NEWLY DISCOVERED ISLANDS, &C. 87 



As the quantities of sugar in the moulds were determined 

 by equal dips of a ladle only, there may have been some in- 

 accuracy; but it' the result in practice should give a saving of 

 twenty per cent, or even less, the manufacturer will be amply 

 repaid for changing the form of his moulds; especially as the 

 decrease by breakage might be supplied by the new form, 

 and thus eventually occasion very little additional expense. 



No. XVIII. 



An Account of some newly discovered Islands and Shoals, in the 

 Indian Seas. By Air. Thomas, an Office)- on board the Ameri- 

 can Stop Ganges. 



Read April 1st, 1803. 



SHIP GANGES, Feb. 15, 1802. 



AT 6 P. M. passed between two islands, lying W b N 

 and EbS, per compass, which we supposed to be Egmont and 

 Edgecomb islands, as seen by captain Carteret in the Swallow. 



After running 25 leagues N b E7E, passed by nine small 

 islands entirely covered with wood, lying in a N W and S E 

 direction; in length about 15 leagues. These islands were not 

 seen by captain Carteret, nor are they laid down in the charts 

 which we had, either of Robertson or Dalrymple, nor in 

 any chart I have since seen. Being a breast of the norther- 

 most at noon, had a very good meridian altitude ; — which 

 made us in latitude 9° 44' S. From distances of moon and 

 stars east and west of her, taken 14 hours after leaving the 

 land, I should lay them down in longitude 166° 43' E. 

 Tiicy are of a middling height, may be seen at the distance 

 of 8 or 10 leagues, and have no dangerous rocks or shoals in 

 their vicinity: having run so close in with the shore as to see 

 the natives on the beech, and their huts, with the naked eye. 



Egmont Island is very erroneously laid down by captain 

 Carteret, in 11° 00' S. & 164° 50' E. From my observations, 



