PAiiT 1.] Clark-. On Volcanic Foci of FJruptiou in the Konkan. 71 
while viscid and full of air bubbles, had flowed in the direction of their longer 
axis. Occasionally also the vesicles were bent and twisted, as though the trap 
had flowed over a hard edge, as water over a weir; particulars which seemed to 
afford a clue to the direction, and in some degree to the circumstances, of the 
flow. Following upon these indications, I found that I reached a number of hil¬ 
locks or cones, holloAV in the centre, and often with a gap in one side, and within 
and about these the trap often lay in small streamlets, crossing over or overlap¬ 
ping one another, but all evidently derived from a common source or centre. 
Frequently these streams had flowed for some distance in parallel lines imping¬ 
ing upon each other, not uniting, so as to leave a V-shaped trough between them, 
which again was filled up with other streamlets. All these had assumed various 
shapes according as they had flowed in a trough or over a flat surface, or over an 
obstacle, or had dropped over some accidental step or fissure in the subjacent 
rock. 
These singular, crater-like, hillocks lay very thick together along the course 
of the Bervee river above Kalian, and upon that of the upj^er Kaloo near Bhalook 
and Kinnowlee. Near the ancient temple of Oombornaut are several, some 
with a central hollow a quarter of a mile across and sides from 200 to 300 
feet high. The interior slope is much steeper than the exterior, and the floor 
is usually very hard and undulating, as though it had been in a state of 
ebullition. 
Besides those hillocks are also a number of flattened domes very distinct, 
though of no great height, like huge bubbles of very hard rock, and seen, whore 
the surface is broken, to bo composed of layers like the coats of an onion. 
These are especially frequent in the upper Kaloo approaching the Malsego. 
I observed also in the bed of this river, dry or nearly so when I saw it, tliat in 
many places it was not excavated as by water, but formed by the f)arallcl j unc¬ 
tion of two lava streams, the stream filling up which had been dissected out, 
not eroded, so that the surfaces remained smooth and sometimes almost glazed. 
All these appearances led me to believe that I had lighted on a number of 
foci of volanie or plutonic action, placed along certain lines, and that these 
were the vents—perhaps source,s would have been a safer word—whence the 
traps of the district, and of the adjacent islands on the one hand and of the 
Ghauts and Deccan on the other, were derived. I came also to the conclusion, 
perhaps upon the examination of too limited an area, that the general line of 
the sources lay nearly north and south down the edge of the Konkan, pointing 
towards the islands of Heneri and Kcncri, and that the line towards the MaLsege 
was a sort of spur or lateral axis, accounting probably for the existence of that 
very remarkable bay or indentation into the main lino of the Syhadree range. 
The contrast was remarkable between the irregularity of the streamlets of 
lava in the plain and near the sources above described, and the excessive 
regularity of the beds at higher elevation, and at a distance from those points. 
This indeed was what might have been expected; a regular and uniform diji 
being more likely to be arriv'ed at by a lava sheet at a certain distance from its 
source. 
