342 AUSTRALIAN, SOUTH AFRICAN, AND INDIAN COAL-MEASURES, 



very long interval — are the Gangamopteris beds of Bacchus 

 Marsh, which are said to afford unequivocal evidence of Glacial 

 action, and are at the same time probably related in some way or 

 other to the Newcastle Coal Measures which contain both Ganga- 

 7)i02)teris and Glossopteris. 



These Gangamopteris sandstones and glacial conglomerates of 

 Bacchus Marsh, resting partly on strongly folded and denuded 

 Silurians and partly on the older volcanic rocks, have generally 

 been regarded as the equivalents of the Indian Talchirs, which, 

 as stated above, are supposed to be represented by the Ecca glacial 

 conglomerates and Glossopteris shales in South Africa ; and by 

 the Upper Marine Beds in New South Wales. But Feistmantel 

 with more reason — as it seems to me — places the Bacchus JNlarsh 

 beds above the Upper Coal of N.S.W. (Proc. Roy. Soc. N.S.W., 

 Vol. XIV. 1880, p. 111). 



Above these Gangamo2:>teris beds of Victoria succeed the Car- 

 bonaceous (so called) beds, the last, if not the first, of the 

 Victorian Mesozoic, with two species of Unio^ three of Cycads, 

 and, more important for our purpose, Pecopteris {Aletltoj^teris) 

 Australis, and Tcaniopteris Daintreei. 



Now these same ferns of the Carbonaceous rocks of Victoria 

 occur also together in the Clarence River beds, so that there cau 

 be little doubt of the correctness of their identification, which, 

 indeed, has not to my knowledge been questioned. But the 

 discovery of the Narrabeen beds, and their identification with 

 the lower beds of the Clarence River, involving the claim of the 

 latter to a position intermediate between the uppermost Glossop- 

 teris beds (i.e., the Newcastle Coal Measures), and the Hawkesbury 

 sandstone alters the argument in some respects. Granting, as we 

 must, the correspondence of the Carbonaceous beds with the 

 Clarence River series, including the Hawkesbury sandstones, we 

 must look for a quite difierent horizon for the Bacchus Marsh 

 Glacial beds than that mentioned above. The absence of Glossop- 

 teris in the one case, as compared with its luxuriance and variety 

 in the other, has always presented some difiiculty in the way of 



