222 
tidal estuary similar to that of the Humber. In the maps 
of the Geological Survey it is marked as " ancient warp." 
The very absence of fossil shells is in favour of this view, 
for moUusca appear to be entirely absent from the tidal 
Ouse ; at any rate, I have never succeeded in finding any. 
4. Upon the clay there usually rests a bed of Sand, 
which indeed forms the surface layer over the greater part 
of the area of which I am speaking. Its usual thickness 
is about four feet, sometimes more, but occasionally, as in parts 
of Goole, it is absent, and then the peat rests directly on the 
clay. The base of this sand is sometimes gravelly, but it 
increases in fineness towards the top. Its texture is very 
loose ; when saturated with water, it is a " quicksand," and 
when dry and newly turned up with the plough, it some- 
times drifts in a strong wind almost like snow, so that 
farmers call it " blow-away sand.'' Indeed, from its appear- 
ance, and the resemblance of some of the mounds formed of 
it, as at Riccall and Holme-on-Spalding-Moor, to the dunes 
on the sea coast, I cannot help surmising that the wind may 
have been at least one of the agents concerned in its dis- 
tribution. The colour is generally yellow or ferruginous, 
except when covered with peat, when it is often white. I 
should be inclined to explain this by supposing that the 
decomposing carbonaceous matter reduces the iron which 
gives the sand its colour, from the ferric to the ferrous state, 
and that the excess of carbonic acid formed partly dissolves 
out the ferrous carbonate. The water which drains from 
the sand under the peat on exposure to the air throws 
down a rusty precipitate, which accumulates largely in 
some of the marsh ditches, especially in the district 
drained by the River Foulness, near Holme- on-Spalding- 
Moor. I am informed on good authority, although I have 
not seen them myself, that there are near this village 
the remains of ancient iron-smelting furnaces ; if so, this 
