142 



NATURE 



[March 31, 1910 



In the " useful " liglit alloys of the ternary system, how- 

 ever, such disintegration does not occur. 



The mechanical properties found in the best of the alloys 

 of the ternary system are remarkable. The authors suggest 

 that alloys of copper with aluminium alone are generally 

 contaminated with a certain proportion of alumina, formed 

 when the aluminium is added to molten copper contain- 

 ing more or less oxygen ; by the previous addition of 

 manganese, which acts as a reducing agent, the formation 

 of this alumina • is inhibited and the properties of the 

 resulting' altoys are improved. It has been found that a 

 small percentage of manganese renders the alloys more 

 ductile without reducing their ultimate strength, while 

 larger proportions of manganese increase the strength but 

 lessen the ductility of the alloys. Sand-castings giving an 

 ultimate strength of 36 tons per sq. inch with an elonga- 

 tion of 22-5 per cent, on 2 inches have been obtained, while 

 in the form of rolled bars the best of the heavy ternary 

 alloys reach an ultimate strength of 43 tons per sq. inch 

 with 22 per cent, elongation. Perhaps the most remark- 

 able result, however, is that obtained with one of these 

 alloys in the hard-drawn condition, where an ultimate 

 strength of. 52 tons per sq. inch with 10 per cent, elonga- 

 tion has been obtained. This is probably the strongest 

 alloy known, containing more than 80 per cent, of copper. 

 The remarkable properties of this alloy led Dr. Rosenhain 

 to take up the challenge recently thrown out by Sir Gerard 



;?«?[«< 



mt' 





Fig. 4. — Micro-structure of alloy after quenching from 900° C. 



Muntz in his presidential address to the Institute of Metals, 

 to the effect that modern scientific attainments in metal- 

 lurgy had not yet enabled us to produce a bronze cutting 

 tool. By further cold-working one of these hard-drawn 

 alloys under a powerful testing machine, a material was 

 produced from which a chisel was ground, and with this 

 it was found possible to incise hard stone or — With a 

 different form of cutting edge — to cut wood so easily and 

 cleanly that a lead-pencil could readily be sharpened with it. 



A result of considerable importance has been obtained 

 by means of abrasion tests on certain of these alloys ; the 

 test applied was one of simple wear against hardened steel 

 rollers, the loss of weight being determined. It was found 

 that the best of these alloys possess a very great resistance 

 to abrasion of this kind far surpassing even the harder 

 varieties of steel. This property, together with their very 

 great strength and the fact that they can be machined 

 and finished very readily, should render them of special 

 value for the construction of scientific instruments, par- 

 ticularly for those parts where much mechanical wear has 

 to be met. These alloys also possess a very great power 

 of resisting corrosion, both by fresh- and sea- water, while 

 tests made upon them at temperatures up to 500° C. 

 indicate that they retain their strength up to 300° C. ; 

 these results suggest the possibility of their employment 

 for the blades of steam-turbines working with superheated 

 steam. 



As regards the light alloys, containing more than 95 per 

 NO. 2109, VOL. 83] 



cent, aluminium, the properties of the ternary alloys are 

 not markedly superior to' those of the aluminium-copper 

 alloys, except, perhaps, that in the form of chill-castings a 

 higher tensile strength has been obtained. The presence 

 of manganese, however, appears to protect these alloys 

 from corrosion to a marked extent, the specimens of these 

 alloys exposed to sea-water, for instance, becoming coated 

 with a black patina consisting largely of oxide of man- 

 ganese. Specimens of some of these alloys have retained 

 their original brightness after more than two years' ex- 

 posure in the laboratory cupboards, thus indicating a 

 decided superiority over pure aluminium, which has hitherto' 

 been regarded as decidedly superior to its alloys in regard 

 to corrosion. 



ELECTRICAL DISCHARGES OVER PHOTO- 

 GRAPHIC PLATES. 



T T is well known that when an electric discharge is 

 •*■ allowed to take place over a photographic plate a 

 latent image is formed which can be developed in the 

 ordinary way. When one electrode consists of a metal 

 plate placed at the back, and the other is a wire brought 

 into contact with the middle of the sensitive side, a very 

 great difference is obtained when a single spark is passed 

 from an induction coil or Wimshurst machine according 

 as to whether the wire electrode is the positive or the 

 negative one.. If it is positive, the figure — which is 

 called a positive figure — consists of numerous ramifications 

 suggesting meandering streams, while if it is negative the 

 main lines in the figure change their directions very 

 abruptly, and arc terminated by expansions suggesting fans 

 or palm-leaves. In each case the discharge may or may 

 not pass to the edge of the plate ; if it does, the corre- 

 sponding line is very broad, with a finer, well-defined, 

 intenser line passing midway through it. Such figures 

 were obtained first by Mr. J. Brown, of Belfast, and have 

 since been repeated by many experimentalists. Experi- 

 ments made by Prof. J. A. McClelland and Mr. Campbell 

 Swinton seem to render certain that the latent image is 

 due to the luminosity of the discharge, and not to a direct 

 electrical action. 



I have recently made experiments with the object of 

 extending our knowledge with respect to the formation of 

 these figures, and the results have been communicated to 

 the Rontgen Society, appearing in the society's journal for 

 January. 



My first idea was that if the paths of the discharge 

 represent moving electricity they would be seriously 

 modified in a perpendicular magnetic field. However, the 

 gi-eater part of the figure is apparently quite unchanged 

 in such a field ; the only evidence of change is in the trunk 

 discharges that flow over the edge of the plate. These 

 become still broader, the fine pilot spark, however, remain- 

 ing apparently unshifted, and forming a sharp boundary 

 to the trunk discharge along one edge. The direction of 

 lateral shift is that corresponding to a wire carrying a 

 current from the positive to the negative electrode. 



A blast of air produces the same kind of effect. Indeed, 

 if several radiating trunk disch.irges occur on the same 

 plate, the effect in a perpendicular magnetic field is much 

 the same as if a cyclonic blast of air had circulated over 

 the surface of the plate. 



The fine tracery lines in the fan-shaped expansions which 

 terminate the lines of discharge in the case of a negative 

 spark are very similar to the paths of the separate portions 

 of an exploded projectile. Such paths are easily plotted 

 by superposing a radial component of velocity, following 

 any assumed law, upon the initial translatory velocity of the 

 uncxploded projectile. Owing to this similarity it is sug- 

 gested that these tracery lines represent the actual paths 

 of single ions or of simple groups of ions in the electrical 

 field. 



When the discharge takes place in a partial vacuum 

 very considerable changes occur. In particular, as the 

 exhaustion proceeds a new phenomenon appears, which 

 reaches its most marked stage at about 17 cm. pressure. 

 This is somewhat difticult to describe, and is shown in 

 Fig. I. 



The wire terminal touched the plate nearly at its centre. 

 Besides the trunk discharge ascending the plate are seen 



