1886.] 



On the Thermal Properties of Ethyl Oxide. 



381 



rubidium, potassium, sodium, lithium, barium, strontium, and calcium. 

 A series of similar liquids to those of one of the groups of acids, of 

 equal (not of equivalent) strength to each other, was also included. 



As electrodes he employed pairs of plates of zinc, cadmium, lead, 

 tin, iron, nickel, copper, silver, gold, palladium, and platinum ; and 

 separate ones formed of small bars of iridium. 



He took each group of solutions, and measured in each liquid 

 separately at atmospheric temperature, the " total resistance" at the 

 two electrodes, and the separate " resistances " at the anode and 

 cathode respectively with each metal, and thus obtained about seventy 

 different tables, each containing about thirty-six measurements, 

 including the amounts of "total," " anode," and " cathode resistance" 

 of each metal, and the averages of these for all the metals. 



By comparing the numbers thus obtained, and by general logical 

 analysis of the whole of the results, he has arrived at various conclu- 

 sions, of which the following are the most important : — The pheno- 

 menon of " transfer-resistance " appears to be a new physical relation 

 of the atomic weights, attended by inseparable electrolytic and other 

 concomitants (one of which is liberation of heat, " Phil. Mag.," 1886, 

 vol. xxi, p. 130). In the chemical groups of substances examined, it 

 varied in magnitude inversely as the atomic weights of the constituents, 

 both electropositive and electronegative, of the electrolyte, independently 

 of all other circumstances; and in consequence of being largely 

 diminished by corrosion of the electrodes it appears to be intimately 

 related to " surface-tension." He suggests that corrosion may be a 

 consequence and not the cause of small "transfer-resistance." The 

 strongest evidence of the existence of the above general law was 

 obtained with liquids and electrodes with which there was the least 

 corrosion and the least formation of undissolved films ; those liquids 

 were dilute alkali chlorides, with electrodes of platinum. 



The research is an extension of a former one on " Transfer-resist- 

 ance in Electrolytic and Voltaic Cells," communicated to the Royal 

 Society, March 2, 1885. Further evidence on the same subject has 

 been published by the author in the " Phil. Mag.," 1886, vol. xxi, 

 pp. 130, 145, 249. 



II. " A Study of the Thermal Properties of Ethyl Oxide." 

 By William Ramsay, Ph.D., and Sydney Young, D.Sc. 

 Received May 5, 1886. 



(Abstract.) 



A year ago, a paper was communicated to the Society on the 

 behaviour of ethyl alcohol when heated. A similar study of the 

 properties of ether has been made, in which numerical values have 



