Pottox— On the Extraction of Glucinum from Beryl. 151 
the portion taken, as the anhydrous chloride of glucinum absorbs moisture with 
extreme rapidity. A quantity of the crystals were quickly dropped into a perfectly 
dry weighing-tube, stoppered, and the whole accurately weighed; then dissolved 
in the weighing-bottle, and transferred to a beaker, filtered from any specks of 
carbon, and made up to one litre. The weight taken was 5:6782 grams, and, 
deducting the weight of the carbon and oxide, :0040 gram, this gives 5°6742 
grams of chloride per litre. The ammonia precipitate from two portions of 100 ce. 
each gave ‘1814 gram and ‘1773 gram of ignited oxide, or an average of °1798 
gram GI1O per ‘5674 gram of chloride taken. The silver nitrate precipitate from 
two similar portions gave 2:0114 grams and 2:0130 grams of silver chloride, an 
average of 2°0122 grams containing -4975 gram of chlorine, and equivalent to 
1123 gram of oxygen, taking the atomic weights of oxygen as 16, chlorine as 
35°45, and silver as 107°93. Now, 1793 G1O less :1123 oxygen = 0670 Gl, and 
0670 GI plus -4975 chlorine = °5645 GI1Ch. 
The quantity actually taken was -5674, the difference being due to moisture 
absorbed during transference of the chloride from the tube to the bottle, and from 
thence to the weighing-tube; but this may be neglected, as the analysis shows 
that the ratio of glucinum and chlorine actually present in the compound is as 
0670 to °4975. 
Calculating the analysis from the glucinum and chlorine actually found, this 
gives— 
Expt. Theory. 
Per cent. Per cent. 
Glucinum, ae gee Si 11:38 
Chlorine, ft bee 88:13 88°62 
The theoretical results are calculated on the accepted atomic weight of 
glucinum, 9:1. If the equivalent is calculated from the experimental results, 
we get— 
4975 : 35°45 : : 0670: eq. Gl= 4:77 
Taking the now almost universally accepted view that glucinum is a diad, we 
have for the atomic weight 9°54, a figure considerably higher than that determined 
from the ignition of the sulphate. 
In a number of other analyses of chlorides prepared from glucinum, extracted 
and purified in different ways, variations were observed in the ratio of glucinum 
and chlorine that could not very well be explained on the supposition of the 
presence of aluminium chloride. I intend to investigate this subject farther. 
Anhydrous chloride of glucinum is freely soluble in absolute alcohol, with 
great rise of temperature, and the solution conducts electricity. 
When a little of the chloride is put in a test-tube, and the tube immersed in 
the vapour of boiling sulphur, the chloride melts easily, but does not boil. 
TRANS. ROY. DUB. S0C., N.8., VOL. VIII., PART XI. 2D 
