334 



NATURE 



lAug. 17, 1876 



in the harbour of St. Denis ; and on landing I organised, 

 with MM. Rochefort, Cazin, and De I'lsle an expedition 

 to the volcano in the hope of arriving in time enough to 

 witness the end of the eruption. We set out from St. 

 Benoit on Sept. i, and made for the plain of Palmistes, 

 our first stage. This plain is surrounded on all sides, 

 except the north-east, by perpendicular ramparts, which 

 may reach a height of 200 metres, and whose sides, 

 covered with vegetation, form a semicircular curtain of 

 verdure that shuts out the horizon. 



From the plain of the Palmistes we had to climb to that 

 of Cafres by crossing the rampart of the Grande-Montde, 

 a long and difficult ascent on account of the abruptness 

 of the rampart. We reached the summit about an hour 

 after mid-day, and found the temperature to be 14° C, 

 less than half that of the lower part of the island. The 



plain of Cafres, at a mean height of 1,600 metres, forms 

 a declivity, a sort of saddle-back or pass between the two 

 parts of the island which we have distinguished under the 

 names of Ancient Mass and Recent Mass. It is a very 

 uneven plain, inclined towards the south-west, i.e., in a 

 direction opposite to that of the Palmistes, and formed 

 by a succession of small echeloned plateaus crossed by 

 rounded hillocks covered with vegetation. The soil 

 which results from the disintegration of the lavas is here 

 very argillaceous, as all that savannah presents fresh 

 pasturage during the dry season, and is changed into a vast 

 marsh during the rainy season ; it is about two leagues 

 in length. Night surprised us near the source of the river 

 of the Ramparts before we could reach the end of our 

 journey, and we had to sleep on the bare ground ; the 

 thermometer reached 3°, and during the night sank to — 2°. 



Map of the Island of Reunion (after M. L. Maillard). 



At eleven on the next morning we reached the Cavern 

 of the Lataniers, after having visited the vast crater 

 named Commerson, singularly situated on the very edge 

 of the magnificent escarpments which form the great 

 section at the foot of which flows the river of the Ram- 

 parts ; from thence we directed our steps towards the 

 pass of the sands (2,386 metres), in order to cross the first 

 inclosure of the volcano. The present volcanic cone is, 

 in fact, preceded by two great circles produced by sub- 

 sidences which have given place to veritable circular 

 walls cut perpendicularly for more than 100 metres from 

 the top, and which are named the inclosures. Of the 

 first there remains only a small part ; on the north-east 

 its wall overhangs the river from the east, and on the east 

 the plain of sands ; but on the south it is not so easily 



traced ; it is prolonged on this side of the great section of 

 the river Angevin, the formation of which is later. 



The plain of sands (about 2,300 metres) which thus 

 circumscribes a basaltic rampart, is formed by a black 

 compact lava covered by a layer of small angular very 

 regular fragments of vitreous lava, often two metres in 

 thickness. In the bottom of the little ravines is noticed, 

 moreover, an accumulation of crystals of olivine and 

 augite which come from the disintegration of certain rocks 

 thrown out by the volcano, and composed almost solely 

 of these two minerals. It is intersected by cones of 

 scoriae regular in form, terminating in little craters, the 

 limited overflows from which appeared consolidated on 

 the ledge. We had to pass round many of them before 

 arriving at the ridge of the second inclosure, which had 



