NOVEMBER 8, 1906 
9 
NATURE 43 
THE INTERNAL ARCHITECTURE 
METALS. 
rf T has been 
cynically remarked that to 
successful scientific lecture to a cultured 
is necessary to divide the lecture into three 
first part should be understood both by the 
OF 
deliver a 
audience it 
parts. The 
audience and 
the lecturer; the second part by the lecturer and not by 
the audience; and the third part neither by the audience 
nor by the lecturer. 
If the foregoing dictum were true, the speaker found 
himself in a paradoxical position. The object of the dis- 
course was to make the subject under consideration as 
clear as possible throughout, hence the more nearly this 
object was achieved, the more unsuccessful the lecture. 
The title of the discourse might seem to some far-fetched, 
since, superficially, a bar of polished brass or steel appar- 
ently presented the archetype of a homogeneous solid. 
Any such idea, however, must in a few moments be dis- 
pelled. Taking a section of pure gold, or at any rate of 
gold of a purity of 99-995 per cent., this, when polished 
power of the micro- 
the etching figures of 
in different crystals. 
and etched, presented under a low 
scope large allotrimorphic crystals, 
which exhibited varying orientation 
Hence (see Fig. 1) one crystal might appear black, another 
show the brilliant yellow of gold, and a third exhibit 
middle tone. All these were purely optical effects. In the 
Fig. 2 shows a micro-section of the structure of gold 
to which o-2 per cent. of bismuth had been added Che 
microscope had at once explained the hitherto mysteriou 
action’ of bismuth. It indicated clearly that the 
quantity of bismuth alloyed with a definite amount of gol 
forming a constituent having a much lower freezing poir 
than the main mass. Hence, when crystallisation set in 
during solidification from a_ series of centres, the 
“eutectic ’’ or constituent last fluid was expelled to the 
exterior of each crystalline grain of pure gold, thus envelop- 
ing each crystal in a membrane of gold-bismuth alloy 
having a much higher coefficient of contraction than the 
crystal itself. Hence, during cooling, the gold-bismuth 
alloy, which may be regarded as the mortar of the struc- 
ture, to a Rancideranies extent detached itself from the 
crystalline grains of gold which may be regarded as the 
stones of which the mass is built up. In the micrograph, 
Fig. 2, the stones of tough gold are represented as white, 
whilst the mortar of gold-bismuth eutectic is shown as 
dark, thick, enveloping membranes. These membranes 
become pz ye well below a red heat, and it was proved 
that at 4oo° C. the mass could be powdered in a mortar, 
the crystalline grains of pure gold becoming detached from 
the feeble alloy cementing them together. One of these 
crystalline grains exhibited no signs of the brittleness of 
the mass from which it was thus detached, but was readily 
beaten out into gold leaf in the ordinary manner 
Fic. 1.—Gold. 
black crystal the orientation was at such an angle as to 
reflect the light entirely outside the objective, whilst, going 
to the other extreme, the gold-coloured crystal had a 
molecular orientation which reflected the light entirely into 
the objective. It was well known that the addition of 
ne or two tenths per cent. of the metal bismuth to gold 
produced a surprising mass brittleness which naturally led 
to the enunciation of theories to account for so remarkable 
a phenomenon. 
Twelve years ago the theory which commanded a general 
acceptance, and at that time reasonably so, was that the 
small quantity of bismuth was incapable per se of pro- 
ducing so profound a mechanical change as to convert one 
of the most ductile of metals into a mass possessing an 
‘almost glassy brittleness. Therefore, the metal bismuth 
must act indirectly, its presence determining the mainten- 
ance of the molecules of gold in a brittle allotropic 
modification. 
In 1896 there was published in Engineering from the 
laboratories of the Sheffield College an unambitious re- 
search recording the discovery of eutectic cements, which 
‘to a considerable extent altered the whole trend of metal- 
iurgical thought. 
1 Abstract of a discourse delivered at the Royal Institution on Friday, 
Webruary 23, by Prof. J. O. Arnold. 
NO. 1932, VOL. 75 
Fic. 2.—Gold containing o*2 per cent. of bismuth. 
Passing from gold to brass, it was proposed to diverge 
from the abstract to the concrete, and to show the value 
of the application of the science of metallurgy to practical 
problems connected with mysterious failures in marine 
engineering. 
A notable 
case in point was the explosion of the brazed 
copper main steam-pipe of the s.s. Prodano in calm weather 
off the Kentish Knock at a sure about one-tenth of 
that to which it had been previously tested. In this case 
the microscope was again successful in clearly indicating 
the nature of the electrolytic decay, under certain conditions, 
of brass used in naval architecture. In this connection, a 
familiar phenomenon is the decay of Muntz metal bolts 
exposed to the action of bilge water, Such bolts break 
suddenly, and present a distinctly coppery fracture. <A 
micrographic examination of such bolts usually revealed 
a minor area of undeteriorated brass and a major area of 
deteriorated brass—that was to say, brass which had been 
more or less dezincified, an expression which meant, in 
other words, that the mass had become transformed into 
rotten, spongy copper. 
Brass often consisted of two constituents, namely, a 
ground mass of true brass of formula Cu,Zn and a 
eutectic corresponding to the formula Zn,Cu. Upon a 
mass so constituted a feeble saline electrolyte attacked in 
