A really enormous area of South Central Africa is covered with volcanic rocks, in which are imbedded angular fragments of older strata, possibly sandstone, converted into schist, which, though carried along in the molten mass, still retain impressions of plants of a low order, probably the lowest—­Silurian—­and distinct ripple marks and raindrops in which no animal markings have yet been observed.  The fewness of the organic remains observed is owing to the fact that here no quarries are worked, no roads are made, and as we advance north the rank vegetation covers up everything.  The only stone buildings in the country north of the Cape colony are the church and mission houses at Kuruman.  In the walls there the fragments, with impressions of fossil leaves, have been broken through in the matrix, once a molten mass of lava.  The area which this basalt covers extends from near the Vaal River in the south, to a point some sixty miles beyond the Victoria Falls, and the average breadth is about 150 miles.  The space is at least 100,000 square miles.  Sandstone rocks stand up in it at various points like islands, but all are metamorphosed, and branches have flowed off from the igneous sea into valleys and defiles, and one can easily trace the hardening process of the fire as less and less, till at the outer end of the stream the rocks are merely hardened.  These branches equal in size all the rocks and hills that stand like islands, so that we are justified in assuming the area as at least 100,000 square miles of this basaltic sea.

The molten mass seems to have flowed over in successive waves, and the top of each wave was covered with a dark vitreous scum carrying scoriae with angular fragments.  This scum marks each successive overflow, as a stratum from twelve to eighteen inches or more in thickness.  In one part sixty-two strata are revealed, but at the Victoria Falls (which are simply a rent) the basaltic rock is stratified as far as our eyes could see down the depth of 310 feet.  This extensive sea of lava was probably sub-aerial, because bubbles often appear as coming out of the rock into the vitreous scum on the surface of each wave:  in some cases they have broken and left circular rings with raised edges, peculiar to any boiling viscous fluid.  In many cases they have cooled as round pustules, as if a bullet were enclosed; on breaking them the internal surface is covered with a crop of beautiful crystals of silver with their heads all directed to the centre of the bubble, which otherwise is empty.

These bubbles in stone may be observed in the bed of the Kuruman River, eight or ten miles north of the village; and the mountain called “Amhan,” west-north-west of the village, has all the appearance of having been an orifice through which the basalt boiled up as water or mud does in a geyser.

The black basaltic mountains on the east of the Bamangwato, formerly called the Bakaa, furnish further evidence of the igneous eruptions being sub-aerial, for the basalt itself is columnar at many points, and at other points the tops of the huge crystals appear in groups, and the apices not flattened, as would have been the case had they been developed under the enormous pressure of an ocean.  A few miles on their south a hot salt fountain boils forth and tells of interior heat.  Another, far to the south-east, and of fresh water, tells the same tale.