NA TURE 



385 



THURSDAY, FEBRUARY 24, i! 



THE MAGNETIC CIRCUIT. 



The Magnetic Circuit. By H. du Bois, Ph.D. Trans- 

 lated by E. Atkinson, Ph.D. Pp. xviii + 362. (London : 

 Longmans, Green, and Co., 1896.) 



NO one is better fitted than Dr. du Bois to write a 

 treatise on the magnetic circuit. He is well 

 known for his experimental investigations in magnetism, 

 and abstract magnetic theory is clearer and more com- 

 prehensive for his restatement of its main propositions. 

 His German treatise, " Die Magnetische Kreise," pub- 

 lished some five years ago, attracted considerable atten- 

 tion in this country. It gave in reasonable compass an 

 excellent account of magnetic theory, methods of mag- 

 netic measurement, and of recent magnetic research, in 

 which Dr. du Bois' own contributions to knowledge 

 formed an important part, not merely on account of 

 their intrinsic value, but of their clear and first-hand 

 and withal modest statement. 



A translation of this book, made by Dr. Atkinson with 

 the assistance and revision of the author, and one or 

 two other experts in the subject, was published in 1896, 

 and was thoroughly welcomed by the already consider- 

 able and continually increasing number of experiment- 

 alists in electricity and magnetism. Some apology for 

 so late a review of so important a work is necessary, 

 and this is now offered by the reviewer, whose fault it 

 has been that an account of it has not long before now 

 appeared in Nature. Still "good wine needs no bush,' 

 and the work of Dr. du Bois carries its own guarantee 

 of excellence on the face of it. 



The book begins with a very valuable summary of 

 magnetic quantities and conceptions, in which the fun- 

 damental ideas of magnetic force, magnetic induction, 

 intensity of magnetisation, permeability and suscepti- 

 bility, &c., are fully explained, and the graphical repre- 

 sentation of their values for actual cases is fully illus- 

 trated by curves of induction and magnetisation obtained 

 for actual specimens of iron and steel. 



The question of the effects of the ends of a bar of 

 magnetised iron or steel, or of a narrow crevasse or 

 joint between two parts of a bar or ring, is taken up 

 towards the end of Chapter i., and is continued in 

 Chapter ii. Ewing's representation of the effect of the 

 cut, by a shearing of the curve of magnetisation for the 

 uncut bar or ring, is given, and illustrated, as before, by 

 the results of actual investigations — mostly those of 

 Ewing himself The effect of a cut leads to a further 

 discussion of that of an end, and thence to a discussion 

 of Coulomb's law for the distance action of a magnetic 

 pole, and of the field at a point due to a pair of equal and 

 opposite " point-charges " of magnetic matter. 



Of course this action at a distance is what is actually 

 observed, and there is an advantage in starting with 

 what are the immediate results of experience. Still, if 

 the book, as no doubt it will, should pass soon to a new 

 edition, we should like to see some account of a more 

 " mediumistic " view of the matter included. It is 

 possible, and in connection with the electromagnetic 

 theory of light it is very desirable, to consider magnetic 

 force and magnetic induction not only as both existing 

 NO. 1478, VOL. 57] 



as directed quantities at every point of a magnetised 

 body, but at every point of the electromagnetic field, in 

 ether or other " interferric " material, as well as in iron 

 itself. With electric induction and electric intensity 

 also introduced as the analogues of the magnetic quan- 

 tities, with k for the electric inductivity, and /x for the 

 magnetic inductivity of the medium, whatever it may 

 be, all difficulties as to units disappear if k and /* are 

 regarded as depending upon real physical properties of 

 the medium which, if fully known, would determine the 

 dimensions of these quantities, and i\k\i is see'n in a 

 natural way to be the square of the velocity of propaga- 

 tion. The serious difficulties which students feel with 

 respect to this matter, and which are only apparently over- 

 come by an unsatisfactory process, are entirely avoided. 



In Chapters iii. and iv. we have an Outline of the 

 Theory of Rigid Magnets, and a discussion of Magnetic 

 Intensity and Magnetic Induction. The account of rigid 

 magnetism and of induction here given is based in the 

 main on that source of all theory on the subject. Lord 

 Kelvin's Mathematical Theory of Magnetism, and on 

 the digest of this part of the subject given in Maxwell's 

 great treatise, and includes most of the more valuable 

 results obtained by KirchhofF, and by later writers 

 who have commented on or developed the theoretical 

 principles laid down by these masters. Here Dr. du 

 Bois has himself done good service. He has given 

 solutions of important problems of induction, and 

 generally, as stated above, in certain points improved 

 the presentment of theory ; work of which English 

 readers now reap the full advantage. 



A chapter on the Magnetisation of Closed and of 

 Radially Divided Toroids, which is divided into two 

 parts — ia) Theoretical ; {b) Experimental— closes the 

 first part of the treatise. This chapter is exceedingly 

 important. Kirchhoff's theory of the magnetisation 

 of a solid of revolution is given, and its results for 

 rings of rectangular and circular section are deduced. 

 This investigation of Kirchhoff's gave rise to the ex- 

 perimental investigations of Stoletow and Rowland, the 

 results of which have only confirmed its correctness. 



Results are next deduced for a radially divided ring, 

 which are fully illustrated in the experimental investi- 

 gations described later. Thus we have the subject of 

 leakage of magnetic induction introduced and clearly 

 discussed and illustrated. The great practical import- 

 ance of this part of the subject will be obvious when its 

 application to dynamo machines is considered, and a 

 large part of the experimental investigations which have 

 been carried out have had to do with the question of 

 the distribution of tubes of induction in or near more 

 or less narrow gaps in what may be regarded as closed 

 circuits or rings of iron. 



The great importance of having a magnetic circuit 

 nearly closed was recognised by Faraday, and was more 

 or less present to the minds of makers of magnets, both 

 permanent and electro, down to the beginning of the 

 present active manufacture, on a large commercial scale, 

 of dynamo machines. 



Then it was that Hopkinson came forward with his 

 explicit statement of the laws of the magnetic circuit, 

 and their application to the construction of the field 

 magnets of dynamos, a piece of work for which practical 



S 



