164 RESEARCHES UPON THE CHEMICAL PROPERTIES OP GASES. 



ditions of the experiment. To ascertain as fully as possible the limits of variation 

 in the temperature of oxidation, the following trials were made: 



Expt. 26. — Instead of metal-coated asbestos, a platinum wire one millimeter 

 thick and one inch long was placed in the one-eighth-inch glass tube in the oven. 

 The same mixture of air and ethylene was used as in the preceding experiments. As 

 in all previous work, the oven was allowed to cool down after each trial and before 

 the next one was begun. In eveiy case the temperature was very gradually raised 

 until oxidation, as indicated by the precipitation in the lime water, occurred. The 

 gas stream was carefully regulated, being maintained at a uniform rate in all the 

 trials. Oxidation occurred at the following temperatures : 



(1) 



At 270O 



(2) 



" 290° 



(3) 



" 390° 



(4) 



" 290O 



(5) 



" 310° 



(6) 



" 2890 



0) 



" 2950 



(8) 



" 300O 



(9) A 



t 300° 



(10) ' 



' 2650 



(11) ' 



' 210O 



(12) ' 



' 217° 



(13) ' 



' 220O 



(14) ' 



' 2250 



(15) ' 



■' 2000 



(16) ' 



' 2100 



(17) 



At 2200 



(18) 



" 2550 



(1'9) 



" 2100 



(20) 



" 220° 



(21) 



" 2350 



(22) 



" 2003 



(23) 



" 2100 



(24) 



" 225c 



As no effort had been spared to secure absolute uniformity of conditions, the 

 conclusion seems justified that oxidation of the hydrocarbon ethylene occurs within 

 somewhat wide limits of temperature. 



Exi)t. 27. — Air containing 3.1 per cent, ethylene was passed through a glass 

 tube containing pieces of glass tubing which had been previously ignited. No oxi- 

 dation occurred at the melting point of bromide of silver (427°) during four hours. 



Uxpt. 28.— 



Air 96.9 



Ethylene 3 1 



This mixture was imssed, as before, over a platinum wire and the temperature grad- 

 ually raised until a carbon dioxide reaction was produced. This occurred at 210°. 

 The gas stream was then continued while the temperature was allowed to fall. The 

 lime water was repeatedly replaced, but each time became rapidly milky. Oxidation 

 was continuous until the temperature fell to about 110°, at which point fresh lime 

 water was found to remain clear. Hence, platinum having been sufficiently heated 

 to induce oxidation of ethylene by atmospheric oxygen, retains this power while the 

 temperature is lowered to a point which would have been wholly insufficient to cause 

 such oxidation if the temperature were rising instead of falling. This is true, more- 

 over, when the gas stream flows at the rate of 20 to 50 bubbles per minute ; so 

 slowly, therefore, that there is no possibility that this effect is attributable to an 



