472 REPORTS ON THE STATE OP SCIENCE. 



and Emich, 1 which yielded very discordant results for the ignition tem- 

 peratures of mixtures of the commoner inflammable gases with oxygen, 

 doubtless suffered from the large amount of flameless combustion which 

 occurred before the temperature of the mixture as a whole had been 

 raised to the true ignition-point. 



Quite recently, however, H. B. Dixon and H. F. Coward, 2 using an 

 apparatus in which the combustible gas and air or oxygen were separately 

 heated to the temperature of the enclosure before being allowed to mix, 

 have succeeded in fixing, within narrow limits, the ignition tempera- 

 tures at atmospheric pressure of a number of gases. In the cases of 

 hydrogen, carbon monoxide, and acetylene, the ignition temperatures 

 were practically the same in air as in oxygen, thus : — 



In Air In Oxygen 



Hydrogen .... 580° -590° 580° -590° 



Carbon monoxide (moist) . 644°-658° 637°— 658° 



Acetylene .... 406° -440° 416° -440° 



In most other cases, however, the limiting temperatures observed in 

 air were higher than in oxygen, thus: — ■ 



In Air In Oxygen 



Methane .... 650°-750° 556°-700° 



Ethylene .... 542°-547° 500°-519° 



Cyanogen .... 850°-862° 803°-818° 



Another notable fact is that in an homologous series of hydrocarbons 

 the limiting ignition temperatures appear to fall as the series is ascended. 

 Thus we have (in oxygen) : — 



Methane, 556° — 700° ; Ethane, 520° — 630° ; Propane, 490° — 570°. 



During 1906-07 K. G. Falk, 3 acting on a suggestion by Nernst, en- 

 deavoured to determine the ignition points of various mixtures of 

 hydrogen and oxygen by compressing them in a steel cylinder by means 

 of a weight falling on a piston. Assuming that (1) the mixture was 

 heated adiabatically and uniformly throughout its whole mass until it 

 reached the ignition point, (2) that the whole then detonated instanta- 

 neously, and (3) that the downward movement of the piston was arrested 

 at the moment of ignition, he calculated the ignition temperatures under 

 adiabatic compression shown in the table given below. 



H. B. Dixon, 4 in a recent criticism of Falk's assumptions, contends 

 that whilst (1) may be in certain cases practically true, (2) and (3) 

 cannot be allowed, and also that Falk's results are unreliable on account 

 of his having neglected to stop the descent of the piston the moment the 

 gases were brought to the true ignition point. Adopting this necessary 

 precaution, Dixon has repeated Falk's experiments, with results which 

 compare as follows when y is assumed to be 1'40 in all cases : — 



1 Monatsch., 1900, 21, 1061. 



2 Trans. Chem. Soc, 1909, 95, 514. 



3 Journ. Amer. Chem. Soc, 28, 1517 ; 29, 1536. 

 ' Trans. Chem. Soc, 1910, 97, 674. 



