1887.] PaloeoUthic Finds in South India. 265 



to a depth of 15 or 16 feet, and exposed a dark red surface of hard 

 sandy loam over an area of several acres. On this surface, I noticed 

 some flakes of a brown chert I had seen nowhere in the South, and 

 these flakes led me to search closer, and I soon found several well-made 

 cores of the Jabalpur type. With the chert flakes were a few of limpid 

 quartz, also a foreign stone in that immediate neighbourhood. A 

 number of fragments, mostly small, of antique red pottery accompanied 

 the chert implements and showed strong signs of having had their edges 

 much worn by the action of " sand blast " at some previous time when 

 the teri* was in a moving condition. 



The northern part of the Carnatic had previously yielded me a few 

 Neolithic implements, a small but perfect celt eleven miles south of 

 Nellore (1865) and half of a large ring-stone the drilling of which had 

 never been quite completed. This ring-stone, or perforated hammer, 

 was found in 1875 at Vemavaram ten miles north-east of Ongoze and 

 close to the northern boundary of Nellore district. 



Some four miles to the east of the famous Amravati tope, I found 

 (close to the village of Vayikunthipuram, on the south bank of the 

 Kistna) what appears to be the body of a good-sized celt minus the 

 cutting edge. This was made of a buff and purple mottled sandstone of 

 upper Gondwana age. The only other implement made of this material 

 was a cylindrical fragment (wanting both ends) of doubtful use, 

 unless it may be regarded as part of a prehistoric *' rolling pin." This 

 I got on the site of a Neolithic settlement at Jerlacherru, nine miles 

 W. of Kammamet in the Nizam's territory. The settlement had existed 

 between two granite-gneiss hills, and the fields all round were covered 

 thickly with fragments of high class antique pottery. I could only 

 mourn the hard fate which denied me time to examine this very pro- 

 mising locality. 



Another Neolithic site, on the left bank of the TJmni-ern, 3 or 4 

 miles south of Kammamet, that I could only pass by on the march 

 looked also very promising. 



I must not forget to mention that I picked up the pointed end of 

 a celt just south of the village of Matur, 27 miles south-east of 

 Kammamet. 



Proceeding westward, I came across a small hill of granite-gneiss 

 south of Poolloygooda (21 miles east by south of Bonagiri), where 

 were a considerable number of highly polished deep grooves worn into 

 the hard rock. From their shape and size, I inferred they must have been 



* Teri is the name given by the Tamil people exclusively to the great drifts of 

 red sand whicli form so remarkable a feature along the eastern side of Tiinnevelly 

 District. 



34 



