14 A. L. Du Toit—Pipe- Amygdaloids. 
in other volcanic formations, e.g., in the very ancient lavas of the 
Ventersdorp System in the Divisions of Kimberley and Barkly West, 
in the Bushveld Amygdaloid. of the Transvaal and the Basalts of 
the Lebombo and Zululand (which all overlie rocks belonging to 
the Karroo System), and in the voleanic fissure of the Zuurberg 
in the Cape Colony (probably of post-Neocomian age). They have 
a striking resemblance to organic structures, and Cohen lhkens them 
to certain corals; in some exposures of lava the little pencils of 
white zeolite studding the surface remind one of the casts of worm 
burrows. 
The‘pipe-amygdaloid never occurs in the middle or upper portion 
of a flow, only at the extreme base where the lava rests upon an 
earlier extrusion or upon sandstone or voleanie ash. In Barkly Hast 
pipes have been found in oval masses of lava embedded in sandstone 
or ash and representing bomb-lke masses of molten rock ejected from 
neighbouring volcanic necks. Here too they only appear on the 
undersides of such masses and never on their upper surfaces. 
The rock containing the pipes is usually a vesicular lava, but not 
necessarily so; sometimes the material may be quite compact without 
any other vesicles. 
The amygdales are commonly from four to six inches in length, 
with about the diameter of a quill; sometimes they may reach about 
‘double that length. ‘The vesicles are cylindrical in cross-section 
with occasional small swellings, and as they are traced upwards 
into the flow the tubes are found to unite, giving rise to a branching 
structure (Fig. 1). 
Fie. 1.—Pipe-amyedales in lava from the Drakensberg. In (a) the vesicles are 
inclined to, and in (4) are normal to, the base of the lava-tlow. (Drawn about 
+ nat. size.) 
Larger tubes are also found, and are rather irregular in cross- 
section and constricted at points. The vesicles are filled in with 
calcite and chalcedonic silica in the case of the Ventersdorp lavas and 
with zeolites in the later volcanics. 
Before discussing the mode of origin of these curious amyedales it 
will be of advantage to describe certain allicd structures found in 
‘some of the compact diabasic lavas of the Ventersdorp System of 
Barkly West. i 
