Apbil 5, 1907] 



SCIENCE 



533 



The paper concludes by emphasizing the 

 suggestion made by Hill that possibly the 

 magnetic properties of these alloys at room 

 temperatures are largely determined by 

 the temperatures from which they have 

 been cooled, and that by suitably heating 

 samples of the alloys to different tempera- 

 tures and then chilling them their magnetic 

 properties at these temperatures may be 

 ascertained, just as the structures of other 

 alloys at different temperatures have been 

 investigated in this way by Neville and 

 Heycock. 



On the Magnetic Susceptibilities of Mix- 

 tures of Salt Solutions: J. C. McLennan 

 and C. S. Weight. 



In this paper the authors give some 

 measurements on the magnetic susceptibili- 

 ties of solutions of manganese, aluminium, 

 and copper sulphates in water, and several 

 mixtures of these solutions made with the 

 object of obtaining information which 

 might be of service in explaining the be- 

 havior of the magnetic alloys, recently dis- 

 covered by Heusler. The method followed 

 in measuring the susceptibilities is that 

 suggested by Kelvin, in which a glass cell 

 of the solution investigated is placed in a 

 strong magnetic field and the susceptibility 

 deduced from the pull exerted on the solu- 

 tion by the field. 



The magnetic susceptibility of water 

 was found to be — 7.33 X lO-''. Measure- 

 ments on a series of salt solutions gave the 

 following molecular susceptibilities: 



Ms 

 Manganese sulphate MnSO, + .01491 

 Copper sulphate CuSO. -f- .00153 



Aluminium sulphate Alj( SO.), — .00018 

 Aluminium nitrate AUINOa), +.00002 

 Aluminium chloride Alj(Cl), — .00005 



A set of measurements on solutions of 

 manganese sulphate of different concentra- 

 tions showed that the molecular suscepti- 

 bility of the salt was independent of the 

 concentration. 



On Magnetic Shielding: A. P. Wills. 



Assuming the results of a previous 

 paper, giving the 'shielding ratio' for a set 

 of three concentric spherical iron shells and 

 for a similar set of cylindrical shells, the 

 following problem was discussed: Given 

 the innermost and outermost radii of the 

 system in each case, what values should the 

 remaining four radii have in order that 

 the shielding shall be a maximum 1 Start- 

 ing with the expression giving the 'shield- 

 ing ratio' (the ratio of field impressed to 

 field within innermost shell), derived in 

 the paper referred to above, the conditions 

 for a maximum of the expression, under 

 the conditions imposed, are examined; and 

 it is found that approximately the best con- 

 ditions are obtained when the radii of the 

 shells are in geometrical progression. This 

 holds for both spherical and cylindrical 

 systems. 



Models Illustrati7ig the Motion of a Violin 



String: Harvey N. Davis. 



The function u(x,t) which gives the dis- 

 placement, at the time t, of a point'a; units 

 from one end of a violin string, can be 

 represented graphically by a surface with 

 the X, t plane as a base-plane. If the units 

 were properly chosen, this graph would be 

 the surface which the string would gener- 

 ate if, while it vibrated, it were also 

 carried along in a direction perpendicular 

 to the plane of its vibration; and in any 

 ease, a section of such a surface parallel to 

 the X axis represents, usually on a magni- 

 fied scale, the configuration of the string 

 at some corresponding time, t = tg, while 

 one parallel to the t axis represents the dis- 

 placement of a corresponding point, x=Xa, 

 as a function of the time. 



Five surfaces of this kind, modeled in 

 three dimensions, were shown, represent- 

 ing, one the general Hehnholtzian solution 

 and the other the motion of a string bowed 

 at points 1/5, 2/5, 2/7 and 3/8 of its length 



