July 22, 1897] 



NA TURE 



267 



These mistakes are avoided in the book before us ; but 

 we cannot help thinking that the part of the book which 

 deals with thermodynamics would have been improved by 

 a somewhat different order of treatment. The first real 

 step in thermodynamics should always be a discussion of 

 Carnot's heat-engine and Thomson's definition of abso- 

 lute temperature. Then, all about the behaviour of 

 gases, and the deviation of temperature as defined on the 

 air-thermometer from absolute thermodynamic tempera- 

 ture, could have been told, we think, much more effect- 

 ively. As it is. Joule and Thomson's experiments, which 

 we are informed were made for the purpose of deciding 

 whether there is any sensible attraction between the 

 particles of a gas, are mentioned only once in the book, 

 and that long before the scale of absolute temperature is 

 referred to. The most important part of the significance 

 of this investigation is really lost unless its bearing on 

 the realisation of absolute temperature is fully pointed 

 out. We may have unduly emphasised this omission ; 

 "but we are sorry a book, so generally good as this is, 

 should have in any degree missed an opportunity of 

 insisting on absolute temperature and all that thereto 

 relates, as the first and fundamental thing, we had almost 

 «aid the only thing, in thermodynamics. 



The second and third volumes of Messrs. Nichols and 

 Franklin's book, which deals with Magnetism and Elec- 

 tricity, we have but little space left to deal with. But we 

 ■may say at the outset that, good as the first volume seems 

 to us to be, these seem quite up to the same level. 



In vol. ii. a short and satisfactory account of Galvano- 

 metry is given under the heading of Electrolysis and 

 liatteries. The laws of electrolysis are stated, and some 

 space is devoted to a discussion of the Energy Theory 

 of the Voltaic Cell, though not more than or even quite 

 as much as the importance of the subject deserves. In 

 the sketch of the thermodynamics of the voltaic cell the 

 phrase, " sweeping processes performed by the current," 

 strikes one as quaint, to say the least. The " sweeping 

 process " is not performed by the current at all, but by 

 Messrs. Nichols and Franklin when they draw an indi- 

 cator diagram to represent a certain part of the work 

 done by the current. 



The book is brought up to date by an account of the 

 kathode discharge within a Crookes' tube, and of Ront- 

 gen's discovery. This chapter contains, besides, a very 

 brief account of the discoveries of Hertz. 



Practical Applications are dealt with in a chapter 

 mainly devoted to Electric Signalling. The usual tele- 

 graphic devices are described ; but none of the vexed 

 questions on this subject are gone into. The authors, 

 however, do say that the so-called KR law holds for 

 telephonic signalling as well as for submarine telegraphy, 

 which is, to say the least, a rather inadequate statement. 



A chapter on Mechanical Conceptions of Electricity 

 and Magnetism completes the second volume. The idea 

 adopted is that the ether in a magnetic field has a cellular 

 structure, and that these cells have a rotational motion 

 about axes parallel to the direction of the field at each 

 point, while the lines of electric force are marked by 

 displacement of these ether-cells— the positive in one 

 direction, the negative in the opposite. There are, of 

 course, serious difficulties in this mode of regarding what 

 NO. 1447, VOL. 56] 



takes place ; so much so, that there seems now rather a 

 consensus of opinion in favour of the view that the 

 direction of magnetic force is that of flow of the ether 

 regarded as a perfect fluid. Such questions, however, the 

 authors, probably from want of space, do not discuss. 



Volume iii. begins with a statement of the distinction 

 between light and sound as cases of wave-motion, and 

 a comparison of the methods of determining their veloci- 

 ties of propagation. Then follows an account of wave- 

 motion with the usual theorems on composition of vibra- 

 tions, constructions for wave-fronts of reflected and 

 refracted waves, Huyghens' zones, &c. The succeeding 

 chapters deal with the theory of mirrors and lenses, 

 treated, as they always ought to be in a physical book, 

 by means of considerations of wave-propagation and the 

 principle of equal optical distances. The old rule, con- 

 venient for use in the approximate fashioning of glass 

 lenses, might have been noticed at p. 47, that the radius 

 of the opening of a convex lens is nearly (for index of 

 refraction r5) a mean proportional between the focal 

 length of the lens and its thickness. 



The chapter on Dispersion (prismatic) strikes us as 

 capable of considerable expansion. The conditions for 

 obtaining a pure spectrum do not seem to be stated, 

 except in so far as they are given more or less implicitly 

 in connection with the spectroscope. In preparation for 

 this an elementary account of primary and secondary 

 focal lines might be added to the chapter on lenses, and 

 the reason for placing the prisms of a spectroscope in 

 the position for minimum deviation explicitly brought 

 out. 



The account of double refraction we find unexpectedly 

 short. Refraction in uniaxal crystals is alone treated, 

 Huyghens' construction is given for the single case of 

 Iceland-spar, and there is no notice of the other typical 

 case, that of quartz. 



But the most serious omission in this volume, perhaps 

 in the book, is that of any adequate discussion of 

 magneto-optic rotation. Article 754 is devoted to 

 " Rotation of the plane of polarization ; the sacchari- 

 meter," and Art. 755 (rather more than half a page) 

 deals with magneto-optic rotation. The latter subject is 

 not discussed in vol. ii., and we had expected to find in 

 vol. iii. an account of phenomena, of the very important 

 absolute determinations of Verdet's and other constants 

 that have been made, and of the applications of the 

 knowledge so obtained in magnetic research. As it is, 

 six lines are given to rotation of the plane of polarised 

 light in bisulphide of carbon, and the remainder to a 

 photo-chronograph, which, though ingenious and valuable, 

 hardly in the circumstances should have had so relatively 

 large a part of the available space devoted to it. 



In our account of this work we have referred to a 

 number of points in which, as it appears to us, it could 

 be improved and amplified without seriously adding to 

 its bulk. The number of such points may seem rather 

 large, but in this it is like every other treatise on its 

 first appearance ; and we hope that the authors will 

 believe that the statements above are not made in any 

 carping spirit, but rather as marking appreciation of what 

 is, within its scope and purpose, really an excellent 

 work. A. Gr.w. 



