72 Transactions of the South African Philosophical Society. 



ciably affect the qua-qua-versal dip of the Beaufort beds due to the 

 main vent. 



As fuller details of the various tuffs and agglomerates — using the 

 former term for the fine-grained and the latter for the coarse-grained 

 rocks — have been given elsewhere, * we shall not repeat them here, 

 but we may summarise the main characters of these remarkable 

 breccias. The coarse agglomerates are made of a sandy matrix with 

 angular lumps of sandstone, shale, mudstone, and occasional pieces 

 of quartzite, granite, mica-schist, and dolerite of the Karroo type ; 

 amongst the smaller constituents are felspar (orthoclose and micro- 

 cline), ilmenite, brown mica, and hornblende. The last three 

 minerals are similar to those in the Matjes Fontein and Silver Dam 

 pipes. It is remarkable that no fragments of granulites or eclogites 

 that are such conspicuous constituents of the Silver Dam breccia 

 have been found at Saltpetre Kop, nor have olivine, augite, melilite, 

 or perofskite been seen in specimens of the Saltpetre Kop agglo- 

 merates. The coarse parts of the agglomerate are sometimes rather 

 cavernous, and the spaces may be lined with small quartz crystals. 

 The agglomerates show no signs of bedding, but in the south-eastern 

 portion of the large vent, which is occupied by rather fine-grained 

 tuff, the rock is distinctly bedded, although the dips vary greatly 

 within short distances. This tuff probably fell back into the vent 

 after an explosion or was washed into it from the surface of the 

 crater. Some of the quartzite fragments are of a remarkable shape ; 

 they looked like ordinary water-worn pebbles which have been 

 battered ; they are covered with shallow depression which are not 

 sufficiently large to interfere with the general oval form of the 

 pebbles. Similar quartzite pebbles were found in the pipes on 

 De Vrede. Many parts of the agglomerates and tuffs were impreg- 

 nated with silica and hydrated ferric oxides. When there is a 

 moderate amount of iron present these silicified rocks have a striking 

 resemblance to some of the surface quartzites of the southern and 

 western districts of the Colony, but the similarity is, so to say, 

 accidental, for there can be no two fragmental rocks with more 

 dissimilar origins than these. Their resemblance consists in the 

 silicification of detritus largely composed of quartz grains. The 

 specific gravity of an average specimen with much ferric oxide is as 

 high as 3* 4. This accession of silica and ferric oxide has taken place 

 to a greater extent in the peripheral parts of the agglomerate than 

 elsewhere, and it also affects the sedimentary beds near the vents 

 and dykes. Carbonates of lime, magnesium, and iron are also 

 present in the agglomerates and tuffs, as well as barytes. The 

 * Ann. Rep. of the Geol. Comm. for 1903. 



