J. W. Mallet on Sipylite, a new Niobate. 399 
was 4°60. By determination of the Siesne se erbium first as 
oxides and then as sulphates it was found that the mixture con- 
tained almost exclusively the latter metal, of which the absorp- 
tion spectrum is obtainable with great distinctness from the crude 
solution. Iron and uranium were proved to exist as ferrous 
a uranous compounds. ‘The following results were obtained : 
le, oe Sree ee aE Gos So f 48°66 
WoO,. PRBS EEE RUE AF 5 "16 
SWAN so ie cgmhing whe onus Came eee 08 
GEO... Sie he des S age va pug oe 2°09 
LE ee 
CeO F080 Sena en), ogee 1°37 
OR Suey ee ee 3°92 : 
THOS. aus 4°06 
wae Sous hwae eke oenes ee 3°47 
MnO soot ier A ioe es trace 
PeO ac soca oes Sa ic 04 
BeO petuesy 62 
MgO 05 
CaO ii je sot sg ane apace abs 2°61 
LA.OF oo oc bios trace 
Nat) cient wipe ena oe "16 
BO oo 06 
BF ns oan Bue ees baa trace 
FS lives + amen ee ene eeee 3°1 
100°48 
Throwing together, as ema age has done in his valuable 
paper** on 1 the natural tantalates and ai the acid oxides 
of niobium, saritatier raerigwethe tin and zirconium, reducin @ br 
the basic oxides present to the equivalent acactatits of dyad 
oxides, and leaving out the water, we have from the above 
gures the ratio, 
R’0: MYO, = 221:100 
oe to the formula R’’,M)O, 4 R”,M?O,, or, applying the 
mon phosphate nomenclature, a single atomic group of 
obi niobates with four of pyro-niobates; while samars 
according to the calculation of Professor O. D. Allen++ from his 
analysis, contains one to one, or is represented by the formula 
* Ta,O; ma wo fin be assumed = Ah 2 per sg 
tos Y.0; may be assumed = 
erous oxide, but Seiomy th Oléve’s formule and atomic weights for this and the 
u 
Containing a rue of Di,O,. | Co: ontatiing a trace of Ce.0s. 
se 
Saal A; August, , 1877, p- 731. 
