THE FELSPARS. 



necessarily tend to increase the percentage of that alkali, and 

 correspondingly diminish the lime. It is much to be regi'etted that 

 Professor Heddle has not made (or published, if he has made) 

 observations on the microscopical and optical properties of the grand 

 series of felspars analysed by him. They would almost certainly have 

 afforded an immensely improved point of departure for argument as 

 to the actual chemistry of the group. 



To determine what felspar we have in a rock is, unfortunately, a 

 very difficult task, at any rate where it occurs in but small-sized 

 crystals. Chemical analysis is very difficult in the case of such 

 small quantities of material as are usually available, and is 

 frequently but uncertain on account of the almost unavoidable 

 admixture of other minerals. But two or three methods have been 

 proposed, and of them I propose to speak, though only very briefly, 

 seeing that to properly elucidate the subject I should have to make 

 more experiments before you than there is time and opportunity for 

 to-night. I the less regret this, as some time ago I read a paper, and 

 showed experiments, on one of the methods mentioned. 



This, which is the one elaborated by Dr. Szabo, professor in the 

 Uuiverisity of Buda-Pesth, is, in essence, a carefully arranged deter- 

 mination of the fusibility in different parts of a Buusen burner flame 

 of particular (or at least invariable) dimensions, the same operations 

 also serving for the estimation of the percentage of the alkalis by the 

 intensity of the flame-colouration. This method, which can be com- 

 pletely carried out in about a quarter of an hour, if all goes fortunately, 

 enables us to decide at once and easily between Orthoclase, Albite, 

 Oligoclase, Labradorite, and Auorthite, but requires more practice 

 and more careful observation to determine accurately the varying 

 proportions of soda in Orthoclase (in eighteen analyses of Scotch 

 felspar, by Dr. Heddle, this varies between 0-53 and 5-5%), the division 

 of Audesine between Oligoclase and Labradorite, and the occurrences 

 between this last and Auorthite, which have been called Bytownite. 

 The best way is to compare the specimen with fragments of known 

 composition, one on one side of the flame, the other on the other. 



Another plan is that of Dr. E. Boricky, depending on the facts that 

 a dilute solution of fluosilicic acid decoinposes silicates, and that the 

 fluosilicates of several of the bases which occur most frequently in 

 minerals crystallise iu characteristic forms, and so can be detected, 

 after the drying up of the drop of reagent, by means of the microscope. 

 Thus Orthoclase leaves beautiful cubic and octohedral crystals of the 

 Potash salt, and a few hexagonal prisms of the soda one. In the case 

 of Albite the x^roportion of the two sorts of crystals is reversed. 



For the purpose of separating portions of felspar for trial by either 

 of these methods it is very convenient to use the heavy solution of 

 iodide of Hg. in iodide of K. which Soustedt proposed, and which is 

 now being much used, especially in Germany, to get out the various 

 constituents of a rock for purposes of analysis. The solution can be 

 got of a sp. gr. of just over three, so that felspars and quartz float 



