sugar is reddish brown, but more clear tasted than any unrefined cane-sugar, resembling mostly brown sugar candy. The syrup seemed to be very wholesome, for though many of our people ate enormous quantities of it, it hurt nobody.
Firewood is very scarce here; to remedy, therefore, that inconvenience as much as possible, they make use of a contrivance which is not unknown in Europe, though seldom practised but in camps. It is a burrow or pipe dug in the ground as long as convenient, generally about two yards, and open at each end; the one opening of this, into which they put the fire, is large; the other, which serves only to cause a draught, is much smaller. Immediately over this pipe circular holes are dug which reach quite down into it: in these the earthen pots are set (about three to such a fire); they are large in the middle and taper towards the bottom, by which means the fire acts upon a large part of their surface. It is really marvellous to see with how small a quantity of fire they will keep these pots boiling, each of which contains eight or ten gallons; a palm leaf or a dry stalk now and then is sufficient; indeed, it seemed in that part of the island, at least, where we were, that the palms alone supplied sufficient fuel, not only for boiling the sugar, but for dressing all their victuals, besides those which are cooked by this contrivance. How many parts of England are there where this contrivance would be of material assistance to not only the poor, but the better sort of people, who daily complain of the dearness of fuel, a charge which this contrivance alone would doubtless diminish by at least one-third. But it is well known how averse the good people of England, especially of that class that may be supposed to be not above want, are to adopt any new custom which savours of parsimony. I have been told that this very method was proposed in the Gentleman's Magazine many years ago, but have not the book on board. Frézier, in his voyage to the South Sea, describes a contrivance of the Peruvian Indians upon the same principles, plate 31, p. 273, but his drawing and plan are difficult to understand, if not actually very faulty, and his description is
