NA rURE 



429 



THURSDAY, MARCH 8, li 



ELECTROMAGNETISM AND DYNAMO 

 CONSTRUCTION. 



A Text-Book on Elcctromagnetism and the Construction 

 of Dynamos. By Dugald C. Jackson, B.S., C.E., Pro- 

 fessor of Electrical Engineering in the University of 

 Wisconsin, &c. Vol. I. (New York and London : 

 Macmillan and Co., 1893.) 



IN this work an excellent attempt is made to present 

 the elements of a very important subject in small 

 compass and in a clear and readable form, without any 

 sacrifice of accuracy. The present volume is only the 

 first instalment of the complete treatise promised by the 

 author, but it contains in nine chapters, covering some \ 

 280 pages, a fairly comprehensive survey of the elements j 

 of electromagnetism, the magnetic properties of iron, 

 magnetic circuits and characteristic curves of dynamos, 

 a discussion of efficiencies, and the action of multipolar 

 dynamos. 



The primary definitions and statements regarding 

 units, which are always a good test of the competency 

 of at least the theoretical treatment in a book like the 

 present, are generally clear and accurate. The author 

 begins by discussing lines of force, and on p. 2 gives 

 the needful caution that such lines "have no material 

 existence, but are merely hypothetical." It might have 

 been added here, perhaps, that this notion of lines of force 

 is a concept corresponding to a state of the field produced 

 by the presence of the magnetic distribution, that is to 

 say that there is some kind of "displacement or motion 

 of the medium" perfectly real though not material, the 

 direction and amount of which is typified by the grouping 

 of the lines of force. 



On the same page it is stated that "when the lines of 

 force in a magnetic field are parallel and of equal number 

 per square centimetre a magnet pole will experience 

 the same force at all points of the field, and the field is 

 said to be uniform." This is, of course, quite correct ; 

 but it ought to be noticed that if the lines of force are 

 parallel throughout any finite portion of the field, they 

 must be of equal number per square centimetre, and 

 vice versa. 



The usual definitions of B and H are given on p. 5, 

 and we do not criticise in any adverse sense the author's 

 procedure in so doing. But we cannot help thinking 

 that it is much more conducive to clearness to give to 

 each medium, whatever it is, a magnetic inductivity, and 

 to consider this as a physical quantity depending on the 

 medium. Thus, denoting the inductivity by \i., we should 

 have the relation B = /iH. Then if the inductivity of a 

 standard medium be fi^ we should have, for the same H, 

 Bo = /ioH. The ratio B/Bq or /n//zo is then properly the 

 permeability Qi the former medium, and might be denoted 

 by ■^. This, it seems to us, would be much more in 

 accordance with Lord Kelvin's original presentation of 

 the matter. The permeability •»• would be in all circum- 

 stances a mere ratio, and therefore of zero dimensions ; 

 while the confusion caused by at one time regarding /x as 



NO. 1271, VOL. 49] 



a mere number, and at another as a quantity having 

 certain dimensions (for example, the dimensions of the 

 reciprocal of the square of a velocity), would be entirely 

 avoided. This procedure was recommended some time 

 ago by Heaviside, and there can be no question of the 

 desirability of its adoption. Instead of the equation 



we should have 



|U = I + 4 TT K 



M = Mo ( I + 4 'T «) 



where k is the magnetic susceptibility, also a pure 

 number. If, as Heaviside strongly advocates, " rational " 

 units be adopted, the 477 must be omitted in these 

 formulas. 



The ordinary mode of dealing with the subject begins 

 by making B a quantity of the same dimensions as H, 

 and later when the energy of the electromagnetic field is 

 discussed the definition of B is virtually altered, so that 

 BH/Stt becomes the energy per unit volume of the 

 medium, and B has not the same dimensions as H. This 

 change may be explained, but it constitutes a sore 

 difficulty to the student. 



In the present case, however, the matter is not so im- 

 portant, as for the dynamo application it is sufficient to 

 regard the relation as that which holds when ^q is put 

 equal to i. There is, however, essentially the same kind 

 of difference between \k and ■»• that there is between 

 density and specific ^gravity. The former depends on 

 the units adopted, the latter does not. 



A very good account is given in chapter iii. of the 

 magnetic properties of iron. The more important recent 

 researches on this subject are summarised, and the results 

 illustrated by curves. The author seems, however, to 

 have missed, or at any rate has not brought out, the point 

 of Dr. Hopkinson's divided bar method, which was to 

 test the total magnetic induction in the bar after certain 

 specified series of changes of magnetic force had been 

 applied. This object would not be attained by having 

 the bar undivided and the coil fixed, and simply reversing 

 the magnetising current, as Mr. Jackson suggests. 



The description in this connection of a ballistic 

 galvanometer as " a galvanometer with a rather heavy 

 needle, and therefore a considerable time of vibration," 

 has the merit of brevity, but is curiously inaccurate. It 

 is no doubt an off-hand careless statement which has 

 escaped correction in proof, but it may mislead a reader 

 into supposing that weight of -needle was in itself an 

 advantage in such an instrument. 



Hysteresis is adequately discussed in this chapter, 

 and Steinmetz's formula for the energy dissipated in 

 a cycle as depending on the quality of the iron and the 

 number of cycles per minute is exemplified by numerical 

 values found by experiment for different kinds of iron. 

 Here we notice the phrases " watts of energy," " energy in 

 watts," improperly used for "joules of energy," " energy 

 in joules." 



With regard to the discussion of energy-losses by 

 hysteresis, it is worth remarking that no assertion can be 

 made as to the disposal of the energy given to the 

 medium (or taken from it) in an unclosed cycle. The 

 ordinary diagram and mode of discussing it easily shows 

 that at certain parts of the cycle more energy is given to 



T 



