188 BUREAU OF AMERICAN ETHNOLOGY [Bull. 177 
other direction, but also by the strength of the Mabaruma Phase in- 
fluence on the pottery of the Abary Phase. In addition to the copying 
of Mabaruma Phase types of decoration already mentioned, there is 
evidence of acculturation in vessel shape. The most striking example 
is the diagnostic Mabaruma Phase bow] with a broad horizontal to 
downsloping flange rim (figs. 45-2, 46-3) which appears as Form 3 
of the Abary Phase Tiger Island Plain (fig. 70-3). Although in- 
fluence is detectable in both decoration and vessel shape, there is no 
indication of any borrowing of Mabaruma Phase varieties of temper, 
suggesting that this may be more resistant to acculturation than 
other aspects of pottery making. 
When the attempt is made to retrace the course of the Abary Phase 
beyond the Northwest District, the connections become more tenuous. 
Projection of the trends in pottery type frequency (fig. 77) back- 
ward in time indicates that a complex ancestral to the Abary Phase 
should be characterized by a predominance of cariapé tempering com- 
bined with a less than 10 percent frequency of sherd-tempered pottery. 
If Mabaruma Phase influence is discounted, Abary Phase vessel shapes 
are reduced to simple bowls and globular jars with direct or exteriorly 
thickened rims. Only about 1 percent of the sherds are decorated and 
decoration is limited to incision, varying from broad lines to crudely 
applied scratches, and modeling in the form of small nubbins, usually 
located on the rim exterior and often embellished with gashes. 
An assemblage of traits almost identical to what would be expected 
in an ancestor of the Abary Phase occurs at the site of Mayo on 
Trinidad. Rouse and Goggin, who did the excavation, generously 
permitted us to examine the sherds and to make a preliminary anal- 
ysis. Five samples, representing several squares and levels, contained 
more than 90 sherds, and classification showed these to be composed 
of 97.6-100 percent cariapé-tempered sherds, 0-1 percent sherd-tem- 
pered sherds, and 0-1.5 percent decorated sherds. Bowl] shapes are 
similar to those of the Abary Phase, with direct or exteriorly thick- 
ened rims, but jars have short necks instead of insloping rims. Dec- 
oration is typically small nubbins embellished with a series of nicks. 
The Mayo and early Abary Phase ceramic complexes are so similar 
that it is difficult to avoid the conclusion that they are related. The 
question of relative chronology, however, presents a problem. Al- 
though we do not have any means of absolute dating, it can be safely 
concluded that the arrival of the Abary Phase in British Guiana took 
place in pre-European times, and probably at least as early as A.D. 
1000 to 1200. If the Mayo complex is ancestral, it should have a com- 
parable antiquity. Unfortunately for our theory, the Mayo site has 
been tentatively identified by Rouse with the historic mission of Mon- 
serrate, which existed between A.D. 1687 and 1789 (Rouse, personal 
