ST. THOMAS AND ST. JOHN


shell tips, part of a bone awl, and four pieces of worked bone. The
vertical distribution of these (P1. VIII) does not suggest any changes
in types of tools during aboriginal occupation of Turtle Point.
Similarly, there were no changes in griddle rims which correlated
with depths in the midden.
 To judge from our test, the midden at the Turtle Point site
was relatively rich. A short but intense occupancy is indicated.

 Test at Cinnamon Bay

 Hatt (1924) mentions the site at Cinnamon Bay (Fig. 2) and
two groups of specimens from there are cataloged in the Danish
National Museum but apparently he did not excavate at this site.
 Our test disclosed a deposit of sand and shells plus artifacts
which was thicker than any we found elsewhere although it was not
as rich in specimens per cubic meter. Specimens are illustrated in
Plates IX-X and the vertical distribution of pottery given in Table
3. Artifacts not listed include part of a flat, ground stone ornament
or tool (P1. X, J) from Level 5, four clam shell scrapers, and four
clay spindle whorls (P1. IX, a, o-p). The last were all in the
highest two levels.
 Of interest in Table 3 are the relatively high provenience of
Botany Incised Casuelas and miscellaneous incised, Bordeaux In-
cised Casuelas, Bay Lip Punctated, and Botany Adorned vessels.
As at Magens Bay, sherds of subtype 5, Botany Adorned, were a little
lower than those of subtypes 1 and 3. These data strengthen those
presented under Turtle Point to indicate that incised casuelas are,
on the average, relatively late. As at Magens Bay, Botany Plain con-
cave rim bowls (Fig. 3) were concentrated in lower levels. Also,
as at Magens Bay, there was no significant difference with depth in
rim variations of griddles.
 The great depth of much of the Coral Plain and of the Harbor
 Plain and micaceous plain sherds should be noted. The amount
 and distribution of sherds of the Coral Series do not prove occupa-
 tion during a Magens I period in the tested area although deposits
 of that period probably are present at the Cinnamon Bay site.
 Levels penetrated by Sleight's test pertain to Magens IIB and IIC
 times.