424 Mr. G. J. Stoney o/i the Penetration of 



of the earth's surface which might be drawn from the great 

 absorptive powder of the air for rays of low refrangibility, such 

 as are largely present in the solar rays, and conclude this paper 

 with a short summaiy of the chief results of my experiments. 



The thermal conductivity of hydrogen and of other gases 

 is far too small to admit of its being proved by the method 

 Magnus adopted. The assumption that the conductivity of 

 hydrogen is similar to that of the metals, if by this statement 

 any thing more is meant than that hydrogen, like solid and 

 liquid bodies, is capable of transmitting heat from molecule 

 to molecule, is therefore not justified. 



On the other hand, hydrogen possesses a diathermancy 

 closely approaching that of a vacuum. 



Dry air absorbs from 50 to 60 per cent, of the rays of heat 

 which it receives from a source heated to the boiling-point of 

 w^ater. 



The absorptive power of moist air surpasses that of dry air 

 by a trifling percentage, but by no means to such a degree as 

 hitherto had been assumed by several physicists. 



Rock-salt is not absolutely diathermanous for the so-called 

 dark rays of heat ; its thermal colour rather resembles that of 

 dry air. 



Giessen, March 30, 1876. 



LVI. On the Penetratio7i of Heat across Layers of Gas. By 

 G. Johnstone Stoney, M.A,, F.E.S., ^c, Secretary Royal 

 Duhlin Society*. 



Pakt I. Theory. 



1. T"| EAT W'ill pass between bodies at different tempera- 

 JLi_ tures by direct contact, by radiation and absorption, 

 or by contact wdth a fluid and convection through it. That 

 heat may be transported in these several w^ays has long been 

 knowm ; and the laws of the transfer have been made the sub- 

 ject of repeated and careful investigation by experiment and 

 by the deductive method. And last year two papers f were 

 published by the author of the present memoir, in which it 

 was shown that heat will also escape, under new conditions, 

 across a Crookes's layer, if the layer be restricted in width. 

 In those papers the mechanical actions that arise, and upon 

 which Mr. Crookes had made many experiments, were made 

 the subject of study ; and the present communication aims at 



* From the Scientific Transactions of the Royal Duhlin Society for 

 1877. Communicated hy the Author. 



t See Phil. Mag. for March and April 1876. 



