The fissure of the cone on the north-west side was large and deep, and extended into the Atria del Cavallo, about 300 metres. No mouth opened along the cleft of the cone itself; all the lava issued from that part which extended into the Atria. From previous experience I should have expected to have seen the formation of adventitious cones along the widest part of the fissure, which is never that most elevated, and these discharging from their summits æriform matter frequently mixed with projectiles, and from their base lava; but on this occasion no cone appeared at the widest part of the fissure, but a long hillock was formed like a little chain of mountains, one point of which was elevated about fifty metres above the plain beneath, and bearing no resemblance to a cone. Another fissure opened in the cone on the south side, which did not extend to the base, and lava issued from this and flowed in the direction of the Camaldoli. Streams of less importance furrowed the cone in other directions, but the largest quantity of lava proceeded from the fissure in the Atria del Cavallo, below the hillock or miniature chain of collines just described. This lava stream was for some time restrained within the Atria del Cavallo, among the holes and inequalities of the lavas of 1871, but these being filled up and overcome, it divided into two branches--the smaller one flowing through a hollow which separated the lavas of 1867 from those of 1871, and made its way over the lavas of 1858, threatening Resina, but stopped as soon as it reached the first cultivated ground; the larger branch precipitated itself into the Fossa della Vetrana, occupying the whole width, about 800 metres; and traversing the entire length of 1,300 metres in three hours. It dashed into the Fossa di Faraone; here it again divided into two streams, one overlying the lava of 1868, on the Plain of the Novelle, partially covering the cultivated ground and country-houses; the other flowing on through the Fossa di Faraglione, over the lava of 1855, reached the villages of Massa and St. Sebastiano, covering a portion of the houses, and thence continued its course through the bed of a foss or trench which, contrary to my advice, had been excavated after the eruption of 1855, in the expectation of diverting the course of that lava. I did not fail to observe that the rains which previously descended through these steep channels, would in future be kept back to filtrate through the scoriæ, without ever reaching the new channel.