Ball — On Ancient Stone Implements in India. 401 



ternal causes on the -manufacturing capabilities of the people. At the 

 ■same time each of the regions is so vast and the mineralogical resources 

 are so varied, that the specialised characters of the implements appear 

 to be all the more remarkable, since the materials for greater diversity 

 are not wanting. In Burmah, however, according to Mr. Theobald, 

 the implements are often of schist or basalt, which are quite unlike 

 anything to be found in the areas where the implements occur. 



Referring again to the maps, we find that implements belonging to 

 Class A (the chipped quartzites) occur throughout a vast area of India 

 which extends in a north and south direction from Saugor to Madras, 

 and east and west from Raniganj in Bengal to Neemuch in Rajpo- 

 tana. 



This area overlaps the others to some extent, or it may be other- 

 wise stated has outliers within their limits, as in Chutia JSagpur, and 

 the Central Provinces, and Rajputana. 



In far distant Java, implements of somewhat similar character ap- 

 pear to have been met with. This is a fact of considerable interest, 

 pointing to a pre-historic connexion. 



The distribution of the flakes and cores which constitute Class B is 

 limited to the area which extends north and south from Kerowlie, in 

 Rajputana to Peyton, on the Godaveri in Bombay, and east and west 

 from Singhbhum in "Western Bengal to Sukkur on the Indus, in Sind, 

 and still further even to Gwadar in Beluchistan. 



The principal known outliers from this area are at Rajmahendry 

 on the lower Godaveri, and in the Andaman Islands. 



The polished celts, &c, whose varieties make up the sub-divisions 

 of Class C, occupy an area which extends from Upper Assam in the 

 north-east to Singhbhum in Bengal, and from the Irawadi Valley in 

 Burmah to Jabalpur in the Central Provinces. The doubtful case of a 

 polished celt from Coorg is the only example, so far as I know, of one 

 being found outside these boundaries.* 



Such being the rough limits of the three areas of distribution, it is 

 obvious that in Western Bengal and the Central Provinces, *'. e. in the 

 most central parts of the peninsula, there is considerable mutual over- 

 lap. It now only remains to make an attempt to offer some rational 

 explanation of this fact. Two theories have suggested themselves to 

 me. According to the first, we may regard these central tracts as in- 

 cluding a radiating point, from which successive waves of emigration 

 may, at different stages in the civilisation and progression of the 

 people, have spread, as the rising peninsula enlarged the area accessi- 

 ble. We cannot say with any degree of certainty whether the flake- 

 makers or the quartz-chippers were the more ancient. The former, 

 however, on this supposition, spread in directions to the west and 

 north-west, while the latter found their way southwards to Madras, 

 and even to Java in the south-east, where they met with the manu- 

 facturers of polished celts. These last, according to the same theory, 



* See Note added in the Press, p. 413. 



