156 BOARD OF AGRICULTURE. 



drift of the North, that there is no escape from the conclusion 

 that even these tropical regions were once under the ice sheet. 

 But I will not here dwell upon a topic which is foreign to the 

 subject on which I desired to speak, but only state that there is 

 this remarkable resemblance between the loose soil of the North 

 and the loose soil of the tropics ; and whatever may have been 

 the plough which has ground our rock to make it fertile soil for 

 man to dwell upon, that same agency has been at work upon the 

 plains of the valley of the Amazon, and has produced the same 

 results. Everywhere it is drift which is the most fertile soil all 

 over the surface of the earth. 



There is only one tract of land which has hills of some strik- 

 ing dimensions, and that is between the mouth of the Xingu 

 and the mouth of the Rio Trombetas. On the north side espe- 

 cially, there are remarkable hills, and on the southern side 

 somewhat lower ones, but equally striking in appearance. 

 These hills are particularly impressive because of their form. 

 All mountain ranges everywhere have undulating crests, have 

 inequalities on their summits, and changes in their direction ; 

 but these hills of Obidos and Almeirim are flat-topped, with 

 sides as even as the gable of a house ; and these flat-topped 

 mountains extend considerable distance. They were for some 

 time supposed to be spurs of the table land of Guiana, or the 

 table land of Brazil, and had never been accurately measured. 

 It was during my journey that, for the first time, I suppose, the 

 means for measuring them were brought to their summits, and 

 we ascertained that they are much lower than they appear. 

 The impression is, when looking at those mountains from the 

 centre of the river, that they are many thousand feet high ; 

 they stand so abruptly, that the eye is deceived, and the impres- 

 sion is so grand that you might suppose that there was a very 

 high mountain range rising on the borders of the horizon. But 

 when actually measured, they were found not to be above nine 

 hundred feet high. It is very curious how the eye, in attempt- 

 ing to measure dimensions on so large a scale, is deceived. It 

 would at first appear, that owing to the very extensive plains, 

 high mountains might produce the impression of low hills. 

 There it is exactly the reverse. Hills not reaching a thousand 

 feet in height, make an impression as great as that of the Alps. 

 The scenery, in fact, recalled to my mind the familiar views of 



