62 ABSOLUTE ATOMIC WEIGHT. 



Here the atomic ratio is r = o.92 593. 



To secure ready comparisons, we shall always print these 

 ratios by leaving an n - space between the second and the 

 third decimal. 



In this manner the five-place decimal becomes easily 

 readable, the first tivo decimals represent the per cent., 

 the last three decimals represent the tenth, hundredth and 

 thousandth of per cent. 



We know of no chemical work where the sixth decimal is 

 actually determined, or significant. Hence we never shall 

 give more than five decimals. 



In the first determination by Erdmann and Marchand, 

 they obtained 75.9347 grammes of metallic mercury from 

 82.0079 grammes of mercuric oxide; accordingly, we obtain 

 the analytical ratio, by dividing the first by the second 

 weight, a = 0.92 594. 



Eviden^ 1 v, in this their first determination, the analytical 

 excess is / high according to our mode of expression ; for 

 the last decimal of the analytical ratio is 4, while in the 

 atomic ratio it is 3. 



In this manner, every statement of fact is reduced to the 

 simplest possible form, and easily grasped by the mind. 



Extremes and Range. 



We shall also have to specially consider the extremes and 

 the range of the analytical ratios of any series. We shall 

 invariably give the highest first, then a dash as minus sign, 

 followed by the lo-west value observed. A semicolon followed 

 by the range completes the statement. Having to give a 

 multitude of results, brevity and uniformity of representa- 

 tion become very important. 



Since in good series of determinations there are no 

 changes in the first two decimals, it would be absurd to 

 incumber the record therewith; hence we only print the 

 last three decimals of the extremes. 



In the case of mercuric oxide, Erdmann and Marchand 

 found the highest analytical ratio 0.92 606 and the lo-west 

 0.92 594. 



Hence we record simply thus: Extr. 606 594; 12. 



