40 ON THE ALLOYS OF COPPER AND ZINC. 
Now since many of the impurities of copper are eliminated when it unites 
with zinc, being thrown up on the surface of the melted metal as a crust or scum, 
which is removed by the workmen,* and as no two samples of copper contain 
the same kind or quantity of impurity, the amount of it removed, or, what is the 
same thing, the amount of copper left to unite with a given weight of zinc, T must 
continually vary. 
From this it may easily be seen how very difficult, if not impracticable, it would 
be to obtain anything like constant results, unless some test were found by which the 
workmen might ascertain at least approximately the composition of his alloy. + 
That it is of the first importance that every alloy used for sheathing shall 
possess some one constant composition, so that it may endure equally well on 
all parts of the vessel, and that no galvanic action may occur between the metal 
of different sheets, is a point too obvious to be mentioned. 
It might still be urged against yellow-metal, that the admixture of other metals — 
of kinds capable of mingling with it — which may have been derived from the copper 
used in its preparation, is a serious objection to its use. This is true, and the remark 
applies with equal or even greater force to all the alloys used for sheathing; 
it must always depend upon the condition of copper metallurgy. It would apply 
more forcibly, for example, to the low-grade brass which Bobierre has proposed 
to substitute for the ordinary sheathing-metal; for since the “test” by means of 
which founders are enabled to prepare the alloy — yellow-metal — of constant com- 
position cannot be obtained if metals are present which are incapable of mixing 
in all proportions with this alloy, in which event small particles of the foreign 
substances would be found irregularly disseminated upon the fractured surface of 
the test ingot, manufacturers are enabled to ascertain at once whether or no a given 
sample of copper is suitable for the preparation of yellow-metal ^ Whenever it is 
* This explains the remark of Karsten (loc. cit., S. 386), that perfectly pure copper can take up from 
1.5 to 2.5 per cent more zinc than impure, and still afford a product of better color, more tenacious and 
more malleable. 
f Much zine also is lost in this case, both by alloying with the foreign metal and by uniting with 
any oxygen which may have been combined with the latter. As the workmen say, impure copper 
* burns up" a great deal of zinc. 
f It must be well understood, however, that as a rule manufacturers of xn of copper and zinc always 
use the best copper they can obtain, since it is generally more economical for them to do so. For the 
purer this metal is, so much the greater will be the total amount of alloy obtained by the use of a given 
weight of it, or, in other words, the loss, mentioned in the preceding notes, which would occur from 
elimination of impurities, will be smaller. 
