82 RESEARCHES UPON ATOMIC WEIGHTS. 



During this time the quartz tube containing the salt showed no perceptioie 

 change in weight. The third determination also was made with a sample of 

 salt which had been exposed to bright light for 3 weeks in a dry state. In the 

 second determination the salt was exposed to light under water for one week. 

 Two facts show that the presence of the small proportion of the residue in the 

 arsenate could have had no important effect upon the results. In the first place, 

 the formation of the insoluble matter under the influence of light is not attended 

 by change in weight. In the second place, the silver content of the residue is 

 very near that of silver arsenate. Nevertheless, care was taken to protect the 

 arsenate as far as possible from exposure to light. 



DETERMINATION OF MOISTURE IN DRIED SILVER ARSENATE. 



T. W. Richards ^ and others have already drawn attention to the fact that it 

 is not possible, without fusion, to dry completely a substance formed in aqueous 

 solution, owing to the mechanical retention of liquid in pockets within the 

 solid. In the case of silver arsenate, although it is possible to fuse the salt, the 

 temperature necessary is so high that decomposition of the salt takes place to 

 some extent. Hence the loss in weight on fusion can not be used as a true 

 measure of the water content of the salt. Since decomposition of the salt could 

 produce only easily condensible substances and oxygen, the difficulty was over- 

 come in the present instance by fusing weighed quantities of the salt in a current 

 of pure dry air and collecting the water vapor in a weighed phosphorus pentox- 

 ide tube. Of course great pains were taken to treat the salt used in the water 

 determinations in exactly the same way as that used in the analyses for silver. 



The procedure was as follows: A sample of salt very nearly as pure as that 

 used in the silver analyses was weighed out in a copper boat which had been 

 previously cleaned and ignited in the blast lamp to remove organic matter. 

 The boat was placed in a hard-glass tube and was heated for between 7 and 8 

 hours at 250° C, in a current of dry air. In these experiments, before passing 

 through the drying towers the air had first been passed over hot copper oxide 

 in order to oxidize any organic matter it might contain. Furthermore, the con- 

 centrated sulphuric acid in the drying towers had been heated with a small 

 quantity of potassium dichromate. One end of the hard-glass tube was con- 

 nected to the apparatus for supplying pure air, by means of a well-fitting ground 

 joint upon which no lubricant was used. The other end was sealed to a small 

 hard-glass tube which was surrounded with a damp cloth during the fusion of 

 the salt in order to facilitate condensation of any silver or arsenic compounds 

 vaporized from the salt. As a matter of fact, very little sublimation actually 

 took place. 



In order to fuse silver arsenate within the hard-glass tube it was necessary 

 to use the hottest flame of the blast lamp, the tube being covered with a semi- 



* Zeit. physik. Chem., 46, 194 (1903). 



