228 PKESSURE OF THE ATMOSPHERE. 



from it as much as possible, and a piece of glass tube connected 

 by an india-rubber tube with the generating bottle is inserted 

 into the mouth. A thread wound loosely round the neck of the 

 balloon and not too firmly tied, holds the balloon sufficiently firm 

 to the glass tube for the purpose of filling it with gas. When quite 

 distended by the latter the balloon is detached by unwinding the 

 thread and, if necessary, by pushing it along the tube by a gentle 

 pressure of the fingers. When detached it rises to the ceiling and 

 remains there, until by the escape of hydrogen and the influx of 

 air the weight of the balloon exceeds that of the displaced air. It 

 is not advisable to tie thread around its neck in order to prevent the 

 escape of the hydrogen from the balloon and to maintain it longer 

 in the air, for the balloon is easily damaged by it. Collodion 

 balloons rise also when filled wrth coal gas, provided they are not 

 too small. 



Hot -air is lighter than cold air, as will be seen further on ; a 

 balloon may hence be made to rise by filling it with hot air. Such 

 a balloon must, however, be of rather large size, and have a wide 

 aperture -at the lower part, where a fire is kept up, usually a spiril 

 flame, or in very large balloons by burning straw, in order to heat 

 the air in the interior. Such balloons are usually made of thin 

 paper, and have a size of one or more metres ; they cannot be well 

 used in a room, and in the open air they are very dangerous if they 

 should come into contact with combustible objects. 



25. Atmospheric pressure. The Barometer. If a 

 tumbler with a smooth edge is filled with water and 

 covered with ,a piece of stiff paper, it may be in- 

 verted, mouth downwards, and no water will run out 

 If the paper touches the edge all round very closely 

 the experiment will often succeed when the tumbler it 

 simply inclined rather gradually and finally inverted 

 but it is safer to press against the paper with the 

 fingers of one hand .spread out, or better with a flat 

 body such as a small board or a plate, until the moutl 

 of the tumbler is turned downwards; the hand or other 

 object used for holding the paper is then removed, ancj 

 it will be found that the paper adheres, and that n< 

 liquid, or at most a small quantity only, runs out. 



