THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[FIFTH SERIES.] 



JULY 1882. 



I. Researches on Melting-point. By Edmund J k Mills, D.Sc.) 

 F.R.S., " Young " Professor of Technical Chemistry in 

 Anderson's College Glasgow*. 



IF we desire to ascertain the purity of a chemical substance 

 — in other words, to establish its species — two lines of 

 investigation are open to us. We may (1) by analytical 

 means determine the composition, or (2) measure by physical 

 methods some natural property of the body. Both modes of 

 research require that what is determined — whether composi- 

 tion or natural property — shall be constant over a fair range 

 of genetic condition. Both modes also involve a considerable 

 amount of inference ; but, of the two, far less is demanded by 

 the methods of physical science, which are, as a rule, distin- 

 guished by their greater certainty because of their experi- 

 mental directness. To the property of melting, these charac- 

 teristics are preeminently applicable. 



The accurate ascertainment of melting-point, in terms of 

 the air-thermometer, supplies us with physical constants of 

 considerable importance. While the readings of the mercurial 

 thermometer are subject to grave correction — its zero, in 

 particular, being never stationary, — it is most unlikely that the 

 melting-point of a substance will vary in any ordinary interval 

 of time, under any common change of pressure, or with trans- 

 ition to a distant latitude. Actuated chiefly by these consi- 

 derations, I undertook the researches of which an account is 

 comprised in the following sections. 



* Communicated by the Author. 

 Phil Mag. S. 5. Vol. 14. No. 85. July 1882, B 



