July 31, 1919] 



NATURE 



423 



misslewsky read Ostromisslensky. " Roberite " 

 (p. 108) is usually written " roburite " — at 

 least on this side of the Atlantic; and "Anal- 

 len " in the list of references should be "Anna- 

 len. " These are blemishes which should not occur 

 in a book written by an instructor in general 

 chemistry in so important a school as the Car- 

 negie Institute of Technology. 



THE PRINCIPLES OF RADIO- 

 COMMUNICATION. 

 The Principles of Electric-wave Telegraphy and 

 Telephony. By Prof. J. A. Fleming. Fourth 

 edition, revised. Pp. xvi + 707. (London: 

 Longmans, Green, and Co., 1919.) Price 42s. 

 net. 



IT is no easy task to keep a large treatise on 

 electric wave telegraphy and telephony abreast 

 of the advances that ha\e been made during the last 

 few years. In this fourth edition the author has 

 been very successful. He does not load up the 

 book with descriptions of all kinds of technical 

 apparatus, or attempt to describe all the methods 

 used in practice. There is now quite a small 

 library of highly technical works which give the 

 required detailed information to the engineer, and 

 to these Prof. Fleming refers the reader. What 

 he does give is a comprehensive view of the 

 subject, particularly on its scientific side. He 

 also dwells fully on quantitative measurements 

 and their theory, which he himself has done so 

 much to develop. 



In part i. a discussion is given of electrical 

 oscillations. The mathematical proofs given are 

 rigorous and straightforward, and will be appre- 

 ciated even by those who are beginning to forget 

 their knowledge of the calculus. The mathe- 

 matical formulae for the high-frequency resistance 

 and inductance of circuits are given, and the 

 author's successful experimental methods of test- 

 ing these formulae described. When the wires 

 are curved, the mathematical difficulties in the 

 way of computation are so great that engineers 

 will welcome the experimental verification of the 

 formulae. 



The tables given of spark-voltages between 

 spherical electrodes are somewhat antiquated. 

 The reviewer had occasion recently to 

 analyse the experimental results on spark 

 voltages published in the standardisation rules 

 of the American Institute of Electrical En- 

 gineers (1918). The Institute used spheres vary- 

 ing in diameter from 3.125 cm. up to 50 cm., 

 and the spark voltages go up as high as 400 kv. 

 The results plainly show that when the voltages 

 of the electrodes are equal and opposite at the 

 instant of discharge, the maximum electric stress, 

 R, at this instant, the temperature being 25° C. 

 and the pressure 76 cm., is given by 

 R = 27-4+ i4-i/v/a 



kilovolts per centimetre, where a is the radius 

 of either spherical electrode in centimetres. It 

 was of interest, therefore, to see whether the 

 NO. 2596, VOL. 103] 



I results (p. 188) obtained by Heydweiller in 1893 

 on spheres of 05, i, 2, and 5 cm. were in agree- 

 ment with this formula. We find that when we 

 j correct for temperature and pressure they are iti 

 I most excellent agreement. It follows, therefore, 

 that if V be the spark voltage when the distance 

 between the electrodes is x, we have 



V = (R//)x, 

 where / is a mathematical factor depending on x 

 and a, a table of the values of which is given on 

 p. 145 of this book. This is at least true for 

 spheres varying in diameter from 0*5 up to 

 50 cm., values of x less than about a millimetre 

 being excluded. Various theories have been 

 given to explain why R should be the sum of two 

 terms as shown above, but in the reviewer's 

 opinion those theories which neglect the effects 

 of the convection currents of air flowing round 

 the "ectrode prior to the discharge are wrong. 



Ill part ii. electric waves are discussed, and a 

 very complete account is given of methods of 

 measuring and detecting them. The remainder 

 of the book is devoted to radio-communication. It 

 begins with a short history of the subject. Then 

 various long-distance telegraphic stations are 

 described, the salient points of the systems of 

 working being pointed out. The theory of trans- 

 mission is given. Full descriptions are given of 

 several of the latest developments of the art, such, 

 for instance, as the Marconi military set for wire- 

 less apparatus, and the Marconi sets used in air- 

 craft. 



The final chapter treats of radio-telephony, and 

 is perhaps the most interesting in the book. It 

 is extraordinary what rapid progress is being 

 made in this branch of the subject, and how 

 different are the methods employed by the various 

 experimenters. Marvellous results have been 

 obtained by the Fleming oscillation valve. The 

 almost incredible sensitivity attained in receiving 

 apparatus by the use of thermionic amplifiers 

 warrants the most sanguine hopes for the future 

 of radio-telephony. 



We ought to mention that, although additions 

 have been made to the book to bring it up to 

 date, yet by the deletion of antiquated matter and 

 the use of smaller type the total bulk of the 

 volume has been reduted. This is an advantage, 

 as the earlier editions were beginning to get 

 unwieldy. We can recommend this book to every- 

 one interested in radio-telegraphy. To the scien- 

 tific radio-telegraphist it is a necessity. 



A. Russell. 



GEOGRAPHICAL ASPECTS OF WORLD 

 POLITICS. 

 Democratic Ideals and Reality. A Study in the 

 Politics of Reconstruction. By H. J. Mac- 

 kinder. Pp. 272. (London : Constable and 

 Co., Ltd., 1919.) Price 7s. 6d. net. 



THERE is no lack of ideas in this book. From 

 beginning to end it is full of striking con- 

 j ceptions which arrest attention even if some of 



