218 Mr. W. H. Steele's Thermoelectric Diagram 



is to be regretted that observations should be published, 

 and conclusions drawn which are affected by errors easy to 

 avoid by the use of pure vapours from liquids boiling under 

 constant known pressures as a means of obtaining constant 

 known temperatures. A little trouble taken at the beginning 

 of a research in purifying the liquids employed as jackets is 

 well-rewarded in the avoidance of erroneous observations 

 during the progress of the research ; and inasmuch as many 

 mathematical physicists are endeavouring to find some ex- 

 pression which will better fit facts than van der Waals' simple 

 but approximate formula, it appears a pity that untrustworthy 

 observations should be offered as a groundwork for their 



analysis. -rr r -r, 



William Ramsay. 



Sydney Young. 



XVIII. Thermoelectric Diagram for some Pure Metals. 

 By W. Huey Steele, M.A., Melbourne* . 



SINCE 1873, so far as I am aware, no work on this subject 

 has been published. In that year Professor Tait, in the 

 Transactions of the Royal Society of Edinburgh, published 

 " A First Approximation to a Thermoelectric Diagram,'''' and 

 stated that it was only an approximation and that other re- 

 sults would follow. The results in the paper above have since 

 been accepted as the most reliable. In his paper he gives 

 very little experimental detail, such as the method of observa- 

 tion, sensitiveness of galvanometer, &c, and the greater part 

 of the paper is taken up with the behaviour of the iron line. 

 In making some thermoelectric observations I had reason to 

 suspect that his diagram was not quite right, and further 

 measurements showed that the copper line was far from its 

 right position. I determined therefore to construct a diagram 

 for as many pure metals as I was able to obtain. 



Whatever method of observation is adopted, one essential 

 in thermoelectric observation is an exceedingly delicate gal- 

 vanometer, that one- being the best which indicates the least 

 electromotive force applied to its terminals, not the one 

 indicating the least current. Of the various sensitive galvano- 

 meters at my command, the best was a one- coil, low-resistance 

 (about half an ohm), astatic instrument made by Elliot 

 Bros. The needles, about 45 millim. in length, were 



* Communicated by the Author. 



