12 
Records of the Geological Survey of India. 
strata to Harigaon; but I could not have attempted this without long preconcerted 
arrangements with Lieutenant Williamson, having had repeated official warnings not to ven¬ 
ture into the Garrow country unsupported. 
The configuration of the. Garrow Hills differs considerably from that of the Cossiah 
Hills; while the corresponding features in each are determined by the same geological 
structure. The high table-land range of the Cossiah Hills rises almost abruptly from the 
plains of Silhet, and is formed by undisturbed strata belonging to the cretaceous and 
nummulitic formations, resting upon a basis of crystalline (mclamorphic and granitic) rocks. 
There is under the Cossiah Hills only a very narrow hand of much lower fringing ridges, 
formed of those same, or of younger, stratified rocks, hut greatly broken from their original 
horizontalitv. The outermost of these are very low, being composed of the comparatively 
soft rocks of age later than the nummulitic; while the ridges of intermediate position and 
elevation are for the most part determined by the nummulitic limestone, or by the cretaceous 
sandstone. The higher ridges of the Garrow Hills, on the contrary, are scarcely visible from 
the plains of Mymensing in ordinary weather. Passing westwards from Cherrapoonji, the 
zone of distal-bed rocks and lower ndges intervening between the table-land and the plains 
becomes wider, the boundary of the main hills having taken a W. N. W. course, while the 
outer limit of the minor hills maintains its E. W. direction. It is found throughout that the 
great change in the form and elevation of the hills is consistent with the appearance of the 
crystalline rocks : on the Sumesurri, the gneiss comes in close to the north of Seju; and 
Tara mountain (upon which Lieutenant Williamson has just established his head quarters) 
is the western termination of the great plateau of Shillong and Cherra, and is formed 
of gneiss with the cretaceous sandstone resting on its S. W. base. In the same continuation, 
up to the edge of the Bramahpootra at Singmari, the crystalline rocks appear almost con¬ 
tinuously, under a thin covering of the same sandstone. No formation seems to be introduced 
in the western expansion of the lower hills, that is, not more or less represented in the shorter 
but steeper sections to eastwards. Just about Laour, in Silhet, the lowest ridges of the 
outer hills are altogether cut away for a considerable length; and the nummulitic limestone 
rises abruptly from the alluvium. The supra-nummulitic strata, however, soon re-appear to 
the west, in force: on the Sumesurri they are ten miles across; and further west, on a 
line S. W. from Tura, they are probably much wider. 
There is a contrast in the features as well as in the extent of these minor hills in the 
Garrow and Cossiah regions. To the east they present much difference of elevation, and 
variety of outliue; whereas from tile summit of T ur a one overlooks, from its base to the 
plains, a monotonous expanse of insignificant hills. This seems due to several circumstances, 
influencing the results of denudation; there is a gradual diminution westwards of disturbance 
in the strata, accompanied by a less induration of the older rocks; and especially is the differ¬ 
ence of features due to the almost total extinction westwards of the limestone, a rock always 
remarkable for its picturesque forms of weathering. 
1 have said that the coal of the Garrow Hills has been examined in two widely separate local¬ 
ities. Although, on the Sumesurri, its position alone would preclude the profitable extrac¬ 
tion even of an otherwise valuable coal, it will be well to describe this locality, as the section is 
much more distinct than that of the more westerly region, and the boundaries of the several 
groups of rocks can be easily fixed approximately. The Rajah of Shushung, who lays claim 
to the sovereignty of a large t ract of hills, although to all appearance his authority is unre¬ 
cognised by the resident Gan-ows, has had search made for coal along the hanks of the Sume- 
surri. In this way two outcrops have been opened out; and a little coal extracted for trial. 
Both outcrops belong, I believe, to the same seam, repeated at the surface by contortions 
of the strata. It occurs near the base of the whole stratified series, within a few yards of 
the underlying crystalline rocks. 
The section on the .Sumesurri is as follows:—At the outermost skirts of hills, below 
Bijessur, we find the sands and subordinate clays of the group next above the nummulitic. 
Here they dip at 40° or 50° to southward, being very much more disturbed than at any point 
higher up the river. They rapidly settle down to a very small slope and even to horizontality; 
so much so that through the long windings of the river above and below Salagaou, the very 
same beds are traceable near the water line. The most distinctive rock of this group is a 
fine, soft, greenish-yellow or gray sandstone, generally massive and falsebedded, but also in 
very regularly laminated layers. On the hue of the Lekong nuddi, these soft rocks rise 
