HORNBLENDE AND BYTOWNITE FROM HYPERSTHENE GABBRO. 2138 
cleaned with distilled water, and dried in an air oven at 
about 80° C. for two hours. This portion was crushed in 
an agate mortar and further cleaned by separation in the 
Brogger funnel with methylene iodide. It was then washed 
with benzol, dried, and examined with a lens before being 
ground for analysis. 
The lower fraction from Klein’s solution consisted of 
hornblende, diopside and hypersthene, the separation of 
which presented some little difficulty. Klein’s solution at 
the density required for separation was too viscous, but 
the use of methylene iodide proved that the specific gravi- 
ties of these minerals were so close that separation by 
heavy solutions was impossible. The electromagnet also 
failed to effect a separation. Eventually the hornblende 
was handpicked by scattering the lower fraction from the 
first separation in Klein’s solution on a sheet of clean white 
paper and picking out the grains with a moist fine camel’s- 
hair brush, with the aid of a large lens of low magnification, 
all composite grains being rejected. The hornblende was 
distinguished by its very dark colour and the fact that it 
was opaque in transmitted light. Also the cleavage of the 
hornblende was better developed than that of the pyroxene, 
with the result that the grains obtained were mostly 
cleavage pieces, the grains of pyroxene being more rounded. 
The hornblende was then carefully washed with distilled 
water and dried in an airloven at 80° C. It was left exposed 
to the air for several days before being used for analysis. 
HORNBLENDE. 
(a) Chemical Composition. 
The hornblende obtained by the separation just described 
was analysed in duplicate, the mean of the two very con- 
cordant results being given in column I. 
