624 THE POPULAR SCIENCE MONTHLY. 



mass, and a second workman attaches the end of his iron rod or 

 " punty " to the free end of the ball. The blower remains station- 

 ary, while the second man walks away from him, carrying his 

 punty with him. In this way the mass of glass is drawn into a 

 long tube, perhaps fifty feet long, the bubble of air preserving a 

 fine opening throughout the entire length of the tube. In the 

 better thermometers, the tube is somewhat flattened, so as to make 

 the thread of mercury more visible, and a background of opaque 

 white glass is added for the same purpose. These modifications 

 are made more easily, perhaps, than one would imagine. By flat- 

 tening the ball of glass before it is drawn into a tube, the ellipti- 

 cal cross-section is secured, while a string of opaque glass welded 

 on to the still plastic ball becomes elongated into a thin plating 

 on one side of the tube. 



It is impossible in this way to secure tubes of absolutely uni- 

 form bore, but the inequalities are much less than one would sup- 

 pose. For ordinary instruments the variation may be neglected. 

 The tubes are then cut into convenient length and sent to the 

 workshop of the thermometer-maker. One can readily pass a 

 whole morning in the little room where he works, for there is a 

 certain interest attaching to so individual a task as this which is 

 not found in more wholesale production. The instrument-maker 

 sits on a high stool before his work-table, his principal tool being 

 a conveniently arranged blowpipe. This is not the hand and 

 mouth tool used by mineralogists and jewelers, but is a perma- 

 nent blowpipe fed by gas and operated by a blast of air. 



The first operation is that of forming the bulb. In the better 

 instruments this is made out of a separate piece of glass and is 

 then attached to the tube. In this case the bulb is made cylin- 

 drical in form, so as to afford large capacity without too great 

 diameter. In the less expensive thermometers, the bulb is formed 

 directly on the end of the tube itself. The glass is first fused in 

 the blowpipe flame until the end is entirely sealed. A short 

 rubber hose with a small rubber ball on the opposite end is then 

 slipped over the open end of the thermometer-tube. The sealed 

 end of the tube is again softened before the blowpipe, and then, 

 by simply pressing the rubber ball, the air forces the plastic glass 

 into a symmetrical bulb. It is a pretty little operation, for the 

 glass responds so delicately to the thought of the workman. 



It is found that glass undergoes a slow contraction during a 

 period of two or three years, and, where great accuracy is desired, 

 the tube must be put away for that time to season. 



The bulb and tube are now to be filled with mercury. The 

 tube is much too fine to allow the mercury simply to be poured 

 into the bulb. Indirect means must be used. The open end of 

 the tube is softened and quickly blown into a large bulb, while 



