263 



Water and made ready for use. After introducing a definite (quantity of 

 the standard sol. of potass, permanganate from a burette, graduated to j^^^ 

 cc, and a deiinite amount of the acid or alkali, the tubes were sealed, then 

 cooled gradually to anneal them. When cold, the tubes were well shaken 

 to mix the fluid and then introduced into a copper bath 4 in. wide and 15 

 in. long. When working at 100°C. pure water was used to fill the bath ; 

 for higher temperatures aniline was used. We found aniline preferable to 

 paraffin because the heat diffused through it more readily and uniformly. 

 Many conditions were tried to determine if possible all the factors which 

 enter into the problem. From the data obtained there seem to be four 

 primary factors which influence the spontaneous decomposition of perman- 

 ganic acid. 1. The quantity and kind of acid or alkali used. 2. Time of 

 heating. 3. Temperature. 4. Dilution. This is shown by the following 

 data : I. Quantity and kind of acid used : 



By using 4-6-8cc of acid with constant time, temp, and dilution, we find 

 the per cent, decomposition is almost exactly proportional to the quantity 

 of acid used. It may be represented by the curve A. 



By using constant acid and dilution and temp, the per cent, decomposition 

 is approximately proportional to the time of heating. Its curve is therefore 

 the same as that for the acid decomposition B. with XaOH. 



