Jan. 31, 1878] 



NATURE 



261 



The Radiometer and its Lessons 



Prof. Osborne Reynolds arranges his last letter (Nature, 

 vol. xvii. p. 220) ■ under four numbered heads, and in the reply 

 which I appear called on to make I will follow this division. 



I. In the first section he says, " There is nothing in ray earlier 

 papers that is 'admittedly erroneous.' If there is error in these 

 papers I am not aware of it." This is strange. In his first 

 paper {Proceedings, Royal Society, vol. xxli. p. 40) Prof. Rey- 

 nolds declares' its object to be "to point out and to describe 

 experiments to'prove that these effects (the motions observed by 

 Mr. Crookes) are the results of evaporation and condensation." 

 Now they are not the results of evaporation and condensation : 

 and it might have been seen, ab initio, that evaporation 

 and condensation could have had nothing to do with them ; 

 for evaporation and condensation can only produce a tempo- 

 rary force, ceasing so soon as the distillation is complete, 

 and cannot therefore be any part of the cause of a per- 

 sistent force such as that detected by Mr. Crookes, which lasts 

 for any length of time during which the heat is applied. In the 

 same paper Prof. Reynolds further says, "The reason why Mr. 

 Crookes did not obtain the same results within a less perfect 

 vacuum [than that of the Sprengel pump], was because 

 he had then too large a proportion of air, or non-con- 

 densing gas, mixed with the vapour, which was also not 

 in a state of saturation," All tliis is manifest error. But 

 this is not all, for the whole of the theory of those papers 

 is erroneous ; neither condensable vapour nor residual 

 gas acts in the way described by Prof. Osborne Reynolds. In 

 investigating the force arising frO'H evaporation and condensa- 

 tion, he has overlooked the circumstance that the evaporation 

 from the disc will keep back part of the vapour which would 

 otherwise have reached it, and in investigating the effect of con- 

 densation he tacitly assumes that it does keep it back. Now 

 in both cases the reverse of the assumption is what takes 

 place, and he actually arrives at the absurd result that "if 

 the opposite sides of a pith ball in vapour weie in such 

 different conditions \i.e., one surface evaporating, the other con- 

 densing] the ball would be forced towards the colder side" 

 (p. 404). His conclusion amounts to this : that the recoil of a 

 cannon would be doubled if it were struck from behind by a 

 missile at the same moment that it discharges an equal mass with 

 equal velocity forwards ! If he had not made these mis- 

 takes he would have got out only the forces which result 

 from " the perceptible motion of the vapour," which he 

 states " would be insensible" (p. 403), along with altera' ions in 

 the general tension of the vapour which would act equally on 

 both sides of the disc. Those errors vitiate the whole of his 

 mathematical reasoning, so that the value for/" which he gets, is 

 not, as he supposes, " the force arising from evaporation," and 

 his law connecting it with the heat falls to the ground. I have 

 all along supposed, from Prof. Reynolds's having long ceased to 

 mention his theory of evaporation and condensation, that he was 

 aware of some of its errors. 



The same error vitiates his reasoning in reference to the action 

 of residual gas. If the error is corrected, and if, as he assumes, 

 the gas coming up to the disc had been unpolarised (i.e. had 

 brought to the disc equal numbers of molecules, and with equal 

 velocities from all directions in front), his investigation would 

 only have given him an increase in the general pressure of the 

 gas, acting, as I pointed out in paragraph 5 of my first paper 

 \Phil. Mag., March, 1876, p. 179), equally on the front and back 

 of the disc, except during the almost inappreciable instant of 

 adjustment. Prof. Osborne Reynolds therefore ivholly missed the 

 source of the persistent force with which Mr. Crookes' s experiments 

 deal. 



2. Prof. Osborne Reynolds next says that his second paper 

 "does not conclude with his own expression of op'nion that 

 residual gas is not the cause of the force observed by Mr. Crookes." 

 In reply I have only to quote the concluding words of that paper 

 (Phil. Maq., November, 1874, p. 391). After passing under 

 leview the two agencies (condensable vapour and residual gas), 

 which he supposes are to be considered, he decides in favour of 

 the former in the following words : " hence in such cases [i.e., 

 under the conditions which he supposed to prevail in Mr. 

 Crookes's experiments] it seems to me that the effects must be 

 due to the forces of condensation." 



3. In the next section of his letter Prof. Reynolds states that 

 Clausius and Maxwell "established the law that the only condi- 

 tion of thermal equilibrium in a gas is that of uniform temf era- 

 ture." I am not aware that they have ever established this law. 



