Vol. 50.] GEOLOGY OF MATTO GEOBflQ. 99 



Alluvial Deposits. 



The lowland plains and swamps are composed entirely of alluvium, 

 though penetrated here and there by elevations of the older rocks. 

 The rivers bring down from the hills a vast amount of detrital 

 material, especially in flood-time. When a stream overflows its 

 banks, the luxuriant vegetation on the margin catches up and 

 entangles most of the matter held in suspension, so that a strip 

 adjoining the river is raised higher than the ground farther away. 

 In some localities remote from the course of the principal rivers 

 permanent lakes of considerable area occur. In the time of high 

 water, which lasts for about six months in the year, these lakes 

 appear to be merged in the extensive sheets of water (xaraes or 

 pantanaes) x which unite the courses of the Paraguay and of its 

 tributaries the Sao Lourengo, the Cuyaba, and others, and cover 

 almost all the alluvial plains. These facts point to the former 

 existence of a great lake or lakes (comparable to those of Equatorial 

 Africa or North America) which have been subsequently filled up 

 by alluvium. 2 



The alluvial deposits are in some places cut into by the rivers, 

 and sections are thus afforded. These occasionally contain thick 

 shell-bands of a large Ampullaria, apparently identical with one 

 that now inhabits the adjoining swamps. 



VIII. Unclassified Rocks. 



At Uruciim, near Corumba, I found, overlying the ancient igneous 

 rocks and probably also the Corumba Limestone, deposits almost 

 entirely composed of oxides of iron and manganese. They form a 

 hill some 600 metres (1950 feet) high, and 5 or G kilometres (3 or 3| 

 miles) long. Some of the neighbouring hills appear to be of similar 

 composition, although it is said that they do not contain so much 

 manganese. The strata form a gentle synclinal. They seem to be 

 fragments of former more extensive deposits (probably lacustrine), 

 and to owe their preservation to the accumulation at this point — 

 the centre of a synclinal — of the above-mentioned oxides, which 

 have proved more durable than the rest of the deposits. 3 



These rocks appear to be younger than the Corumba Limestone, 

 but are probably older than the Chapada Sandstones, for they have 

 suffered more from earth-movements. They may, perhaps, be a 

 local phase of the Matto Shales. 



Massive siliceous iron ore, with geodes lined with quartz, occurs 

 near Coimbra, but I have not seen it in place. 



1 [Perhaps here .raracs = charcos, and pantanaes =pantanos. — Ed.] 



2 A similar process, on a much smaller scale, is now going on in the district of 

 the Norfolk Broads, {21) p. 353. 



:1 From the description given me at Uacurisal (a locality on the Paraguay 

 already referred to) of the rocks in the hills near Gahyba, farther north, it 

 seems not improbable that they may be similar to those at Uruciim. 



