292 
DR. TEMPEST ANDERSON ON THE ERUPTIONS OF THE SOUFRIERE IN 
but the north wall being higher than the south the plane of measurement was inclined 
in that direction, so that on a level the shape was really somewhat oval, with the long 
diameter from the W.N.W. to E.S.E. The lake was also oval, with its diameter in 
the corresponding direction. Its length on February 6 was 540 metres (1772 feet), 
its breadth 340 metres (1116 feet), and the height above the sea-level 585 metres 
(1919 feet), # which agreed almost exactly with that given on the chart (1930 feet). 
The water was boiling in the centre and also at the south-east corner. The district 
between the Somma wall and the two craters was deeply covered with ejecta. The 
new crater was filled up almost Hat. It had a small shallow lake (about 70 metres 
(230 feet) in diameter) towards its eastern side, with water marks which showed that 
it had been bigger. The saddle, which previously separated the new from the old 
crater, had disappeared, probably owing to landslides. Fumaroles were found in the 
crater rim in the gap leading down to the Larikai valley and also one somewhat more 
to the east above the head of the Rozeau valley. 
Professor Lacroix made the ascent along with Madame Lacroix, Ensign Deville, 
Professor Hovey, and the Rev. T. Huckerby of Chateau Belair, on March 3, 1903, 
i.e., before the last considerable eruption, and has published his observations in the 
‘ Annales de Geographie.’f The following is an abstract of the parts of his paper 
which relate to matters that have occurred since the May eruption. He mentions 
that banks of hot ash and terraces still remain in the Wallibu valley, and that steam 
explosions still occasionally take place from them.| These banks render the river 
valleys somewhat narrower than before. He saw from his boat, when off the mouth 
of the Richmond River, an explosion of hot mud take place from the crater of the 
Soufriere, and he observed the sudden descent of a torrent of mud in the high Rozeau 
valley, probably caused by a shower. At a height of about 600 metres the ground 
was covered by large lapilli, the product of the eruption of September, and ejected 
blocks and bombs were common. They were often partly buried in the lapilli into 
which they had fallen. He mentions the gap at the head of the Larikai ravine, and 
was informed by Mr. Huckerby that it had been enlarged by the eruption of 
October, 1902. Avalanches were falling from the walls of the crater, and there was 
a talus at their base. Large ejected blocks of old rock were more common at the 
rim of the crater, particularly at the east, than they were lower down. At first, 
when M. Lacroix gained the edge of the crater, the water of the lake was 
tranquil, and had the appearance of yellow steaming mud. The least agitation 
of any part caused it to assume a more grey colour. Suddenly from the centre of 
the lake there rose a mass of mud of inky blackness, entangling blocks of rock. 
* These measurements apjjear to have been carefully made. Hovey gives 600 metres; most of the 
other observers also give figures, but as they are merely estimates I have not quoted them. 
t “Les Dernieres Eruptions do Saint-Vincent,” Mars 1903, A. Lacroix, ‘Annales de Geographic,’ 
tome xii., 1903, No. 63 du 15 Mai 1903. 
| They appear to have ceased about this time. 
