202 
PROCEEDINGS OF THE NATURAL [APRIL, ’89. 
He passes the current through Al 2 0 8 , using copper as 
negative electrode on the bottom of a carbon crucible and 
a carbon rod as positive electrode, and extending down 
from the top nearl\ r or quite touching the copper. A 
powerful current fuses the Alo Oy , which, at this high 
temperature, in presence of the carbon and copper, is 
reduced, and forms with the copper A1 bronze. That 
there is no electrolysis I am convinced from the following 
experiments: 
I passed a current of 80 amperes for one-half hour 
through A1 contained in a carbon crucible having copper 
m the bottom. Sometime after fusion took place, the cur¬ 
rent was broken and a sensitive galvanometer showed no 
deflection of the needle when put in the alumina current. 
Theie seemed to be no electrolytic conduction, inasmuch 
as the current did not pass with 50 volts if the carbon 
electrode did not form at least a loose contact with the 
copper or was not connected by particles of carbon. Fur- 
er, the direction of the current made no perceptible differ¬ 
ence with the amount of A1 set free to unite with the cop- 
P o ortn bronze. I also used an alumina crucible, with 
copper and platinum electrodes passing through the cru¬ 
cible nearly horizontally and forming loose contact in the 
cento of the mass of alumina. No alloy of copper was 
A ‘ an<1 "° could be ob- 
‘ IO “ e ' '»as complete, and of coarse, 
w,th ^ the copper and platinum 
S e„To a ” d “ ir ' r re partim, " 1 y to present in the pre- 
d N L C ‘ 
erodes and more especially lead. 
Lead (Pb), Tin (Sn), Zinc (Zu) 
Antimonv ^ C_adrniuni (Cel), 
read,rallo? ^ithtn “To (M) ™ed. Thev all 
greater part of the No whin rta~ d “"° yS *"* ” P 
1 aced m melted cryolite, and 
