538 



NATURE 



[May 20, 1909 



with the definition of which little serious fault is, in 

 most cases, to be found, though there may often 

 be room for difference of opinion. 



Perhaps the most valuable feature is the section 

 which describes the distribution of igneous rocks in 

 the British Isles, and the maps, mostly taken, by 

 permission, from well-known papers, with which it 

 is illustrated. 



The work may be safely recommended as a text- 

 book for students, but they should be warned against 

 the employment of the numerous little-known and un- 

 necessary rock-names to be found in its pages. In 

 almost every case the same idea can be more happily 

 expressed by prefixing a word or phrase to a well- 

 established name. Their presence, however, undoubt- 

 edly increases the value of the book as a work of 

 reference. J. W. E. 



Catalogue of the Lepidoptera Phalaenac in the British 

 Museum. Vol. vii. Catalogue of the Noctuidae in 

 the Collection of the British Museum. By Sir 

 George F. Hampson, Bart. Pp. xv+709; plates 

 cviii-cxxii ; 184 text-figures. (London : Printed by 

 Order of the Trustees, 190S.) 

 In no group of animals and plants is the enormous 

 increase in our knowledge more conspicuous than in 

 insects. Thus, at the time of the publication of the 

 twelfth edition of Linn^'s " Systema Naturse " (1767), 

 we find only 112 species described under Noctuae. Sir 

 George Hampson now divides the family Noctuidee 

 into fifteen families, of which the first three are 

 Agrotinse, Hadenin®, and CucuUianae, the species be- 

 longing to each being described in vols, iv.-vi. of 

 the general " Catalogue of Moths " respectively, and 

 vol. vii., now before us, forms the first of three 

 volumes intended to be devoted to the fourth sub- 

 family of Noctuidae, the Acronyctince, and includes 

 descriptions of species numbered from 2748 to 3590, 

 a considerable number of which (and also many 

 genera) are described as new by the author. 



It is possible that all the remaining families of 

 Noctuidae may not require a whole volume apiece, 

 and it would be difficult to estimate the total number 

 of Noctuidae which the present work is likely to 

 contain when completed, but it can scarcely be less 

 than 20,000 species, and may well be 30,000, or even 

 more, as against the 112 species which were all that 

 were known to Linn^, the most learned entomologist 

 of his time, in 1767. 



We notice no alteration in the general arrange- 

 ment of the work, and the usual high standard of 

 letter-press and illustrations is fully maintained in the 

 present volume. 



Physikalische Musiklehre. Eine Einfiihrung in das 



Wesen und die Bildung der Tone in der Instru- 



mentalmusik und !m Gesang. By Dr. Hermann 



Starke. Pp. viii+232. (Leipzig : Quelle and 



Meyer, 1908.) Price 3.80 marks. 



This little work on the physical theory of the nature 



and production of musical sounds is almost entirely 



free from mathematics, and may be regarded for the 



most part as an abstract of the simpler portions of 



Helmholtz's great classic, " The Sensations of Tone." 



The text is, however, freely illustrated by cuts, many 



of which, the author acknowledges, are borrowed 



from other books ; thus at every few pages may be 



found an old and familiar figure.' 



The treatment is divided into five parts or chapters. 

 Of these the first and second are occupied with the 

 origin and propagation of waves and sound, while 

 the third describes musical tones, intervals, and 



^ Perhaps it is this pr.nrtice which has led to the representation of a metal 

 s/r!/> vibrating like a string (p. 22), for the same error occurs in Tyndall's 

 "Sound "(p. 128), 1895. 



NO. 2064, VOL. 80] 



scales. The fourth chapter consists of four parts, deal- 

 ing respectively with (i.) stringed instruments, (ii.) wind 

 instruments, (iii.) vibrating bodies with inharmonic 

 overtones, and (iv.) human speech and song. The last 

 chapter is devoted to consonance and dissonance, and 

 after giving Helmholtz's theory concludes with a 

 risumi of more recent work on the subject. This 

 part includes notices of intermittence and variation 

 tones, and of the work and theories of C. Stumpf. 



To those who wish for a bright, readable treatment 

 of this borderland between music and physics, free 

 from mathematics, but with the opportunity of im- 

 proving their conversance with German, this book is 

 heartily recommended. E. H. B. 



LETTERS TO THE EDITOR. 

 [The Editoi does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can lie undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.] 



Electrons and the Absorption of Light. 

 On the theories of dispersion given by Drude and 

 Lorentz, an absorption band in the spectrum corresponds- 

 to the free period of an electron, and, if we assume that 

 only one electron in each molecule is concerned with an 

 absorption band, it is theoretically possible to calculate 

 e/m for this electron from the values of the coefficient 

 of extinction throughout the band. I have made this 

 calculation, apparently for the first time, using the 

 formula 



-- = 1-297 "« \ 3 " . 



which may be derived on both the above theories. R i> 

 the maximum value of the coefficient of extinction, X„ the 

 position of the maximum, X, the wave-length, for which 

 the coefficient of extinction has a value equal to half its 

 maximum, and y the index of refraction. The following 

 table gives some results : — 



For the anilin colouring matters e/m is of the order 

 lo', whereas for the glasses and inorganic salts it is of 

 the order lo' and under, showing that in the one case we 

 are dealing with electrons and in the other with ions. A 

 calculation made by Drude from the dispersion of solid' 

 cyanine in the neighbourhood of its band gave 

 c/in = S-5X 10*. If there are two electrons for each of the 

 original molecules of the colouring matter the values of 

 cjm should be halved, or if there is only one electron for 

 two molecules the value of e/m should be doubled. 

 According to Kalandek, corallin probably undergoes some 

 change in solution. This may account for the low value 

 of e/m. 



