162 IRVINE H. PAGE. 



Removal of the calcium as carbonate is unsatisfactory because 

 of the very marked tendency of CaCOs to form supersaturated 

 solutions. Gentle heating of the solution will, however, bring 

 the carbonate down. 



Oxalate precipitation seems quite satisfactory followed by 

 phosphate for the magnesium precipitation, but here two steps 

 are required and there is slight danger of carrying some oxalate 

 along until the thiosulphate titration. 



Using the above slight modification we have been able to get 

 most accurate check analyses of known solutions of NaCl and 

 CaCl 2 . Kramer and Gittleman's method is a most useful and 

 unique technique and seems to have a wide applicability for the 

 direct determination of sodium. 



As in the case of potassium we feel that the direct determination 

 of sodium as compared with the determination by difference, as 

 the combined sulphates, etc., has certain advantages. 



Chlorine was estimated by the AgNO 3 -KCrO 4 titration of Fr. 

 Mohr. 



Sulphate was precipitated as BaSO 4 , ignited and weighed. 



Phosphate was determined in 5 cc. samples of water by pre- 

 liminary oxidation of organic matter with H 2 SO 4 and H 2 O2 as 

 described in more detail by Page (3), followed by the Benedict- 

 Theis (4) technique. This technique consists essentially in the 

 colorimetric estimation of a stable blue color developed by 

 molybdic acid, hydroquinine and sodium bisulphite in the 

 presence of a large excess of H 2 SO 4 . The following table gives 

 the average of 4 analyses taken during the month of August 1926 

 from the tank water of the laboratory. 



TABLE I. 

 ANALYSIS OF TANK WATER OF MARINE BIOLOGICAL LABORATORY. 



Grams per 1000 cc. sea water at 20 degrees C. 

 Specific Gravity = 1.0180 (20 degrees C.). 



J&Stimoles. 



Sodium 8.80 0.3826 



Potassium 0.412 0.0105 



Calcium 0.428 0.0107- 



Magnesium 1.3004 0.0534 



Chlorine 18.350 0^5912 



Sulphate 2.615 0.0272 



Phosphate 0.002 



