Section III, 1917 [51] Trans. R.S.C. 



The Determination of the Specific Gravity of Mineral Fragments by 



Heavy Liquids. 



By R. P. D. Graham. 



Communicated by H. T. Barnes. 



It not infrequently happens that a determination of the specific 

 gravity affords one of the readiest means of identifying a doubtful 

 species, or of ascertaining its composition if it be an isomorphous 

 mixture, such as a plagioclase feldspar. In many such cases, as for 

 instance where minute feldspar crystals are distributed through a rock, 

 it may be a long and tedious process to isolate, in a pure state, a suffi- 

 cient quantity of material for the pycnometer, or any other method 

 involving weighing, while the employment of such methods becomes 

 impossible when only one or two minute crystals or fragments of the 

 mineral are available. At such times, the employment of the heavy 

 liquid method becomes advantageous or imperative. The usual proce- 

 dure is to immerse one or more fragments of the mineral in the liquid 

 or solution selected, add a suitable diluent to the latter until the mineral 

 remains exactly in suspension, and then find the specific gravity of the 

 liquid by introducing substances of progressively increasing or decreas- 

 ing density from a standard set. As a rule, however, there are frequent 

 wide gaps between successive indicators in such sets, and unless these 

 can be bridged by addit-ional test fragments, it becomes necessary, 

 if the result is to be at all accurate, to make an actual determination 

 of the specific gravity of the heavy liquid by the Westphal balance 

 or some other method. 



To overcome this difficulty, SoUas^ suggested the diflfusion column, 

 which he obtained by placing some methylene iodide in a tube, care- 

 fully adding benzene, and then allowing the mixture to stand for 

 some time. In this way there results a column of liquid whose density 

 increases in a uniform manner from the surface downward to the 

 bottom. The desired specific gravity of any fragment is then found 

 accurately from its position of suspension relative to the positions of 

 two substances of known specific gravity, one lighter and the other 

 heavier than the unknown. 



The writer's experience with the diffusion column has not been 

 altogether successful, and, in place of it, he has found the simple and 

 rapid burette method described below to give good results. 



1 "Nature," February 26th, 1891. 



