CHEMICAL, CHANGES IK CALCIUM ARSENATE. 5 



the absorption of carbon dioxid, to a mixture of calcium carbonate 

 and dicalcium arsenate. It appeared that any increase in water- 

 soluble arsenic oxid in commercial calcium arsenate during storage, 

 therefore, would probably be the result of absorption of carbon 

 dioxid from the air. 



Accordingly, the authors decided to determine both carbon dioxid 

 and water-soluble arsenic oxid in all of the samples and subdivisions 

 at successive intervals. Soluble arsenic oxid was determined by the 

 official method of the Association of Official Agricultural Chemists 

 and carbon dioxid in a specially designed apparatus similar to that 

 described by W. H. Chapin.^ This method did not give extremely 

 accurate results but it is very rapid and sufiiciently accurate for this 

 work. Duplicate determinations seldom differed by more than 0.10 

 per cent. The fact that the results here vary more than this from 

 the curves drawn may be explained on the basis of uneven distribu- 

 tion of the material in the barrels and the method of sampling. The 

 subsamples were always thoroughly mixed, but they were taken from 

 the barrels by means of a trier, and, since carbon dioxid must work 

 its way through from the outside of the package, it is reasonable to 

 suppose that three or four trierfulls might not always remove a 

 strictly representative sample. Taken as a whole, however, the 

 results leave no doubt of the fact that carbon dioxid is absorbed 

 under certain conditions. 



The results of these examinations are reported in Table 2 and are 

 also shown graphically in Figures 1 to 46. The results on all the 

 samples obtained from one manufacturer are presented before those 

 of another are shown. Within each of these groups the order is as 

 follows: Unsubdivided sample kept at Washington; unsubdivided 

 sample kept at Tallulah; portion of subdivided sample kept at Tal- 

 lulah in original container; subdivided samples at Tallulah repre- 

 senting the different container tests. The last subdivisions are indi- 

 cated by figures following the letter designation as follows: 1, sheet- 

 iron drum; 2, unlined sugar barrel; 3, paper-lined sugar barrel; 4, 

 heavy hardwood barrel; 5, unlined veneer drum; 6, paper-lined 

 veneer drum. 



In every graph the arsenic oxid values are represented by a solid 

 line and the carbon dioxid values by a broken line. The curves 

 are drawn in what seemed to be the most probable positions without 

 attempting to calculate them. In order to bring out more clearly 

 the relationship between their variations, the ordinate scale of arsenic 

 oxid (AsjOg) values has been magnified to five times that of the car- 

 bon dioxid (CO2) values. The horizontal spaces on all the plots 

 indicate intervals of three months, the initial ordinate representing 

 in each case December 1, 1919. 



7 J. Ind. Eng. Chem., 10 (1918): 527. 



