Descending to the crater of Iatsifitra, of which I obtained a photograph, we observed that the lava rocks, which had issued from it to the north, were black, sharp and fresh, as if they had been broken yesterday. On its eastern side was a plain, a mile square, covered with heaps of broken lava, like stone cottages, fortresses, and ruined palaces. I counted thirty greater piles: and noted numberless smaller ones. It was clear that, like the Phlegrœan fields in Italy and the neighbourhood of Mouna Roa in Hawaii, the entire plain had at some time been on fire; and that a hundred jets of fire and flame and molten lava had spurted from its surface, hurling their blazing rockets into the sky. The heaps were now old and moss-grown: but one of the peasantry informed Mr. Sewell, that there was a kind of tradition amongst the people, that their ancestors had seen these flames bursting forth. They called the fire _áfo-to_: and the lava _kitroka_. Near to Betafo in a cutting, ten feet deep made by a stream, I observed a specimen of the strata formed by the successive eruptions of these volcanoes. At the bottom were lumps of lava, somewhat small. Above these came several bands of brown earth: then one of black earth: above this several strata of ashes, and of lava gravel: above these again were strata of earth. Similar strata we had seen in a cutting in the Mándridráno: and subsequently saw others under the hill of Ivohitra in Sirabe. I need not give the details of our inquiries and their results on each day of our journey. Travelling farther east, across Sirabe and round the southern end of Ankaratra we had the lava still, with long tongues, cinder heaps, and old craters again and again. We ascended one of the noblest hills in the country, the conical peak of Votovorona, a most important station in our survey: and we found it to be volcanic. Another fine cone to the east of it, Ihankian, was volcanic also. In this district we found sixty of these cones and craters. Altogether in this important journey we saw and counted a hundred extinct craters, extending over an arc of ninety miles, not reckoning the central mass of Ankárat, the finest of all, round one side of which this great arc bends. Even these do not exhaust the tale. The volcanic belt appears once more to the northward in the lofty hills with which the island terminates. The peak of Mataola is volcanic. Mount Amber is volcanic: the renowned fortress close to it is an ancient crater. Nosibe is an island full of volcanoes. Mayotta and Johanna in the Comóro isles are full of craters, lava peaks and lava streams. What a mighty volcanic force must have been exerted over this enormous area! Does Java itself show a more wonderful volcanic field? If human eyes could have beheld and appreciated them, what a scene of indescribable grandeur must have been presented when these volcanoes were active: when the land was rocked with earthquakes, and the mighty hills of gneiss and granite were