proceedings: philosophical society 77 



to calculate and may be of considerable magnitude. It is therefore 

 important that bars, which are to be used as standards for comparison 

 of different methods, or whose properties it is desired to measure with 

 high accuracy, should be examined for uniformity. A method was 

 described which clearly indicates the position, nature, and magnitude 

 of non-uniformities. which exist in a test bar. Curves were given show- 

 ing the effects of non-uniformities on the accuracy of magnetic measure- 

 ments. Such results naturally lead to a consideration of the possibility 

 of examining ferrous material for soundness and the detection of flaws. 

 Curves were given showing some of the possibilities in this direction. 



Mr. J. H. Dellinger then spoke on Rationalization of the magnetic 

 units. There are two distinct sets of magnetic units in common use. 

 The first set consists of the eg s units, which are consistent with the 

 ordinary electromagnetic equations. The second set involves the use 

 of the ampere-turn, which introduces certain changes into the equations. 

 In the first set of units, the current use of the name "gauss" both for the 

 unit of induction and the unit of magnetizing force is questionable. The 

 two quantities are sometimes looked upon as physically the same. 

 There are preponderating reasons, however, for considering them to 

 have an essentially different nature, induction corresponding to the 

 magnetized state of the medium, and magnetizing force being the 

 agency tending to produce that state. The double use of "gauss" is an 

 inconvenience in practice. There have been various proposals from 

 time to time to rationalize the units, i.e., to use units such as to redis- 

 tribute the factor 4ir in the equations. The first proposal, that of 

 Heaviside, had much to recommend it but required a radical change of 

 all the electric and magnetic units. Subsequent proposals have in- 

 volved less change of existing units, but have all had the disadvantage 

 of incorporating 4x in the value of permeability of vacuum. It is 

 pointed out herein that the use of the ampere-turn leads to a rationalized 

 set of units, without either of these disadvantages. 



Discussion: Messrs. Swann, Silsbee, and Rosa discussed this com- 

 munication. Such proposals for rationalization are of distinct value 

 to emphasize the sometimes illogical designation of units; the advisa- 

 bility, however, of radical changes from the existing systems is to be 

 questioned since it is difficult to revolutionize practice. Although the 

 two systems of magnetic units sometimes cause ambiguity, simplifi- 

 cation as proposed would lead probably to greater confusion, particu- 

 larly so because of radically different usage by investigators. The in- 

 corporation of the factor 4tt is quite natural since it has a physical 

 significance in distribution of force over a spherical surface. Mr. Rosa 

 suggested that an International Committee on units and nomenclature 

 is desirable. 



INFORMAL COMMUNICATIONS 



Mr. F. E. Wright exhibited a graphical device for solving equations 



of the general form A = B • C. (Published in full in Journ. Wash. Acad. 



Sci. 6: 1-3. 1916.) 



J. A. Fleming, Secretary. 



