OP LA PER0USE. I43 



combuftion of fhrubs growing in the fhade of 

 thefe forefts, the tree would have fuffered in 

 every part of its circumference. Thefe excava- 

 tions in trees ferve as a place of fhelter to the 

 natives, who come hither to make their meals : 

 we found here remains of lobtfers and other 

 fhell-fifh which they had been eating : we iaw 

 alfo, in feveral of them, the afhes of fmall fires 

 which they had kindled to drefs their food. The 

 favages are not in the greater! fafety under thefe 

 large trees, partly undermined by fire ; a firong 

 wind may blow them down ; nor do they lie 

 within them too much at their eafe, for the 

 ground is often very unequal, and we perceived 

 no means taken for diminifhing its hardnefs. 

 Anderfon fpeaks of fire-places of clay made in 

 thefe trees by the natives. The clay which I 

 faw there did not appear to me to be fafhion- 

 ed by the hand of man ; it is fometimes to be 

 found naturally attached between the boughs at a 

 greater or lefs height. Befides, the favages,, as 

 we fhall fee in the fequel, do not conftrucl fire- 

 places ; they make their fire limply on the 

 ground, and it is on the coals that they dre& 

 their victuals. 



Some of the thickeit trees, excavated by fire 

 throughout their whole length, formed a fort o[ 

 chimney ; yet they did not the lefs continue to 

 vegetate. 



. Several 



