410 Proceedings of the Royal Irish Academy. 



melting point of a particular specimen, once the curve is got, is the 

 following : — First a tiny speck of the mineral is put on the ribbon, 

 then the micrometer screw is adjusted till the point is just in 

 contact with the spring of the arm ; the galvanometer will then 

 oscillate if the head be moved the slightest bit back. The contacts 

 can be made so that a movement of the 20,000th of an inch of the 

 point of the screw will make or break the circuit through the galva- 

 nometers. This forms a very delicate means of reading the expansion 

 of the ribbon. "WTien the screw is thus adjusted the trough is raised 

 so as to surround the ribbon, and then the microscope can be placed 

 so as to have the q^eck in the field of view. This is difficult, 

 and it was found convenient to use a pointer of a little piece of 

 platinum wire to find the speck. If the mineral is expected to melt 

 before a temperature of 1000° is reached, a ray of light should be 

 thrown on the speck by means of the mirror previously mentioned ; 

 the current is then turned on, and the resistance in the circuit 

 decreased till the mineral is observed to show signs of melting ; the 

 screw is kept following the expansion of the platinum all the time, 

 till the substance is observed to melt, then stopped, and the expansion 

 calculated as previously shown. The first determination is generally 

 too high, and at least four or five trials are made before the melting 

 point is satisfactorily arrived at, but generally ten to fifteen are found 

 necessary when dealing with a viscous body. If one knows about the 

 temperature at which the body may be expected to fuse, five or six 

 are, however, generally sufficient. The expansion at each trial is 

 generally less than the previous one, and thus a point can be arrived 

 at, which is the lowest at which the mineral under observation is 

 observed to be fused. One can in a similar way determine the lowest 

 point at which a substance is soft, as when soft it can be seen falling 

 about on the ribbon. 



The following example may illustrate this. The mineral was a 

 specimen of Diallage. When cold the ribbon was 383 inches long, 

 the head reading 1812 divisions : — 



1st 



trial, 



2276 



was 



reading of head 



2nd 





2284 







yrd 





2263 







4th 





2267 







5th 





2260 







6th 





2264 







