There are some small islands in the Caspian, and its waters are much less saline than those of the ocean; storms are here very dangerous, and large vessels are not used in it for navigation, because it has many sand banks, shoals and rocks scattered under the surface of the water. Pietro della Valle says, "The largest vessels employed on the Caspian Sea, along the coasts of Mazanda in Persia, where the town of Ferhabad stands, although they are called ships, appear smaller than our Tartanes. Their sides are high, and they draw but little water, having a flat bottom. They give this form to their vessels, not only because this sea is shallow, but because it is filled with shoals and sand banks; so that if the vessels were not fabricated in this manner they could not be used with safety. Indeed, I was astonished, why at Ferhabad they fish only for salmon, which are found at the mouth of the river, some poor sturgeons, and other sort of fresh water fish, of little value: I attributed the cause of it to their ignorance of the arts of fishing and navigation until the Cham of Esterabad, whose residence is at a sea port, informed me that the waters are so shallow 20 and 30 leagues from shore that it was impossible to cast the nets with the chance of taking any fish, and that it was for this reason they gave the above-mentioned form to their vessels, which are not mounted with any cannon, as but few corsairs and pirates ever visit this sea." Struys and other travellers have asserted, that in the neighbourhood of Kilan, there were two gulphs wherein the rivers of the Caspian were ingulphed, and carried afterwards by subterranean canals into the Persian Gulph. De Fer and other geographers have even marked out these gulphs in their maps, nevertheless we are assured by the people sent by the Czar that they do not exist.[B] [B] See Mem. Acad. Sciences, 1721. The circumstance of willow leaves being seen in great quantities on the Persian Gulph, and which are supposed by the same authors to come from the Caspian Sea because there are no such trees on the Persian Gulph, is fully as improbable as their subterraneous gulphs, and which Gemelli Careri, as well as the Muscovites, asserts are entirely imaginary: in fact, the Caspian is near one third smaller than the Black Sea, which last also receives much more water by rivers than the former: the evaporation therefore is sufficient to carry off all its water, nor is it necessary to suppose subterraneous gulphs in the Caspian any more than in the Black Sea.