SCIENCE AND INDUSTRY. 
so-called residual copper is always redeposited copper. Dezincification 
is therefore only apparent and not real, and the term is always sub- 
sequently employed. by the authors in this sense. 
The layer type of dezincification, which is characterized by disin- 
tegration of the brass tube and redeposition of copper over large areas, 
has been found to oceur under both marine and fresh water conditions. 
Several ways in which such action may occur are indicated. Under 
fresh water conditions, it is often associated with acid water, par- 
ticularly if the acid is not too dilute and the access of oxygen is not 
very easy. 
Type IV., Plug or Local Dezincification—This type of local corro- 
sion may be regarded as a form of deposit attack, as it always proceeds 
beneath a deposit, and is stimulated by the presence of foreign bodies. 
It differs, however, in many ways from Type II. So far as is known, it 
occurs only in sea water—or diluted sea water—and is always associated 
with adhering white salt, consisting of colloidal zinc oxychloride, also 
containing some carbonate. It is readily reproducible in the laboratory, 
as it occurs spontaneously on the surface of 70 : 30 brass tubes after im- 
mersion in sea water at elevated temperatures—40 deg. to 50 deg. cent. 
—for a few days; and is hence rather more amenable to experimental 
study than the previous types of corresion. It is shown that the pro- 
duction of the right concentration of zine in the liquid layer adjacent 
to the corroding brass surface plays a large part in the formation of 
the characteristic white oxysalt, and that interference with the pro- 
duction of this condition, e.g., by lowering the zine contact of the brags 
or by raising the zine content of the sea water, is sufficient to prevent 
its occurrence. The dezincifying action is thought to be due to a 
small concentration of hydrochloric acid contained within the colloidal 
white salt. The difference in behaviour of different batches of 70 : 30 
tube—some always showing local dezincification, others never—persists 
after annealing or pickling, or both, and the reason for the difference 
in behaviour is still not clear. 
Type V., Water-line Attack.—In the case of a brass tube only par- 
tially immersed in sea water, increased corrosion—compared to that of 
the immersed portion—takes place, not at the water line as is commonly 
supposed, but above it, and sometimes as much as 2 em. above the 
air-sea-water surface. Further, the attack.is not uniform, but is con- 
centrated at areas where salt deposits have formed and is coterminous 
with the area covered by the deposits. Narrow bands of salt connect 
the sea water with the deposits. This type of corrosion is obviously 
a special form of deposit attack, taking place under the most favorable 
conditions, inasmuch as the attack beneath the deposits is much more 
severe than in any of the previous types of corrosion. This type of 
attack may occur at the inlet end of condenser tubes when entangled air 
clings to the surface of the tube and is prevented by eddying effects from 
being swept away by the water flow. 
The fourth section of the report contains an account of preliminary 
work on the electrolytic protection of condenser tubes. The particular 
question investigated was that of the efficiency of electrolytic protection 
in preventing deposit attack, 7.e., attack by cupric chloride solution. 
A piece of 70 : 30 brass tube made cathode to strip iron ina normal — 
358 
