CusACK — On the Melting Points of Minerals. 409 



This substance is, however, very viscous, whereas CuO has a very 

 definite point of fusion, so this explanation will not account for the 

 case of CuO. The roughness of the fragment of CuO, and the move- 

 ment and regularity of the halo, appear to negative any explanation 

 depending on the reflection of light. In fact I have not been able to 

 arrive at any satisfactory explanation of the phenomenon. 



The meldometer, as will be seen from this Paper, can be used as 

 a high temperature thermometer up to the melting point of platinum, 

 reading with ease to 2° C. In the present instrument no attempt 

 has been made to magnify the expansions of the ribbon ; this could 

 best be done by lengthening the arm, especially for a short range, 

 say of 300°, and thus obtain closer readings. 



A far more interesting use the meldometer could be put to is that 

 of analysis in place of the blowpipe. Here it would be of enormous 

 advantage — one would have no waste, and could always have the use 

 of a microscope, and use only very small quantities of the minerals. 

 In the meldometer, not only can the reduction or oxidisation of a 

 mineral be observed under the most favourable circumstances, but 

 also the temperature at which such changes take place can be recorded. 

 The meldometer is also much cleaner to work with (and one never 

 gets a red-hot spark into one's eye as is sometimes the case with the 

 blowpipe), and it is very much easier to handle in every respect than 

 the blowpipe. Sublimates can also be very easily obtained with the 

 meldometer, as any one who has read Dr. Joly's Paper,^ on the 

 "Melting points of Minerals," will have seen. He there describes 

 many of the sublimates he obtained, and also the means used for 

 obtaining them. 



It is interesting to consider what the melting points of minerals are 

 influenced by, whether it depends entirely on the chemical composition 

 of the mineral, or if the molecular structure influences it also. 



As an example : both riitile and brookite fuse at the same tempera- 

 ture, though of different forms of symmetry ; but topaz and kyanite, 

 though nearly of the same composition (some of the oxygen in the 

 kyanite being replaced by fluorine in the topaz), show in the case of 

 kyanite a fusing point of 1090°, whereas topaz is quite infusible. Can 

 the infusibility of topaz be accounted for by the presence of fluorine ? 

 It would appear as if the laws of fusion were very complicated. 



The following list of the minerals whose melting points have been 

 determined may be of interest. The procedure in arriving at the 



^ Loc. cit., p. 38. 



2 F2 



