Dec. 13, 1888] 



NATURE 



165 



petrographical description of the rocks and their metamorphoses. 

 A sketch-map and five plates of microscopical sections accom- 

 pany the paper, which is well summed up in German.— Geo- 

 logical observations on the Yug River, by B. Polyenoff. — On the 

 Spaniodon barbotii deposits of Crimea and the Caucasus, by 

 N. Andrusoff (also summed up in German). — Zoology and 

 Physiology: — Notes on the ichthyology of the basin of the Amur, 

 ')y N, Warpachowski and S. Hertzenstein, being an elaborate 

 i ascription of forty species, and their connection with kindred 

 i'ccies in neighbouring regions (diagnoses in latin, and the 

 A hole summed up in German). — On the Vertebrate fauna 

 r the Balkhash depression, by M. Nikolsk)-, The mcst in- 

 cresting conclusions of the author as to the recent geological 

 history of the Balkhash depression have already been mentioned 

 in Nature. Now we have a full description of the fauna {39 

 mammals, 226 birds, 21 reptiles, 3 amphibians, and 16 fishes), 

 together with an elaborate inquiry into the connection of this 

 fauna with those of the neighbouring regions. — The Percarina 

 and Beuthophihts of the Sea of Azoff, by J. Kuznetsoff.— The 

 minutes of proceedings contain several papers of great interest — 

 namely, on a journey to Dutch India, by A. Korotneff ; on the 

 ornithology of Caucasus, by K. Rossikoff; on the fossils of the 

 Nijne-udinsk cave, by I. Tcherski ; on a journey to Turkestan 

 and Bukhara, by S. Lidsky, &c. 



SOCIETIES AND ACADEMIES. 

 London. 



Royal Society, November 22. — "On the Magnetization of 

 Iron and other Magnetic Metals in very Strong Fields." By 

 J. A. Ewing, B.Sc, F.R.S., Professor of Engineering in 

 University College, Dundee, and William Low. 



The large magnet of the Edinburgh University, kindly lent 

 by Prof. Tait, was used throughout the experiments, and 

 allowed the authors to effect a high concentration of the 

 magnetic force by using bobbins, the necks of which had a 

 cross-sectional area of (in some cases) only ysVir of the cross- 

 sectional area of the magnet cores. By this means the induc- 

 tion 38 was raised to the following extreme values :— 



In wrought iron 



,, cast iron 



,, Bessemer steel , 



,, Vickers's tool steel ... . 



., Iladfield's manganese steel 



,, nickel 



,, cobalt 



C.G.S. 

 45,350 

 31,760 

 39,880 

 35,820 

 14.790 

 21,070 

 30,210 



The induction was measured by means of a coil consisting of a 

 single layer of very fine wire wound upon the central neck of 

 the bobbin. Outside of this coil, at a definite distance fiom it, 

 a second coil was wound, and the magnetic force was deter- 

 mined in the annular space between the two. In a paper com- 

 municated to the Manchester meeting of the British Association, 

 the authors showed that if the force so measured could be proved 

 to have the same value as the magnetic force within the metal 

 neck itself, it would follow that the intensity of magnetism 3 

 had begun to diminish under the action of excessively strong 

 fields, in the manner which Maxwell's extension of the Weber- 

 Ampere theory of molecular magnets anticipates. In the present 

 paper the authors discuss at some length the question of how far 

 the magnetic force within the metal is fairly measurable by the 

 magnetic force in the ring of surrounding air, and they show 

 that, with the form of cones originally used, the force within the 

 metal must have been less than the f .rce cutside, by an amount 

 probably sufficient to explain the apparent decrease of 3- The 

 form of cone suited to give a uniform field of force with sensibly 

 the same value in the metal neck and round it is investigated ; 

 and experiments are described m which the condition necessary 

 for a uniform field was satisfied. The results of these experi- 

 ments are conclusive in showing that no considerable change 

 takes place in the value of 3 (in wrought-iron) when the mag- 

 netic force is varied from about 2000 to 20,000 C.G.S. units. 

 Throughout this range of force, the intensity of magnetism has a 

 sensibly constant value of about 1700 C.G.S. units, which is to 

 be accepted as the saturation value f jr wrouj^hi iron. The term 

 saturation may be properly applied in speaking of the intensity 

 of magnetism, but there appears to be no limit to the degree to 

 which the magnetic induction may be raised. 



