108 



tures; and to determine, if possible, the lOAvest temperature at which any 

 chloride would be formed. The temperatures varied in the different 

 experiments between 90° and 200°, and the time of lieating from one 

 hour to 21 days. Chlorides were found in each of the 20 experiments, 

 and the amount varied somewhat regularly with the increase of temper- 

 ature and the time of heating. At 90°-93°, the lowest temperature used, 

 the amount of chloride formed in 14 days was .22 per cent, of that theo- 

 retically possible. 



In order to show whether the pure chlorate would decompose at all 

 under these conditions some of it was heated in the same manner as 

 that described above. The heating was continued for nine days at 106°- 

 109°. But not a trace of a chloride was produced. 



It is interesting to note that decomposition begins 200° below that 

 at which it is sufficiently rapid to be easily observed. But this is in 

 line with the modern idea that the velocity of an action is a function 

 of the temperature. And this observation has its parallel in the fact 

 that 200° below its ignition point hydrogen combines with oxygen in 

 quantities sufficient to be determined. 



It has been found also that mixtures of manganese dioxide and 

 potassium perchlorate produce oxygen at a temperature much lower than 

 that necessary to decompose the perchlorate alone. The amount of 

 oxygen is quite appreciable at 310°, but does not become rapid at 360° — 

 a temperature below that at which the perchlorate begins to evolve 

 oxygen. 



In order to compare the action of other catalytic agents at low tem- 

 peratures mixtures of potassium chlorate and platinum black were 

 heated at two temperatures; one sample for 6 days at l-t5°-150°, the 

 other for 7 days at 95°-100°. Both tubes lost in weight and both gave 

 evidence of considerable amoimts of chloride produced. I hope soon to 

 get results at higher tempera tiu-es. But at these temperatures manganese 

 dioxide and platinum l)lack are almost identical in their effect on the 

 decomposition of potassimn chlorate. 



In the near futin-e the study of the action of other oxides at low tem- 

 peratures will be undertaken in order to get comparative results. 



At the beginning of the investigation on catalysis it was believed 

 that many of the actions would prove to be of a purely chemical nature. 

 At the present time there is no evidence that such is the case; but 

 rather that we are dealing with cases of true contact action. 



