1893.] D. Praiu — Flora of Narcondam and Barren Island. 53 



of the northern half, shows a second snbconcentric ridge separated from 

 the true rim by a gorge that debouches on the east side of the island. 

 Gorge and ridge owe their origin, however, — like the ridge and ravine of 

 the same nature, but of more imposing proportions, that occur at the 

 south end of Karcondam — to subaerial denudation, not to volcanic action. 



The excentric position of the newer cone, with the lesser relative 

 height, and the steeper seaward slope of the northern half of the origi- 

 nal crater, seems to point to subsidence of that half. Perhaps 

 the explosive eruption which effected the breach to the west may have 

 had some connection, direct or indirect, with this subsidence. The vol- 

 cano represented by the outer cone was doubtless at one time much 

 higher than it is now. 



At the landing-place in the breach there is a hot spring on the 

 beach ; the temperature of this spring is steadily falling, and at the 

 time of the writer's visit was 106° F.* The spring doubtless only 

 represents percolation of rain water through the heated newer mate- 

 rials — the inner cone and lava streams — contained within the circuit 

 of the ancient crater. f 



The anchorage in the bay at the breach is of the most uncom- 

 fortable description ; the safest anchorage is opposite a small bay with 

 a sandy beach, a Pandanus sea-fence and a line of Coco-nut trees, on the 

 south-west side of the island. Landing by boat is, however, usually 

 quite easy on the beach at the hot spring to the north of the point where 

 the lava stream falls into the sea ; the surf that rolls into Anchorage 

 Bay must make it impossible, as a rule, to land there. 



At Landing Bay the boulders and stones on the beach, bathed by 

 the water of the hot-spring, are covered by a species of Galotlirix which 

 occurs in considerable quantities. Another, Alga, also a Galotlirix, was 

 obtained from bare rocks in one of the gorges ; no marine Alga? were seen. 

 On the beach itself, behind a small bed of drift, are some examples of 

 Ipomoea hiloba ; the drift contained, in addition to fruits and seeds of 

 species noticed in the island, fruits of Barringtonia speciosa and of 

 Seritiera littoralis.% Close to the beach and to the lava flow is an ex- 

 ample of Pongamia glabra ; a little further inland to the north of the 

 lava is a considerable grove of Flueggia microcarpa, with quantities of 

 Mitreola oldenlandioides, in the sandy soil beneath. Beyond this grove is 



* Prain : Proceedings As. Soc, Bengal, 1891, p. 84. 



f Mallet : Memoirs of the Geol. Survey of India, xxi, 274. 



% Barringtonia speciosa occurs in Narcondam, and it may possibly also occur at 

 some of the bays on the south-west and south of Barren Island, where the surf 

 made landing impracticable. But Heritiera littoralis, the fruits of which were col- 

 lected in Narcondam also, does not seem to occur in either island. 



