CLEANING AND FILLING THE TUBE. 



necessary to render the tube perfectly clean on its inner surface. 

 It generally happens that tubes exposed to the air, always more 

 or less damp, have a film of moisture upon them. It is necessary 

 therefore to expel this. After cleaning the tube by internal friction, 

 it is warmed over the flame of a spirit lamp from end to end, so as 

 to evaporate any moisture which may remain upon it and render 

 it perfectly dry. 



5. Mercury is then poured into it by means of a funnel with a 

 very small aperture, until a column of about ten inches has 

 entered. However pure this mercury may be, and however clean 

 the tube, it will be more or less mixed with air, which will enter 

 with it as it passes from the funnel. To dismiss this air the tube 

 with the mercury in it is heated over a spirit lamp, until it is 

 raised to a temperature higher than that of boiling water. By 

 this process the air combined with the mercury, or adhering to the 

 inner surface of the tube, being expanded by the heat, escapes 

 from the tube, as well as any moisture that may have entered 

 with the mercury. 



Mercury is again introduced in the same manner, and again 

 heated, and the process is repeated until the tube has been com- 

 pletely filled. 



In this process it is usual to heat the mercury to nearly the 

 same temperature as that of the tube before pouring it in, since 

 .otherwise there would be some danger of cracking the tube, by 

 the expansion or contraction of the glass consequent on the sudden 

 change of temperature. 



6. When the tube is in this manner completely filled, the 

 open end is finally stopped with the finger, and being inverted 

 it is plunged in a small cistern of mercury (fig. 1, p. 177). When 

 the finger under the mercury in the cistern is withdrawn, the 

 column in the tube will subside until it falls to the altitude which 

 will be balanced by the atmospheric pressure. 



If several tubes be prepared in this manner, it will be found 

 that the columns of mercury sustained in them at the same time 

 will be exactly equal. 



7. In adapting such an apparatus to indicate minute changes 

 in the pressure of the atmosphere, there are several provisions to 

 be made. 



The height to be measured being that of the surface of the 

 column in the tube above the surface of the mercury in the 

 cistern, it is not enough to ascertain the position of the surface in 

 the tube, unless the surface in the cistern have a fixed level. 

 Now it is evident, that whenever the surface in the tube rises, the 

 surface in the cistern must fall, and vice versa, inasmuch as 

 whatever mercury enters the tube must leave the cistern, and 



2 179 



