242 Barns and Iddings — Electric conductivity ohs, rrrd 



the air was admitted to the atomizer under a pressure of about 

 one-third of an atmosphere above the ordinary. 



Certain experiments made to determine whether ferric 

 chloride or chromic acid might be used in the purifier instead 

 of dilute nitric acid showed that the superficial effect of these 

 reagents upon the extremely finely comminuted mercury pre- 

 vents the reunion of the particles even when submitted to 

 considerable pressure. When dilute nitric acid is employed 

 no such difficulty arises. 



Art. XXXII. — Note on the change of electric conductivity 

 observed in rock magmas of different composition on pass- 

 ing from liquid to solid / by Carl Barus and Joseph P. 

 Iddings. 



[Published by permission of the Director of the U. S. Geological Survey.] 



§ 1. Introdxiclion. — The chemical composition of rock mag- 

 mas and the process and results of their crystallization and 

 solidification have been studied with more or less thoroughness ; 

 but direct investigation of their physical properties has not been 

 systematically undertaken until recently. 



Conceptions of the physical nature of molten rock magmas, 

 however, have been obtained by observing the phenomena of 

 their crystallization, and by pointing out analogies between the 

 latter and the behavior of aqueous solutions, and of metallic 

 alloys. 



Evidences that a chemico- physical differentiation of molten 

 magmas was the cause of chemical differences in igneous rocks, 

 which evidences have been reviewed by one of us in a paper on 

 the origin of such rocks,* have led to the further analogy 

 between this process of differentiation and the concentration 

 and partial dissociation of salts in aqueous solution. They also 

 suggested the possibility of testing the fact of dissociation by 

 electrolytic investigation. 



The method employed was the determination of the electric 

 resistance encountered by a current in passing through fused 

 rock magma, whose temperature was gradually lowered. The 

 investigation, though of a preliminary character has been pro- 

 ductive of such definite results that we have embodied them 

 in the present paper in order that they may become known as 

 early as possible. 



§ 2. Observations. — The following table contains typical ex- 



* J. P. Iddings. The origin of igneous rocks, Bull. Phil. Soc. Washington, 8°. 

 Washington, 1892, vol. xii, pp. 89-214. 



