THE BAROMETER. 235 



force of gases they exert a pressure upon the mercurial column, 

 making the height of the latter appear less than that really due to 

 atmospheric pressure. The tube should be filled by means of a 

 small funnel of folded paper to about 2 cm from the open end ; this 

 is firmly closed by the finger, and the tube is slowly inclined up 

 and down, thus moving the large air- bubble which has been 

 purposely left up and down along the sides of the tube ; the small 

 bubbles along the sides are swept off by this operation and join the 

 larger one ; finally the tube is completely filled to the open end. 

 It is thus possible to remove, if not all, at least the larger bubbles. 

 In a wider tube, of about l cm width, the operation is more successful, 

 and such a tube has also the advantage that the action of capillarity 

 produces a scarcely sensible error, while in narrow tubes it causes a 

 depression of the mercurial column below its real height. A tube 

 80 cm long and l cm wide requires, however, more than 850s r of 



FIG. 162 (\nalsize). 



mercury, and as a quantity of mercury is required for the vessel in 

 which the tube dips, it will be more expedient to make use of a 

 narrower tube. Fig. 162 is the section of a very convenient shape for 

 the vessel required in this experiment. It should be made of iron 

 3r stone ware, and will be found useful for other experiments. 



The height is measured by means of a wooden scale, divided into 

 centimetres. Obtain from a joiner a rectangular bar, made of dry, 

 mrd wood, l m long, l cm broad, and 5 mm thick, and paint it white 

 >n one side. "When thoroughly dry, the whole painted length is 

 livided by pencil lines with the help of a metre rule into square 

 Centimetres. The first, third, fifth, and all odd-numbered squares 

 are left white, every tenth square is painted red, and all the other 

 i'ven-numbered squares are painted black. In fig. 163 the red 

 3 indicated by oblique lines across the square. A scale of this 

 S:ind is very convenient where great accuracy is of less importance 

 >han distinctness of the graduation. The difference in the colours 

 enders any number of centimetres discernible even at some distance, 

 without figures on the scale. Small quantities of the necessary 

 'amts may be bought at a very little expense, ready for use, of any 

 oil and colourman.' Or the student may purchase the materials 



