ON THE ACTION OF AIR AND WATER UPON IRON, 259 
The primary object of these two series of experiments was 
to determine, in all its generality, the question as to the pre- 
servative or non-preservative power of brass or gun-metal to 
iron in sea water, a statement affirmative of which, it will be 
recollected, was made at the meeting of the British Association 
at Liverpool. This question has been pretty fully discussed in 
my previous report, and it was therein shown that neither brass 
nor gun-metal, as commonly so called, had any protective 
power (of an electro-chemical character) over iron in water, 
but, on the contrary, promoted its corrosion. The few experi- 
ments on which this limited conclusion was made, were tried 
on alloys of uncertain, or at least non-atomic constitution; it 
was desirable not merely to set the question of protective power 
finally at rest, but to establish a set of practical data for the 
engineer as to the actual amount of increment or decrement of 
corrosion of iron due to the presence of various alloys of the 
orders brass and gun-metal, when immersed in sea water. 
245. It is obvious that this question is only a particular case 
of a much more general one, namely, if there be three metals, 
A, B and C, whereof A is electro-positive, and C electro-nega- 
tive with respect to B, and capable of forming various alloys, 
A+C, &c.; then if B be immersed in a solvent fluid in pre- 
sence of A, B shall be electro-chemically preserved, and A cor- 
roded, and vice versd. If B be so immersed in presence of C, 
B will be dissolved or corroded, and C electro-chemically pre- 
served, the amount of loss sustained in either case by the posi- 
tive metal being determined according to Faraday’s general law 
of volta equivalents. 
But now let various alloys be formed, having atomic constitu- 
tions, as 2 A+C, A+C, A+2C, &c., and let B be exposed 
tothe same solvent in presence of each. Query, what will be the 
electro-chemical relation of the metal B to each alloy, in respect 
to preservation, or amount of loss by corrosion? and what will 
be the nature and amount of the reactions of several such alloys 
upon an acid or saline solution, of a third metal, or of either 
_ of those constituting the alloys? thus, 
246. When the metals, zinc and lead, and their alloys, having 
_ the compositions (4Zn + Pb), (3Zn+ Pb), (2Znu-+ Pb), (Zn-+ Pb), 
- (Zn +2 Pb), (Zn+3 Pb), (Zn+4 Pb), are immersed under simi- 
lar circumstances in a solution of acetate of lead, it would be 
presumed that the decomposing power of every alloy would be 
in proportion to the quantity of zinc enter ing into its composi- 
tion. The result is not so, however. The zinc and the alloys 
(4 Zn + Pb) and (3 Zn + Pb) at once reduce the lead of the acetate 
of lead to the state of metal, ang as rapidly as zinc alone; after 
s 2 