The following are probable values of the intensity of mag- 

 netism when saturation is reached in the particular metals 

 examined : — 



Saturation 

 value of J. 



Wrought-iron 1700 



Cast-iron 1240 



Nickel (with 075 per cent, of iron) 515 



Nickel (with 0-56 per cent, of iron) 400 



Cobalt (with I "66 per cent, of iron) 1300 



Experiments were also made with specimens of Vickers's tool 

 steel, and other crucible steels, Whitworth's fluid-compressed 

 steel, Bessemer steel, Siemens steel, and Hadfield's manganese 

 steel. This last material, which is noted for its extraordinary 

 impermeability to magnetic induction, was found to have a 

 constant permeability of about I '4 throughout the range of 

 forces applied to it— namely, from 2000 to nearly 10,000 C.G.S. 



Physical Society, November 24.— Prof. Reinold, President, 

 in the chair. — Captain Abney read a paper on the measurement 

 of the luminosity of coloured surfaces, which was illustrated by 

 experiments. In a communication to the Royal Society, General 

 Festing and the author have described a method of comparing 

 the intensity of the light of different parts of the spectrum, 

 reflected by various pigments, with that reflected from white, 

 and luminosity-curves have been constructed, the areas of which 

 give comparative measures of the total luminosities. This 

 method of comparison is accurate, but requires considerable 

 time, and the author has devised a more rapid process. The 

 coloured surface whose luminosity is to be compared with white 

 is placed beside a white patch within a dark box. A direct 

 beam of light passes through an aperture in the box, and a 

 black rod casts a shadow on the coloured patch ; another beam 

 from the same source is reflected at an angle, and forms a 

 shadow of the same rod on the white patch, the junction of the 

 two shadows coinciding with that of the two surfaces to be com- 

 pared. In the path of the direct beam is placed a rotating disk 

 with angular openings, adjustable whilst rotating by a simple 

 lever, and by this means the white patch can be made to appear 

 too light and too dark in rapid succession. By gradually dimi- 

 nishing the range of oscillation of the lever, a position of equal 

 luminosities can be found. The coloured surface is now replaced 

 by a white one, and the adjustment again made ; and from the 

 angular apertures required in the two cases the relative luminosi- 

 ties are determined. Comparisons made in this way (the num- 

 bers relating to which are given in the paper) with emerald green, 

 vermilion, French ultramarine, &c., gave results in close agree- 

 ment with those deduced from the luminosity-curves obtained 

 by the spectrum method. In reply to questions, Captain Abney 

 said the spectrum method was the more accurate, and could be 

 relied on to i per cent. The new method gave results within 2 

 per cent., showing that the eye is very sensitive to small changes 

 of luminosity when such changes take place in rapid suc- 

 cession. — Prof, Riicker made a communication on the sup- 

 pressed dimensions of physical quantities. In arranging a 

 system of dimensional equations for thermal quantities, the 

 question arises as to what are the dimensions of temperature. 

 A degree of temperature, as measured , by the ordinary arbi- 

 trary method of the mercurial thermometer, is not affected 

 by changes in the units of length, mass, and time ; but the 

 numerical values of thermal quantities (J, for instance) depend on 

 the scale of temperature adopted, say Centigrade or Fahrenheit. 

 Two courses seem open, either of which renders a complete 

 system of thermal dimensional equations possible : (l) temper- 

 ature may be considered as a measure of energy, as in the 

 kinetic theory of gases, and may be expressed as the energy of 

 translation of a standard number of molecules (say that number 

 contained in i cubic centimetre of air at standard pressure and 

 temperature) ; or (2) temperature may be considered as a second- 

 ary fundamental unit. If the first be adopted, the dimensions 

 of specific heat become M"^, and the temperature of o" C. is 

 expressed by 1-5207 x lo^ ergs. If a practical unit correspond- 



10' 

 ing to - ergs be adopted, this new unit of temperature will 



coincide with the Centigrade degree to about i part in 3000. 

 The chief objection to such a definition of temperature is that 

 the above relation between temperature and energy is not yet 

 proved to hold for liquids and solids.- If the second course be 

 adopted, the dimensions of all thermal quantities may be ex- 

 pressed in terms of M, L, T, and tf, where tf is the unit of 



