294 a F. Hartt on Amazonian Drift 



Witb the last species Mr. Coffin otbained another, which is 

 not well enough preserved to ^dmit of accurate determination. 

 It may be a species of Amour oucium, or another Macroclumni. 

 It forms broad incrustations, 2 or 3 inches across and about -25 

 thick in the center. The marginal portion is thin and covered 

 with adherent grains of sand ; the central part is smooth and 

 glabrous, slightly convex. The zociids are about -30 long, 

 with a long post-abdomen, which is not pedicellate, resembling 

 that of A. glahrum. 



Prof. Agassiz has claimed that the structure of the Ama- 

 zonian valley is very simple ; that, from one end to the other, 

 there was laid down in the fresh water lake formed by the dam- 

 ming up of the mouth of the valley, by a glacial moraine, beds 

 of sands, clays and sandstones, filling the valley to the height of 

 several hundred feet above the present level of the river ; that, 

 fi-om the bursting of the barrier and the draining of the lake, 

 the uppermost bods were dcundod down, and swept off from 

 over a very great par: of the \ aUoy, leaving only isolated out- 

 lines as a series of taMr-toppcd liills or serras, like those of Al- 

 meirim, Paraiiquara, Vdhu Pohrc ; that the lake, reduced to a 

 lower level, then threw down over the denuded portion, a sheet 

 of clav, the river iinally cutting its charmel through this clav 

 and the underlying sand, &c. He did not claim to have Ibiui'l 

 grooved or striated surfaces in the Amazonas, and the oni} 

 instance of supposed erratics he gives is that of the dionte 

 found on the flanks of the serras of Erere. Prof Agassiz gave as 

 a type of the table-topped hills the serra just mentioned. Last 

 summer, in making an examination of the lower part ol the 

 Amazonian valley, I visited Erere and spent a month m its vi- 

 cinity going over the ground on foot in every direction, and m 

 the most detailed wav. , , .^i 



The serra does not belong to the system of table-topped Jius 

 of which Parauquara and Velha Pobre are examples. Ihese 

 last and the range of highlands on the south of the Amazomaa 

 valley are composed of soft materials ; they are perfectly m ou 

 top and descend on all sides by a steep even slope; moreo\e. 

 the strata composing them are horizontally disposed. ^ .^ 



Erere is a short, narrow, irregular lidge about 800 lee 

 height, several hundred feet lower than the table-topped niu^. 

 the top being anything but horizontal. The strata composing 

 are principally a heavy-bedded coarse white sandstone exa 

 lugly hard, with bands of indurated feldspathic clay ^^ ^ 



