194

R, C. BOSANQUET AND F. B. WELCH

the upper and lower part of it can hardly be nail-holes; the two faces are
carved alike, so it is not likely to have been affixed to another surface: the
lower part was not meant to be seen and may have been mortised into some
piece of furniture, c.g. the back of a chair. Found in G 3 :4 at a depth
of 1 90 (early period of the Third City).

A piain cylinder, XL. 22, very ill-preseryed, '029 long, '03 in diameter
perforated lengthwise.

§ 5.—Flint.

Numerous chips and pebbles of the white Mint which is found in abund-
ance along the north-east coast of Melos and still supplies the inhabitants
with strike-a-lights: XL. 13, 18, 19 ; an arrowhead of yellow, and a Hake of
amethyst-coloured flint.

Five ' sickle-teeth,' oblong flakes, with one serrated edge, formed out of
Hat pieces of flint to which the rough outer coating still adheres. XL. 11,12,
15—17. The teeth are polished with use. They were probably set side. by
sidc in a wooden sickle ; sec Petrie Illahim vii. 27, and Kahun ix. 22, for
discoveries of such sickles in Egypt. One of ours, XL. 17, has an incised
line above the teeth which might give a better hold to the resin used in
securing it. Similar sen-ated flints were found at Troy (Ilios, Fig. 665,
Dörpfeld, T.v.I. Fig. 362), and perhaps in Therasia (Perrot and Chipiez, vi.
p. 145).

XL. 17 differs from the others in having a semicircular back ; it may
have been hafted and used separately; cf. Ilios, Fig. 1342, and Dörpfeld,
T.u.l. Fig. 361 c. XL. 10 is a piece chipped into the same form but without
teeth.

These toothed flints have considerablc sawing power ; experiment shows
that it is possiblu to cut wood with them. Mr. Hogarth in B.S.A. iv. p. Ii,
describes them as from ' the surface earth '—consequently Mycenaean.

§ 6.—Marble.

Idols of Cycladic type.—(a) Seven more or less fiddle-shaped figures
rlat pieces of marble with rudimentary head and arms ; length '04 to '08; all
found in the eastern region where the excavation was carried below the
Mycenaean level. XXXIX. 5 is the rudest; though unsymmetrical it has
been polished with carc ; the neck is broken. XXXIX. 3 and 7 have a
waist; the Scratches on the latter represent folded arms (cf. some figures
from Paros in the Athens Museum, especially No. 4873); from Squares K 2,
depth not recorded and H 2, depth 1-50—2 80, respectively. XXXIX. 4
and 8, and another broken figure like them, form a distinet and uniform
group, perhaps of local manufacture; the type has not been found elsewhere ;
from H 2, depth unknown : H 3, depth 2-40: and G 3, depth 1 '60. There is