264 REPORT—1840. 
mentally obtained in column five, do not follow the ratio of the 
amount of copper, increasing as it increases, although their 
general tendency is towards this; the greatest perturbations 
take place in the interpolated series(17 Zn + 8Cu)to(23 Zn + 8Cu). 
These specific gravities are taken on the alloys just as they were 
cast, and suddenly cooled in the cast-iron mould; but on sub- 
mitting some of them to lamination, very variable amounts of 
condensation took place. Hence it is probable that sudden 
cooling produces an effect analogous to tempering in cold water 
on the alloys of copper and zinc. Dussaussoy found that as 
the latter became soft and malleable by tempering (trempe), 
their specific gravities were reduced in variable proportions ; 
and it matters not whether this tempering be effected in water, 
or by sudden cooling in a metallic mould; the alloys now in 
question have therefore been submitted to this process ; and as 
no link at present exists connecting the density of such an alloy 
with its specific gravity after lamination, the densities now given 
in Tables XIV. and XV. will be found, in most instances, not 
to correspond with those occasionally given in books, and 
which have been chiefly made on alloys submitted to compres- 
sion, or cast and cooled in various ways. The densities given, 
however, are I believe close approximations to the truth; and 
all the alloys having been cast in the same way, at the same 
temperature, and cooled at the same rate, these specific gravities 
must be relatively correct. 
258. Table X. contains the results of the action of sea 
water on cast iron, in presence of copper and tin, or their alloys. 
What has been said of the preceding explains the general na- 
ture of this Table also. It is therefore only necessary to re- 
mark here, that as copper and tin are each singly electro-negative 
with respect to iron*, they both, together with every alloy in 
the Table, increase or accelerate the rate of corrosion of cast 
iron in a solvent, though in every variable degree. 
The maximum increase is produced by tin alone ; which in- 
dicates that tin is more powerfully electro-negative to cast iron 
than copper, contrary to the opinion previously held. The in- 
crease of corrosion produced generally by alloys of copper and 
tin is far greater than by those of copper and zinc: hence the 
important practical deduction, that when submerged iron-works 
must be in contact with either alloy, common brass, or copper 
and zinc is much to be preferred to gun-metal, although con- 
trary to general practice amongst engineers. 
The losses on these alloys in column seven, as in the former 
case, do not apparently follow the law of volta-equivalents. 
* Report, § 96. 
