276 THE CANADIAN NATURALIST. [Vol. vi. 



decided is the result as regards magnetite, which is invariably 

 observed to be the oldest formed mineral in the more recent 

 eruptive rocks, all the crystalline constituents of which enclose 

 it. The felspars contained in trachytes, basalts, dolerites, and 

 melaphyres, and the augites and hornblendes of the same rocks, 

 all found the magnetite ready formed when their developement 

 began, and enclosed it as their growth progressed. Even leucite 

 and olivine, which are ordinarily free from foreign enclosures, are 

 found to contain magnetite. On the other hand magnetite is 

 seldom enclosed by quartz, but it is to be remembered that 

 rhyolites very seldom carry the former mineral. In the matrices 

 of many basalts, melaphyres and trachytes, which, in an unde- 

 composed condition, present under the microscope a mass of 

 microlites, the magnetite is found inserted between the needles 

 and determining their limits. The andesite of Lowenburg in 

 Siebengebirge shews, under the microscopCj many of these phe- 

 nomena clearly and distinctly. 



In considering the observations that have been made on this 

 subject one cannot avoid remarking that magnetite, tourmaline, 

 and other basic accessory minerals, appear to have been the 

 first to separate from the solidifying magma of crystalline rocks. 

 After the very basic minerals the essential constituents seem to 

 have been formed somewhat in the following order : 1st. Mica ; 

 2nd, Hornblende ; 3rd. Felspar ; 4th. Quartz. It would, there- 

 fore, seem possible to recognise the operation of a definite law in 

 the order of the separation of these minerals from their mother 

 magma, namely, that the minerals of original rocks have crystal- 

 lised out in the order of their basicity. Some facts, in bupport 

 of the existence of such a law, are observable in connection with 

 the composition of porphyritic rocks. Not unfrequently the 

 felspar crystals found in these, and which we must suppose, in 

 accordance with facts stated above, to have been produced pre- 

 vious to the solidification of their matrices, have a more basic 

 composition than the latter, or, what amounts to the same thing, 

 the composition of the matrices is more siliceous than that of the 

 whole rock including the crystals. Thus, according to Laspeyres, 

 the felsitic porphyry of Muhlberg, near Halle, enclosing colour- 

 less sanidine, oligoclase, quartz and a little mica, contains 72,24 

 p. c. silica, while the dark greyish green matrix contains 74,41 

 p. c. Again, the porphyrite of Ganse-Schnabel, near Ilfeld, con- 

 taining triclinic felspar aud other crystals has a silica contents 



