610 REPORT — 1887. 



7. On the Physical Properties of a nearly Non-Magnetisahle (^Manganese) 

 Steel. By Professor W. F. Barrett. 



Early in 1884 Messrs. Hadfield and Co., steel founders of Sheffield, exhibited 

 at the Institute of Mechanical Engineers specimens of steel which they had 

 recently manufactured, containing from 10 to 13 per cent, of manganese. Contraiy 

 to the general belief at the time, this steel was found to be extremely tenacious 

 and tough. At the Aberdeen meeting of the British Association Mt. J. T. Bot- 

 tomley drew attention to the fact that this steel was almost unmagnetisable. His 

 experiments showed that the intensity of magnetisation that could be imparted to 

 it was from 3,000 to 7,700 times less than that which could be given to ordinary 

 Steel. The author of the present paper has, through tlie kindness of Messrs. 

 Hadfield, succeeded in obtaining this steel drawn into wire, but only after re- 

 versing the ordinary annealing process ; quenching the manganese steel rods in 

 cold water rendered them ductile, and thus lengths of wire were drawn of No. 

 13 and No. 19 S.W.G. Tlie wire was of two kinds, hard and soft, the latter being 

 as flexible as soft iron wire. This steel contained 13'75 per cent, of manganese, 

 a,nd had a density of 7'81. The hard wire easily scratched steel, not hard tem- 

 pered. Exposed to the air, it rusts rather more quickly than ordinary steel, but 

 not so quickly as iron. The modulus of electricity (Young's modulus) was found, 

 the mean of numerous observations giving 1,680 x 10" grammes per square centimetre 

 for the hard wire, and 1671 x 10'^ grammes per square centimetre for the soft wire. 

 These numbers are lower than either iron or steel. The breaking strain of the 

 No. 19 soft manganese steel wire was found to be 48'8 tons per square inch with 

 18 per cent, elongation. The hard wire of the same gauge had the enormou.s 

 breaking strain of 110 tons per square inch, but snapped with scarcely Jiny 

 appreciable elongation. Steel pianoforte wire is the only material with which 

 the author is acquainted that exceeds this tenacity. The electric resistance of the 

 wire was found to be 78 microhms per cubic centimetre. This is more than six 

 times the resistance of iron and three times the resistance of German silver. The 

 resistance temperature coefficient was found to be 0'136 per cent, for 1° C. for 

 a range of 200° 0. This is much lower than iron, which has a temperature 

 coeflicient of 05 per cent, for 1° 0. ; but it is higher than German silver, which 

 gave only 0'04 per cent, for 1° 0. Hence for resistance coils for electric lighting 

 manganese steel wire may be useful. The magnetic susceptibility of manganese 

 steel was also carefully examined and found to be extremely low : in similar power- 

 ful magnetic fields, if iron be taken as 1,000, manganese steel is less than 3. 

 The enormous magnetic change wrought in this material by the alloy of 

 12 to 13 per cent, of manganese is very remarkable, and indicates a valuable 

 application of this material for the bed plates of dynamos and for iron-plated 

 vessels. An iron-clad built of manganese steel would not only be of great strength, 

 but would have practically no deviation of the compass. 



In conclusion the author pointed out that manganese steel wire does not 

 exhibit the anomalous expansion on cooling and recalescence which is found in 

 ordinary iron and steel wire. This affords new evidence of the connection between 

 these peculiar molecular phenomena and the magnetic state of the body. 



8. On the Application of the Centi-ampere cr the Deci-ampere Balance for 

 the Measurement of the E.M.F. of a Single Cell. By Professor Sir 

 William Thomson, F.B.S. 



For the purpose of measuring the E. M. F. of a single cell the centi-ampere 

 or the deci-ampere balance is put in circuit with a battery of a sufficient number 

 of cells, a rheostat, and a standard resistance, in the manner shown in the diagram. 

 The current measured by the balance is then varied by means of the rheostat until 

 the difierence of potential between the ends of the standard resistance is exactly 

 equal to the potential of the cell. This equality is tested by placing the cell in 

 series with a mirror galvanometer or a quadrant electrometer in a derived circuit. 



