THE "RED-HILLS" OF CANVEY ISLAND. 61
It is of interest to note that here, again, the average apparent
density figure in each group tends to approximate to a distinctive
figure. A closer agreement with an ideal type could hardly
be expected, in view of the fact that the pieces may well have
come from widely separated centres of manufacture, though
they may be the products of a technical process similar in its
intention.
It was hoped that closer agreement with an ideal type
might be found to exist in the case of pottery manufactured
from raw material dug on the same site and fired in the same
kiln. Through the kindness of Mr. Hull, Curator of the Museum
at Colchester, a number of Samian sherds were obtained from
a kiln at Sheepen, near Colchester. The sherds included parts
of rims, bases and odd pieces, derived from different pots and
dishes, but all fired—so far as could be seen—in the same kiln.
The pieces varied in character, and were roughly classified by
Mr. Hull as follows:—
(a) "Overtired"—8 pieces.
(b) "Defective paste"—8 pieces.
(c) "Normal Colchester Samian"—8 pieces.
The Average Apparent Density for the 24 pieces was 1.81,
and the corresponding Per Centage Figure 62, results which
suggest that the classification adopted by Mr. Hull needs
substantial modification.
Careful examination of the sherds, taking into account their
hardness, colour and the condition of the glaze, lead me to
classify the pieces provisionally in five groups:—
Group A. One piece, soft, uniform red, slightly fissured,
unglazed.
Q. Made from different materials and underfired.
App. Dens. 1.47
Group B. Nine pieces, soft, red, paler towards centre, glaze
defective or absent.
Q. Underfired. App. Dens. 1.68
% Figure. 90
Group C. Five pieces, hard, uniform red, glaze intact.
Normal Colchester Samian Ware.
App. Dens. 1.80
% Figure 100