48 Tlie American Geologist. J^'J'- 1^02. 
ally by comparison with a solution of chemically pure EroOg 
of known strength. Knowing the amount of Kr^O^ present, 
and also the equivalent weight of the mixture of the earths, 
it was a very easy matter to determine the amounts of YoOj 
and Gd.O,.' 
Spectroscopically we had found 24.6;/ Er^O^ in the mix- 
ture of the earths. 
Xow if, 
x=:percentag-e of YoO;. and 
yzizpercentage of Gd.O.,, then 
we have, 
x--y 4-24.6— 100. 
Also ^^ + -^ + -J:^ = ±^ 
226 360 380 254.5 
From which we obtain 
x=72. i>'YoO.., 
y=3-3'^Cxd;03. 
It may be further mentioned tliat the mixture of the earths 
was very carefully tested for cerium and didymium. and that 
even the most sensitive reactions failed to show a trace oi 
these elements. Lanthanum is certainly not present for the 
spectrum, obtained by the method already mentioned, did not 
show any brilliant lines characteristic of oxide. The re- 
sults of the quantitative analysis are as follows : 
I. II. III. 
S03=6.i37r not det. 6.13% 
P2O3=33-50 ' 33-52% 33-51 fYsOs 
R203=6o.2i 60.28 60.24 ^Er203 
Fe203notdet. 0.20 0.20 (Gd203 
100.08% 
I, first analysis; II, second analvsis ; III, average of I. 
and II. 
The earths are present as follows : 
( Y,0 3=43-43;^ 
R.,03^o.24= - Er.,0,-14.82 
( Gd,03=i.99 
Taking 254.5 as the molecular weight of RoO.., the follow- 
ing molecular ratios result for SO^ : P0O5 : R0O3. 
I. II. " - " jj- 
SO3— 0.0765 not det. 0.0765 
P.O5-0.2359 0.2360 0.2360 
RoO.. =0.2365 0.2369 0.2367 
