August 3, 19 16] 



NATURE 



46) 



LETTERS TO THE EDITOR. 

 [The 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.] 



Productive Work and Classical Education. 



At this time people are awakening to the mischief 

 that has been done to this country by the neglect of 

 science as a part of education, and there seems a 

 danger of the pendulum of opinion swinging too far, 

 and of classics being looked upon as something to be 

 completely eliminated from the educational curri- 

 culum. In relation to this, I think a short personal 

 anecdote may be instructive. In 1868 I had the privi- 

 lege of working with the late Prof. Willy Kiihne as 

 his only student in his laboratory in Amsterdam, and 

 the friendship which began there continued up to the 

 time of his death. 



Prof. Kiihne was a most remarkable man. He 

 was, I think, one of the greatest physiological chemists 

 of last century, and was quite half a century in ad- 

 vance of nearly all his contemporaries. Belonging to 

 a rich banking family, he could go where 

 he pleased, do what he pleased, and obtain 

 any optical or other apparatus he needed, 

 regardless of cost. He accordingly elected to work 

 with Claude Bernard, and used the chemical and 

 microscopical skill which he acquired to such advan- 

 tage that at an age when most men are only thinking 

 of beginning university life he had produced a mono- 

 graph on protoplasm and contractilit}^ (" Ueber Proto- 

 plasma und Contractilitat "), which was not only far in 

 advance of anything then in existence when it was 

 written, but still remains unrivalled half a century 

 later. 



His great ability led to an invitation to become pro- 

 fessor of physiology at Amsterdam. After some years 

 he was invited to occupy the chair at Heidelberg 

 rendered vacant by the transference of Prof. H. von 

 Helmholtz to Berlin. This invitation he accepted, and 

 remained at Heidelberg until his death. 



Such a career seems ample vindication of the claim 

 that classics is unnecessary to education, more 

 especially if it be borne in mind that Kiihne was an 

 exceptionally good linguist, si>eaking three or more 

 languages with perfect ease, that he had travelled 

 much in Euroj>e, and was a perfect encyclopaedia of 

 knowledge and criticism in painting and sculpture. 

 Yet there was one bitter drop in his cup of know- 

 ledge and honour. The nature of this was confided 

 to me as a strict secret by our mutual friend. Prof. 

 Hugo Kronecker, when we were discussing together 

 some data for a short life of Kuhne which Kronecker 

 thought of writing. As both Kiihne and Kronecker 

 are dead, there is no further reason for preserving the 

 secret, which I for one never could have suspected. 

 It was that Kiihne had felt deeply the scorn with 

 which some people had regarded him because he had 

 never taken a classical degree. Fools they were no 

 doubt, but their attitude probablv indicated the mental 

 attitude of the mass of German graduates to whose 

 devotion to a scientific education we are now inclined 

 to attribute much of Germany's success. 



Lauder Brunton. 



1 De Walden Court, New Cavendish Street, 

 London, W., July 15. 



Gravitation and Temperature. 



Dr. P. E. Shaw's striking experimental result 

 (Phil. Trans., 1916) as to a variation of gravitational 

 attraction with temperature of the large mass, and 

 NO. 2440, VOL. 97] 



that of Poynting and Phillips as to no variation in 

 attraction with temperature of the small mass, may 

 seem reconciled satisfactorily by the formula put for- 

 ward by the latter collaborators, and quoted by Dr. 

 Shaw in Nature (July 13), viz. : — 



Mm 



^=K-4r:r)^ • • • <■> 



where T and t are the absolute temperatures of the 

 masses M and m respectively, placed at a distance r 

 apart. But it seems desirable to notice that this 

 formula does not in general allow of the derivation of 

 the attraction of a finite mass from the attractions of 

 its component particles in the usual ^ay by vector 

 addition. 



Thus, for a pair of particles, each of mass m, at 

 temperatures T and t, and placed r apart, we have 

 as the attraction : — 



F,=c(.+Jf')-' .... (a) 



Again, the attraction of two particles, each of mass 

 m, close together, and at temperature T, on a single 

 particle of mass m and temperature f at a distance r, 

 would be : — 



F, = G{i+.iniy^, ... (3) 



Hence, F, is not, in general, equal to 



2F, (4) 



For the effective temperature of the system varies be- 

 tween those of the particles, according to their relative 

 masses, just as the position of the centre of mass of 

 a system varies among those of its particles accord- 

 ing to their masses. 



Accordingly, the component attractions do not sum 

 to their resultant in the usual way. 



Of course, this is no disproof of the formula, but 

 must be regarded simply as a somewhat grave conse- 

 quence involved by the formula. It is indeed a conse- 

 quence that may well give us pause before accepting 

 the formula, pending either (a) a rigorous deriva- 

 tion of the formula theoretically, or (b) some crucial 

 experimental evidence that it is preferable to other 

 formulae. 



Suppose, instead of formula (i), we tr\- the follow- 

 ing :— 



F^G(i+a^^(-^^y»+^). . (la) 



where, as before, T and t are the absolute tempera- 

 tures of the masses M and m, and 6 is the mean, or 

 effective, temperature of the ^ace, whether vacuous 

 or not, between the masses. 



It is to be noted that, with Max Planck's theory of 

 entropy, a temperature is now theoretically assign- 

 able to a vacuous space which is a field of radiation. 

 Using this different formula for the cases already 

 considered, if one particle at temperature t is attracted 

 by one or two particles at temperature T, we have 

 the relations : — 



Fi = G(i+o^(i+/3T)(i+)5/)^* . (m) 



F,=G(i+a<>)(i+^T)(i+/3/)?'«' . (3a) 



So here, F,=2Fi {4a) 



And, however we vary the mass at temperature T, 

 provided the temperature 6 remains unchanged, the 

 attraction on the single particle would vary in direct 

 proportion to the attracting mass. 



This new formula, then, restores the validity of die 

 vector addition of the component attractions. It 

 seems, however, at first sight to have lost the power 



