ON CERTAIN RECENT CHANGES IN THE CRATER OF STROMBOLI. 125 
which cone in that case it would have completed. As it is, the summit 
of the mountain is about 3000 feet above the sea-level, while the 
eruptions take place about 600 feet lower, i.e. at a height of about 
2400 feet. 
When I first visited the mountain in 1888 (see Plate II., which is 
from a photograj^h taken by me at the time), the crater was an oval 
hollow, which I afterwards roughly estimated at 60 or 70 yards in 
diameter, but which, on examining the place again in 1904, I believe 
to have been much larger. I made a second ascent to check this, but 
was prevented by bad weather from completing the observation. Prof. 
Judd’s estimate of 400 feet was probably nearer the mark. In 1888 
there were two small cones, one on the edge of the Sciara near its 
western end, from which the explosions took place, another in the 
hollow of the crater near its eastern end, which only gave out vapour. 
In 1904 the accumulation of materials thrown out by the volcano had 
filled this crater and continued the slope of the Sciara upwards, so that 
the explosions no longer took place from the hollow of the crater or 
even from its lip, but from mouths on the slope of the Sciara, I have 
no reason to think that the main axis of eruption has shifted materially 
down the Sciara towards the sea, notwithstanding the appearance to 
the contrary, which is, I believe, due to the piling up of fragments to 
the south-east of the vents on the site of the crater seen in 1888. It 
is interesting to note that in 1888 the fumarolic activity was chiefly at 
the north-east of the crater, then a hollow, and the explosions took 
place from the western portion. In 1904 the relation still continues; 
the fumarolic area, now much raised into a mound, as shown in Plate IV., 
is still at the north-eastern portion of the eruptive area, while the main 
explosions take place from the western parts. It is, however, note¬ 
worthy that the flow of lava in 1891 appears to have proceeded from 
the eastern part of the crater. These facts will have to be taken into 
account by any one who frames a new theory of the mechanism of 
Stromboli. 
Since my return home, I have endeavoured to collect photographic 
and other information so as to fill in the changes which have occurred 
in the interval, and present them in a connected form. 
Prof. Judd, f.r.s., has kindly placed at my disposal several papers 
from the Atti della Societci Italiana di Seienze Naturali, by Prof. Mercali, 
of Naples, from which the following information is condensed. 
On February 4-5, 1879, a severe eruption of Stromboli took place, 
and such a quantity of pumice was thrown out that the sea in the 
neighbourhood is stated to have looked like dry land. There is no 
mention of any lava being discharged.* 
* Mercali, Atti della Soc. Ital. di Sci. Nat., vol. 24. Milano, 1881. This number 
contains a good list of the ancient eruptions of Stromboli and a discussion of Mallet’s 
theory. 
