September 29, 1923] 



NATURE 



491 





With alloys which solidify over a range of tempera- 

 ture, the slower the rate of solidification the lower is 

 the density, and when they are prepared in cylindrical 

 chill moulds, they are less dense in the centre than 

 at the outside. When prepared in chill they are 

 cicher on the outside in the component of the lower 

 '" elting point. — A. H. Mundey and C. C. Bissett : 

 he effect of small quantities of nickel upon high- 

 ade bearing metal. Nickel is now added to the 

 Veil-known bearing metal consisting of tin 93 per 

 cent., antimony 3-5 per cent., copper 3-5 per cent. 

 Tensile, compression, and hardness tests gave no 

 indication of improvement. The comparison of hard- 

 ness at varying increased temperatures exhibited no 

 improvement. In the case of the alloy with no nickel, 

 the hard copper-tin constituent is very marked in its 

 characteristic crystalline formation as seen under the 

 microscope. Tlie presence of nickel even in small 

 quantities results in a great diminution of this 

 crystalline structure. — Hikozo Endo : The measure- 

 ment of the change of volume in metals during 

 solidification. In the casting process it is very im- 

 portant to know to what extent a change of volume 

 occurs during solidification. In 1888, Vincentini and 

 Omodei calculated the change of volume of some 

 fusible metals during solidification from the change 

 of density at the melting point. E. Wiedemann, Paul 

 Pascal, and Louis Hackspill also used this method. 

 M. Toepler studied the change of volume by means 

 of a dilatometer ; he suggested a relation of the 

 change of volume of a metal at melting point to its 

 atomic weight. K. Bomemann and F. Sauerwald 

 measured the density of metals at various high tem- 

 peratures, using the principle of Archimedes, by means 

 of a mixture of sodium and potassium chlorides as 

 liquid. The method of investigation now used for 

 metals having melting points up to iioo°C., which 

 was suggested by Prof. K. Honda, consists in the 

 measurement of the change of buoyancy of a metal 

 suspended in an inactive liquid during its solidifica- 

 tion or melting by means of a thefmobalance. 



September 12. — Marie L. V. Gayler : The constitu- 

 tion and age-hardening of the quaternary alloys of 

 aluminium, copper, magnesium, and magnesium sili- 

 cide. Alloys containing up to 6 per cent, copper, 



4 per cent, magnesium, and 4 per cent, magnesium 

 silicide were used. When copper, magnesium, and 

 magnesium silicide arc present in aluminium, any two 

 of these components have a marked effect on the 

 solubility of the third and ultimately CuAl.^ and MgjSi 

 are both thrown out of solution. If copper and 

 magnesium are present in a ratio greater than 12 to 



5 approximately, then the alloys when quenched from 

 high temperatures age-harden at room temperature, 

 owing to the difference in the solubility of MgjSi at 

 the quenching and ageing temperature. Age-harden- 

 ing of alloys of the " Duralumin " type is due primarily 

 to MgjSi, and the addition of magnesium and copper 

 is important since both reduce the solubiUty of 

 MgjSi at high and low temperatures and consequently 

 reduce the maximum age-hardness due to MgjSi. — • 

 Ulick R. Evans : The electro-chemical character of 

 corrosion. There are two main types of corrosion : 

 (i) that accompanied by evolution of hydrogen is 

 characteristic of reactive metals placed in acid solu- 

 tions, but the velocity varies greatly with the degree 

 of purity of the metal ; (2) slower corrosion, deter- 

 mined by the diffusion of oxygen to the metal, and 



omparatively independent of the purity. When a 

 metal is immersed in a solution of potassium chloride, 

 ill kali is produced at the cathodic portions, the 

 chloride of the metal at the anodic portions, and the 

 liydroxide is precipitated where these meet. The 

 electric current produced accounts for the greater 

 part of the corrosion actually observed. Generally 



NO. 2813, VOL. I 12] 



the cathodic areas are those to which air has free 

 access, while the anodic areas are those protected 

 from aeration. Corrosion usually proceeds most 

 rapidly at the comparatively unaerated places — hence 

 the intense corrosion observed in " pits " and over 

 areas covered up by porous corrosion-products. — 

 Douglas H. Ingall : Experiments with some copper 

 wire ; cohesion a function of both temperature and 

 cold work. Five samples of copper wire were used : 

 soft annealed and four degrees of cold work given by 

 25, 40, 50, and 75 per cent, reduction of area by 

 drawing. The cohesion at high temperatures was 

 determined by placing given loads on the wire at 

 atmospheric temperature, heating the wire and deter- 

 mining the temperature at which it broke. All the 

 samples gave similar graphs in which with rise of 

 temperature the cohesion decreased along a straight 

 line to a constant critical temperature of 350° C, 

 beyond which the cohesion was represented by a 

 sharply descending curve. The equations to the 

 straight Unes C = a - 6T and to the curves TC" = k 

 (where C = cohesion and T - temperature) showed that 

 the percentage increase of the constant h and the 

 percentage decrease of the constant n were repre- 

 sented by the corresponding percentage reductions for 

 any given cold worked wire, with the exception of 

 75 per cent, reduced wire. At the critical inflection 

 temperature the material was comparatively ex- 

 tremely fragile. — D. Hanson, C. B. Marryat, and 

 Grace W. Ford : Investigation of the effects of im- 

 purities on copper. Pt. I. — The effect of oxygen on 

 copper. The effect of oxygen, up to a concentration 

 of 0-36 per cent., on pure copper, was investigated. 

 The mechanical properties are not much affected by 

 small quantities of oxygen, and copper containing as 

 much as o-i per cent, differs very slightly from pure 

 copper. The electrical conductivity does not fall 

 rapidly, and values exceeding 100 per cent, of the 

 International Standard are obtained in all annealed 

 materials containing less than o- 1 per cent, of oxygen. 

 This is due to the low solubility of the oxide in solid 

 copper. The oxygen-bearing metals can be con- 

 sidered as a heterogeneous mixture of pure copper 

 and finely divided particles of cuprous oxide. There 

 is a soft ductile copper matrix, in which harder 

 particles of cuprous oxide are distributed so as to form 

 a mechanical mixture. — Hugh O'Neill : Hardness tests 

 on crystals of aluminium. Brinell tests showed that 

 at low loads the different crystallographic planes 

 resist penetration to different degrees, and give in- 

 dentations of different shapes. In the Brinell sense 

 the (no) face is the " hardest " and the cube (001) 

 face appears to be the " softest." But the load 

 required to immerse the ball is apparently the same 

 in all cases. Crystal boundaries are without any 

 appreciable effect in increasing the resistance of 

 aluminium to penetration. — H. I. Coe : The behaviour 

 of metals under compressive stresses. Compression 

 tests carried out on small cylinders of metals show 

 that with successive increments of loads plastic flow 

 occurs, after the elastic limit has been exceeded, at an 

 increasing rate. At a certain load the rate of flow 

 changes abruptly, metals such as tin and lead be- 

 coming perfectly plastic, harder metals becoming 

 more plastic than under preceding loads and im- 

 mediately succeeding loads. The term " critical 

 plasticity " is used to indicate the change in the rate 

 of plastic deformation which most metals exhibit at 

 a pa rticu lar load . Annealed metals flow at a compara- 

 tively low load and the rate of flow increases up to 

 the load corresponding to critical plasticity ; when 

 worked, they arc more resistant to compressive 

 stresses until they approach the load corresponding 

 to a critical plasticity, when they suddenly collapse 

 and a marked temporary flow occurs. — Albert M. 



