eyes against all light except that which comes through 

 the set of mirrors. If the box is held against a white surface, the mir- 

 rors gain greatly in sharpness, and smaller differences may be in this 

 way detected when comparing mirrors obtained in an analysis. The 

 lower part of the plate shows the set of standards when viewed by this 

 device. 



Analysis of Solutions containing Antimony. 



To test the availability of the standard mirrors, a series of seven 

 analyses was made of solutions in which the amounts of antimony 

 were unknown to the analyst. The solutions were weighed to the 

 second decimal place in a side-neck test tube of about 30 c.c. capacity. 

 After the apparatus had been running for about ten minutes, a few 

 drops of the solution were added to the reduction flask. If no mirror 

 appeared within ten minutes, a larger portion of the solution was added, 

 and if this again gave no mirror, the addition was continued. After 

 the appearance of the mirror, the run was continued for thirty or forty 

 minutes, until there was no probability of further deposit. By re- 

 weighing the test tube after the addition of the portion or portions 

 which produced the mirror, the amount of solution taken was deter- 

 mined. The mirror obtained from this amount was then compared 



" The suggestion of Panzer (Clieni. Centralbl., 74 (1),821 (1903), to seal stand- 

 ard arsenic mirrors with pliosphorus pentoxide could probably be applied with 

 advantage to the set of antimony mirrors. 



