64 Transactions of the South African Philosophical Society, 



not arranged along certain lines, but are distributed in a most 

 irregular manner. They mostly occur in groups, but are sometimes 

 isolated from one another. In their lack of any definite arrangement 

 they resemble the Carboniferous volcanoes of Central Scotland, the 

 Permian vents of Fifeshire, those of the Eifel, the Swabian Alps, 

 and the Auvergne. 



There is no necessity to give a description of any of these 

 volcanoes, and we may proceed to an account of the material 

 which issued from them. 



According to all the descriptions available, the Volcanic beds 

 in Griqualand East, Basutoland, and Natal, are almost entirely basic 

 lavas. 



In Barkly East and west of Jamestown beds of sandstone and 

 volcanic ash occur in the midst of the igneous flows. The volcanic 

 ash may pass over insensibly into a sandstone by the addition of 

 sedimentary material, or vice versa ; the thickness of such beds is 

 very variable, but may reach as much as 350 feet. The usual type 

 of ash is a reddish or bluish-green rock, with fragments of shale 

 sandstone and grit derived from the underlying Stormberg beds, 

 boulders of Cape and pre-Cape rocks, and portions of lava from 

 minute lapilli to great masses several feet across. These beds of 

 fragmental and siliceous material afford a vast amount of minute 

 information with regard to the volcanic history of the areas in which 

 they occur. Where the lavas alone occur the information to be 

 gleaned from them is very meagre. 



One of the most peculiar types of igneous rock is that known as 

 pipe-amygdaloid, the lava being penetrated by long, thin branching 

 pipes, about the thickness of a lead pencil, now filled with secondary 

 minerals. These pipes are found only at the base of a lava flow 

 where it rests on a bed either of lava, ash, or sandstone, and never in 

 the centre or upper part of the flow. There can be no doubt that 

 the pipes were produced by steam generated below the lava as it 

 flowed over a wet surface. 



If the molten material were very mobile the steam bubbles would 

 ascend rapidly to the top of the mass, but if it were very viscid 

 long elongated bubbles would be formed, which would ascend very 

 slowly. Two or more of these bubbles would tend to coalesce as 

 they expanded upwards, and hence the branching of the vesicles. 

 In many cases the movement of the lava is shown by the bending of 

 the pipes so that they are inclined forward in the direction of 

 motion. In a few examples the deflection from the vertical was 

 found to be as much as sixty degrees. The common length of a pipe 

 is from four to six inches, but some were occasionally met with of 



