ON THE CORROSION OF ALUMINIUM. 
239 
TABLE No. 30. 
Seven Bottles each Garden Tank and Peaty Matter, and Garden Tank and 
Ammonia Solution. 
Temperature 
(Fahrenheit) . 
— 
Garden Tank 
and l’eaty Matter. 
Garden Tank 
and Free NH 3 . 
Morning. 
Evening. 
Nos. 
Date. 
Free NH 3 . 
Album. NH 3 . 
Free NH S . 
Album. XH S . 
1876. 
73° 
75° 
1 
Dec. 11 
*72 
T9 
2*8 
*20 
77° 
78° 
2 
„ 12 
1*12 
*04 
2-80 
*20 
71*5° 
73° 
3 
„ 13 
1*12 
T3 
2-72 
*23 
76° 
78° 
4 
„ 15 
DOB 
T2 
2*70 
*21 
80° 
78° 
5 
„ 16 
•03 
*04 
*64 
TO 
74° 
75° 
6 
„ 19 
•02 
•03 
*56 
•07 
76° 
77° 
7 
„ 20 
•01 
•01 
*20 
•01 
The self -purification of the above samples is very remarkable. 
2.— ON THE COEEOSION OF ALUMINIUM. 
By A. LIVERSIDGE , M.A., F.R.S., Professor of Chemistry , University of Sydney, 
New South Wales. 
In order to ascertain the effects of the weather npon ordinary 
sheet aluminium, two shallow dishes were made of 1 mm. or ¥ yd nc li 
gauge metal, and exposed on the roof of the chemical laboratory of 
this University from 23rd November, 1893, to 7tli December, 1S94, or 
fifty-four weeks. 
The metal was made into basins, so as to catch rain-water and to 
give the salts, &c. f which it might have in solution a longer time to 
act upon the metal. 
The metal very soon lost its brilliancy, and became somewhat rough 
and speckled with grey spots mixed with larger light-grey patches ; 
it also became rough to the touch. The grey parts could be seen to 
distiuctly project above the surface, and under the microscope they 
presented a blistered appearance. This incrustation is held tenaciously 
and does not wash off, neither is it removed on rubbiug with a cloth. 
The raised parts are probably due to the formation of a hydrated 
oxide, but I am leaving the determination of the composition of this 
until I have a larger quantity at my disposal. 
Contrary to my expectations, the cups had not lost weight, but 
had even increased. 
One weighing 13*91 grammes had increased by *104 gr., and the 
other, weighing 13*S(35 gr., increased by *080 gr. After boiling in 
water for some hours and rubbing, the first still showed an increase of 
'077 gr., and the second of *055 gr. 
To ascertain the effects of common salt, a plate of the same 
metal, 3x4 inches, and weighing 19*829 gr., was repeatedly dipped 
in a solution of sodium chloride, and allowed to dry ; the alternate 
dipping and drying was repeated almost daily for three months ; the 
plate lost *019 gr., and after washing and rubbing dry *059 gr. 
