72 



NATURE 



[November i6, 1911 



able that in selecting a commentator and literary 

 executor the choice should fall on one who possessed 

 the power of removing what was obscure in the theory 

 and of placing the scheme in the most advanta^'eous 

 light. Admiral de Horsey has nothing to recommend 

 him for the ofTice he has undertaken but an unstinted 

 admiration for the original author and a loyal desire 

 to secure his recognition as a profound thinker. We 

 respect and admire the sincerity of his conviction and 

 his resolute effort to uphold the reputation of his 

 departed friend. The struggle he has made is 

 pathetic, but we regret to say he has only succeeded 

 in darkening the issue. 



(ieneral Drayson was dissatisfied with the theory 

 of precession. He could not accept the explanation 

 of the change of coordinates as due to the revolution 

 of the earth's pole about that of the ecliptic, while at 

 the same time the obliquity of the ecliptic was con- 

 tinually varying. He did not admit that the circular 

 motion was a close approximation to the truth, and 

 that greater accuracy was obtained by making the 

 radius of the circle vary. Yet the device is a very 

 usual one in the explanation of a recondite subject. 

 It is often found that a broad general truth requires 

 a minute degree of qualification. "The geometrical 

 absurdity of a circle with a movable centre " seems to 

 have presented a difliculty that the gallant General 

 never mastered, and he therefore devised another plan 

 for computing precession. Owing to the slow motion 

 of the earth's pole, there is no difficulty in contriving 

 an arithmetical process, by which the results when 

 confined to a limited number of years shall be similar 

 to those obtained by the ordinary formula. General 

 Drayson 's plan was to make the earth's axis revolve 

 in a circle of radius 2q° 25' 47*, about a point 6° from 

 the pole of the ecliptic, and situated near the solstitial 

 colure. The annual motion of the point marking the 

 origin of longitude (apparently not precisely coincide nt 

 with the first point of Aries) is 40'89'', consequently 

 the cycle of precession is about one-quarter longer than 

 that assigned by astronomers. 



Admiral de Horsey 's contribution in support of his 

 friend's theory has been to compute the precession of 

 many stars by this method, and to compare the results 

 with the Nautical Almanac values. The agreement 

 is satisfactory, but if this proved anything one would 

 think it proved the Nautical Almanac correct ; but that 

 view does not commend itself to the Admiral. Partly 

 perhaps because in Drayson 's method the obliquity 

 of the ecliptic will vary in the course of a cycle between 

 23° and 35°, and thereby the glacial theory, provided 

 that geologists could be satisfied with so short a period 

 as 15,000 years, might be satisfactorily accounted for. 

 The author also claims that some difficulties he 

 imagines to exist in the reckoning of time can be 

 removed by this means of explaining precession. He 

 is not, however, very fortunate when he puts the late 

 Mr. Stone in the witness-box to prove an anomaly in 

 time reckoning. This may be a small matter, but 

 when the author confuses precession with aberration 

 \ve feel that, with the best intentions of serving the 

 interests of his lost friend, he is scarcely fitted for the 

 task. 



NO. 2194, VOL. 88] 



77/E MEASUREMENT OF ILLUMINATIOS 



Illumitiation: its Distribution and Measurement, l.v 

 A. P. Trotter. Pp. xvii + 292. (London: Mac- 

 millan and Co., Ltd.. 191 1.) Price 8s. 6d. net. 



DLDICVIKD to Piern- Bouguer, the father of 

 photometry, this bt)ok is the first really scien- 

 tific attempt to put illuminating engineering on a 

 proper basis, and is the outcome of the work wl 

 has been done in .America and England of late y< 

 to break away from the haphazard methods of li;. 

 ing which have so long been in vogue, and to repl 

 them by arrangements of the sources of light wl 

 shall lead to a satisfactory distribution of light < 

 the area to be illuminated. 



No one more fitted to undertake this work t 

 Mr. A. P. Trotter could have been found, and 

 experience he has gained since 1879, when he wor 

 out his dioptric system of uniform distribution 

 light, has enabled him to produce a book which \\ ill 

 prove invaluable to those who realise that the u 

 statement of the candle-power of a light offer- 

 guide to its lighting effect, and that fifteen can 

 burning in different parts of a room give a \ 

 different illumination from one fifteen candle ga- 

 burning in the chandelier. 



The book very wisely is confined to the metl 

 of distribution and measurement of illumination, .1 ■. 

 the portions dealing with photometry are more 

 especially amplified in this direction, whilst all 

 scriptions of systems of lighting have very proj 

 been omitted. 



The first chapter deals with the units and standard'; 

 of candle-power, from the much-abused candle to 

 impracticable VioUe melted platinum unit, but si 

 Mr. Trotter is a little unjust to the former whf 1 

 says "the so-called English Parliamentary cand! 

 spermaceti was not more scientific and hardly i: 

 accurate than the barleycorn of which three went 10 

 the inch." There are many photometricians of the 

 old school who could assure him that the Sj 

 standard candle, as made by Miller, when its 

 was guided bv common-sense rather than by depart- 

 mental directions, fell short of the modern standard 

 in little else than convenience. 



In the second chapter the author discusses " illu- 

 mination and derived units," and it is pleasant to 

 find due credit given to Sir William Preece, who "" 

 early as 1889, recognised the necessity for a me: 

 of illumination, and adopted the carcel-metn 

 which unit he gave the name "lux," a name ;» 

 wards applied by the Geneva Congress in 1896 t 

 the bougie-metre. The latter part of the chapter ; 

 devoted to a clear enunciation of the laws of ligh 

 flux of light, brightness, quantity, and reflection. 



The distribution of illumination, more especial 

 over a plane, occupies the next two chapters, and i 

 the fifth photometers received full attention, and th 

 chapter is of special value, as Mr. Trotter has intr(J 

 duced into it so much of his own work. It would 

 have been even more interesting if he had criticised 

 the various photometers from the point of view ol 

 the personal equation, as many observers would have 

 liked to know his opinion of the Referees' table photo- 



