Table Lands of South India. 309 



J . A poor gravelly soil intermixed with angular pieces of 

 quartz lying upon pegmatite and porphyries. 



2. Light red soil, with few pieces of quartz lying upon 



sienite. 



3. Red soil lying upon greenstone, with associated kunkur. 



4. Dark brown red soil, very ferruginous, intermixed with 



angular pieces of ferruginous claystone, lying upon trap. 



In none of these soils are pebbles ever found, the fragments 

 of rocks being always sharply angular, Near the beds of 

 rivers and water-courses, water-worn pebbles are frequently 

 found in the fields, but I have never in any part of India, 

 seen pebbles in any other situation. 



Superposed upon the surface of the " schistose series" 

 deep beds of earth are sometimes found of several miles in 

 extent, and sometimes 20 feet in depth. They are charac- 

 terized by seldom containing fragments of rocks, and never 

 any pebbles. That they are not alluvial is proved by blocks 

 of granite being sometimes found in them, which are traversed 

 by veins of quartz, which also run into the earth, as men- 

 tioned in describing the granite formation. As similar beds 

 of earth never occur below schistose rocks, I am inclined to 

 consider them as a separate formation, which may be genera- 

 lized under the term terraceous series. 



Layers of sand, as found in alluvium, never occur in these 

 beds. 



The soil is generally red, or sometimes an arenaceous cal- 

 careous loam, and generally portions of ramose kunkur 

 traverse them in all directions, both vertically and horizon- 

 tally, and the pieces are all in connection from the surface 

 down to the depth of even 10 feet. A remarkable formation 

 of this kind is found in the valley which runs north and 

 south-west of Balcondydroog in the Barramahal, which is 

 10 miles in length by 4 in width. 



The beds of the Regur soil, or cotton ground of the Salem 



and Barramahal, I consider as belonging to this series. The 



2 R 



