62 D. Prain — Flora of Narcondam and Barren Island. [No. 2, 



of two — a western tertiary ridge most marked in the north and tailing 

 off towards the south, and an eastern volcanic ridge most marked in the 

 south and dwindling into insignificance northwards. 



The question whether the line in which Narcondam, Barren Island, 

 and Flat Rock are situated consists of a series of isolated peaks, or if 

 these peaks are only the sub-aerial portions of a continuous ridge, remains 

 to be considered. Such evidence as there is appears to indicate that 

 they are situated on a ridge : it is not, however, at all complete. It has 

 already been remarked that the soundings on a line passing north-north- 

 east from Narcondam are relatively shallower than those on any other 

 line. This has been explained by Carpenter as perhaps indicating that 

 the deltaic shelf of the Irrawady extends as far out as Narcondam.* 

 It may be anticipated that this will not be found a sufficient explana- 

 tion of the phenomenon. It will be observed that the soundings gradu- 

 ally deepen for a space of 9| miles, till the bottom carries 362 fathoms, 

 and that beyond this point it gradually shallows till the coast of Pegu 

 is reached. If Narcondam were situated on the edge of a delta-shelf, 

 one would expect that the soundings w r ould not show so great a dip within 

 its margin, and would further expect that soundings on lines carried at 

 right angles to the line under discussion would give some indication of 

 a more or less level area. Yet what we do find is that before four miles 

 to the east or three miles to the west of the island have been reached, 

 greater depths have been obtained than the deepest sounding on the 

 north-north-east line. This appears to indicate that Narcondam is not 

 so much on the edge of a shelf, as at the end of a ridge that runs towards 

 and into the Pegu coast-line. That this ridge is overlaid by the deltaic 

 mud to within ten miles of Narcondam, and that the presence of this 

 mud explains the gentle slope from its deepest point upwards to the 

 Pegu coast is no doubt true ; but the steady rise during the last ten 

 miles towards Narcondam, coupled with the more abrupt dips to the 

 east and to the west, indicate the existence of a ridge. The matter is 

 capable of direct demonstration : a few lines of deep-sea soundings co- 

 ordinate to the line of soundings taken towards the north-north-east, 

 will disclose the true state of matters. It would also be equally easy, 

 by making a line of borings along the continuation of its line, and a few 

 co-ordinate lines acix>ss in the mud of the Irrawady delta, to demonstrate 

 whether the supposed ridge passes subterraneously into Burma. 



The same comparative shallowness is indicated by the line of 

 soundings to the south-south-west of Barren Island, and to explain the 

 fact Malletf suggests the possibility of eruptions of ash distributed 



* Carpenter : Records of the G-eol. Survey of India, xxi., 48. 

 f Mallet : Records of the G-eol. Survey of India, xxi., 47. 



