438 



Mr. G. J. Stoney on the Penetration oj 



everywhere below the upper line of the figure, and probably 

 lies considerably above the curve representing the loss by con- 









—I 



Fic^A 



OU,rA<Jes,ant of htat imthhydr 





■i 

 i 



uler, 



1 







^\ ' 



1 



\\ 1 liilipilijlllilllll 



wmmm Ml 



cLcross 1 -T 





■j 



'the c¥' 



j 

 nder- 





' 



1 ! iiljllllllllllljlill i 



hydrogen 



1 1 n r 11 - - 





p^-"' 



] 









« 



f 1 fPfor'^ 



i?«e to 



!_____ — 



! 



ece^rer. 



-— 





; 





coni/-e.otioii. 



















c 



) 100 200 300 400 500 600 It 



; • ^ .i 



?^ JV«5Z 



ons 







■ . J , ; ■ ■ 1 ■ -i 



_J 



vection in air ; it therefore occupies some such position as the 

 dotted line of the figure. The shaded interval between this 

 dotted line and the upper line of the figure represents the 

 enormous escape of heat to be attributed to the penetration of 

 heat across two centimetres of hydrogen. 



16. The observations previously made by Dulong and Petit 

 on the rate of cooling in hydrogen come here to our aid. The 

 interval between the walls of their receiver and the bulb was 

 12 or 14 centims. ; and they found that, after allowing for 

 radiation, the rest of the escape of heat within this receiver 

 charged with hydrogen was proportional to p^'^^ ; whereas in 

 the other gases examined it was proportional to p in a power 

 which differed slightly from one gas to another, but was in 

 none far from \. The escape of heat with which they were 

 here dealing, Dulong and Petit supposed to be wholly due to 

 convection ; but we have now reason to suppose that, when 

 the gas was hydrogen, penetration contributed largely to it, and 

 was the reason why the index of p in Dulong and Petit's empi- 

 rical formula was so different with hydrogen from its value 

 with other gases. Accordingly, if we plot down the curve 



