On a Metliod of Determining Vapour-Densities. 4G3 



of the vapour-density becomes impossible in the case of a 

 number of bodies. 



It must appear rather surprising that no attempt has been 

 made to extend to Dumas's method the same artifice which 

 Hofmann has so successfully applied to that of Gay-Lussac — 

 namely, to make the determination of the density of the vapour 

 under diminished pressure. I have made such an attempt, and 

 believe that I have succeeded by very simple arrangements in 

 proving that it is possible : — 



(1) to make a determination of a vapour-density by Dumas's 

 method with less than a gram of the substance ; 



(2) to recover without difficulty the greater part of the 

 substance employed, free from impurity ; 



(3) to effect the determination at a temperature which need 

 never exceed the boiling-point of the substance. 



In the execution of the experiments about to be described, 

 which were made with bodies whose boiling-points extended 

 up to 250°, there were needed, in addition to the glass balloon, 

 a moderately large thick- walled bulb-tube, an aspirator with 

 a mercury manometer, a thermometer, a barometer, and an 

 oil-bath — -things, therefore, which are hardly wanting in any 

 laboratory. 



It is scarcely necessary to say any thing with respect to the 

 glass balloon ; for if suitable for Dumas's method, it is equally 

 applicable to the modified process. It may conveniently have 

 a capacity of from 250 cubic centims. to 300 cubic centims., 

 although (as the examples show) it may be considerably less. 

 The neck of the balloon, which should previously be carefully 

 cleaned and dried, is drawn out to a capillary tube 2 millims. in 

 diameter, of suitable length, and as thick as possible in its wall. 

 The capillary tube is twice bent, in a manner which will be 

 perfectly evident from the figure (p. 465). The position of the 

 two bendings is dependent upon the size of the balloon and 

 the diameter of the vessel used for the oil-bath ; matters 

 should, however, be so arranged that w^hen the balloon is 

 sunk in the bath the second bending just appears above the 

 opening of the vessel and near its edge. At this bend the 

 capillary tube constituting the neck is to be further con- 

 stricted to a diameter of 1 millim ; the open end of course 

 can be closed before the blowpipe. The bulb-tube employed 

 as the receiver is also bent to an obtuse angle. The portion 

 of the tube containing the bulbs is so drawn out at the end, 

 that the thick caoutchouc tube of the pump can be conveni- 

 ently pushed over it ; whilst the end of the shorter arm is 

 adapted, by a short length of sufficiently stout caoutchouc 

 tube, to the balloon. 