The converse of it is obviously true, and has often been used, 

 and the law itself has sometimes been assumed, but has never, 

 so far as I know, been proved. I am, however, disposed to 

 concur with those who think that it is probably true, and the 

 conclusion in my paper on penetration (which is the reverse of 

 that attributed to me by Prof. Reynolds) is in conformity with it. 

 My conclusion is expressed in the following words (Phil. Mag., 

 December, 1877, § 4) : — " Hence there must, in the cases that 

 really arise, be some escape of heat which may be small but 

 cannot vanish." And, I may remark, there will, according to 

 my view, be two other sources of escape of heat, viz., conduction 

 by diffusion, which was excluded from my investigation ; and 

 conduction by radiation, which was excluded both from Clausius's 

 investigation (Phil. Mag., June, 1862, p. 422, footnote) and from 

 mine. 



Prof. Osborne Reynolds a second time objects to my having 

 excluded conduction when investigating the penetration of heat. 

 As he attaches weight to authority he will perhaps be reconcded 

 to my doing so, by the example of Clausius, as cited above, and 

 by his justification of it in the following words (loc.cit.) : — " In 

 any case, however, it is allowable to consider separately each of 

 these two ways in which heat moves." 



Before passing from this subject I wish to take the opportunity 

 of stating that Dr. Schuster's letter (Nature, vol. xvii. p. 143) 

 has satisfied me that I have hitherto erred in my estimate of the 

 relative efficiency of penetration and conduction as agents for 

 conveying heat. I am now convinced that penetration is 

 usually feeble compared with conduction, and, in the figures re- 

 presenting De la Prevostaye and Desains' experiments, is to be 

 sought in those portions of the curves which slope steeply down- 

 wards. The second part of my paper on penetration, that in 

 which I apply the theory to experiment, will accordingly require 

 considerable modification ; and some of the statements which I 

 made in my papers on Crookes's force will need amendment. 

 The corrections that are required do not, however, affect any of 

 the material parts of my theories of Crookes's force and of pene- 

 tration, which depend essentially on the fact that there is a layer 

 in the gas extending to a limited distance from a heater or cooler, 

 throughout which the effects of the discontinuity in the gaseous 

 motions at the surface will be felt, and that within that layer the 

 stresses and the communication of heat follow special laws. 



4. I have to express my great satisfaction at the explicit 

 admission made by Prof. Osborne Reynolds in the fourth section 

 of his letter, in the following sentences : — "There is one state- 

 ment in Mr. Stoney's letter which is not erroneous. He says, 

 * I cannot find anywhere in Prof. Osborne Reynolds's writings an 

 explanation of the thing to be explained, viz., that the stress in 

 a Crookes's -layer is different in one direction from what it is at 

 right angles to that direction.' I [Prof. Osborne Reynolds] do 

 not at all admit that this is ' the thing to be explained,' and I 

 am quite sure that Mr. Stoney would find no explanation of it 

 in my writings." This admission disposes finally of all con- 

 troversy as to priority between us. 



I need hardly, after this admission, follow Prof. Osborne 

 Reynolds through the rest of his letter. His supposed invariable 

 law " that the [Crookes's] force always tends to drive the vanes 

 or bodies in the direction of their colder faces," does not seem 

 to be true. A familar exception occurs when a spheroidal drop 

 is supported over a platinum dish. The Crookes's force actii.g 

 upon the platinum dish is equal to the weight of the drop, and 

 acts downwards, i.e. in the direction of ih.Q hottest surface of the 

 dish. 



In applying his hypothetical case of a heater and cooler, a 

 and B, within an envelope of intermediate temperature, to prove 

 that "the force that causes the motion in the bodies cannot be 

 due " to the stresses of my theory, he' has overlooked the very 

 obvious circumstance that the envelope, as well as B, is a cooler 

 in reference to A, and the envelope, as well as A, is a heater la 

 reference to B. 



Prof. Reynolds observes that I have not defined polarisa- 

 tion. I described the kind of polarisation that exists in 

 radiometers in my first two papers, and I will give a formal 

 definition of the term as applied generally to gases in an article 

 which I am preparing, and which I hope will be admitted into 

 the pages of Nature, giving as clear an account of my theory as 

 I can, compatibly with brevity and the omission of mathematics. 



The way in which Prof. Reynolds has excluded polarisation 

 from his explanation is by assuming that the state of the gas 

 close to the heated disc may be adequately represented by un- 

 polarised gas of one temperature coming up to the disk, and 

 unpolarised gas of another temperature leaving it, i.e., by mole- 



