LATERITIC FORMATIONS. 51 



series to whicli so much attention has been drawn in foregoing pages. 

 Other gneissic rocks have also furnished a few pebbles, and a few stray- 

 ones of chert and one or two of agate were noted as well. 



Besides the concretionary pellets of earthy haematite with a smooth 



Concretionary and ac- ^""^ "^^'^^ ^^^^'^ '^^"^^^^ ^^^"^^^ ^'^ ^f COmmon OC- 



cretionary ferruginous currence wherever the sands or gravels are richly 

 ferruginous, there is another form of pellet of 

 very frequent occurrence even in the less richly ferruginous beds. This 

 differs from the other in consisting of a mere aggregation of grains of 

 sand by a ferruginous cement which shows no concentric arrangement. 

 This form of pellet is almost always rough on the surface from the pro- 

 jection of numerous grains of sand, and the gravel it forms is invari- 

 ably due to deposition of ferruginous matter by water at various levels. 

 Where this action has been long enough continued the pellets are 

 aggregated into a quasi-conglomeratic mass, the real origin of which is 

 sometimes not very easily discernible. 



As might be expected, the gravels in the more easterly part of this 

 tract are much less coarse tbaa those occurring at higher levels and 

 Pale gravels at Abi- further inland. In the extreme south-easterly 

 ramam. p^^,^ ^^^^ Abiramam (Abramum) the gravels are 



but very slightly iron stained, and the pebbles of rolled granular quartz 

 rock, which form fully 95 per cent, of the whole, are of a pale cinnamon 

 colour tending to pale ochrey yellow. 



The last and most southerly tract which I have called the Parnalli 



tract (No. 9) shows like the foregoing one a much 

 The Parnalli tract. 



smaller development of the ferruginous forms of 



lateritic rock than do the tracts north of the Vaigai river. This tract 



differs from all the others in that its surface is to a very great extent 



masked by a thick and, in the southern part espe- 

 Great spread of regur. _ '^ ^ 



cially, almost unbroken sheet of regur or cotton 

 soil. Owing to this extremely thick covering and the great paucity of 

 sections, the area of the tract as shown on the map must be received as 

 only a very rude approximation to the truth. 



In the northern part of the tract the pall of cotton soil is wanting, and 



( 51 ) 



