April 14, 192 1] 



NATURE 



20T 



Letters to the Editor. 



\T\ie Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous communications.] 



"Space" or "^ther"? 



Your readers are indebted to Mr. Bonacina's letter 

 in Nature of April 7 for a very clear statement of a 

 fundamental pomt in the relativity controversy, and 

 it is important that the views held with regard to it 

 should be clearly understood. Ihe issue is stated con- 

 cisely in the sentence "the relativists seem now . . . 

 to indicate that space, instead of being conditioned by 

 matter, is itself the foundation of matter and physical 

 forces." Now it seems clear that if any relativist 

 expresses himself in terms like these he cannot be 

 regarding space as mere emptiness or as the arbitrary 

 co-ordinate system of the pure mathematician ; for 

 him it is the substitution of matter, light, and electric 

 force — that is to say, it is the thing which most of us 

 call aether. Since it is not matter, it has not (and 

 we ought not to expect it to have) the material pro- 

 perties of density, elasticity, or even velocity ; but it 

 has other dynamical attributes, measured by tensor- 

 expressions, which stand in much the same relation 

 towards it that mass and strain do towards matter. 

 It is, in short, a physical medium. It is sometimes 

 stated that the relativity theory does away with the 

 aether; the defence of this statement must be left to 

 those who make it; I do not think it is the view of 

 Prof. Einstein. It seems more reasonable to say that 

 relativity has added to the importance of the aether 

 bv enlarging its functions. 



" But it must not be thought that the whole issue 

 reduces to a question of terminology. It will 

 naturally be asked : How can those who believe in 

 a physical aether regard gravitation and electro- 

 magnetic phenomena as the " outcome of the geo- 

 metrv of the universe "? The phrase is Prof. Weyl's, 

 and "reference to his book, " Raum, Zeit, Materie," 

 shows that he believes in a physical aether, and does 

 not mind saying so. "We shall use," he says, "the 

 term ' condition of the world-aether ' as synonymous 

 with ' metric ' in order to intimate the real character 

 of the metric." We must recall that the geometrical 

 quantitv called "distance" is none other than the 

 materia'l or aethereal attribute of "extension," as Mr. 

 Bonacina admits. Thus experimental geometry, 

 which comprises the study of distances, is the science 

 of the aether so far as its attribute of extension is 

 concerned. The sentence then means that not only 

 the phenomena immediately recognised as spatial, but 

 also mechanical and electrical phenomena, fall into 

 place in a complete development of the theory of 

 extension — a truly remarkable discovery. They do 

 not introduce any other attribute of the aether. I 

 think it is because physical science is confined to this 

 one attribute of the substratum of the universe that 

 such qualities as beauty lie outside its scope. 



The statement that the phenomena of mechanics 



are the outcome of the geometry of the world implies 



P the complementary statement that the phenomena of 



experimental geometry are the outcome of the 



mechanics of the world. Either form expresses the 



central truth of the generalised relativity theory, but 



the great advance lies not so much in the conception 



1^ of the idea as in the discovery of the key to this 



B unification of geometry and mechanics. The unifica- 



» tion leaves us with a redundancy of names, and ap- 



B parently there is some divergence of view as to the 



B NO. 2685, VOL. 107! 



right name for the fundamental substratum of every- 

 thing. Since it is the medium the condition of which 

 determines light and electromagnetic force, w^e may 

 call it aether ; since it is the subject-matter of the 

 science of geometry, we may call it space ; sometimes, 

 in order to avoid giving preference to either aspect, 

 it is called by Minkowski's term world. 



A. S. Eddington. 

 Observatory, Cambridge, April 11. 



"Absolute" Temperatures in Meteorological 

 Publications. 



In a note in Nature of March 31 referring to one 

 of the publications of the Meteorological Office occurs 

 the remark: "The normal constant for absolute 

 temperature given is 200°. With a normal constant 

 of 273° the resulting values would be in ordinary 

 degrees Centigrade, a system adopted by niany 

 meteorologists on the Continent and by some at home. 

 To the uninitiated it gives a reading more easily 

 coniprehended, although . . . some of the values 

 would be given with the negative sign." Whatever 

 may be the meaning of the "normal constant" for 

 any scale of temperatures, may I express my dis- 

 agreement with the opinion, and give a reason for 

 doing so? Premising that, when dealing with the 

 upper air, it is not the few, but the great majority of 

 readings that have the negative sign on the Centi- 

 grade thermometer, my reason is that, to the un- 

 initiated, negative values are not an aid, but a terrible 

 obstruction to comprehension, because their use im- 

 plies a process of thinking in two directions, upwards 

 and downwards, at the same time, and keeping the 

 two trains of thought distinct. 



Looking into a well-known historical work a few 

 days ago, I came across a perfect analogy of that im- 

 perfect system of measurement, one which expresses 

 the difficulty very clearly. The author wrote of some- 

 thing as taking place "at the end of the third cen- 

 tury B.C." He was counting time as your annotator 

 would have us count temperature. I understand 

 the time-reference to mean "towards 200 B.C.," the 

 end of the third century being the beginning, not of 

 the fourth, as the ordinary process of measurement 

 would suggest, but of the second. If you substitute 

 "the third degree Centigrade below zero" for "the 

 third century B.C.," you have the same difficulty. It 

 is obvious that in dealing with a single degree, as 

 for a single century (for purposes of estimation of a 

 fraction, for example), you may have to think 

 upwards ; but in dealing with a number of degrees 

 or centuries you think downwards. This does not 

 make for easy comprehension, and the only possible 

 excuses for exposing an uninitiated reader or observer 

 to the risks of such a system are either that there is 

 no alternative — or that fractions of a degree do not 

 really matter anyway. 



In order to make comprehension easy you have, 

 in fact, to become initiated in the practice of standing 

 on your head ; and no doubt after years of practice 

 it becomes easier to stand on your head than to alter 

 the zero of your own thermometer. But the un- 

 initiated ought not to be prayed in aid of the practice. 

 They will not find it anything like so easy as a hoary 

 initiate like myself. 



Incidentally, let me say that I know no meteoro- 

 logists at home who habitually use the Centigrade 

 scale. Many physicists do so; but, being "initiated," 

 they skip quite lightly into the absolute scale when 

 they want to deal with thermodynamics or radiation 

 or any other of the applications of physics that go 

 bevond the mere quotation of a temperature ; they 

 skip back again just as easily to Centigrade when 

 the job is done. Skipping from one svstem of Units 



