370 Prof. Liveing, On a fall of temperature [Feb. 1, 



dipped into the oil or paraffin and gave the temperature of the 

 bath. The bath was heated by a Bunsen burner beneath it. 



The oxalic ether and the potassium cyanide were mixed in 

 proportions intended to correspond to two molecules of the latter to 

 one of the former, so as to react according to the following identity, 



(C 2 H 5 ) 2 C 2 4 + 2KCN = 2C 3 H 5 N + K 2 C 2 4 . 



Commercial cyanide was employed, which was afterwards found to 

 contain a great deal of carbonate. The effect of this was equivalent 

 to having an excess of oxalic ether. 



As the bath was heated all the three thermometers rose gradu- 

 ally until that in the paraffin indicated 150°, when the thermo- 

 meter in the mixture stood at about 108° and that in the vapour 

 at about the same degree. The mixture at this temperature boiled 

 rapidly, and the thermometer immersed in it shewed rapid fluctua- 

 tions between 90° and 110°, which were no doubt due to convection 

 currents, and to the return of the condensed distillate to the flask. 

 We seemed to have a substance evaporating which had a boiling 

 point not far from 100°. The boiling point of ethyl- cyanide or 

 propio-nitrile, one of the two isomers which have the composition 

 C 8 H 5 N, has been given by different observers at various degrees 

 between 82° and 98°"5, and it seems probable that the higher 

 figure is the more correct. Gautier, using a carefully purified 

 sample, found its boiling point to be 96°"7 (Bull, de la Soc. Chem. t. 

 IX. p. 4). Oxalic ether does not boil below 180°. It is therefore 

 most likely that the vapours evolved at this stage consisted prin- 

 cipally of propio-nitrile. 



When the heating was continued and the paraffin reached 168° 

 the temperature both of the mixture and of the vapour above it 

 fell suddenly to 85°, and that of the vapour soon dropped to about 

 74°. As long as the temperature of the bath was kept at about 

 170° the mixture remained at a temperature between 85° and 90° 

 and the vapour at about 74°. 



When the lamp under the bath was now turned out, the 

 paraffin of course gradually cooled, and when it had reached 150° 

 the mixture had risen to 101° and the vapour to 90°. On again 

 heating up the bath the fall in temperature of the mixture and 

 vapour was observed to occur as before when the paraffin reached 

 168°; and the same phenomena recurred several times when the 

 bath was alternately cooled and heated. 



It seems, at first sight, strange that the rise of temperature of 

 the bath, and consequent increase in the supply of heat, should 

 cause a fall in the temperature of the mixture. But I have no 

 doubt that it is due to a difference in the chemical changes which 

 occur in the mixture at the different temperatures. When the 

 bath is at about 170° we seem to have a substance formed which 



