86 Scientific Intelligence. 



about 45. The authors believe that the new elements of the air 

 should be placed before group I in the periodic system, so that 

 helium = 4, argon =39, and krypton = 59 come before lithium, 

 potassium, and copper. It is their opinion that air contains less 

 than 0*00002 per cent of krypton, but if this is so it is evident that 

 the 750 cc of liquid air of Ramsay and Travers, in which they dis- 

 covered the element, must have been previously subjected to a 

 great amount of concentration by evaporation. — Ghem. News, 

 lxxxi, 205, and lxxxii, 209. h. l. w. 



3. The Combustion of Gases. — It is well known that sufficient 

 dilution of an explosive mixture of gases with any other gas will 

 prevent explosion. Thus Bunsen found that a mixture of 25*79 

 per cent of detonating gas (hydrogen and oxygen) with 74*21 

 per cent of carbon dioxide cannot be kindled, while 8*72 parts of 

 detonating gas and 91*28 parts of oxygen behave in the same 

 way. Such facts are satisfactorily explained by assuming that 

 the dilution lowers the temperature of combustion below the 

 kindling point. S. Tantar calls attention to the fact that where 

 the dilution is made with certain hydrocarbon gases the effect is 

 caused by so little of the latter that the usual explanation is not 

 sufficient to account for it. He found that 11-12 per cent of 

 propylene, CH 2 : CH . CH 3 , or of trimethylene, (CH 2 ) 3 , is suffi- 

 cient to prevent the explosion of detonating gas, while methane, 

 CH 4 , prevents the explosion when present in twice the amount, 

 or 22-24 per cent. He found that when a mixture of 5 CC of pro- 

 pylene with 45 cc of detonating gas was exploded only a small 

 change of volume, at most a diminution of 5 CC , took place instead 

 of the expected disappearance of the detonating gas. If 5 to 

 15 cc more of detonating gas was taken than in the preceding 

 experiment the contraction increased to the extent of this addi- 

 tion. An analysis of the resulting gases showed that propylene 

 was no longer present, and that no noticeable amount of carbon 

 dioxide was formed, particularly when the proportion of gases 

 was such that the limit of combustibility was reached. The 

 oxygen was completely used up in the combustion, so that the 

 mixture of gases consisted essentially of carbon monoxide and 

 hydrogen. It is remarkable that the combustion of propylene 

 does not take place under these conditions until there is just 

 enough oxygen present for its complete combustion to carbon 

 monoxide and water, for 5 CC propylene require 15 cc oxygen or 

 45 cc detonating gas, which is the smallest amount that will cause 

 an explosion. Methane requires half as much oxygen, and com- 

 bustion actually takes place upon the addition of 25 cc of detonat- 

 ing mixture to 5 CC of the gas. The results show a remarkable 

 selection which oxygen exhibits in combining with hydrocarbons 

 in preference to hydrogen. Berthelot's thermochemical principle 

 of greatest work does not apply here, for the reaction 6H 2 '-h30 2 = 

 6 H 2 produces more heat than C 3 H 6 -f-30 2 =:3CO + 3H 2 0. It is 

 the author's opinion that the only possible explanation of the 

 phenomenon is that the rapidity of the action of oxygen with 



