TRANSACTIONS OF SECTION C. 579 



marine shells. The question is, are the marine sheila or the mammalia to be 

 accepted as deciding the age of the beds ? 



I agree so far with M. Gaudry, who at once accepts the evidence of the shells 

 and considers the beds Pliocene. But he thinks the mammalia were of Miocene 

 age, and that they died after the close of the epoch ; their bones being subsequently 

 washed down into the Pliocene formations. In this view I cannot concur, because 

 bones decay when exposed to the atmosphere ; and I can only conclude that the 

 animals as well as the beds are of Pliocene age. 



If these views be admitted, it will be seen that the evidence afforded by land 

 mammalia as to geological age cannot be accepted with the same confidence as that 

 of marine moUusca. It has long been found that the evidence of fossil plants is 

 very liable to mislead ; but in their case t)ie difficulty is greater, because the remains 

 do not always furnish equally clear proof of the relations of the fossils. There is 

 however, some reason to believe that the age of certain mesozoic beds in India, con- 

 taining land or fresh-water reptiles and fishes, is not the same as that of strata 

 characterised by the same animals in Europe ; and generally land and fresh-water 

 organisms do not seem so characteristic of age as marine fossils. 



The explanation of the occurrence of so many Miocene forms in the Siwalik and 

 Pikermi beds is very possibly the migration southwards of the palfearctic Miocene 

 fauna. Some of the relations between the Siwalik animals and those of the penin- 

 sula of India at the present day, and the close connection between the fauna of 

 Pikermi and that now found in Africa may be due to a fm-ther southern mio-ratory 

 movement in Post -pliocene times.* ° 



10. On the Sandstones and Grits of the Lower and Middle Series of the 

 Bristol Coalfield. By Edward Wetheebd, F.G.S., F.G.S. 



The Bristol coalfield is noted for its series of grits and sandstones, and these pro- 

 bably have their equivalents, in the South Wales and Forest of Dean coalfields as 

 well as in that of Somersetshire. They serve as stratigraphical landmarks ; and it 

 was the object of the .paper (1) to compare the grits of the above coalfield with 

 one another, with a view of ascertaining whether there were distinguishino- features 

 which might enable them to be respectively determined, and assist in coi-relation. 

 (2) To examine the chemical and physical conditions. (3) To note chano-es 

 which occur when rocks are in contact with carbonaceous matter. The first point 

 raised was the application of the term grit and sandstone. The author confirmed 

 the statement of Mr. Sorby, in his presidential address to the Geological Society 

 in 1880, to the efiect that the Carboniferous sandstones were composed of angular 

 grains. Of those examined by the author, the grains of the millstone grit were 

 the least angular. It was also pointed out, that as rocks show such variation of 

 coarseness_ in the same deposit, this could not be taken into consideration as a 

 test for grit. It was therefore suggested that the term grit should be confined to 

 those rocks which show angularity of grains, irrespective of coarseness ; and the 

 term sandstone to those which are composed of rounded grains (i.e. from which 

 the angularity has been removed). In any case, the term grit must be more gene- 

 rally applied to Carboniferous rocks than has been the case hitherto. 



Reference was then made to ' duns,' which was defined as those Carboniferous 

 beds intermediate between grit or sandstone and clay. In mining operations 

 where the 'driving' of branches was by contract, questions arose between employer 

 and employed, in the case of ' hard duns,' as to whether it is ' stone ' or ' duns ' 

 double price being paid for worldng in the former. It was also important for 

 geological pm-poses, in the construction of sections, that there shoidd be an easy 

 and ready test for this determination. The author suggested that the scratchiiio- 

 of glass would be a suitable one, which would represent a hardness of 7 for 

 " stone " (that which scratches glass). 



' For a fuller discussion of the subjects here mentioned, see Mamial of the 

 GeoliHjy of India, chap, xxiv, 1879. 



p p 2 



