18 KING : COASTAL REGION OP THE GODAVAKI DISTRICT. 



third of this length of outcrop in the middle of the field ; and this is agaia 

 overlapped by the higher one, which extends from the Tammiler or 

 Ellore river to within a short distance of the Godavari. The latter is 

 also, as far as can be made out, continued on the Rajahmimdry-Vizaga- 

 patam side of the country, but without any association with other sand- 

 stones, where it lies directly on the gneiss. 



The three sub-divisions are fossiliferous at rare intervals ; the lower 



only showing vegetable remains, whereas the other 

 Fossils. . .'*'., „ , 



two give marine animals, one of them also having' 



a few associated plants. The Gollapili sandstones, with their typically 



Rajmahal flora, have given one of the most definite geological horizons 



on the Coromandel, while the Ragavapuram shales and Tripati sandstones 



are brought into correlation by their marine fossils, the former with the 



previously-known fossiliferous patches of shales further south in the 



Nellore-Kistna area and at Sriperumbudur (Sripermatur) near Madras, 



and the latter with the Umia beds of Cutch. 



Gollapili Sandstones. — About 12 miles west of Ellore, the large 

 village of Gollapili stands on a series of dark-brown and yellowish- 

 red sandstones grits and conglomerates which have yielded numerous 

 plant remains of Rajmahal age. At the northern edge of the field, 

 near Somavaram, these sandstones are lying on, and overlap, other 

 coarser but softer and more vari-coloured felspathic sandstones and 

 sandy shales of Kamthi age, with the characteristic glossopteris and 

 vertehraria. 



The fossils of these Rajmahal beds were found only at a. few localities 



late in the progress of the survey ; and as there is 

 Palaeontology. 



often a very close resemblance between these sand- 

 stones and those of the Kamthis, it was thus some time before I was able 

 to separate this group from the larger area of rocks which W. T. Blanford 

 had approximately sketched in as of the latter series during his prelimi- 

 nary traverse between Niizvid and Ellore. 1 The complete collection was 

 finally determined by Dr. Eeistmantel, the palaeontologist of the Survey, 



1 Records, Geol. Surv. of India, Vol. IV, p. 49 ; V., p. 23. 

 ( 312 ) 



