312 BUREAU OF AMERICAN ETHNOLOGY [Bull. 191 



crudely made saucer-shaped miniature vessel. As nearly as we could 

 tell, there were 23 pots in all. 



The entire collection filled six large potato sacks and it required 

 three round trips for our two men to transport them to the boat. 

 Two oUas were too large for the sacks and were carried separately. 



Two of the ollas had traces of red paint and one small vessel had 

 a simple geometric pattern in black painted on the interior. The 

 pottery varies in color from light buff to brick red, and is very hard. 

 Considering the size of the majority of the vessels, the tempering 

 is not coarse. The wall thickness of the large ollas generally varies 

 from 5 to 13 mm., but the bases on some are 26 mm. thick. Approx- 

 imately one-half are decorated with rather haphazard combing, 

 probably done with a scallop shell. The remainder are smooth and 

 plain. The typical shape is that of a round-bottomed rather squat 

 olla with outfiaring rim. One large and one small vessel had a bevel 

 around the widest part of the body. The small pots were made of 

 the same type ware and were similar in shape excepting for the two 

 saucer-shaped miniatures placed inside small vessels as described 

 above. 



The large ollas varied in height from 30 to 40 cm. The largest 

 orifice measured 33 cm. across (pis. 55, 56). 



It is evident that the pottery deposits were burial offerings, un- 

 doubtedly for bodies brought from Taboga. Bones apparently were 

 not placed in the urns, which presumably contained food. It is 

 probable that Uraba was uninhabited in prehistoric times but may 

 have supported limited cultivation by the people of Taboga as it 

 does today. 



TABOGUILLA ISLAND 



On Taboguilla Island we excavated three sites. Two of these are 

 midden deposits on the west side of the island. Taboguilla-1 is 

 located at an elevation of about 115 meters above the sand beach 

 where, at the time of our visit, a few unoccupied shacks were located. 

 Taboguilla-2 is at an elevation of about 100 meters almost directly 

 below Taboguilla-1. Since the two deposits seemed to contain the 

 same type of material, we excavated a test trench in TaboguiUa-2 

 to check on this fact, and concentrated our efforts on Taboguilla-1, 

 which was considerably larger and more productive. The refuse 

 of TaboguHla-l , consisting of black earth mixed with shell, was 

 somewhat more than 1 meter in depth. Potsherds were very abundant 

 and in a good state of preservation. We conducted the excavations 

 stratigraphically, but our statistics showed no change in the type or 

 proportions of the ceramics. We concluded that the midden repre- 

 sented a single occupation over a not very considerable length of time. 



Taboguilla-3 is a rock shelter about 200 meters south of TaboguiUa-l. 



