2a MEDLICOTT, SHILLONG PLATEAU. 



than the glauconite-sandstone at a much lower level under Cherra. 

 There are lithological and stratigraphical local facts which would 

 support such a partition of the bottom conglomerate; and the association 

 of the several portions with the beds contiguous to each: the sand- 

 stone matrix of the conglomerate varies with the nature of the ad- 

 joining sandstones — in the southern region it is often coarse and 

 granitiC; in the middle region it is often fine grained and purely siliceous, 

 and to the north it has all the characters of the beds of the Cherra 

 band. Another point in favor of this view is that in all positions there 

 is more or less of transition, by interstratification, between , the con- 

 glomerate and the sandstone immediately overlying. Both of the 

 characters here noticed may be well seen in the section below Cherra 

 village on the north, where the fine grained and calcareous band (the 

 second highest of the sub-divisions that are recognisable hereabouts) 

 passes into the conglomerate. 



The only special fact I observed that would suggest the entire in- 

 dependence of the bottom conglomerate and of its immediate coverings, 

 :and therefore its most distinct separation from the Cherra band where 

 in the northern sections, they appear to coalesce, was the very frequent 

 symptoms of carbonaceous deposits in the bottom rock. Along the Liam 

 .^len south of Cherra there occurs a fine tough sandstone just over the con- 

 glomerate full of impressions of gnarled wood; these are generally mere 

 impressions, but also often with a coating of coal, and associated with 

 threads of coal. The very same thing occurs under Mamluh ; and here I 

 found in the same position, under the cliffs to west of the village, a bed of 

 carbonaceous shale full of obscure plant remains. It is exactly on the 

 same horizon, overlyiag the conglomerate, that the cretaceous coal 

 occurs at Maobelarka. I have no sufiicient observations as to whether 

 this character is constant between these two positions ,• but even if it 

 were so, it would after all be very incomplete evidence for the unity 

 of the band : were it possible to discriminate the plants in the two 



( 172 ) 



