140 PROCEEDINGS OF SECTION B. 



•general use in water analysis, A small room is devoted exclusively 

 to the work, so as to avoid the fumes of ammonia and hydro- 

 chloric acid prevalent in the general laboratoiy. 



The total dissolved matter are estimated by evaporating 500 

 c.c. of the water in a platinum dish on an iron hot-plate, dust, 

 etc., being prevented from entering by covering the dish with a 

 large funnel. The final drying is effected in a water bath at 

 100°C. 



Niti'ogen existing as ammonia compounds is determined by 

 placing 500 c.c. of the water, with a few drops of sodic carbonate, 

 in a retort connected with a Liebig condenser, the same having 

 been previously freed from ammonia by the steam from distilled 

 water; 150 c.c. are distilled over, and the ammonia determined 

 by Nessler's well known colorimetric test. The Nessler solution 

 is frequently sensitised with mercuric chloride, so as to give an 

 indication with .2 c.c. of standard ammonic chloride solution, 

 equivalent to .002 milligrammes of nitrogen. The distilled water 

 used in tlie testing is specially prepared in a large copper still, by 

 distilling over two-fifths and collecting the next fifth part which 

 is found to be practically free from ammonia. 



The albuminoid nitrogen is determined in the water left after 

 the removal of the ammonia, by distilling with 50 c.c. of an 

 alkaline solution of potassic permanganate, which has been 

 previously boiled, collecting the first 150 c.c. and applying 

 Nessler's test as before. 



Nitric acid is estimated by converting it into ammonia by 

 means of a copper-zinc couple and determining the ammonia 

 evolved with Nessler's test, 250 c.c. of the water being used for 

 the purpose. The copper-zinc couple consists of a cylinder of 

 perforated zinc 3 x f inch, which has been immersed in a five 

 per cent, solution of cupric sulphate until a firm coating of 

 copper is obtained. The action is allowed to continue for twelve 

 hours at a temperature of about 80° F. 



For testing the presence of nitrous acid an acid solution of 

 metaphenyldiamine is employed. 



Chlorine is determined volumetrically by a standard solution 

 of silver nitrate, 1 c.c. of which is equivalent to .005 grms. of 

 chlorine, potassic chromate being used as an indicator. The 

 water used, 250 c.c, is reduced by evaporation to about 50 c.c. 



Sulphuric anhydride is estimated in the w^ater, concentrated to 

 about 50 c.c, by precipitating with baric chloride in the presence 

 of hydrchloric acid. The quantity employed varies with the 

 amount of the rainfall, 500 c.c. being the minimum. In oi^der to 

 avoid error with the small precipitates obtained, acid extracted 

 filter paper only is employed. 



For ascertaining the reaction of the water rosolic acid is used. 



Rain being formed by the condensation of the moisture 

 evaporated from the suiface of the earth, would, at the time that 



