570 Notices respecting New Boohs. 



strength of the swinging needle produced by altering the strength 

 of the controlling field. 



It will be noticed that although all the experiments agree in 

 showing that tjX is not a constant, yet the law of its variation 

 is different in different galvanometers. It is evident when we 

 consider what is the cause of this variation, that this might 

 he expected ; for the law must he dependent on the con- 

 struction of the galvanometer and the conditions under which 

 it is tested. The problem is further complicated by the fact 

 that astatic combinations are generally used. The variation in 

 the ratio will depend upon the alteration in strength in each of 

 the magnets of which the combination is composed : and since 

 some of these magnets may be nearer the controlling magnet 

 than others, they will be more affected by altering the posi- 

 tion of the controlling magnet. It is, moreover, possible 

 that one of the magnets in the astatic combination may be 

 more favourably situated, by being near metal parts of the 

 instrument, for example, for producing eddy-currents, in 

 which case any changes in its strength will have a propor- 

 tionately larger influence on the total change in the eddy- 

 current damping. In addition, the change in the strength of 

 any particular magnet will depend upon its initial magneti- 

 zation, on the steel of which it is made, and also to a certain 

 extent on its previous history. When we remember, further, 

 that the percentage change in the value of the ratio t/X 

 depends upon how much of the damping is due to eddy- 

 currents and how much to the other causes, viscosity of the 

 air and the suspension, we see that we may expect the ratio to 

 vary in a very great number of different ways. 



My thanks are particularly due to Mr. T Mather for advice 

 as to carrying out the experiments, which were made at his 

 and Prof. W. E. Ayrton's suggestion. 



LVI1I. Notices respecting New Books. 

 Ueber die Mathematische Theorie der Elelctrodynamisclien Induction, 

 von Riccardo Eelici. Uebersetzt von Dr. B. Dessau. Heraus- 

 gegeben von E. Wiedemann. Leipzig: Wilhelrn Engelmann, 

 1899: pp. 1-121. 



THE publishers of Oslwald's Klassiker der Exakten Wissenschaften 

 certainly deserve the gratitude of all students of physical 

 science for bringing within their reach many of the classical memoirs 

 which are not always easily accessible. The volume before us 

 forms No. 109 of the series, and consists of a translation of Feliei's 

 three papers on the mathematical theory of current-induction. As 

 the editor justly remarks, the title is somewhat inadequate, for the 

 memoirs contain not only an attempt to formulate the theory of 



