Apr., 1889. 
SEPARATION OF ROCK CONSTITUENTS 
98 
probably be found that a separation has occurred; some of 
the constituents have fallen to the bottom of the beaker, the 
rest float. In the former fraction we have the ores, such as 
magnetite, ilmenite, pyrites, augite, and olivine. The floating 
portion consists of the felspars, quartz, with some of the 
subordinate minerals. This lighter fraction may be partially 
skimmed off with a glass or platinum spoon, and then the 
liquid is poured off the heavy part, which may be done pretty 
completely, usually leaving a ring of the lighter particles 
adhering to the sides of the beaker, interrupted, however, 
at the spout by a gap where the liquid was poured out. 
Through this gap, which should be made as wide as possible 
by gentle swaying from side to side while pouring, the heavy 
portion is washed into a porcelain dish by means of a wash- 
bottle. It is washed bv decantation and boiling with water 
and a little solution of iodide of potassium, and then dried. 
The addition of water, drop by drop, to the liquid on which 
the lighter part of the rock is floating, soon effects a separation, 
and, if this is continued till another “ indicator ” just sinks, we 
know to what specific gravity we have reduced it. In order to 
get the substances quite pure a repetition of the process is 
required. This may be suitably performed in a tube closed 
below by a stopcock, through which the heavier fraction may 
be run off. 
To determine the specific gravity of a mineral, a fragment 
is floated on the solution and water added, drop by drop, with 
careful and thorough stirring, until the fragment rests 
indifferently anywhere in the liquid. Then fill a small 
specific gravity bottle or tube, and determine the specific 
gravity by weighing. A more simple plan, and one which is 
quite accurate enough for almost any purpose, is to use the 
principle of Westplial’s balance, and weigh a plummet, of 
which the weight in air and the specific gravity (which must 
be greater than that of the liquid) are known, immersed 
in the liquid. The plummet for this purpose may be made of 
a glass tube containing mercury, and its specific gravity is, 
of course, previously determined by weighing in air and in 
water. 
The determinations should be repeated, and it is obvious 
that, when the mineral fragment tends neither to sink nor rise 
in the solution, the specific gravity of the two is identical. 
For the suggestion of the potassium iodide fluid we are 
indebted to Sonstadt, for that containing barium iodide to 
Rohrbach, and for the borotungstate of cadmium to Klein. 
