e 
G. J. Brush on Durangite. 181 
In No. 1, the decomposition was effected by sulphuric acid, 
the arsenic acid was reduced to arsenous acid by sulphurous 
arsenic acid may be due to the fact that. the soda fusion was 
made over a gas blast-lamp, and possibly a portion of arsenic 
acid was reduced to arsenous acid and volatilized ; or the loss 
may be attributed to other accidental causes. 
0. 3 was a qualitative-quantitative analysis in which only 
the alkalies were estimated ; the soda and lithia were first 
weighed as sulphates, then converted into chlorids, and_the 
alkalies separated by dissolving the chlorid of lithium m a 
mixture of absolute alcohol and ether. The alkalies examined 
hp the spectroscope showed only the lines of sodium and 
ithium, 
The oxygen ratio in No. 1, between the protoxyds, sesqui- 
oxyds and arsenic acid is 3°74 : 11°07 : 19°16, or, very nearly, 
t7o: This involves the assumption that the small amount 
of manganese exists as protoxyd, of which we have no proof, 
but it seems probable that this ratio is the true one, although 
it is impossible at present to determine the exact state of oxyd- 
ation of the iron and manganese. This composition and ratio 
Suggest an analogy between the new mineral and amblygonite, 
a fluo-phosphate of alumina, lithia and soda, and it may be 
viewed as an arsenic-acid amblygonite, containing a small per- 
centage of iron and manganese and more soda than lithia. _ 
he formula may be written @R* +38) Ms, and, as in ambly- 
gonite, a portion of the oxygen is replaced by fluorine. A 
number of hypotheses showing the possible relation of the fluo- 
Tine might be suggested, but until inv! tion these 
‘ 
