ST. THOMAS AND ST. JOHN them could almost as well have been classified as Hull Plain. The Botany Wide Handled sherd, also extremely sandy in temper, ex- hibited the raised dot, typical of this type. Occupation started late in the Magens I period and continued, undoubtedly intermit- tently, into the Magens IIB period. The last two sites on the north coast, those at Cinnamon and Francis Bays (Fig. 2, 7-8) have been discussed in the previous sec- tion. East and South Shore Sites The next site, at Haulover Bay (Fig. 2, 9) was the only place where we were able to find any Indian pottery along the long eastern peninsula of St. John Island. Here, at a gap in the moun- tainous ridge, was a path leading from that part of Coral Bay known as Round Bay to Haulover Bay on the Atlantic Ocean. Along this trail there was no evidence of Indian occupation but troweling in the ground disclosed two Botany Plain sherds. Apparently, it was visited, at least once, during Indian times. The large site near Coral Bay (Fig. 2, 10) has been dlic u-d. Except at Yawzi Point (Fig 2, 11), we were unable to locate any sites along the southeastern peninsula. The Yawzi Point col- lection includes twenty-two Botany Plain, one Botany Griddle, one Botany miscellaneous incised, one Bordeaux Plain, nineteen Hull Plain, and three incised sherds of Hull paste (P1. XIV, a-b). Dating of this collection in terms of the Magens sequence is difficult. The incised sherds may be from casuela vessels and so suggest a rela- tively late date of occupation but the relatively large amount of Hull Series sherds would argue otherwise. Perhaps, Indians visited Yawzi Point during two different time periods. Hatt (1924: 30) mentions some sherds but not a midden neai Reef Bay. We found thirteen Botany and Hull Plain sherds there (Fig. 2, 12) but no midden. Aboriginal visits to Reef Bay in an early Magens II period seems likely. Two small sites were located at Fish Bay (Fig. 2, 14-15). One produced eight Botany Plain, the other one Hull and five Botany Plain sherds. Temporal placement is extremely dubious but use in post-Magens I times may be suggested.