38 MBDLICOTTj SHILLONG PLATEAU, 



head of the Liam glen ; showing' that the trap, at least locally and at 

 the lower levels, could be highly intrusive, even penetrating for some 

 short distance into the hard metamorphic rock. This is the more 

 remarkable, as it is only in this position, near the contact, that pheno- 

 mena of intrusion have been observed with this trap. At still lower 

 levels, in the Theria river, earthy highly amygdaloidal and well- 

 bedded trap predominates, and I did not notice there a single case of 

 later intrusion. But my observations are too few and too scattered to 

 suggest a presentable explanation of these peculiarities. 



It will have been noticed from the foregone' remarks that there is 

 much variety in the rocks of the Silhet trap formation. Some of the 

 gray and dull claret-coloured amygdaloids with green-earth and olivine 

 are very like common forms of the great Deccan trap. It seems pro- 

 bable too that a detailed examination in the field will disclose some 

 order in their arrangement. The two or three hundred feet of soft 

 ashy beds noticed just south of Mamluh seem to be to some extent charac- 

 teristic of the upper part of the series ; and from the absence of these 

 beds in other places, as along the southern boundary, the formation seems 

 to have undergone much denudation prior to the commencement of the 

 cretaceous period. 



A trap formation of such great thickness must have had a corres- 

 pondingly wide horizontal range. We seem to have here its original 

 northern limit. There is no surmising how far it may stretch to the 

 south beneath the present delta and the bay of Bengal, being still 

 farther buried beneath the remains of the cretaceous and tertiary 

 deposits which seem to have also extended in a southerly direction. 

 To the west the trap has not been noticed beyond the region of the 

 Umblai, but this is manifestly due to concealment beneath the sedimen- 

 tary series. To the east it seems to become concealed in a similar 

 manner, as if still overlapped by the cretaceous beds. There is certainly 

 no trace of it under these rocks on the Kopili or the Namba, no more 

 ( 188 ) 



