200 



NATURE 



[April 15, 1920 



similar to those of the Sixth Cataract, about forty 

 miles north of Khartum, where the volcanic rocks are 

 certainly older than the Nubian Sandstone. In the 

 northern parts of Dongola basalt intrusions occur in 

 the Nubian Sandstone, and there is a hot spring at 

 Akasha, about eighty miles south-west of Haifa. 

 Turning to more distant regions, one of the solitary 

 landmarks on the White Nile is Jebel Ahmed Aga, in 

 latitude ii° N., consisting of the remains of a vol- 

 canic cone formed of basaltic scoria and evidently of 

 comparatively recent age. Towards the east there are 

 the plateau basalts of Abyssinia, with outliers extend- 

 ing into the Sudan. Along the Langeb Valley, north 

 of Kassala, there is an interesting suite of acid and 

 intermediate volcanic rocks, but we are still in doubt 

 about their age. Similar rhyolites certainly occur 

 farther north among the Red Sea hills. The western 

 parts of Kordofan have been traversed geologically 

 without revealing the existence of volcanic rocks on 

 the continuation of the line referred to by Mr. Camp- 

 bell Smith. Farther west Darfur appears to be full 

 of recent volcanic rocks, principally of scoriaceous 

 types. 



The N.E.-S.W. features seen by Dr. Chalmers 

 Mitchell may have been to some extent due to erosion 

 by sand driven from the N.N.E. by the prevalent 

 wind. The direction of strike among the meta- 

 morphic rocks is another factor to be borne in mind. 

 It is not constant over these large areas, but it is 

 very often N.E.-S.W., and would then account for 

 some of the features seen from the air. In these 

 circumstances caution appears desirable in basing wide 

 structural theories on rather scanty data. 



G. W. Grabham. 



Box 178, Khartum, March 25. 



The FitzGerald-Lorentz Contraction Theory. 



In the discussion on relativity at the Royal Society 

 on February 5 (Nature, February 12), Mr. Jeans stated 

 that the FitzGerald-Lorentz contraction theory pre- 

 sented grave difficulties in the case of a wheel rotating 

 about a fixed axle, so that the circumference would 

 contract while the radius would not. Surely these 

 difficulties are not so grave as would appear at first 

 sight? Let us adopt the point of view of the old- 

 fashioned non-relativist to whom space is rigid and 

 Euclidean, even though his measuring instruments may 

 change and so introduce errors in his measurements. 

 A scale is not a rigid invariable unit of length. Its 

 length, even if its orientation is unchanged, depends 

 on its temperature and the tensile or compression 

 stresses to which it is subjected. If we change its 

 temperature, keeping the stresses constant, its length 

 (as measured by a standard scale at fixed temperature) 

 varies. But we may by suitable means prevent the 

 variation of length, in which case the change of tem- 

 perature will cause a change of stress. Similarly, on 

 the FitzGerald-Lorentz theory, turning the scale to a 

 different orientation relative to the supposed aether 

 stream causes a change in the electric forces to which 

 the cohesion of the molecules is ultimately due, so 

 that if the temperature and the external stresses 

 remain constant, the length changes. In this case, 

 however, we cannot detect the change airectly, as it 

 would be necessary to turn our standard scale also, 

 and it^ too, would change. If for any reason the 

 change of length is prevented, the FitzGerald-Lorentz 

 effect causes a change of stress. 



Now in the case of the rotating wheel the ratio of 

 the circumference to the radius must remain con- 

 stant, sp that any, tendency of one to change its dimen- 

 sions will affect the other, with the result that both 



NO. 2633, VOL. 105] 



circumferential and radial stresses will be set up, 

 and any changes of length caused must be compatible 

 with the constancy of ix. These stresses would in any 

 actual case be almost vanishingly small compared 

 with those due to centrifugal force, so that the only 

 effect of the FitzGerald-Lorentz contraction would be 

 to alter the latter stresses to an utterly negligible 

 extent. Horace H. Poole. 



Physical Laboratory, Trinity College, 

 Dublin, March 19. 



Moseley Memorial. 



The fund founded in the University of Manchester 

 for the provision of a memorial to the late H. G. J. 

 Moseley (killed in action at Gallipoli, 1915), and 

 originally proposed as a private memorial from 

 Moseley 's personal friends and fellow-workers in 

 Manchester, has now been extended in order to give 

 other scientific bodies, both ip England and abroad, 

 an opportunity of participating. This extension has 

 been made at the request of a number of scientific 

 men interested in Moseley 's work, but not personally 

 connected with him, and it is in order to reach this 

 wider public that you are asked to publish this letter. 



The scheme of memorial proposed is (i) the provi- 

 sion of a memorial tablet in the physical laboratory 

 and (2) the foundation of a Moseley prize or medal for 

 physics in the University of Manchester. 



The fund is administered by a committee consisting 

 of Sir Henry A. Miers (chairman). Profs. W. L. 

 Bragg and H. B. Dixon, Sir E. Rutherford, and Dr. 

 E. J. Evans. 



Subscriptions, which should be made payable to the 

 "Moseley Memorial Fund," and crossed' "Williams 

 Deacon's Bank, Ltd.," may be sent to either of the 

 hon. secretaries, Mr. C. G. Darwin, Christ's College, 

 Cambridge, or Dr. H. Robinson, Physical Laboratory, 

 University of Manchester. 



About 170Z. has already been received, comprising 

 donations from Great Britain, Canada, the United 

 States, and France (including contributions from the 

 Soci^t^ Fran^aise de Physique and the Soci^t^ de 

 Chimie-Physique). 



It is desired to close the fund in July of this year. 

 Henry A. Miers, 



Chairman. 

 C. G. Darwin, 

 H. Robinson, 



Hon. Secretaries. 



The Aurora of March 22-23. 



I had a fine view of this superb display at Working- 

 ton between midnight and i o'clock a.m., in a clear 

 and bright starlit sky. The whole sky was filled with 

 the light except a small area in the south-east. I 

 could detect no colour except creamy-white, the general 

 intensity being, to my mind, at times equal to full 

 moonlight. Curtains of light surrounded a point just 

 east of the zenith, which seemed to mark the "hub" 

 of the display. The bright star (a) in Canes Vena'tici 

 almost exactly marked this point, and filmy sheets of 

 light seemed to dash upwards from the south-west 

 and north-east horizons and merge together at this 

 star. The only display I have ever seen to equal 

 this was on 1907 February 14 at Motherwell, in the 

 previous sun-spot maximum period. It was the fact 

 that I could see the great sun-spot train on March 22 

 without telescopic aid that made me expect and look 

 out for the aurora that night. 



W. B. HOUSMAN. 



Seaton Cottage, Workington, April q. 



