TRANSACTIONS OF THE SECTIONS. 33 



On the other hand, when the temperatiu'es were made to vary, the result was 

 found to be proportional to the absolute temperature. 



These experiments do not show that there is no such thing as fethereal friction — 

 that is to say, friction from somethinjj which fills all space and is independent of 

 air; but we may argue from them that such an jethereal friction must either have 

 been nearly insensible in these experiments, or it must, as weU as the friction from 

 the gas, have -\-aried with the absolute temperatures, in which case the two frictions 

 would not be separable from one another by the method of the experiment. 



Prof. Tait and myself have made some experiments upon the heating of a disk 

 by rapid rotation in vacuo. In these experiments we found a mere surface-heating 

 due to air, which vaiied not only with the quality but also with the quantity of the 

 residual gas ; and we also found a surface-etiect (more deeply seated, however, than 

 the former) which appeared to be a residual effect, and which it is possible may be 

 due to sethereal fi-iction. We made no experiments at varying temperatures ; but 

 we made use of various residual gases, and found that' the heating-effect for 

 carbonic acid was perhaps a tride less than for air, while that for hydrogen ap- 

 peared to be about four times less than that for air. Now, comparing Prof. Max- 

 well's experiments with ours, we ha^•e in the former a stoppage of motion, which is 

 rather less for carbonic acid than for air, and about half as large for hvdrogen as 

 for air. In the latter, again, we have a heating-effect rather less for carbonic acid 

 than for air, and only about one fourth as large for hydrogen as for air. Thus it 

 appears that the stopping effect of hydrogen in Prof. Maxwell's experiments is re- 

 latively greater in comparison with a'ir than is its heating-effect in our experiments 

 when compared with that of air. The effects of these various gases would bear to 

 one another more nearly the same proportion in both experiments if we might 

 suppose that in Prof. Maxwell's experiments there was mixed up with gaseous 

 friction a very sensible ajthereal friction; but in that case it would be necessary to 

 suppose that the nethereal friction was proportional to the absolute temperature. 



During the M eeting of the British Association at Edinburgh, I brought before the 

 Association reasons for imagining that if we have a body in visible motion in an 

 enclosure of constant temperature the visible motion of the body will gradually be 

 changed into heat. The nature of the argument was such as to render it probable 

 (although not absolutely certain) that in such a case the rapidity of conversion will 

 be greater the higher the temperature of the enclosure. 



1 will now refer to some experiments of Prof. Tait, which formed the subject of 

 the last Rede Lecture. These experiments were suggested to Prof Tait by an 

 h_^T3othesis derived from the theory of the dissipation of energy, which led him to 

 think that the resistance of a substance to the conduction of electricity, and also of 

 heat, would be found proportional to the absolute temperature. Matthiessen and 

 Von Bose in the case of electricity, and Principal Forbes in the case of heat, had 

 already proved that, as a matter of fact, the law was not very different from that 

 imagined by Prof. Tait. The result of these experiments has"^ been to confirm the 

 truth of this law. 



The following considerations, also connected with the dissipation of euergv, point 

 to the same conclusion.^ Perhaps we may regard the nsthereal medium as "that 

 medium whose office it is to degrade all directed motion and ultimately convert it 

 into univei-sally difliised heat, and in virtue of which all the visible ditierential 

 motion of the universe will ultimately be destroyed by some process analogous to 

 friction. 



Now in order to imagine the way in which aether may possibly act in bringing 

 about this result, let us imagine some familiar instance of directed motion, as, for 

 instance, a railway-train in motion. The train, let us suppose, and the air in it, 

 are both in rapid motion, while the air outside is at rest. Now as the train pro- 

 ceeds, suppose that a series of cannons loaded with blank cartridges are fired 

 towards the train. A series of violent sounds will go in at the one window and 

 out at the other of each carriage. Each sound will push some air from the stratum 

 of air at rest into the carriage on the one side, and it will push some air from the 

 carriage into the stratum at rest on the other side. Now in this operation it would 

 seem that part of the visible motion of the train must be taken from it. To make 

 a comparison, it is as if a series of individuals were jumping into the train at the 



l'^73. 3 



