CusACK— Ow the Melting Points of Minerals. 401 



in place of the carbon rods. An additional self-'^'orking rheostat was 

 introduced for convenience. This rheostat was formed of a carbon 

 rod, about 2 feet in length, enclosed in a glass tube in an upright 

 position ; mercury is allowed to flow in from an adjoining vessel and 

 surround the carbon, thereby reducing the resistance. Without any 

 attention being paid to the rheostat, the flow of mercury can be 

 regulated so that the resistance alters either slowly or rapidly as is 

 convenient to the observer. 



It would be as well, perhaps, to explain the operation of fixing 

 the ribbon between the forceps, and also how the curve for the 

 expansion of the ribbon is arrived at. 



The ribbon used was supplied by Messrs. Johnson & Marthey, and 

 weighs 0"0073 grammes per centimetre, 3'80 inches of which were 

 taken, and clipped at each end at about 30°. The ends thus clipped 

 were fixed in the forceps, and adjusted, so that when a suitable current 

 was passed through, the entire ribbon was uniformly heated. 



The ribbon, when adjusted so that it is heated uniformly, is raised 

 to a bright red heat, and left thus for a few minutes ; the current is 

 then cut off, and the whole apparatus allowed to cool before calibra- 

 tion is commenced. To calibrate a ribbon the milled head of the 

 micrometer screw is turned until the point of the screw comes in 

 contact with the spring projecting from the other arm of the movable 

 forceps from that to which the ribbon is attached ; the number of 

 divisions through which the head has moved are then read off ; a 

 speck of silver chloride (the melting point of which is assumed from 

 the determination by Carnelly) is then placed on the ribbon, the 

 current is turned on, and the resistance to the current is reduced by 

 the rheostats in the circuit, till the AgCl is seen to melt, a micro- 

 scope being used to aid the eye. I may here observe that a small 

 concave mirror was found very convenient for illuminating the sub- 

 stance under observation. The ribbon is not sufficiently luminous of 

 itseK until the melting point of cupric oxide is reached. The expan- 

 sion of the platinum ribbon should be carefully followed with the 

 micrometer screw till the substance melts, and should then be instantly 

 stopped. The point of the screw can be kept in contact with the 

 spring by use of the galvanometer in the eyepiece ; this becomes 

 quite easy after a little practice. The number of divisions moved 

 through is again read on the head, and this reading, miaus the previous 

 reading, gives the expansion of the ribbon for AgCl. This expansion 

 is then marked off to scale on an ordinate, the temperature at which 

 AgCl is known to melt on another ordinate at right angles ; and 



