596 



NA TURE 



[November 4, 1922 



mately correct results in nearly every case. We 

 notice that the author adopts Kennelly's names for 

 the absolute unit of electric quantity, capacity, and 

 pressure. We thus get the abcoulomb, the abfarad 

 and abstatfarad, and the abvolt. As they have never 

 been recognised, even in America, by any technical 

 society or institution, they are " technically irregular." 

 It has to be remembered, however, that all good new 

 names are introduced in this way. 



We do not like the phrase " dielectric constant " ; 

 it surprises, at least mathematicians, to find that this 

 •• constant " varies with temperature. Many excellent 

 methods of measuring capacity are given, but beginners 

 would appreciate some little guidance as to which one 

 to adopt in special cases. 



Dr. Rayner's article on dielectrics will be appreciated 

 by engineers, and Mr. Melsom's article on direct-current 

 indicating instruments will be most helpful in the 

 test-room of every factory. Mr. F. E. Smith writes 

 an authoritative article on systems of electrical measure- 

 ments which will be of great value for reference by 

 subsequent writers. Dr. Allmand gives a concise and 

 excellent description of the technical applications of 

 electrolysis. Any one reading the fascinating account 

 of the electron theory and its application to spectrum 

 analysis, by Sir William Bragg, will find it difficult to 

 believe that Rutherford and Bohr's theories of the 

 atom are not substantially correct. In his description 

 of "electrons and the discharge tube" Dr. Crowther 

 is also very convincing. 



Mr. F. E. Smith gives an illuminating account of 

 galvanometers, including very helpful rules for choos- 

 ing a galvanometer for a particular purpose. Mr. 

 Butterworth writes a valuable account of the formulae 

 used for measuring inductance and gives an excellent 

 table for computing the mutual inductance between 

 coaxial circles. We miss, however, his own formula 

 and that of Mr. H. L. Curtis for computing the high- 

 frequency inductance of parallel cylindrical wires. 

 Mr. Campbell gives a very complete account of methods 

 of measuring inductance and gives some 200 references 

 to papers on the subject. The method of compensating 

 for the inductance of a coil shown on p. 402 is not as 

 accurate as the author states, a term having been left 

 out in the algebraical reduction shown on this page. 

 Mr. Dye, in " Magnetic Measurements and Properties 

 of Materials," has produced a very complete account 

 of modern methods. 



The General Electric Company gives a very brief 

 account of " incandescence " lamps, which is excellent 

 so far as it goes. Most readers would like to have 

 had further data on tungsten vacuum and gas-filled 

 lamps. The Americans are not so reticent. We note 

 that the temperature of the tungsten filament is about 

 NO. 2766, VOL. I IO] 



" 2300 K." We take it that this is in the absolute 

 Centigrade scale and that the K. refers to Kelvin. 

 This is "technically irregular"; but there is a real 

 demand by engineers and by some physicists that 

 the absolute Centigrade scale, which is the one they 

 use, be called the Kelvin scale, and we hope that this 

 nomenclature will be adopted. 



Prof. Honda gives a thoughtful article on the molec- 

 ular theories of magnetism, and we have an account 

 of Ewing's latest model. Dr. Chree writes a thorough 

 and interesting account of the observational methods 

 used in terrestrial magnetism, and Dr. Chapman 

 describes some of the theories of terrestrial magnetism 

 and how far the solar agent is responsible for magnetic 

 storms and auroras. Positive rays are described by 

 Dr. Aston. A description is given of his mass-spectro- 

 graph and typical mass spectra are shown. 



The lengthy article on radio-frequency measure- 

 ments by Mr. Dye will be appreciated by workers in 

 many research laboratories, as these methods are often 

 of great value. We are doubtful whether it is legiti- 

 mate to assume that the capacity of a coil can be 

 represented by supposing that the coil has no capacity 

 and that a condenser of a certain size is placed across 

 its terminals. The formula for the high-frequency 

 resistance of a round wire at an infinite distance away 

 from other wires is given, but the formulas found 

 recently for more practical cases are not given. 



Very complete accounts are given of switch-gear, 

 telegraphy, and telephony, which will be appreciated 

 by electrical engineers. We have only space to 

 mention the valuable articles on vibration galvano- 

 meters, wireless telegraphy, and thermionic values, by 

 Mr. Campbell, Dr. Eccles, and Prof. Fortescue. Finally, 

 the articles by Prof. Richardson and Dr. Wilson on 

 thermionics, Mr. Smith-Rose on the use of thermionic 

 valves, Mr. Melsom on direct - current meters, and 

 Dr. Crowther on X-rays, are of great value. 



Under units it is stated that, at the International 

 Electrical Conference held in Paris in 1900, the Gauss 

 was defined to be the C.G.S. unit of magnetic force, 

 and the Maxwell was defined as the practical unit of 

 magnetic induction. It is also stated that if we take 

 the permeability of air to be unity and to be a pure 

 number, the value of the Maxwell is the same as that 

 of the Gauss. But the Maxwell is not the unit of 

 magnetic induction density. Hence we should read 

 in this case that a Maxwell per square centimetre is 

 the same as the Gauss. There are several definitions 

 of self-inductance given : that appearing on p. 727 is 

 wrong, as the self-inductance is the linkage of the 

 magnetic induction, and not the magnetic force, with 

 the current. In no case is it explained how the linkages 

 inside the conductor have to be computed. We have 



