'286 PHILOSOPHICAL TRANSACTIONS. [ANNO 1782. 



some occasions, to have 2 or more pieces, according to convenience, put in toge- 

 ther at first, that they may be successively cooled in water, and the degrees of 

 heat examined at shorter intervals. It will be unnecessary to say any thing fur- 

 ther on precautions or procedures which the very idea of a thermometer must 

 suggest, and in which it is not apprehended that any difficulty can occur, which 

 every experimenter will not readily find means to obviate. 



12. It now only remains, that the language of this new thermometer be 

 understood, and that it may be known what the heats meant by its degrees really 

 are. For this purpose a great number of experiments has been made, from which 

 the following results are selected. The scale commences at a red-heat, fully 

 visible in day-light; and the greatest heat that I have hitherto obtained in my 

 experiments is l6o°. This degree I have produced in an air-furnace about 8 

 inches square. Mr. Alchorne has been so obliging as to try the necessary expe- 

 riments with the pure metals at the Tower, to ascertain at what degrees of this 

 thermometer they go into fusion; and it appears, that the Swedish copper melts 

 at 27, silver at 28, and gold at 32. Brass is in fusion at 21. Yet., in the brass 

 and copper foundries, the workmen carry their fires to 140° and upwards: for 

 what purpose they so far exceed the melting heat, or whether so great an addi- 

 tional heat be really necessary, I have not learnt. The welding heat of iron is 

 from QO to 95°; and the greatest heat that could be produced in a common smith's 

 forge 125. Cast iron was found to melt at 130°, both in a crucible in my own 

 furnace, and at the foundry; but could not be brought into fusion in the smith's 

 forge, though that heat is only 5° lower. The heat by which iron is run down 

 among the fuel for casting is 1 50°. 



As the welding state of iron is a softening or beginning fusion of the surface, 

 it has been generally thought that cast iron would melt with much less heat than 

 what is necessary for producing this effect on the forged; whereas, on the con- 

 trary, cast iron appears to require, for its fusion, a heat exceeding the welding 

 heat 35 or 40°, which is much more than the heat of melted copper exceeds the 

 lowest visible redness. Thus we find, that though the heat for melting copper 

 is by some called a white heat, it is only 27° of this thermometer. The welding 

 heat of iron, or 9O , is likewise a white heat; even 130°, at which cast iron is 

 in fusion, is no more than a white heat; and so on to 1(J0° and upwards is all a 

 white heat still. This shows abundantly how vague such a denomination must 

 be, and how inadequate to the purpose of giving us any clear ideas of the extent 

 of what we have been accustomed to consider as one of the 3 divisions of heat 

 in ignited bodies. 



A Hessian crucible, in the iron foundry, viz. about 150°, melted into a slag- 

 like substance. Soft iron nails, in a Hessian crucible in my own furnace, melted 

 into one mass with the bottom of the crucible, at 154°: the part of the crucible 



