Heat across Layers of Gas. 443 



ported with but a slight difference of temperature between it 

 and the liquid on which it rests. In this respect the spheroidal 

 state on liquids differs from that in which the drop rests upon 

 a heated solid. The difference of behaviour is probably due 

 to the deformation of its natural spherical shape to which a 

 drop is compelled to submit when it rests on a rigid surface. 

 Owing to this constraint the surface-tension over the drop 

 will force some parts into closer contact ; and, moreover, the 

 vibrations which always arise in this case must tend to a 

 similar result. On the other hand, when the drop is resting 

 on a liquid, it settles tranquilly into a beautiful concave socket 

 that can be seen by looking at the surface of the fluid from 

 beneath. This socket allows the globule to retain a nearly 

 spherical and therefore unconstrained form ; and, accordingly, 

 the opposed surfaces come within an approximately equal 

 distance of one another throughout a large arc. And it is 

 evident that as the whole pressure arising from the molecular 

 motions in the air would support a column of spirits of wine 

 11^ metres high, it needs only a very moderate Crookes's 

 polarization of the stratum of air to furnish that slight pre- 

 ponderance of momentum in a determinate direction which is 

 required to support the few additional millimetres that cor- 

 respond to the weight of the drop. We must remember, too, 

 that a drop may, under favourable circumstances, outlive the 

 difference of temperature, because it would take a sensible 

 time for so slight a pressure as the weight of the drop to 

 squeeze the film of air, when once established, out of its narrow 

 chink. Even a heavy metal proof-plane will float over another 

 proof-plane, upon the stratum of air entangled between them, 

 for a considerable time. 



22. Professor Barrett has called my attention to another 

 unexplained phenomenon, of which we can now see the cause, 

 viz. the mobility imparted to a very fine powder (as, for 

 example, magnesium carbonate or precipitated silica) by heat- 

 ing it in a metal dish. When the dish is disturbed the powder 

 glides about as if floating ; and it is in fact floating on the 

 compressed Crookes's layer, which will spring into existence 

 whenever the powder is able by radiation to maintain a lower 

 temperature than the dish. 



23. The communication of heat by penetration is a very 

 familiar phenomenon ; for when surfaces at different tempera- 

 tures are brought into what is commonly called contact, there 

 is usually a thin intervening stratum of air, except at special 

 points ; and, accordingly, the greater part of the transfer of 

 heat, so long as the difference of temperature is considerable, 

 must be effected, not by contact, but by penetration across a 

 Crookes's layer. 



