On the Eruptions of the SoufrUre and, Mont Pelee. 435 



following the steepest descent. It clung to the valley bottoms and 

 coursed along them in a manner which somewhat recalls a raging 

 torrent in a river. The streams in these valleys after descending the 

 first part of the hill turn sharply at a right angle towards the coast, 

 deflected by the opposing mass of the Morne Garu. The hot blast 

 mostly followed these valleys, and in them it piled up enormous 

 deposits of sand, but part of it swept up the shoulders of Morne 

 Garu, and tore "up the heavy timber which was growing there. The 

 direction in which the fallen trunks point shows that the blast was 

 split into two parts — one taking the east and one the west side of the 

 mountain, rushing upwards obliquely from below. The mountain 

 protected the country behind, and the line of demarcation between the 

 burnt and the green forest almost corresponds with the dividing ridge. 

 The south side is green; the north side towards the Soufriere is 

 devastated and burnt. 



The effect of even comparatively small ridges in deflecting the blast 

 and protecting the country behind them is still more noticeable near 

 Chateaubelair. Between the Rivers Wallibu and Eichmond there is a 

 high dividing ridge. The northern valley (the Wallibu) is filled with 

 ash and utterly burnt up, that to the south (Richmond Valley) is in 

 large part green. One side of the dividing ridge is blasted ; on the 

 other the arrowroot is again putting out its green leaves. Another 

 ridge separates Richmond Valley from Chateaubelair. This ridge has 

 been in many places scorched, but the country behind it has been 

 perfectly protected, and, though covered with the rain of ash, has 

 resumed its normal appearance. There can be little doubt these ridges 

 served to direct the path and intercept the violence of the hot blast. 



For some days after the eruption no rain fell, and the first to visit 

 the district were able to observe the effects of the eruption unmodified 

 by the erosive action of running water. But on May 25, 5J inches of 

 rain fell. On the previous day the rainfall had been 2£ inches, and 

 the rainy season now set in in earnest. The effect of these deluges 

 acting on loose material lying on steep slopes was phenomenal, and by 

 the time we reached the island the surface of the sheets of ash had 

 been sculptured into innumerable furrows and runnels. They cut 

 down through the incoherent sand to the layer of burnt vegetation on 

 the old soil beneath, or even into this, forming new channels, which 

 varied from a few inches to many feet in depth. To one fresh from a 

 temperate climate and unaccustomed to the power of tropical rains the 

 rapidity of denudation under these conditions was astounding. On 

 the upper part of the Soufriere beautiful feather patterns of rain rills 

 converging towards a central main axis everywhere characterised the 

 surface. The knife edges between the valleys were the only parts 

 retaining the original smooth surface, and they formed excellent paths, 

 as the sand was firm, except near the summit of the hill. 



