220 SCIENCE PROGRESS. 



different lengths of time (12). The solutions obtained may 

 be tested for characteristic reactions. 



The results produced by etching a rock-slice can be 

 brought out by subsequent staining with some aniline dye, 

 which takes effect especially on the gelatinous silica liberated 

 by the acid from some of the silicates. The presence of 

 nepheline, for instance, can often be made evident in this 

 way ; and, by adjusting the time allowed for the etching 

 process, other minerals may be more or less clearly brought 

 out. Becke (22) has successfully used staining to dis- 

 tinguish between quartz and felspar ; for instance, in the 

 fine-textured aggregates in some crystalline schists, which 

 are apt to present difficulties by purely optical methods. 

 He even finds it possible to distinguish thus to some extent 

 between the different felspars, the plagioclases taking the 

 colour much more readily than orthoclase, while quartz 

 remains quite colourless. Among the plagioclases, those 

 rich in lime more readily become stained than those rich 

 in soda, and the zonary banding so common in the crystals 

 of many rocks can thus be rendered apparent. 



When we speak of " microchemical methods," however, 

 we refer rather to chemical reactions, carried out actually 

 under the microscope, resulting in various recognisable crys- 

 talline precipitates. This method of examination, initiated 

 in 1877 by Boricky, has been improved by Behrens, 

 Streng, Haushofer, Klement and Renard, McMahon, 

 and others ; and, exercised with due precautions, it is 

 capable of giving valuable assistance to the petrologist. 

 It can be made to yield results in some measure quantita- 

 tive ; to the extent, for instance, of distinguishing felspars 

 having different relative proportions of soda and lime. A 

 great variety of tests have been devised. In many cases 

 it is found convenient to bring the bases sought into the 

 form of sulphates, these being readily obtained and having 

 often highly characteristic crystal-habits. As a rule, the 

 form of the minute crystals is sufficient to indicate their 

 nature, but McMahon (23) has shown that their optical 

 properties may also be usefully employed to control the 

 conclusions. Since the recently published translation of 



