26 ANNUAL OF SCIENTIFIC DISCOVERY. 



nut. At the bottom end the orifice is much contracted, so that any 

 fluid entering can easily ascend in the larger portion. On this small 

 orifice is placed a cap of soft metal, fusible at a low temperature, and 

 the tube being screwed tight down compresses this fusible cap against 

 the top of the flue. Now, as long as the cap remains at 212, the 

 temperature of the boiling water, it will keep its place ; but if the sup- 

 ply of water fails, so that the surface of it in the boiler gets below the 

 cap, the flue becomes red-hot, and the fusible metal melting off from 

 the small tube, the steam instantly rushes out, and thus gives notice 

 to the engineer. More water is of course then let into the boiler, and 

 the pipes are stopped with plugs of wood till at the first opportunity, 

 when the steam is down, a new cap is screwed on. It would seem, 

 therefore, that though this may answer for one occasion, yet if the wa- 

 ter gets too low twice during the same trip, the boiler may be blown 

 up without any warning being given of the want of water. 



PATENT STEAM AND WATER GAGE. 



THESE different gages are the invention of Mr. A. S. Lyman, of 

 St. Louis, and may be separately described, though intended to go to- 

 gether. The steam-gage consists of an iron tube, of any length, 

 which is attached at one end to the top of the boiler, while the other 

 is curved in the form of an inverted siphon ; connected with this is a 

 strong glass tube, sealed at the top, and placed in an upright position 

 beside a scale graduated to show the number of pounds of steam in 

 the boiler to the inch. The longer leg of the siphon is in part filled 

 with mercury, above which is a column of water to protect it from 

 the heat of the steam, which is forced from the boiler through the iron 

 tube until it comes in contact with the water. The pressure of the 

 steam on the water compresses the air in the glass tube, so that double 

 the pressure drives the air into nearly half the space. The correct- 

 ness of the gage depends on the fact that the water is a good non-con- 

 ductor, and cannot transmit heat downwards. Air is expanded by 

 being heated, and it requires a greater pressure to force it into a given 

 space when warm than when cold. Eight degrees of temperature 

 make a difference of a pound in the indication of the gage, and it is for 

 this reason that the thermometer is added, and the scale is made when 

 the mercury stands at 72 degrees. At this temperature the scale is 

 strictly correct, but for every 8 degrees more or less you add or sub- 

 tract one pound. 



The water-gage is constructed in the same way as the steam- 

 gage, except that, instead of being connected with the top of the boil- 

 er, it is connected with a copper box, hermetically sealed, lying in 

 the boiler and on the flue. This box is filled with water, so saturated 

 with salt as to prevent freezing, and it has no outlet except through 

 the water-gage. The indications of this gage vary from those of 

 the steam-gage for several reasons, the principal of which is that 

 more heat is required to produce a given pressure from salt than from 

 fresh water. As soon as the water falls too low, or leaves the surface 

 of the flues bare, they receive extra heat, and so the pressure in the 



