APPENDIX.

XXXV of their daily progress. And therefore we suppose that some cause is to be looked for, besides that of dulness of sailing.

“ That this had a considerable share in the delay is evident, by the circumstance mentioned by Pliny (if we may depend upon the numbers), of the Roman Ships sailing no more than about forty -four geographical miles per day across to open sea between Arabia and India, in which we cannot suppose them to have absolutely stopped at night, as in their coasting voyages and in soundings.

“ We may reckon, at a medium, from thirteen to fourteen hours of daylight throughout the year in that parallel ; so that three miles per hour for the daylight makes up the whole sum (bating three or four miles), which is a very slow rate of sailing before the brisk monsoon that prevails in that sea, and leaves little or nothing for the night ; and although it is possible, and even probable, that they may have lain to during this interval, yet ten or eleven hours drift must amount to something. Hence we shall not lay so great a stress on this instance, being a solitary one of the kind, as on the others in the coasting navigation. In these it appears almost certain that the ordinary mode of sailing was confined to daylight ; for without a compass, or a substitute for it, great danger must have been incurred in the night, when a small error in the angle of the course would be fatal. Lighthouses, on prominent parts of the coast, would doubtless direct them ; but this could not be a general arrangement, and must have been confined to particular coasts only. Notwithstanding, sailing by night was doubtless practised occasionally, as in clear moonlight, or at other times when necessity pressed, or the nature of the shore (as at the mouths of the Tigris and Euphrates) prevented his coming to an anchor or landing. Sailing by night is also implied, when Scylax admits nights as well as days in his calculation of the distance between Carthage and the columns of Hercules ; a navigation by no means intricate, and perhaps assisted by lighthouses or signal fires. It must also have been occasionally practised in the Euxine. (See Note to p. 678, article 6.)

“ In effect, then (Major Rennell continues), we must suppose a rate