486 



BELL SYSTEM TECHNICAL JOURNAL 



Suppose, for example, it is desired to determine traces of heavy 

 metals such as copper and nickel, in a certain plant ash. Because of 

 the disproportionality between the quantities of the elements sought 

 and the size of the apparatus, the errors introduced are large when 

 ordinary methods are employed. Mechanical losses incurred in the 

 many manipulations and transfers of material, over-dilution with 

 resulting incomplete precipitation, contamination both by dust and 

 by substances dissolved from the glass are almost unavoidable. Since 

 the heavy metals represent only a few hundredths of a per cent of the 

 ash, it is obvious that a large sample, perhaps twenty-five grams, is 



Fig. 1 — A corner of the microanalytical laboratory. 



required, if an ordinary balance sensitive to 0.1 mg. is used. It would 

 therefore be necessary to start with a kilogram or more of the fresh 

 plant to obtain results which are sufficiently precise. 



Other disadvantages of the usual scale of operations are great time 

 consumption, explosion hazards, and costliness of chemicals and 

 apparatus. On a greatly reduced scale, these difficulties frequently 

 tend to vanish. For example, the precipitation of the sulphides of 

 hea\y metals is ordinarily avoided wherever possible in quantitative 

 analysis. This is because they are slimy and difficult to filter as 

 ordinarily precipitated with hydrogen sulphide. To filter and com- 

 pletely wash a gram of lead sulphide might be a matter of several 



