In many cases the anorthosites which exhibit a perfect foliation may be traced step by step into the massive variety, the gradual development of a foliated structure in the rock being accompanied by a progressive granulation of the constituents, most beautifully seen under the microscope. The change, however, differs from any hitherto described in that it is purely mechanical. There are no lines of shearing with accompanying chemical changes, but a breaking up of the constituents throughout the whole mass, though in some places this has progressed much further than in others, unaccompanied by any alteration of augite or hypersthene to hornblende, or of plagioclase to saussurite, these minerals, though prone to such alteration under pressure remaining quite unaltered, suffering merely a granulation with the arrangement of the granulated material in parallel strings. This process can be observed in all its stages, and there is reason to believe that it has been brought about by pressure acting on the rocks when they were deeply buried and very hot.[3] The anorthosite areas, of which there are about a dozen of great extent with many of smaller size, are distributed along the south and southeastern edge of the main Archean Protaxis from Labrador to Lake Champlain, occupying in this way a position similar to that of volcanoes along the edge of our present continents. Curiously enough precisely similar occurrences of this anorthosite have been found in connection with similar gneissic rocks, supposed to be of Archean age, in Russia, Norway and Egypt. These anorthosite rocks being intrusive, may be left out of consideration in endeavoring to work out the succession of the Archean in this great area. The whole Laurentian system, including the anorthosites, is in many places cut by numerous dykes of large size, which can often be traced for great distances. These are of several kinds, the principal series consisting of a beautiful fresh diabase often holding quartz in considerable amount in micro-pegmatitic intergrowths with plagioclase. Other sets of dykes and eruptive masses consisting of augite and mica syenites, quartz-porphyries and other rocks are also known to occur but have not as yet been carefully studied.