Sec. 3.] detailed descriptions. 99 



Further west a similar small exposure of syenite was seen, but its 



S enite inlier relations were equally obscure. Possibly, other 



small patches might be found if close search were 



made. The position of the two seen along- the same line, is rather in favour 



of their being- part of the same area of metamorphic rock, brought up 



by a fault striking east-north-east. 



The traps north of the Dhar forest are perfectly horizontal, and. 

 Traps north of Dhar ^^ "^'^6 lower flows at least, very massive and 



forest, T 



columnar. 



A remarkable columnar trap dyke occurs about half a mile north of 



the small village of Pullasee, and about eight miles 

 Colunmar trap dykes. 



west of Kataphor. It forms a ridge-shaped rise 



of no very great extent, the top being a mass of columns inclined 

 slightly to the north. They lie perfectly loose, having been rendered 

 so by weathering, and are very regular in shape, mostly pentagonal 

 or trapezoidal, rarely hexagonal. They average 1 foot to 1 foot 3 inches 

 in diameter. Columns 6 and 8 feet long without a crack abound. 



A still more extraordinary dyke intersects the Vindhyans to the 

 north of Dharee, and may be mentioned here. It may be traced for 

 several miles across the country forming a series of elevations. The 

 columns are equally perfect with those in the Pullasee dyke, . but they 

 vary in direction, being generally nearly horizontal, but sometimes 

 inclined at various angles, sometimes even perpendicular. 



Between the traps and the Bijawurs, along the north and west of 

 Sandstones along the ^^^ Dhar forest, peculiar soft sandstones and con- 

 north edge of Dhar forest, gioj^e^-ates intervene, which increase in amount 



to the westward. They are not observed to the east of Rattee Talao, 

 thouo-h a closer search may shew that they extend farther in that direc- 

 tion. Some sandy travertine-like beds which were Indistinctly seen south 

 of Kataphor, and about | mile north of the village of Pora, may perhaps 

 belono- to this series, but it appears more probable that they are 



( 261 ) 



