ON THE CHEMICAL NATURE OF ALLOYS. 45 
ordinary state; hence the great evolution on the addition of the first portions 
of zinc to the melted copper, for it only requires a small quantity of zine to 
convert copper into its allotropic modification. Person has already proved that 
heat is evolved when melted lead is added to melted bismuth, and he explains 
it by assuming that the specific heat of the alloy is less than that of the com- 
ponent metals. The great evolution of heat might possibly be an indication 
of the existence of chemical compounds in the liquid alloy ; but of this we 
have at present no data upon which we can go, as very little is known as to 
the behaviour of these alloys in a liquid state. 
Storer in the same paper* states, ‘“ Upon the assumption that the crystals 
of the alloys of copper and zinc belong to the regular system, as well as upon 
the fact that none of the crystals have been found to contain any larger 
quantity of the component metals than was contained in the remainder of the 
molten liquid from which they were separated, I have based my conclusion 
that the alloys of copper and zinc are isomorphous mixtures of the two metals ; 
on this hypothesis it is of course presumed that both copper and zine are 
capable of crystallizing in the regular system.” And, further on, “ Indeed 
these fibres, although described by Calvert and Johnson as prismatic crystals 
indicating that the alloy CuSn is a definite chemical compound, are evidently 
nothing more than a collection of octahedral crystals, similar to those which 
form the fibres of sublimed sal-ammoniac and of several metals.” 
This answer, respecting the existence of chemical compounds in these alloys, 
is sufficient to prove their non-existence, more especially when it has been 
shown, which I have already pointed out in this Report, that alloys of a de- 
finite crystalline form are not necessarily chemical compounds. 
Storer, as will be seen from the above, has arrived at nearly the same con- 
clusions with regard to the chemical nature of these alloys as I have done, 
the main difference being that he has not taken into consideration the marked 
change in most of the physical properties of copper when it is alloyed with 
traces of zinc. 
It has been shown that the action of reagents on these alloys is different to 
their action on the two metals in contact with one another, and that this 
is an indication of the existence of chemical compounds. By the above 
hypothesis, this behaviour may be explained by assuming that the action of 
reagents on the allotropic modification of copper is not the same as their action 
on ordinary copper; when, therefore, we try the action of different reagents 
on alloys, we cannot expect to find the same results as when experimenting 
on the metals unalloyed in contact with one another. Another reason for 
the different action of reagents on alloys and on the metals in contact with 
one another is, that in every case one metal is attacked more easily than the 
other, so that after a certain time the mass of the alloy becomes covered with 
a coating of the more difficultly soluble metal, whereby it is protected from 
the further action of the reagent (gold-assaying). 
What was said of the lead-zine alloys may also apply to those of bismuth- 
zine ; for when these metals are fused together, they do not mix with each 
other, but separate into two layers, the upper one (zinc) containing 2-4 per 
cent. bismuth, and the lower one (bismuth) with from 9 to 14 per cent. zine. 
Here then, again, we have another example of mechanical mixture ; for if these 
metals were mixed together in a liquid state, such a mixture would be one of 
a solution of the allotropic modification of bismuth in zinc, and one of zinc in 
the allotropic modification of bismuth ; or if the mixture were cooled rapidly, 
it would be a mechanical mixture of a solidified solution of the allotropic 
* Memoirs of the American Academy, vol. viii. p. 29. 
