THVEMAL CONDUCTIVITY. 31 



made to pass to keep it at an uniform temperature. The 

 lower one is allowed to radiate off heat to the air. It gets 

 heat by its proximity to the cone above. A certain amount 

 of liquid is sucked up from a vessel below into a tube which 

 is in connection with the lower cone, and this lower cone 

 thus acts the part of the bulb of an air thermometer, and 

 if it be heated very much the liquid in the tube will be 

 forced down. If it be heated only a little it will be forced 

 clown very slightly, so that you have a measure of the 

 temperature of this lower cone. The interval between 

 these two cones is filled with a liquid to be examined ; 

 then the temperature of the upper cone being always 

 100 C. the conductivity of the different liquids is 

 measured by the depth to which the liquid is driven 

 down, that is to say, by the temperature acquired by the 

 lower cone. The greater the flux of heat through this 

 liquid the greater will be the temperature of the lower 

 cone, and the temperature is measured by the height of 

 the liquid in the tube. There is a scale at the top for 

 measuring the distance between the two cones. If that 

 distance is kept constant as we employ different liquids we 

 shall be able to arrange the liquids which we examine in the 

 order of their conductivity, those which are the best con- 

 ductors being placed above, and those which are the worst 

 being placed below. From these experiments it appears 

 that water is one of the best conductors among liquids. 

 Convection plays a very slight part there, or none at all, 

 because the upper surface is the one which is heated. 

 Despretz made some experiments with a large cylinder of 

 water which he heated from the surface, and arrived at a 

 rough approximation to the conductivity of water, but one 

 which it is very desirable should be repeated. In fact, 

 experiments have been made since then, but they have not 

 been published, though it is very desirable that we should 

 have a knowledge of the absolute conductivity of water. 



