it is perfect. But still the scale must be in the main correct, for it can be applied to the rock succession in far-distant parts of the earth. More probably the Bokkeveld-Witteberg gap is due to the fact that the Bokkeveld sediments, which were laid down in deep water, became less and less as the sea floor sank and the shore line retreated, till finally sedimentation ceased over the area now covered with Bokkeveld Beds. It must be remembered that detritus from land, even the finest muds, is not carried beyond 200 ml. from the shore. When the sea floor began to rise again, the succeeding Witteberg sediments would follow the previous ones conformably; but there might thus be a considerable time interval between the last of the Bokkeveld Beds and the first or lowermost of the Witteberg Beds. Such a conformity, in which succeeding beds skip a period on the time scale, may be called a per saltem conformity.
General Arrangement of the Rocks in South Africa. — Africa, as a whole, is a vast plateau land, distinguished from Europe by the fact that there the main geological features are folds which have crumpled up the rocks, whereas in Africa the folds which occur are of extremely ancient date, and have been almost obliterated: and the predominant features are faults of vast extent. Rhodesia on the north belongs still to the main plateau, but in the Transvaal we enter into a new order of things. The main area of the Transvaal is occupied by a gigantic laccolite, which has invaded the ancient beds and pushed them southwards; and at a later date a similar laccolite, or spread of dolerite dykes having the effect of a laccolite, has invaded the Triassic and earlier beds of Cape Colony, and has again pushed them southwards. There
