240 



Prof. T. Carnelley and Mr. W. Mackie. 



sample of air collected by pumping out the contained air with a small 

 bellows, and allowing the air to be examined to flow in. 50 c.c. of the 

 standard permanganate are next run into the jar, which is then tightly 

 stoppered and well shaken up for at least five minutes. 25 c.c. of the 

 permanganate are afterwards withdrawn by a pipette and placed in a 

 glass cylinder holding about 200 c.c, 25 c.c. of the standard solution 

 being placed in a similar cylinder for comparison. Both are next 

 diluted up to about 150 c.c. with distilled water, and allowed to stand 

 for ten minutes, after which the tints in the two cylinders are com- 

 pared. Standard solution is then run in from a burette, until the 

 tints in both cylinders are of the same intensity ; usually from \ to 

 6 c.c. are required. 



The amount of solution added from the burette is a measure of the 

 bleaching effected by the known volume of air on half the perman- 

 ganate employed. This multiplied by 2 gives the total bleaching. 



The results may be expressed either in terms of the number of c.c. 

 "N 



of the Jqqq bleached by 1 litre of air,'or, as we prefer, by the number 



of volumes of oxygen required to oxidise the organic matter in, say, 

 1,000,000 volumes of air ; e.g., 25 c.c. of solution from a 35 litre jar, 

 in which 50 c.c. had been used, required 3 c.c. of the permanganate to 

 bring it up to the standard, or the whole 50 c.c. would have required 

 3x2 = 6 c.c. This represents the number of c.c. of standard perman- 



ganate bleached by 3500—50 = 3450 c.c. of air; consequently 



l - 74 c.c. is the bleaching effected by 1 litre of air. But 1 c.c. of 

 KMnO 4 =0-0000056 litre of oxygen; . ■ . T74 c.c. KMnO 4 =0"0000056 

 X 1 , 74=0 , 0000097 litre of oxygen is required to oxidise the organic 

 matter in 1 litre of air, or 9*7 volumes of oxygen to oxidise the organic 

 matter in 1,000,000 volumes of air. 



Correction for temperature was not considered necessary, as it falls 

 within the limits of experimental error. It requires about twenty 

 minutes to collect the samples and complete the analysis. Scrupulous 

 cleanliness is, of course, necessary in all the operations. 



We have examined many hundreds of samples of air by this method, 

 and in this large experience of it have found it to be a very con- 

 venient and ready process. We believe it to be as accurate as is 

 possible in the present state of the knowledge of this subject. Dupli- 

 cate analyses of the same air give very concordant results, as evidenced 

 by the examples given below. We are nevertheless fully conscious 

 that objections may be taken to the method on the following grounds : 

 — (1.) That it does not directly estimate the organic matter, but only 

 measures the amount of oxygen required to oxidise either the whole, 

 or more probably only a portion of it. (2.) That the permanganate 

 acts upon various matters in the air, besides the organic matter, 



