been none before, as did also several mineral springs, of which be fore there was not a yestige. Commonly, however, the fountains became more swelled and more copious, and emitted a larger volume of water than usual. The waters of some fountains were also observed to be troubled, and to assume a whitish or yellowish colour, according to the countries through which they passed. For a long time before the earth shook, the sea appeared considerably agitated, so as to frighten the fisherinen from venturing upon it, without any visible winds to make it so. Our volcanoes too, as I am confidently assured, emitted no eruptions for a considerable time before; but there was an eruption of Etna in the first earthquake, and Stromboli showed some fire in the last. [Phil. Trans. 1783.] CHAP. XXV. CAVERNS, GROTTOS, AND OTHER NATURAL EXCAVATIONS. No man who has attentively perused the preceding chapters—who has carefully contemplated the mode in which the solid crust of the earth appears to have been formed and aggregated on its emerging from a state of chaos, and broken, undermined, decomposed, and recombined into new shapes by the posterior agency of subterranean gasses, volcanoes, and earthquakes, of confined air, vapours, and water, can be much surprised at finding that its structure is in many instances to be met with rent asunder into chasms of awful and tremendous depth, or scooped out into hollows of the most grotesque and romantic appearance. From an almost incalculable variety of phænomena of this description, our limits will only allow us to select a few of those that are of the most singular character; and in doing this we shall strictly con. fine ourselves in the present place to those that are of natural origin alone: reserving our account of those that have been produced solely or chiefly by human labour for that division of our work which will take a general survey of the wonders and curiosities of ART. SECTION I. Brief Survey of Foreign Caverns. THERE are few countries that have not to boast of a variety of natural excavations, which from their extent or structure have in all ages become objects of popular attention, and have not unfrequently been interwoven into the traditions or superstitions of a remote and doubtful antiquity. Near the city of Heraclea, whose walls rose on the shores of the Euxine, was situated a cave of this description, which was visible in Xenophon's time, and of two furlongs depth. It was through this that Hercules, to whom the city of Heraclea was dedicated, was feigned to have descended into the infernal regions, and to have dragged to light the three-headed Cerberus, The vestiges of the city still remain, and are shewn by the Turks under the name of Eregri: the cavern, however, has long been de⚫ stroyed, or at least closed up to a considerable extent at the mouth. The Averno, in the Phlegrean fields, offers an excavation of a similar kind, and is equally celebrated in traditionary story for its being an entrance to the infernal regions; but as this cavity is still more famous for its sheet of water and the deadly vapours that are said to issue from it, we shall describe it more particularly under the chapter upon Lakes, The volcanic country about Rome is peculiarly diversified by natural cavities of great extent and coolness, on which last account we are told by Seneca, that the Romans were accustomed to erect their country seats in their vicinity, to enjoy their refreshing chillness in the summer season. He gives a particular account of two such grottos belonging to the villa of Vatia; and it was in a place of this kind that Tiberius was nearly destroyed whilst at supper; for during this banquet its roof suddenly gave way and buried several of his attendants in its ruins; which so fearfully alarmed the rest, that every one fled away and abandoned the emperor except Sejanus, who stooping upon his hands and knees, and covering the body of Tiberius with his own, received all the stones that fell from the roof, so that, though he himself sustained considerable injury, the emperor escaped unhurt *. *Tacit. Ann. lib. iv. cap. 69. We meet with a singular grotto or cavern in the rocky island of Antiparos, the ancient Oliaros in the Levant, which lies about two miles to the west of the celebrated Paros, and is not more than sixteen miles in circumference, but has a port accessible to small barks. This extraordinary grotto appears to be about forty fathoms high, and fifty broad: the roof forms a pretty good arch, and every where entertains the eye with an infinite variety of figures of a white transparent crystalline marble, representing vegetables, marble pillars, and a superb marble pyramid, as regular as if formed by art. Every one has heard of the labyrinth of Crete or Candia. This, which is a natural curiosity, runs under a small hill at the foot of Mount Ida, in a thousand intricate aud irregular meanders, something like that of Egypt, which, as being a product of art, belongs to another division of the present work. The entrance is an aperture seven or eight paces broad, but in some places so low that it cannot be entered without stooping. The ground is rugged and uneven, but the top is level, and consists of an horizontal layer of stones. The principal passage, in which there is less danger of a person's being bewildered than in the others, is about twelve hundred paces in length, and reaches to the end of the maze, where are two spacious chambers. The most dangerous part of the mainwalk is about thirty paces distance from its mouth; for if a person happens there to take a wrong course, he is soon bewildered amidst the innumerable windings which occur, and his extricating himself is very difficult, and even uncertain; but travellers always provide themselves with guides and torches. There is little appearance of this ́place having ever been a stone quarry. Tournefort considers it as effected by human labour, which, in that case, must have been very great; but it is most probable that human exertions have greatly extended what nature had formed. Some persons assert this ca vern to be the ancient labyrinth of Dædalus; but Pliny distinctly informs us, that even in his time not a vestige of this celebrated maze was to be seen. Milo, anciently called Melos, latitude 36° 41′ north, longitude 25° east from Greenwich, lies near sixty miles north of Candia, and is fifty miles in circumference. It consists almost entirely of one hollow, porous rock, which is macerated, as it were, by the seawater; and the heat of a continual subterraneous fire is felt, on introducing the hand into the holes of the rock: a place in the island |