422 Art Account of the Discovery ' ^ 



obtained before and afterwards, at the house of Major Dantas/ 

 called Camuciata, near Itapicuru. 



The rapidity of growth in plants is wonderful in the neigh- 

 bourhood of the Bendego, although the main granite rock is 

 so near the surface as to protrude in many jjlaces ; and what 

 lies on it is chiefly a coarse gravel, consisting of rolled fragments 

 of quartz, felspar and granite of the size of eggs, together with 

 smaller pebbles and sand, which contains, of course, a great deal 

 of mica, but hardly anv vegetable earth. 



At about forty leagues to the southward, tire found hills of 

 yellow and red sandstone, in which organic remains have not 

 been found ; while to the nortliward there is a formation of si- 

 milar hill-i, in which are oliserved most beautiful impressions of 

 whole fishes and remains of vegetables. 



Between the Bendego aiul the sandstone hills to the south- 

 ward, 1 observed a great deal of what I certainly take to be ba- 

 salt. I met with balls from the diameter of two inches to that 

 of upwards of three feet, and numberless prisms, with three and 

 with six faces, scattered about ; all of these small, that is to say, 

 about three or four inches in length, and two or three in dia- 

 meter. 



To the southward of the sandstone hills is a sandy plain, al- 

 most barren, extending many miles, perhaps sixty or eighty, 

 east and west to the sea, but not twenty in breadth, where I 

 crossed it. Small conical hillocks are scattered over it, some 

 of which, the largest, have flat tops, and appear all to be of the 

 same height, about twenty fathoms. 



Appearances impressed me with the idea, that they were the 

 remains of a plain which formerly extended over the one on 

 which I then stood, but v/hich had been washed away in a tu- 

 multuous manner by a violent current running nearly in an 

 easterly direction. The larger hillocks appear to he stratified, 

 but they consist of loose sandy materials, except in so far as they 

 contain beds of a dark-red iron ore, containing imbedded mi- 

 nute crystals of magnetic iron ore : the thickness of these beds is 

 about two inches, and they are exactly similar to those which 

 are found in the clay hills of Bahia. 



The smaller hillocks consist of confused heaps of gravel and 

 loose stones, interniixed with a very large quantity of the same 

 iron ore in rragments, and lumps of manganese, very compact, 

 and of a steel-gray colour, containing arsenic, but apparently no 

 iron. 



The dreary appearance of this plain is increased by the nu- 

 merous nests of ciipim (white ants) standing upright like so 

 many tombstones. On being viewed nearer, they are conical, 

 rather CGmj)ressed; so that the base is elliptic. All those which 



I examined 



