i 9 o8] CROCKER AND KNIGHT— CARNATIONS 273 



of the gas constituents with bromin) was separated, washed with a 

 weak solution of potassium hydrate, and later with distilled water. 

 This oil was then dried with fused calcium chlorid and later frac- 

 tionated. In the first distillation all the portion boiling between 

 129 and 134 C. was saved. This was later redistilled and the 

 fraction boiling between 103 and 132 C. saved as representing the 

 ethylene dibromid, since this compound boils at i3i°C. About 

 3 per cent, of the dried material absorbed by bromin boiled between 

 130 and 13 2 C; a small portion boiled at 129 C. or below. From 

 this it rose up quickly to 13 1° C, where it again gave a considerable 

 fraction. Then it rose rapidly to 139 C, where a considerable 

 fraction distilled. In one trial, 208 liters of gas at 27 C. and under 

 pressure of 745. 5 mm of mercury gave i30 gra of dried oil; of this 

 44.2 gra boiled between 130 and i32°C. After correcting for pres- 

 sure and temperature the following equation equals the percentage 

 volume of gas that is ethylene: 



22.4X760X44-2X300 



— -^ — ^f— : — — =2.o+ per cent. 



208X745-5X178X273 y 



In a second determination 138 liters of gas at 27 C. and 745 . 5 mm 

 pressure gave 3i.6 gm of oil boiling between 130 and 132 . Cor- 

 recting for temperature and pressure, the following equation gives 

 the percentage volume of ethylene in this case: 



22.4X760X31.6X300 



— — - — - — =3.2 per cent. 



138X745-4X178X273 6 F 



It must be stated, however, that according to Winkler (17) the 

 absorption of ethylene by bromin is not complete, and farther that 

 considerable ethylene dibromid is necessarily lost in washing, dry- 

 ing, and distilling, so that the percentage is probably considerably 

 higher than here obtained. It must be urged also that the presence 

 of other oils with boiling points rather near that of the ethylene 

 bromid tends to make this fractionation less accurate. 



4. General 



It is of great interest to know that the most delicate chemical test 

 for illuminating gas in the atmosphere falls far short of detecting 

 amounts that work havoc with the flowers of the carnation. The 

 tests for carbon monoxid are those used for detecting illuminating 



