1884.] between the Fish and Juja Rimrs, South Africa. 5 



5tli — The dykes do not interleave or intrude themselves horizontally 

 into the stratified rocks they pass through till they reach an elevation 

 of upwards of 2,500 feet above the present sea level. Above this 

 height,, in many places they have thrust themselves horizontally 

 between the strata at different levels — where they form extensive 

 sheets — protecting those portions of stratified rocks covered by them 

 from rapid denudation, and causing the form of many of our tea-caddy 

 and table-shaped mountains seen in the divisions of Queenstown, 

 Cradock, and the Tarka. And absence of these horizontal beds in the 

 lower regions has produced the clustered hummock-like and rounded 

 outlines that form so distinctive a feature of the country lying below 

 the 2,500 feet level. 



6th — They have caused very little disturbance or displacement in 

 the stratified rocks, through which they have forced their way — a fact 

 of some importance which may be of value in determining the condi- 

 tions under which the dykes were formed. The corrugations seen in 

 the section on plate or map 3 are the results of a lateral compression, 

 subsequent to the formation of the dykes. The greatest disturbance 

 that I have observed (within the range of the maps) occurs at dyke 

 No. 29, between the Kei Bridge and Toleni. A cutting in the new 

 road gives a good section of the stratified rocks and this dyke, which 

 is here about 70 feet wide. The south side of the stratified rock is not 

 disturbed ; on the north side it is tilted about 30°, and this shades o:ff 

 into the general bedding of 5^* N.W. at 200 yards from the dyke. In 

 Griqualand East, however, a tilting up of 70° in a ripple-marked shale 

 has been observed near Ficani's Kraal, on the Umzimkulu river. 



7th. — That though the smaller dykes (100 feet and under in width) 

 are sufiiciently homogeneous to be comprehended under one term, as 

 dolorite, diorite, &c., as the case may be, the same does not hold good 

 with respect to the larger dykes, some of which are upwards of a mile 

 in width. In the larger ones every variety of erupted trap (not vesi- 

 cular or amygdaloida) may be seen, shading gradually off into each 

 other, so that it is impossible to point out where one begins or the 

 other ends. In illustrating this I will particularly refer to dyke No. 

 9, being the one which my opportunities of studying have been 

 greatest, having had the platform between high and low water marks 

 laid out like a map before me daily for a great number of years. The 

 eastern edge of this great dyke consists of a coarse rough trachyte, 

 light coloured, with glassy felspar. At 100 ft. to westward this passes 

 insensibly into a dark hornblendic rock with crystals of albite. This is 

 the only portion of the dyke that decomposes into globular masses 

 and concentric layers with a solid kernel — the so-frequently mentioned 

 but never-explained cheeses, cannon balls, &c., of authors. The con- 

 centric layers are studded with hexagonal plates of a bright yellow- 

 coloured mica l-20th of an inch in diameter, which are not to be found 

 in the solid kernel or in those portions of the rock not undergoing 

 decomposition, though slices have been subjected to scrutiny by 

 microscope. They are therefore assumed to be a secondary combina- 

 tion, resulting from the decomposition of the original rock. This 

 portion of the dyke is also traversed by a multitude of parallel lines 

 of a jet black hornblende, from the finest line to a quarter of an inch 

 in thickness ; the lines follow the general direction of the dyke, and 

 cut clean through the crystals of felspar and hornblende which com- 



