194 Historical Review of Experiments 



barometer tube which was soldered to a small hollow 

 tube or point projecting outwards from the globe. This 

 projection extended downwards when the thermometer 

 fastened to the globe was in its natural upright posi- 

 tion. 



As soon as the vacant space inside of the globe and 

 around the thermometer bulb, as well as the barometer 

 tube (thirty-six inches in length), was filled with mercury, 

 the end of the tube was dipped into a vessel of mer- 

 cury ; the tube was then inverted and brought into a 

 perpendicular position, so that the globe in which the 

 thermometer was fastened was at the top. 



Since the instrument was converted in this way into 

 a true barometer, the mercury in the globe and in 

 the upper part of the barometer tube fell until the 

 upper surface of the mercury in the tube was twenty- 

 eight inches above the surface of the mercury in the 

 vessel, where it remained at rest, being kept at this 

 height by the pressure of the outside air. A lighted 

 wax-candle was now held at the upper part of the 

 tube where it entered the globe, and where the diame- 

 ter of the tube had previously been contracted, and 

 the flame was directed, by means of a blow-pipe, against 

 that part of the tube which it was desired to melt 

 together. 



As the glass was softened by the heat, the pressure 

 of the outside air immediately forced the walls of the 

 tube together ; the whole operation was successful. 



The barometer tube was then detached, and the bulb 

 of the thermometer was now surrounded on all sides by 

 a vacuum, as may be seen from the figure on the oppo- 

 site page.* The thermometer was filled with mercury, 



* See also Vol. I., Plate to p. 404, Fig. i. 



