74 



NATURE 



[May 28. 1908 



We should have thoug-ht that on this explanation 

 the velocity of the electron, when sensiblv outside 

 the metal, would have been about zero instead of 

 nearly that of light. 



We regret the choice of electrostatic units, while 

 thoroughly approving the adoption of a single kind 

 instead of the mixed electrostatic and electromagnetic 

 units which are so often met with. 



Many data emplo^'ed are those relating to an atom 

 or an electron. There is doubtless gain as well as 

 loss in making use of these instead of the correspond- 

 ing data for (say) unit volume or unit mass. We 

 think, however, that there is preponderating gain in 

 writing all formula? so as definitely to indicate to 

 what extent the value accepted for the number of 

 particles in a cubic centimetre influences the numerical 

 result. Thus in the expression for electrical con- 

 ductivity, 



40 r 



a is the gas-constant referred to an electron. Now 

 a/e is known with more certainty than either a or 

 c, and there is gain in writing e\ as a product, so 

 that the formula would be written — 



(r=-_ , . In. 



Probably the most daring speculation in this book 

 is in connection with the problem of aberration and 

 aether drift. Our author desires to remove the diffi- 

 culties in connection with this problem by denying 

 the existence of the aether altogether. The scientific 

 man, in accepting the a?ther, " has fallen into the 

 most glaring errors of the crudest nominalism." Our 

 author's explanation is based upon Faraday tubes. 

 May we suggest that these Faraday tubes seem 

 suspiciously like an sether, but with special properties 

 attributed to it? 



In many other respects we think that the author is 

 too dogmatic in his assertions; his exposition would 

 gain if the overbearing tone were modified. The 

 reader also resents the too colloquial character of 

 some sentences, such as, " My own vote is cast, for 

 what it is worth, for the latter." Science does not 

 advance by the mere casting of votes, whatever they 

 may be worth. 



Although we do not find ourselves in agreement 

 with everything in this book, it is undeniablv a very 

 invigorating study of the subject. The publishers are 

 to be congratulated on securing it, and also on the 

 care taken in producing it. There are exceedingly 

 few typographical errors; as proper names are im- 

 portant, we mention that Spender (p. 216) should be 

 Spencer. 



OCEANIC TIDES. 

 Scientific Papers. By Sir George Howard Darwin, 

 K.C.B., F.R.S Vol. i., Oceanic Tides and Lunar 

 Disturbance of Gravity. Pp. xv + 463. (Cam- 

 bridge : University Press, 1907.) Price 15X. net. 

 ' I 'HE syndics of the Cambridge University Press 

 -»- are bringing out in four volumes the collected 

 papers of Sir George Darwin. The first volume is 

 before us, and contains a list of about sixty papers 

 NO. 2013, VOL. 78] 



vi-ritten between 1875 and 1906, that will be distributed 

 over the four volumes. 



Sir George Darwin's papers being easily separated 

 into well-defined groups, the collected papers will not 

 be in chronological order, but will be classified accord- 

 ing to subject as follows : — Vol. i., Oceanic Tides and 

 Lunar Disturbance of Gravity; vol. ii.. Tidal Friction 

 with Astronomical Speculations; vol. iii.. Figures of 

 Equilibrium of Rotating Liquid; \-ol. iv.. Periodic 

 Orbits. 



The height of the tide at any point at a variable 

 time t must be e.xpressible as the sum of a number 

 of periodic sine or cosine terms, the arguments moving 

 uniformly with the time, and the coefficients being- 

 constant. The periods are mostly forced periods, and 

 there is little difficulty in pointing out what they are. ' 

 In addition there are the free periods. " A dvnamical 

 problem of this character," writes Sir George Darwin 

 (p. 350), cannot be regarded as fully solved unless we 

 are able not only to discuss the " forced " oscillations 

 of the system, but also the " free." Hence we regard 

 Mr. Hough's work as the most important contribution 

 to the dynamical theory of the tides since the time 

 of Laplace." 



The coefficients of the forced oscillations are in- 

 dicated by theory to some extent. We have first an 

 equilibrium theory, and then a dynamical theory. 

 (" The problem of the tidal oscillation of the sea is 

 essentially dynamical," p. 349.) By laborious quadra- 

 tures the effects of continents may be roughly taken 

 into account. Finally we are driven back upon the 

 empirical determination of coefficients from observ- 

 ation. This is dealt with in the sixth paper of the 

 present volume. The work is theoretically easy, but 

 most laborious in practice. 



In the ninth paper the author concludes from tides 

 of long period that the rigidity of the earth is about 

 that of steel. 



Tide prediction follows naturally, when the coeffi- 

 cients of the various superimposed tides have been 

 obtained. An ingenious machine at the National 

 Physical Laboratory, near Teddington, traces the com- 

 bined effect of twenty-four different tides upon a chart. 

 In two hours the curve that represents the tides of 

 one year can be traced. It afterwards takes a com- 

 puter a few days to measure the times of high and 

 low water. The machine is used for forty different 

 ports, and, it may be remarked, is therefore not over- 

 worked, as its services are required for eighty hours 

 each year. 



On p. 5 there is a schedule of notation of the 

 principal tides, with the speed attached. The speeds 

 are combinations of simple multiples, positive and 

 negative, of the earth's rotation, and the mean motions 

 of the sun, moon, and lunar perigee. The speeds 

 are given numerically on pp. 20,21, &c. On p. 139 we 

 find tables of coefficients for Port Blair, which 

 quickly and clearly indicate the relative import- 

 ance of the various tides. On p. 116 we note that 

 an attempt to detect the nineteen-yearly tide failed. 



" The actual change of sea-level between 1870 and 

 1873 [at Karachi] was nearly 0^25 feet, and this is 

 just about nine times the range of the nineteen-yearly 



