358 REPORTS ON THE STATE OF SCIENCE.—1912. 
sists of a saturated solid solution, rich in thallium, whilst more pro- 
longed heating causes the appearance of a second layer, composed 
mainly of the compound Bi,Tl,. The thickness of these two layers 
increases at higher temperatures at the expense of the original metals. 
The diffusion in this case is recognisable by means of the microscope 
after a year at atmospheric temperature. Lead and thallium behave 
similarly, as far as the evidence of heating curves goes, but a micro- 
scopical examination was not carried out. A determination of the 
electrical conductivity at different intervals shows that in this case also 
diffusion proceeds at a measurable rate at the ordinary temperature. 
The diffusion of tin into copper at 200°, and of zine into copper at 
400°, was also observed by more than one method. 
The last-mentioned process, the diffusion of zinc into copper, lends 
itself well to microscopical observation on account of the formation of 
compounds differing in colour from the original metals. The super- 
ficial conversion of copper into brass by the action of zine vapour has 
been known since the experiments of Spring, or even longer, and has 
been made the basis of a technical process.°* The object to be coated 
is packed in zinc dust, mixed. with zine oxide or carbon to prevent 
eaking, and heated at 250° to 300° in a rotating drum. The process 
has found a more extensive application in the coating of iron. with 
zine as a substitute for wet galvanising, and is known as ‘ Sherardis- 
ing.’ The zine vapour unites with the metal (iron or copper) to form 
a superficial alloy, but the subsequent penetration of zine into the 
interior takes place by true diffusion. 
The manner in which diffusion takes place in such complex cases is 
conveniently studied by immersing a rod of copper in molten zinc and 
withdrawing it after a short time. An adherent layer of a white brittle 
alloy is formed, and microscopical examination of a transverse section 
shows that this is separated from the copper by an intermediate layer 
which is readily identified as the # constituent.*® There is a sharp 
boundary between this layer and the unaltered copper. If the metal is 
heated for several hours at 700° and slowly cooled, it is found that 
zine has diffused inwards, the white outer layer disappearing, whilst 
the zone of the B constituent has greatly increased in breadth, and 
has formed a new layer, composed of the a constituent, between it and 
the unchanged copper core. 
The most recent, and in some respects the most satisfactory, experi- 
ments are those of G. Bruni and D. Meneghini.7° These authors 
employed metals of high melting-point, which form solid solutions 
without chemical combination, and they followed the progress of the 
diffusion by means of the change in the electrical conductivity. A 
nickel wire, 0°5 mm. diameter, was coated electrolytically with copper 
until the increase of weight corresponded with 59 per cent. copper and 
41 per cent. nickel. The compound wire was then heated to 1,000° 
in hydrogen, and the conductivity was determined from time to time. 
68 §. Cowper-Coles, Electrochem. and Metall., 1904, 8, 828. 
69 ©, H. Desch, ‘ Metallography ’ (London, 1910), p. 221. 
7 Atti R. Accad. Lincei, 1911 [v.], 20, i. 671, 927; Intern. Zeitsch. Mctallographic, 
1912, 2, 26. 
