280 SCIENTIFIC AMUSEMENTS. 



illustrates another way of heating the top part of a vertical tube 

 containing water by means of an outer cup, which is filled with 

 hot oil. 



Expt. 327. Another Illustration. Provide a small differential 

 air thermometer (Expt. 319), and weight it with lead so that it 

 will sink in water ; place it inside a large bell jar turned upside 

 down and filled with water, the jar being so large that when the 

 thermometer is turned on its side, with one bulb near the bottom 

 of the jar and the other higher up, the upper bulb is still some 

 inches under water. Preferably the differential thermometer may 

 be constructed with a U shaped glass connecting tube, the legs of 

 which are of unequal length. Now place a large red hot poker in 

 the upper part of the jar so as to heat up the top layer of water. 

 It will be seen, on repeating the application of the hot poker 

 several times, that the upper layer of water will ultimately become 

 very hot, far too hot to bear the hand in ; but the heat will travel 

 downwards by conduction so slowly that very little difference of 

 temperature between the two bulbs will be evident, the index 

 column of fluid moving but little, if at all. Obviously, if the 

 heat travelled readily, the upper bulb would become much more 

 rapidly warmed than the lower one, and a considerable difference 

 of temperature would be indicated. 



It is somewhat difficult to carry out experiments with gases in 

 reference to their power of conducting heat, on account of their 

 mobility and the ease with which convection currents (Expts. 318, 

 320) are set up in them. When these are prevented by employ- 

 ing loose highly porous substances (such as eider down, feathers, 

 loosely piled cotton wool, &c.) the conducting power of the entire 

 mass is found to be extremely low ; so that, as the passage of what 

 little heat does pass in such cases is partly due to the solid matter 

 present, and partly to the not entire absence of convection currents, 

 it is obvious that the conductivity of gases by themselves is very 

 small. 



Unless a current of air (wind, &c.) be set up, the abstraction of 

 heat from a well-clothed person by even very cold air is but slow, 

 so that intensely cold frosts and Arctic climates can be borne 

 without any particular inconvenience when the air is still ; but a 

 much more moderate degree of cold is far less bearable if the air 

 is in motion. At extremely low temperatures contact of the bare 

 skin with substances of high conductivity, such as metals, produces 

 local frostbite almost instantly, and corresponding injury to the 

 animal tissues thus affected ; although mere contact with the air 

 or other nonconducting substances at the same temperature pro- 

 duces no such result (vide Expt. 328). 



