October 15, 1908] 



NA TURE 



599 



and auxiliary machinery and boilers in a modern 

 steamship involve great responsibility, and an in- 

 teresting section of the book is devoted to the dis- 

 cussion of the preservation and repair of boilers, the 

 adjustment of machinery, and the duties of the 

 watch-keeping engineer. A comparatively brief 

 sketch is given of the modern theory of the resist- 

 ance experienced by ships when moving through 

 water, and of the conditions influencing the efficiency 

 of propellers. In all cases the author illustrates his 

 conclusions by modern instances and recent experi- 

 ments, showing himself to have been a diligent 

 student of published data. This is a distinctive fea- 

 ture of the book throughout. 



The final section describes " recent developments " 

 in marine engineering, including marine steam 

 turbines and the applications of internal-combustion 

 engines to ship propulsion. The Parsons type of 

 turbine naturally receives most attention, haviny 

 been applied so much more extensively than any other 

 type, and the descriptions and illustrations are excel- 

 lent. The arrangements of turbine machinery de- 

 scribed include those of battleships, cruisers, the 

 latest Cunarders, and certain small, swift vessels of 

 the destroyer class. A summary of the results ob- 

 tained on contract trials and actual service is also 

 given. It is interesting to note how rapidly the 

 Parsons system has made its way abroad as well as 

 at home. So far, its only rival — and that at a very 

 great distance — is the Curtis turbine, which has been 

 successfully applied in the United .States scout- 

 cruiser Salem, of which the contract trials took place 

 subsequently to the completion of the book. 



In regard to internal-combustion engines the 

 author gives much information, indicating the fea- 

 tures in which they must still be regarded as experi- 

 mental, as well as those in which they promise a 

 possibility of further advances in speed and fuel- 

 economy. 



The volume is well produced, has a good index, 

 and contains about 400 illustrations. It deserves and 

 will secure a good reception from all who are in- 

 terested in the subjects of which it treats. The author 

 has the courage of his opinions, and, in not a few 

 instances, exception may be taken to his conclu- 

 sions; but in all cases the materials for judging 

 independently are given, and readers can claim no 

 more. W. H. White. 



r//E MOON'S MOTION. 

 The hiequalities in the Motion of the Moon due to 

 the Direct Action of the Planets. By Prof. E. W. 

 Brown, F.R.S. Pp. xii-1-93. An Essay which 

 obtained the Adams Prize in the University of Cam- 

 bridge for the Year 1907. (Cambridge : University 

 Press, IQ08.) Price 6^. net. 



PROF. BROWN is much to be congratulated on 

 having at length written the word " Finis " to 

 his lunar theory. His achievement has been a very 

 great one, for he has completely solved the problem 

 that he had proposed to himself, viz. the motion of 

 the moon under the attraction of known bodies ; he 

 has pushed his solution sufificiently far bevond the 

 NO. 2033, VOL. 78] 



standards required by observation to cover any prob- 

 able increase in the accuracy of observation during 

 the near future ; his mathematics have been elegant, 

 and his numerical computations performed under 

 systems of check that command, not onl}' his own 

 confidence in their accuracy, but that of his readers. 

 -Vt last, therefore, we are entitled to say that any 

 discrepancy between theory and observation must be 

 attributed to fresh causes and not to imperfect calcu- 

 lation. A similar remark has somewhat readily been 

 made before after the completion of other lunar 

 theories, but a degree of numerical accuracy far 

 beyond Hansen or Delaunay may safely be claimed 

 for Prof. Brown's theory. 



The memoir especially under review is the investi- 

 gation of the direct action of the planets, which was 

 recently awarded the Adams prize in the Uni- 

 versity of Cambridge. The subject was unknown to 

 Hansen, whose tables are still in use. In 1876 Prof. 

 Xewcomb discovered an empirical term in the moon's 

 motion. .Shortly afterwards Mr. Nevill attributed 

 this term to the action of Jupiter. Some years then 

 elapsed, and Dr. G. W. Hill gave a computation of 

 the new term, and a little later Radau computed a large 

 number of planetary terms in the moon's longitude. 

 It is remarkable that both Hill and Radau gave o"'go 

 as the coetificient of Newcomb's term, and both of 

 them were 20 per cent, in error. Radan's results are 

 in other respects free from sensible error, and it is 

 unfortunate that the term which started the whole 

 subject should have been the one most difficult to 

 calculate with accuracy. Quite recently Prof. New- 

 comb and Prof. Brown have published their 

 researches. It is clear that the latter has reached a 

 higher order of accuracy, but the former's memoir is 

 probably amply good enough for comparison with 

 observation. They agree in an increased coefficient 

 of i"'i for Newcomb's empirical term. It is not 

 possible to compare either investigation with the 

 other at any intermediate stage before the conclusion. 



This is perhaps the time to give an answer to the 

 question. How will the actual motion of the moon 

 agree with Prof. Brown's theory? We have 

 already expressed our belief that any want of agree- 

 ment will point to the action of unknown causes. 

 Possibly, therefore. Prof. Brown's work will be even 

 of more importance if his tables fail to predict the 

 motion of the moon than if they succeed. 



As regards short-period inequalities, we believe 

 that Prof. Brown's tables will be practically perfect. 

 We should like, however, to invite the attention of 

 astronomers in thirty years' time to one point. Let 

 every discordance between observation and tabular 

 position be multiplied by the sine and cosine of the 

 moon's longitude and the mean taken. If this be 

 done for the last fifty years, the result is too large 

 to attribute to accidental error; nor will the altera- 

 tion of the moon's parallax and the insertion in the 

 new tables of a Venus term with coefficient o"'7 

 entirely remove the difficulty. Possibly the past 

 observations have been affected by a systematic error, 

 but be the cause what it may, the point is worth 

 remembering and looking into when the proper time 

 comes. 



