46 PUBLIC HEALTH CHEMISTRY 



20KOH + 10I 2 . The liquid turns a yellow colour from 

 the iodine set free. The quantity of iodine liberated is 

 strictly proportional to the amount of unused perman- 

 ganate. It remains, therefore, to estimate the amount of 

 iodine set free, which measures the amount of oxygen 

 unused, and this deducted from the amount known to 

 have been added, gives the amount absorbed. This is 

 done by titrating the yellow solution with the thiosulphate 

 solution until the yellow colour is nearly gone, and then 

 adding 1 c.c. of starch solution to give a more distinct end 

 reaction. The titration is finished when the blue is just 

 gone : 



1 2 + 2Na 2 S 2 3 = 2NaI -f- Na 2 S 4 6 . 



Both the sample and the control are treated thus. In the 

 control presumably no oxidation takes place, so that the 

 number of c.c. of thiosulphate solution required for it, is a 

 measure of the iodine liberated by all the oxygen free to 

 cause oxidation. The amount of thiosulphate solution used 

 for the sample measures the unused oxygen, and the differ- 

 ence between these two numbers gives the proportion of 

 oxygen used up. For example, say that 10 c.c. only of per- 

 manganate were required to be used, and that after adding 

 KI solution the control took 40 c.c. of thiosulphate solution 

 to decolorize, and the sample took 30 c.c. ; then 40 c.c. 

 of thiosulphate measure 1 mgr. of oxygen, and 40 c.c. 

 30 c.c. = 10 c.c. measure 10/40 = 0-25 mgr. O, and 

 this is the quantity absorbed by the 250 c.c. of sample 

 taken. This multiplied by 0-4 gives o-i mgr. O absorbed 

 per 100 c.c, or o-i part per 100,000. 



Waters of great organic purity will not consume more 

 than 0-05 part of oxygen per 100,000 in fifteen minutes 

 at 8o F., and if the amount absorbed exceeds o-i part in 

 fifteen minutes, the sample may be considered of doubtful 

 purity. After four hours' exposure, an absorption of more 

 than 0-3 part must be regarded with suspicion. 



Ferrous salts, nitrites, and sulphuretted hydrogen, if 

 present, vitiate the test. 



Kjeldahl's Process for the determination of organic 

 nitrogen is performed only in very polluted waters. The 

 process is described under Sewage and Sewage Effluents. 



