39U D. M. S. Watson — The Beaufort Beds, SoutJt Africa. 



The most remarkable features of the fossil contents which bear 

 on the problem are — 



1. The extraordinary rarity of MoUusca. During my own rather 

 extended visit I saw no fossil shell of any kind, although they were 

 carefully searched for. Dr. Broom, whose experience of the beds is 

 unrivalled, has, I believe, only found one occurrence ; and Mr. Whaits, 

 in the course of a very considerable examination of the lower zones, 

 has never met with them. 



2. Fish are very rare. The only large group, that from the 

 Caledon River described by Dr. Broom, is contained in bedded 

 sandstones, and judging from specimens in museums those from other 

 horizons are also usually in bedded shales and sandstones. 



3. The great rarity of plant-remains, except in the clay-gall 

 conglomerates, where pieces of wood are often seen. The only plant 

 known to have been found in one of the mudstones that I am 

 acquainted with is a bit of Schizotieura stem from Kuilspoort, District 

 Beaufort West. 



4. The fact that the definitely stratified shales very rarely yield 

 any Tetrapod remains. Dr. Broom states that reptilian bones usually 

 occur just below a sandstone, at the top of a mass of unstratified 

 mudstone. This statement has been controverted by the Survey, 

 but so far as my own observations go they support it, at any rate 

 for the lower beds. 



5. Tetrapod remains occur sporadically, and very large areas 

 are practically free from them, although the rocks may be quite 

 similar to those of the same age which are richly fossiliferous in 

 other areas. 



6. lieptilian remains often occur in associated sets. In museums 

 such groups are very rare, but this is largely because the great 

 majority of fossils are picked up completely weathered out, and the 

 nodule containing the skull is the most obvious part. The very 

 careful collecting of Mr. Whaits from the Endothiodon beds of the 

 Beaufort West flats shows that by picking up all the small pieces 

 of bone within a large area it is often possible to recover a good 

 deal of the skeleton, even of such completely weathered-out specimens ; 

 the individuals are so rare that the chance of mixing the bones of 

 two of theui is comparatively slight. Where the face on which 

 a fossil is exposed is steep, it is difficult to discover how much of 

 a scattered skeleton is present. In my own experience bones 

 found in situ were almost always associated with other parts of 

 the skeleton. A. G. Bain's original manuscript catalogues show 

 conclusively that many of the bones he sent to England were in 

 associated lots. Reptilian remains sometimes occur in small groups, 

 two or three individuals of the same or different individuals lying 

 close together. 



7. The fact that a much larger percentage of the bones found in 

 sandstones belong to Stegocephalia than is the case with those in 

 mudstones. I first noticed this fact in the upper beds of the 

 Burghersdorp district, and examination of the remains in museums 

 seems to confirm it very strikingly. 



The problem of the mode of origin of the rocks which present the 



