On Deneholes. 
57 
usually came along the one great highway, the Thames. 
But the banks of the Thames would, on the other hand, be a 
favoured resort of the native population, attracted both by 
the abundant supply of fish and the means of intercommuni¬ 
cation. And a glance at a geological map will show that 
certain spots only, abutting on the river, make suitable sites 
for villages or towns. South of the Thames and east of 
London we have Greenwich, Woolwich, Erith, Greenhithe, 
and Gravesend, all places offering sites close to the river, 
and yet more or less above the level of the marshes. On the 
Essex side a broad spread of marsh-land borders the Thames 
between the outfall of the River Lea and Purfleet. At 
Purfleet and Grays, however, higher ground abuts on the 
stream, and approaches it more nearly than usual at East 
Tilbury and the neighbourhood of Mucking. Now, as re¬ 
gards Essex, Mr. Spurrell remarks that Deneholes are 
abundant between East Tilbury and Purfleet: and all the 
Essex examples mentioned by him are in the district between 
those places. 
In Kent also we find the Deneholes especially abundant 
near the old settlements, or sites for settlements, on the 
river, at Greenwich, Woolwich, Erith, and Greenhithe. And 
in all these cases, both in Kent and Essex, the position of 
the Deneholes, one, two, or three miles from the river, and 
their concentration in spots about the same distance from 
the natural sites for settlements on the Thames, seem to 
suggest that they were used both as storehouses and as 
places of occasional refuge from pirates who might attack the 
villages on the river-bank. This seems to me the only 
natural presumption in accordance with, and favoured by, 
the evidence afforded by the plan and situation of the Dene¬ 
holes themselves. And when, in addition, we review the 
store of facts collected by Mr. Spurrell, showing the past and 
present prevalence ^ of underground granaries and dwellings 
in all parts of the world, civilized and uncivilized, the fact 
dawns upon us that caves for these purposes have accom¬ 
panied a certain stage of social progress, almost everywhere, 
while those subserving other ends have been local and 
