584 REPORT—1883. 
The whole of this material has been derived from the adjacent Himalayas, 
representing many feet of the older and higher mountain ranges, and has travelled 
down valleys that had been excavated in pre-tertiary times. This points to a slow 
subsidence of the whole southern side of the mountain mass, deposition generally 
keeping pace with it, broken off by recurring long intervals of re-elevation. Such 
important, well-marked features as these cannot be omitted when treating of a 
mountain system. Many long and instructive pages of its history are written 
on these rocks, with the help of which we may reconstruct some of the outlines 
of its more ancient geography. 
The next most interesting feature connected with the former distribution of 
zand and sea is that these Sub-Himalayan formations are fresh-water, or torrential, 
showing that since nummulitic or eocene times the sea has never washed the base 
of the Himalayas.' In fact, there is no evidence of this from the gorge where the 
Ganges leaves the mountains up to the base of the Garo Hills; pointing to an ex- 
tension northward at that early age of the Arabian Sea, separated from the Bay 
of Bengal by peninsular India. I am led also to believe that from Assam to 
Scinde there once existed one continuous drainage line, a great river receiving its 
tributaries from the Himalayas, partly a land of lakes and marshes, the home of 
that wonderful mammalian and reptilian fauna which Cautley and Falconer were 
the first to bring to light. In pliocene times, before the greater displacements 
commenced, it is not unlikely that the Kashmir basin drained at the north-west 
end into the Kishingunga Valley to Mozufferabad, and that of Hundes and Ladak 
trended towards the same direction vid Dras. 
The southern boundary of this long alluvial plain was formed by the present 
peninsula of India, and probably of the extension of the Garo and Khasi Hills 
westward to the Rajmahal hills.* Depression has been considerable in the neigh- 
bourhood of Calcutta, nearly 500 feet. We know probably only a portion of the 
alluvial deposits. At 380 feet beds of peat were passed through in boring, and 
the lowest beds contained fresh-water shells; the beds also were of such a gravelly 
nature as to indicate the neighbourhood of hills, now buried beneath the Ganges 
alluvium, This is precisely the appearance of the country above Calcutta on 
approaching the present valley of the Brahmaputra. The western termination of 
the Garo Hills sinks into these later alluvial deposits, and along the southern face 
of the range up to Sylhet, the waters of the marshes * during the rainy season wash 
the nummulitic rocks like an inland sea, and point to the very recent depression of all 
this area. The isolated granite hill-tops jutting up out of the marshy country from 
Dhoobri to Gwalpara and on to Tezpur all testify to the same continuous depres- 
sion here. It is exactly north of this that we find the Sivalik formations absent 
at the base of the Himalayas, and we have the evidence of exclusively marine 
conditions in pliocene times at the base of the Garo Hills.5 We find also a large 
development of marine beds above the nummulitic limestone in the Jaintia coun- 
try,° passing up conformably into a great thickness of upper miocene sandstones of 
the Burrail range. In such sandstone north of the Munipur valley the only fossils I 
found were marine forms. 
This gradual depression of the delta of the Ganges, the relative higher level of 
the water-parting and shifting of the Panjab rivers westward, appear to be only the 
last phase of that post-pliocene disturbance which broke up the Assam Sub- 
Himalayan lacustrine system draining into the Arabian Sea. Zoological evidence 
which I cannot here find space to quote is also in favour of this former connection 
of the now separated waters of the Ganges and Indus basins, and the hill tracts of 
the Garo and Khasi Hills with peninsular India. 
The ground where the miocene rocks are absent is not where any denuding 
force from the north could have acted with any abnormal intensity. It lies under 
the hills where no great tributary enters the plain, and might have removed the 
* Blanford and Medlicott, Memoirs of the Geological Society of India, p. 393. 
? Loe. cit. p. 31. 8 Loc. cit. p. 397. 
* For a very excellent account see Hooker's Himalayan Journals, pp. 263-265. 
® Colebrooke, Geological Transactions, vol. i. p- 135. 
° H. H. Godwin-Austen, J.A.S.B. 1869, pp. 12 and 152. 
i 
