52 W. BLANI'ORD^ WESTERN INDIA. [PaRT I. 



the different trap flows, and an instance will be shown where, south of 

 Kawat, one flow of trap covers a bed of rounded pebbles, very many 

 of which have been derived from previous trap flows, all accumulated 

 in a hollow wliich to all appearance has been formed by denudation in 

 the interval between two successive eruptions. We have thus in the 

 trappean series a break of a similar character to that which is seen at 

 its base between its lowest members and the Bagh beds. It is perhaps 

 wrong to conclude that the Bagh beds and the traps form one series, and 

 therefore that the lowest traps, at least, are of middle cretaceous age, 

 but it appears difficult to believe that they can be much newer, and highly 

 improbable that they are of so late a date as eocen.e. This subject has 

 already been treated in an earlier paper. (See Memoirs A^'ol. VI, pages 

 158,159). 



It remains to trace out such connexions as may exist between the 



Bagh beds and the great formations to the eastward. 



Relations of Bagh beds ^j^ ^^^^ ^^^^^ ^^^^^^. ^^^^ ^^^^ Mahadevas, the Raj- 



to Manadevas. "^ 



mahals, Panchets and Damudas, have been shown 



by their fossils to be of pre-cretaceous age ; but of the age of the 



Mahadevas nothing has with certainty been determined. They have 



been referred alternately to tertiary and cretaceous periods, but always 



on avowedly imperfect evidence.* 



* The latest opinion on this subject is that expressed by Mr. Hislop in the Journal 

 of the Bombay Branch of the Royal Asiatic Society Vol. VI, p. 200. He there con- 

 siders the Mahadeva cretaceous, because " it underlies the Takli series, which is situated at 

 the very base of the eocene," and because " similar sandstone with stems rests on the 

 lower cretaceous rocks of Southern India." This passage contains several errors. In the 

 first place the beds near Nagpoor, to which Mr. Hislop alludes under the name of 

 Mahadeva, and which I have recently been able to examine, do not belong to that group, but 

 are interstratified with the beds containing glossopteris, which are certainly not newer than 

 the Trias. Then the " Takli series", or intertrappean of Nagpoor, as I have shown, are 

 more probably cretaceous than eocene, and the sandstone with stems resting on the upper 

 (not lower) cretaceous beds of South India has been shown to be probably tertiary. 

 Nevertheless Mr. Hislop's conclvision is very probably correct, although more recent 

 researches have shown that his data requii'ed modification. 



( 214 ) 



