WATER ANALYSIS. 28-!> 



gallons — should invariably be rejected, and tbe bottle should be 

 rinsed with the water, filled nearly full, care being taken to 

 avoid undue aeration, and despatched to the laboratory as 

 quickly as possible. 



The following data should be obtained : — 



Colour. — This should be observed in a layer at least 12 inches 

 in length ; a yellowish-green tint is always suspicious, and 

 points to sewage contamination ; a brownish or brownish-yellow 

 indicates vegetable products, not necessarily harmful, but usually 

 undesirable. A nearly colourless water, with a faint blue or 

 bluish-green tinge, is shown by most good waters. 



Smell. — A small wide-necked bottle is half filled with the 

 water, which is warmed to about 60° C. (140° F.) ; the water is 

 shaken, the stopper removed, and the smell noted. Foul smells 

 show badly polluted waters ; a peculiar sweetish unpleasant 

 odour is often given by waters containing sewage. Few waters 

 are absolutely devoid of smell when tested thus ; for instance, 

 waters from the Oxford clay sometimes smell of petroleum, and 

 a smell of pines is not uncommon in wooded districts. 



Analytical Figures — Total Solids. — 250 c.c. (or 100 c.c.) are 

 evaporated in a weighed basin on the water-bath, and dried 

 to constant weight at 150° C. 



Loss on Ignition. — The residue is ignited over a very small 

 flame ; the smell of the vapours given off should be noted, as 

 polluted waters often give an unpleasant smell. Much blacken- 

 ing indicates a large amount of organic matter ; if nitrates are 

 abundant, red nitrous fumes may be observed. 



Chlorine. — 100 c.c. of the water are placed in a white porcelahi 

 basin, 1 c.c. of a 1 per cent, solution of pure potassium chromate 

 added, and silver nitrate (4'7887 grammes AgNO, per litre) run 

 in till a faint reddish colour is produced. The quantity of silver 

 nitrate required to give a similar tint with 100 c.c. of distilled 

 water is subtracted, and the difierence represents milligrammes 

 of chlorine, or parts of chlorine per 100,000 of water. 



Free and Albuminoid Ammonia. — 250 c.c. of water are 

 placed in a stoppered Wiirtz flask, to the delivery tube of which 

 a condenser is connected ; the condenser must be a good one, 

 and drawn out at the end, so that the diameter of the opening 

 does not exceed 1 millimetre. If the water be distinctly alkaline 

 to methyl orange, nothing need be added ; but if not, a little 

 freshly ignited sodium carbonate must be dropped in. A flame 

 is placed under the flask, and about 125 c.c. of the water distilled 

 and collected in a stoppered bottle. So soon as the flame is 

 placed under the flask, about 250 c.c. of distUled water are placed 

 in a flask and brought to the boU (or nearly so) ; the flask is 

 removed to the bench, 10 grammes of caustic soda added, and, 



