8 THE ANCIENT ROADS AND THE SITE 



Tliey are identical with the querns found in Danebury, 

 near Northampton, and in the Lake Village of Glaston- 

 bury, both of which belong to the Prehistoric Iron 

 Age. They differ from those introduced by the Romans 

 in the fact that the latter are thinner and wider, and disc- 

 shaped, with grinding surfaces frequently grooved, as 

 may be seen from the group (Fig. 2) of six portions of 

 Roman querns from the mill-house in Melandra. These are, 

 with one exception, of Millstone Grit, and were probably 

 made in the district. The exception (the lowest in the 

 figure) is of volcanic rock, and came from the Roman 

 quern factory of Andernach, near Coblentz, from which 

 querns were sent almost over the whole of Roman 

 Europe.^ A fragment of another quern of the same 

 material has also been found. The bee-hive querns are 

 frequently met with on the moors of Yorkshire, and, so 

 far as my experience goes, are not found in association 

 with Roman remains. Whether or no they were used in 

 Roman times is an open question. If they were used 

 they are merely a survival from the Prehistoric Iron Age 

 — like the greater portion of the roads guarded by 

 Melandra. 



In conclusion, we may very well ask why should the 

 roads from Melandra westwards point towards Stockport 

 and Manchester. The answer is to be found in the fact 

 that both these places, as pointed out by Mr. Henry 

 Taylor and Mr. Roeder, were inhabited centres in pre- 

 Roman as well as in later times. Both grew round the 

 fortified rocks which commanded, the one the marshes of 

 the Mersey, and the other the junction of the Irk with 



the Irwell. 



W. Boyd Dawkins. 



9. I have identified these querns in Hod Camp, near Blandford, in 

 Roman Chester, and in Caerwent. 



