TRANSACTIONS OF SECTION O. 607 



across the flank nf the island, forming a curved belt abont three miles long, nearly 

 half a mile wide at its broadest, rising in places to 50 or GO feet, above sea level, 

 and ending sharply, whore it touches the original island, against the lower bare 

 ground, with hardly any 'oulwash.' It consists almost entirely of streaky red 

 clay containing a few scratched boulders, and crowded with marine shells, some 

 broken, but most of them perfect and the bivalves united. The clay has evidently 

 been derived, in the first place, from the red Devonian rocks into which the fiord 

 is cut ; but its more immediate origin was the neighbouring sea-bottom, which has 

 undoubtedly been dragged up in some way by the glacier in its advance. The 

 existing remnant of the glacier was seen to be curiously entangled among the clay, 

 and the presence of smaller masses of ice buried under the moraine was indicated 

 by the crater-like hollows of subsidence by which its surface was pitted. 



These facts show clearly that (1) material similar to some English boulder- 

 clays may form the terminal moraine of a glacier; (2) a glacier advancing over 

 a sea-bottom may incorporate marine detritus abundantly in its moraine; (3) banks 

 of boulder-clay may terminate abruptly and with a sharp boundary on bare un- 

 glaciated land ; (4) material in the path of an advancing glacier may be uplifted 

 sharply on a rising slope. The possibility <>f these events has been frequently 

 doubted but is here proved. 



Joint Meeting with Snb-seclion B (Agriculture). — See p. 585. 



TUESDAY, SEPTEMBER 6. 



The following Papers were read : — 



1. The Cause of Gravity Variations in Northern India. 

 By Professor Sir Thomas H. Holland, K.C.I.E., D.Sc, F.R.S. 



Among the gravity variations indicated by the plumb-line and pendulum obser- 

 vations conducted in India the most conspicuous is the band of high gravity, first 

 detected by Colonel 8. G. Burrard, R.E., E.R.S., stretching from about Kisnapur 

 (latitude 25° 2', longitude 88° 28') in Bengal to a point between Feroze- 

 pore and Montgomery in the Punjab (latitude 30° 50', longitude 74° 30').' 

 The direction of this band of high gravity parallel to the general trend of the 

 silt-filled Gangetic depression, the great Himalayan folds which have developed 

 since Lower Tertiary times, and the pre-Tertiary shore line between Gondwana- 

 land and the great Eurasian Ocean suggests a genetic relationship between 

 these four features. With these may be correlated the general trend of faulting 

 during pre-Deccan Trap times in what is now the Central and Northern part 

 of Peninsular India. A general east-west trend is shown by the faults which 

 affect the Vindhyans and are of older date than the Permo-Carboniferous ; by 

 the later faults which affect the Lower Gondwanas and are of earlier date than 

 the Trias ; and by the still younger faults that disturb the Upper Gondwanas 

 of Triassie and Jurassic Age. In the same area there is a general tendency for 

 the dykes of Deccan Trap Age (Upper Cretaceous) to follow an east-west direc- 

 tion. These faults and also the fissures into which the basic rock was injected 

 were formed throughout the period during which, from Lower Palaeozoic to 

 Upper Mesozoic times, there was a continual denudation, with consequent relief 

 of load, from the Gondwana continent, and a corresponding loading with accom- 

 panying, if not consequent, depression of the adjoining ocean bottom immediately 

 to the north of the line now occupied by the crystalline, snow-covered peaks of 

 the Himalaya. The persistence of this process must have resulted in a stretching 

 of the continent in a general north-south direction, as shown by the transverse 

 normal, or tension, faults. The process culminated with the great out-welling 



* See Burrard, Phil. Trans., A, vol. 205, 1905, pp. 289-318; Lenox-Conyng- 

 ham. Survey of India. Professional Paper No. 10, 1908. 



R R 2 



