6 The Philippine Journal of Science i9ia 



Calcium was formerly determined turbidimetrically by the 

 United States Geological Survey method, but this has been 

 abandoned because of its inaccuracy. 



The qualitative field test for calcium is made by adding enough 

 ammonia to some of the water in a test tube or bottle to make 

 it alkaline to litmus and adding some ammonium oxalate. 



Total solids are determined qualitatively by evaporating 50 

 cubic centimeters of the water in a porcelain casserole to dryness 

 over an alcohol lamp. The solid content is reported merely 

 as "very small," "moderate," "large," etc. The residue is then 

 ignited, and any change in "appearance on ignition" is noted. 

 This may be a browning or blackening due to organic matter, 

 or a deep red-brown coloration due to the oxidation of con- 

 siderable amounts of iron present. The last is of value as a 

 confirmatory test for large amounts of iron. 



Odor is reported, wherever possible, in such a way that both 

 the derivation and the relative amount are indicated, for in- 

 stance, "very slightly sulphuretted," "strongly acid." 



Alkalinity, acidity, chlorides, normal carbonates, and total 

 hardness are determined by the use of tablets, as outlined by 

 Leighton. In brief, this method consists of the use of pellets 

 containing known amounts of reagents, instead of standard solu- 

 tions. The titrations are performed in a small (100 to 150 

 cubic centimeters) , heavily glazed porcelain mortar, a pestle being 

 used to crush the pellets and to stir the liquid. The volume 

 of water used for a titration is conveniently measured from a 

 tall, 100 cubic centimeter graduated cylinder, provided with a 

 double scale, so that both the water withdrawn and the volume 

 remaining can be directly read. What are practically duplicate 

 determinations can be made very rapidly in the following 

 manner : 



A few pellets are crushed in the mortar, and water is added 

 from the cylinder till the end point is reached. The volume 

 used is noted. Several more pellets — preferably the same num- 

 ber as before — are added, followed by water from the cylinder, 

 until the second end point is obtained. In this way not only 

 is it possible to secure more accurate results by taking the 

 mean of the two values obtained than by making a single de- 

 termination, but in addition any gross error that may arise 

 from an unclean mortar, contaminated indicator, or defective 

 tablet can be detected and corrected. 



The following reagents are used in tablets in the various 

 determinations : 



Sodium acid sulphate for alkalinity and normal carbonates; 



