Determining the Vapour Pressures of Solutions. 377 



those in the second. Examples of such circumstances are, barometric 

 pressure, rate of ebullition, height of flame employed, depth of ex- 

 perimental liquid, and many others, and there is no doubt that the 

 chief reason why the boiling-point method has been so little utilised, 

 is the inability to reproduce these conditions. In particular, if the 

 pressure under which the boiling point is to be measured is much 

 less than 760 mm., the difficulties are so much increased that no 

 one seems even to have attempted to make observations under such 

 conditions. 



• § 2. Principle on which success depends in measuring by the Boiling - 

 Point Method the Vapour Pressure of Salt Solutions. 



The difficulty in applying and extending the boiling-point method 

 disappears in a great measure if not entirely, if the apparatus is 

 duplicated, and the difference alone is measured of pure water and 

 solution boiling under similar conditions and under the same pressure, 

 the magnitude of which need not then be known to within a millimetre. 



The difference of temperature referred to may very profitably be 

 measured by the platinum resistance method. A decided advantage 

 of methods which obey the principle above stated, lies in the fact 

 that, whereas the differences of vapour pressure, measured at a 

 common temperature, diminish very rapidly as the temperature is 

 lowered, the difference of temperatures, measured under a common 

 pressure, varies much less rapidly. As an example, suppose that a 

 certain solution under a pressure of 760 mm. has a boiling point 0'52° 

 higher than pure water, then under 360 mm. it will be found to have 

 a boiling point 0*45° higher than that of pure water, a diminution of 

 only 14 per cent. The corresponding diminution in the difference 

 of tension would be 53 per cent. 



The first apparatus employed by me to take the difference of 

 boiling points of pure and salt water was in form very simple, and 

 constructed from ordinary laboratory apparatus. A description of it 

 is here introduced, mainly because it illustrates the principle of the 

 more complex apparatus shortly to be described. 



A current of steam from a boiler divides itself at A, fig. 1, and 

 passes through two precisely similar pieces of apparatus, one con- 

 taining pure, the other containing salt, water. The contents are 

 agitated by the steam, and so raised to their boiling points, and the 

 difference alone of the temperatures in the liquids was found by the 

 two platinum thermometers Ti T 2 . Without giving details of this 

 measurement it may at once be said that, as soon as it was complete, 

 solution could be drawn off at the tap t v and its concentration deter- 

 mined. The remaining taps (t%, £ 3 ) serve to regulate the flow of 

 steam, and the bulbs to prevent a suck-back. The apparatus worked 



