V REVISION OF THE ATOMIC WEIGHT OF ARSENIC. 



83 



cylinder of sheet iron. Furthermore, since at this temperature even the hard 

 glass became very soft, it was found necessary to wrap the tube with asbestos 

 and closely wound iron wire for several inches at the piont where fusion took 

 place. This also served to distribute the heat more evenly and to prevent the 

 tube from cracking during the experiment. 



Just before the salt was fused a carefully weighed U-tube containing resub- 

 limed phosphorus pentoxide was attached to the end of the tube, and beyond 

 this was joined another similar tube which served as a protection against any 

 moisture which might diffuse back into the weighed tube from the outside air. 

 These phosphorus pentoxide tubes were provided with one way ground glass 

 stopcocks lubricated with Ramsay desiccator grease. 



The salt was heated for 25 minutes with the hottest flame of the blast lamp, 

 being then completely fused, and was further heated for 35 minutes at a con- 

 siderably lower temperature in order to make certain that all moisture was car- 

 ried into the absorption tube by the current of air. Finally the phosphorus 

 pentoxide tube was reweighed. 



The pentoxide tube was weighed by substitution with the use of a counter- 

 poise of the same size and weight filled with glass beads. Before being weighed 

 both tubes were carefully wiped with a damp cloth and were allowed to stand 

 near the balance case for one hour. One stopcock in each tube was opened im- 

 mediately before the tube was hung upon the balance, in order to insure equi- 

 librium between the internal and external air pressure. The stopcock of the 

 counterpoise was left open during the weighing. Owing to the considerable 

 length of time required for the tubes to come to equilibrium on the balance, it 

 was considered unsafe to leave the stopcock of the pentoxide tube open during 

 the weighing. As a check on the first weight of the pentoxide tube one stopcock 

 was opened and closed and its weight determined a second time. Ordinarily no 

 change in weight was observed. 



Since it seemed possible that the hard-glass tube itself, when heated nearly 

 to fusion, might give off traces of water vapor, two blank determinations were 

 made by heating the empty hard-glass tube in exactly the same way as in the 

 water determinations. These determinations showed a gain in weight of the 

 pentoxide tube of 0.00022 and 0.00037 gm. respectively, the average being 



