- 
26.2 Ads cea Oceurrence of Alkalies in Beryl. 
Sections of the mineral when examined with the microscope 
suggested itself as an element which might have been over- 
looked and weighed with the alumina. As soon as this possi- 
many respects resemble beryl. more or less perfect basal 
and a very imperfect prismatic cleavage had previously le 
me to believe that the mineral agonal and a qualitative 
analysis proved that beryllium was present in large quantity. 
The analysis was then repeated, using however better material 
which had in the mean time come to hand. 
After finding alkalies in one beryl it seemed to be of interest 
to test others from various localities. The result has been to 
show that, as far as tested, they always contain alkalies, 
although sometimes only in small quantities. Sodium an 
lithium were always present, cesium occasionally, while potas- 
sium and rubidium. were never detected. 
To. prove if possible — the alkalies in beryl replace beryl- 
lium, as well as to find to what extent they are present, a 
series of quantitative ae were made, the results of which 
are given below. regret very much that the time at my dis- 
posal has been limited, and that I could not first have devoted 
and aluminum or for determining the errors which would be 
involved by a certain definite procedure. I will state here that 
I shall certainly improve the first opportunity sn presents 
itself for investigating this subject further, and I shall hope to 
be able to add to the value of this article by giving more cor- 
rect data as to the percentages of beryllium nA aluminum. As 
_ the chief interest seemed at first to lie in the detection and 
determination of the alkalies, I proceeded with : agi 
and I will give here the results of them. I add also a full 
description of the methods employed, so that a proper re may 
be attached to the beryllium and aluminum determinations. 
The mineral was fused with sodium carbonate and the silica 
separated and weighed according to the usual methods ; it was 
in all cases tested with hydrofluoric acid and the small residue 
added to the oxides. The b eryllium, aluminum and iron were 
precipitated together with ammonia, filtered and washed, the 
precipitate dissolved, reprecipitated with ammonia, thoroughly 
washed and weighed. After weighing, the oxides were dis- 
solved in hydrochloric acid and the solution evaporated on the 
water bath till the free acid was driven off ; the chlorides were 
then dissolved in a little water, filtered into a flask containing 
about ae c. c. of a cold saturated solution of ammonium 
carbonate and some undissolved carbonate, and the slight 
Gestdiae: of silica, after washing, weighed, tested as to purity 
