July 2, 1908] 



A\-7 TURE 



come from various members of the Wills family, no less 

 a sum than 135,000!. having been promised toward its 

 realisation by them. Of this amount, Mr. H. O. Wills 

 has promised lOO.oooL, Lord Winterstoke (formerly Sir 

 W. H. Wills) 20,ooo!., and Sir Frederick Wills lo.oooL 

 More money is, however, still required to establish the 

 University in a satisfactory manner ; and it is to bo hoped 

 that other merchant princes of Bristol will follow the 

 magnificent example which the Wills family has given 

 them. 



SOCIETIES AND ACADEMIES. 



London. 



Royal Society, April 30. — "Note on Ihe Represenlation 

 of the Earth's Surface by Means of Spherical Harmonics 

 of the First Throe Degrees." By Prof. A. E. H. Love, 

 F.R.S. 1 



In a previous paper the author had concluded from 

 dynamical considerations that those elevations and de- 

 pressions on the surface of the globe which correspond 

 with continents and oceans should be capable of being 

 represented, in respect of their main features, by means 

 of spherical harmonics of the first, second, and third 

 degrees. A rough spherical harmonic analysis of the 



IB' ISO" ns" no° «>f 9if r^ 60° <5» a6° B* e* 15* 30° m' 



actual elevations and depressions yielded a rather imperfect 

 representation of the surface, which nevertheless offered 

 a general resemblance to the actual distribution of land 

 and water. It had, however, certain definite defects. To 

 alter the computed figure it is necessary to change appro- 

 priately the coefficients of the spherical harmonic expression 

 by which it is proposed to represent the elevation of the 

 surface above the mean level. In the present paper there 

 is recorded the best of many sets of trial coefficients, and 

 the result is shown on the accompanying chart. In this 

 chart the fine continuous line is a rougii outline of the 

 actual land of the globe, drawn in such a way that all 

 degrees of latitude or of longitude have the same value 

 on the map ; the heavy line is the zero line of the surface 

 harmonic with the chosen coefficients ; the dotted line is 

 the contour line along which the computed elevation is 

 equal to one-tenth of its maximum value. 



" The Relation between the Crystalline Form and the 

 Chemical Constitution of the Picryl Derivatives." By G. 

 Jerusalem and Prof. W. J. !?ope, F.R.S. 



By a method which depends upon dividing up the space 

 occupied by a crystalline substance into polyhedral cells, 

 each of which is assumed to be the habitat of but one 

 atom, Barlow and Pope have been able to determine the 

 general relation which exists between crystalline form and 

 chemical constitution. They conclude that the polyhedral 

 cells, each of which represents the domain of one atom 

 and which fit together without interstices to form the 

 crystal structure, possess volumes which are approximately 



NO. 201S. VOL. 78I 



proportional in magnitude to the lowest valencies exhibited 

 by the elements which they respectively contain. The 

 assemblages obtained by packing together the atomic 

 domains mentioned are partitionable into identical units, 

 each of which represents in composition, constitution, and 

 configuration a chemical molecule of the substance con- 

 cerned. The relation between the crystalline form and 

 chemical constitution of a series of substances is con- 

 veniently illustrated by comparing the dimensions of poly- 

 hedra, of which the volumes are represented by the sum, 

 W, of the valencies of the atoms composing the molecules, 

 and of which the relative dimensions are the crystallo- 

 graphic axial ratios a : b : c. The dimensions referred to 

 are termed the equivalence parameters x : y : z, and are 

 calculated from the relations that a:h : c = x -.y : z and 

 .%7: = W. 



In the present paper the above considerations are applied 

 to a number of derivatives of picric acid and allied sub- 

 stances ; it is shown that in this series one of the three 

 equivalence parameters remains approximately constant, 

 and has about the value of z in the equivalence parameters 

 of crystalline benzene, namely, x -.y : 2 = 3-101 : 3-480 : 2-780. 

 The direction in which the dimension z is measured in 

 benzene is that in which columns formed by superposing 

 triangularly arranged groups of three polyhedra, each 

 occupied by a carbon atom, occur throughout the 



assemblage repre- 



3^__Tf so" l"^^' 120° 135° 150' i6s° sontlng "crystallised 



benzene. It is con- 

 cluded that the 

 crvstal structures of 

 the picryl deriva- 

 tives are derived 

 from that of ben- 

 zene by moving the 

 columns of carbon 

 domains apart and 

 packing the substi- 

 tuting groups in 

 between them in 

 accordance with the 

 method already de- 

 scribed by Barlow 

 and Pope. 



" On the Hyster- 

 esis Loss and other 

 Properties of Iron 

 .Alloys under very 

 Small Magnetic 

 Forces." By Prof. 

 Ernest Wilson, 

 V. H. Winson, 

 and G. F. O'Dell. Communicated by Sir William H. 

 Preece, K.C.B., F.R.S. 



The experiments were carried out on two alloys of iron, 

 namely, " stalloy," of which the distinguishing feature is 

 tliat it contains about 3 per cent, of silicon, and " lohys," 

 which is a good sample of transformer plate. The prin- 

 cipal object of the research is to find the magnetic proper- 

 ties of these materials under small magnetising forces, 

 especially as regards hysteresis loss. Information is also 

 given on the specific resistance and temperature coefficients 

 of the materials. 



For the magnetic tests the specimens are in the form 

 of rings composed of stampings, and the ballistic galvano- 

 meter method has been employed. 



Lord Rayleigh found by the magnetometer method that 

 in the case of Swedish iron the permeability was nearly 

 constant when the magnetic force (H) varied from 000004 

 to 0-04. In the present experiments the permeability also 

 tends to become constant, the limiting values being 260 

 for stalloy and 222 for lohys. ks regards the maximum 

 value of the permeability, the results obtained have been 

 compared with published figures for a very pure iron and 

 a good sample of plate rolled from Swedish iron. For 

 these materials the maximum permeabilities are re- 

 spectively 54S0 and 4450, and occur for values of the 

 magnetic induction (B) of 9100 and 7000 respectively. For 

 stalloy and lohys the maximum permeabilities are 4520 

 and 3280, and occur for values of B of 5000 and 5500 

 respectively. 



As regards hvsteresis loss, the following figures are 



