28S. L."Penfield—Oceurrence of Alkalies in Beryl. 
in the determination of the alkalies. The mineral was in al? 
eases decomposed by long treatment with — hydrofluoric acid, 
tions were carried on in platinum. In the last analyses where 
it was known that the amount of alkalies would be very small, 
the beryllium oxide and alumina were precipitated in platinum 
dishes, and the only glass with which the solutions were in con- . 
tact were the glass funnels used in ae off the precipitates ; 
eare was also taken to use water which had not stood long in 
wash bottles. The sulphates of the alkalies after weighing 
were converted into chlorides, and the lithium nis phage by 
means of alcohol and ether and weighed as sulphate. The 
cesium was precipitated and weighed as cesium SlaGars chlo- 
ride, After weighing it was tested carefully with the spectro- 
ee for potassium’ and rubidium, but neither could be de- 
tect 
The loss Py ignition is considerable, as will be seen by the 
analyses. That water is contained in perceptible quantity can 
be readily seen by heating the mineral’ very intensely in a 
closed glass tube. The loss by ignition does not in ail cases 
represent the water chemically combined in the mineral, be- 
cause liquid inclusions are readily seen in the beryl from 
Branchville and may be contained in others. The water reacts 
neutral or only slightly acid to test papers, (the acid reaction 
probably due to carbon dioxide). “It is pa completely given 
off except by very strong ignition. By strong ignition over a 
blast lamp in a covered crucible no sublimate collected on the 
lid of the crucible. The powder usually sinters together to a 
porous cake and this behavior affords an indieation of the 
presence of alkalies, those containing much alkali sintering to- 
gether to a firm _ ous mass, while those containing very little 
as in analyses VI and VII, left a mass which could be readily 
erushed to powder between the fingers 
lorine, fluorine, boracie and phosphoric acids could not be 
_ detected. 
The varieties which have been analyzed are the following, 
arranged according to. the amount of alkali which they contain. 
The specific gravities were taken in water of the ordinary lab- 
oratory pe ads on a chemical balance, and are given 
with the a 
1. Tene, pi eo Maine. The material was taken from a 
fragment of a very much cracked, colorless crystal imbedded 
in lepidolite and was given to me by Professor O. D. Allen, 
who collected it at the locality. Care was necessary in select- 
