37] The text is here obscure; but it would appear that only some of the
men belonging to these two boats remained on board, and the rest
returned to the coast. Not that the Moorish pilots from Mozambique
were here dismissed, as the text of Lichefild's translation seems to
insinuate.--E.

[38] Motta, in the Portuguese East Indian Pilot, places this town in lat.
3 50'S. He says the entrance is much incommoded with shoals, and so
narrow in some places as not to exceed the length of a ship. This city
is said to have once stood on a peninsula, converted into an island by
cutting a canal across the isthmus.--Clarke, I. 469.

[39] This may be understood that part of the inhabitants were unmixed
Arabs, comparatively whites; while others were of a mixed race between
these and the original natives, perhaps likewise partly East Indian
Mahometans, of a similar origin.--E.

[40] This is surely an oversight in Castaneda or his translator, for
_one_ year.--E.

[41] It is difficult to ascertain what place in India is here meant.
Cranganore comes nearer in sound, but is rather nearer Melinda than
Calicut; Mangalore is rather more distant. The former a degree to the
south of Calicut, the latter not quite two to the north; all three on
the Malabar coast. On a former occasion, Castaneda says these
merchants were of Cambaya or Guzerat, above eleven degrees north of
Calicut.--E.

[42] This seems to be the same office with that named Kadhi, or Khazi, by
the Turks and Persians, which is rather the title of a judge than of a
priest, which is named Moulah.--E.

[43] It is probable that this passage should be thus understood, "The
king sent him a pilot, who was an idolater from Guzerate, &c."--E.

[44] The addition to, or observations on the text, inserted in this place
within inverted commas, are from Clarke, I. 486, 487.--E.

[45] In Lichefild's translation this date is made the 22d; but the Friday
after Sunday the 21st, must have been the 26th of the month.--E.

[46] The difference of longitude between Melinda and Calicut is thirty-
four degrees, which at 17-1/2 leagues to the degree, gives only 575
Portuguese leagues, or 680 geographical leagues of twenty to the
degree. Thus miserably erroneous are the estimated distances in old
navigators, who could only compute by the dead reckoning, or the log.
--E.

[47] The course fro