PROFESSOR W. C. ROBERTS-AUSTEN ON THE DIFFUSION OF METALS. 407 
Stodart, # who, in the course of an investigation on the alloys of iron with other 
metals, noted their failure to produce certain alloys by “ cementation,” but consider 
it “ remarkable,” in the case of platinum, that it will unite with steel at a tempe¬ 
rature at which the steel itself is not affected. They also show that solid steel 
and platinum, in the form of bundles of wires, may be welded together “with the 
same facility as could be done with iron or steel,” and they observe that on etching 
the surface of the welded mass by an acid “ the iron appeared to be alloyed with the 
platinum.” Their interest in this singular fact led them to promise some direct 
experiments on “ this apparent alloy by cementation,” that is, by the interpene¬ 
tration of solids. Since this time there have been many more or less isolated 
observations bearing on the subject, and brief reference may be made to the more 
important of them in chronological order. In 1877 Chernoff t showed that if 
two surfaces of iron of the same nature be placed in intimate contact and heated 
to about 650° they will unite. In 1882 Spring| made his remarkable experiments 
on the formation of alloys by strongly compressing their constituent metals at 
the ordinary temperature, w T hile in 1885, O. Lehmann§ suggested, and in 1888 
Hallock|| demonstrated that compression is not necessary, as alloys might be formed 
by placing carefully-cleaned pieces of two constituent metals in juxtaposition and 
heating them to the melting point of the alloy to be formed, which was, in some 
cases, 150° below the melting point of the more fusible of the two metals. In 1889 
Coffin showed, and I have repeatedly verified the accuracy of his experiment, that 
if the freshly-fractured surfaces of a steel rod, 9 '5 millims. square, be placed together 
and heated to below redness, they will unite so firmly that it is difficult to pull them 
apart by hand. The steel is highly carburized and the diffusion of a carbide of iron 
probably pla} 7 s an important part in effecting the union. There must have been 
molecular interpenetration in this case, though the steel was at least 1000° below 7 
its melting point. In 1894 SpringIF proved that if the carefully surfaced ends of 
cylinders of tw T o metals were strongly pressed together and maintained for eight 
hours at temperatures which varied from 180° to 400°, interpenetration v 7 ould take 
place, true alloys being formed at the junction of the two metals. 
In these experiments, which are of great interest, the temperatures at which the 
cylinders were maintained were below the melting point of the more fusible of the twm 
metals. Care appears to have been taken to avoid heating them up to the melting 
point of the eutectic alloy, though it was in some cases close to it. The necessity for 
* ‘ Quarterly Journal of Science,’ vol. 9, 1820, p. 319. 
t 1 Revue Universelle des Mines,’ vol. 1, 1877, p. 411. 
X ‘ Ber. der Deutsch. Chem. Gesell.,’ 15, 1882, p. 595. 
§ ‘Wied. Ann.,’ 24, 1885, p. 1. 
|| Communicated to Phil. Soc. of Washington, Feb. 18, 1888 ; ‘ Zeitschr. Phys. Cliem.,’ 2, 1888, p. 6, 
or ‘ Chem. News,’ vol. 63, 1891, p. 17. 
% ‘ Bull, de l’Acad. Royale de Belgique,’ vol. 28, 1894, p. 23. 
