B d 
Myangan ARCHEOLOGY IN BRITISH GUIANA 181 
reads: 1,156 sherds per 1 by 1 meter equals 100 years. The values ob- 
tained for the Abary Phase sites are given in table E. 
TaBLE E.—Duration of Abary Phase village sites estimated from rate of refuse 
accumulation 
Site area Size of Total Adjusted | Duration 
Site and cut in square cut in sherds total per in years 
meters meters per cut 1 by 1m. 
Toth ae Ste ae ee oe a er EO} SOO Ri aoe ee Ie ee eI ee OU a 
LOATH Lnctee 2 etal os teleost pet ete yal pe eg a ea by 1 3, 276 1, 092 94.5 
OTe ee ee ee eee ee by 1 2, 463 2, 463 213.5 
RENEE ELS SOIR SAIS TOD TRY |e acer lag ae ol late Me A ee i A OR Tae 
yr a ee ee ee eh bee 1 1,159 579 50.1 
ees Samia wancne cease ontatnecenacme ass 2, 480" | See ean ets! |saetcccns see taseecaeeee beneeseeeeas 
(Ch ae ee ee oe eee Lee See by 2 66 14,4 
RC Gee ee re ee ae eee one 2 by 2 7,108 1,777 153.7 
The durations derived from density of sherds in the refuse raise 
several interesting points. One is the contrast between the almost 
equal depth of the refuse deposits at the three sites and the great 
difference in the number of sherds contained in this depth, which leads 
to estimated durations varying from 14.4 to 213.5 years. Even within 
a single site, for example B-1, the variation in sherd density may 
exceed 100 percent, indicating a considerable difference in the amount 
of dirt that may be deposited with the sherds. The even greater 
discrepancy between the two values for B-3 can be explained by the 
fact that the conditions involved in the deposition of the refuse in 
the area represented by Cut 1 were untypical. The small sherd 
sample showed a pottery type distribution that could not be seriated 
into the Abary Phase sequence, implying nonrandom deposition of 
refuse. Since the cut is unreliable in this respect, it cannot be con- 
sidered reliable for other purposes. With the elimination of this 
figure, the estimated durations range from 50.1 to 213.5 years. All of 
these are unusually long in comparison to the village durations typical 
of ethnographically recorded Tropical Forest tribes, but such long 
durations are in keeping with the pottery type seriation. This shows a 
very great amount of ceramic change, with an almost complete altera- 
tion in the plain wares, and must have required a relatively long period 
of time for its realization. 
The total duration for the Abary Phase can be calculated by adding 
the durations of the individual sites, since the pottery type seriation 
shows them to have been occupied sequentially (fig. 77). For this 
purpose, only one duration for each site can be used. In the case of 
B-3, the choice of Cut 2 is dictated by the reasons mentioned above. 
Determination of the best value for B-1 is more difficult. The seriation 
shows that the two cuts cover the same time span, and there is no evi- 
dence to indicate that the difference between the durations reflects 
differential length of occupation of the two parts of the site. Since 
