76 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 92 



and water was extremely unpleasant, and the omnipresent flies and 

 mosquitoes did not help. An amusing though at times painful aspect 

 of the work was furnished by the numerous land crabs lurking in the 

 muck, which felt like highly desirable artifacts but pinched like 

 demons. 



The rock shelter (cave 2) was much smaller than the above and 

 consisted of a niche or ledge 24 feet long and 6 to 7 feet above the 

 level of the swamp (fig. 16). The ledge tapered in width toward 

 both ends, being about 4 feet wide in the center. The roof was formed 

 by the overhanging cliff, and the floor was concave with a niche or 

 shelf some 3 feet higher at the south end. In the concavity was 

 about a foot of brown dust and scaled-off fragments from the walls 

 containing potsherds and a few artifacts. Sam had previously obtained 

 here two or three complete pots, which he had sold. 



Particularly striking was a stalagmitic formation, 5 feet in height, 

 which stood in the center and on the outer edge of the main shelf 

 (fig. 16). This definitely suggested a life-size human form having 

 a rounded head and blocklike body. It was natural in origin, although 

 the smooth, almost polished condition of the head may have been the 

 result of human activity. Mounted well above the level of the swamp 

 against such a striking natural background, the object at once sug- 

 gested some sort of a fetish. Owing to the density of the mangroves 

 at this place, some of our photographs did not come out; however, 

 the field sketch gives a good idea of the object and its situation 

 (fig. 16). Aside from the few artifacts, there were no other signs of 

 human occupation, nor were the walls of the shelter smoke-stained. 



We searched along the cliffs for a considerable distance in both 

 directions and climbed to the top in several places without discovering 

 any further human evidence. Several other clefts or small shelters, 

 as well as the foot of the cliff, were carefully examined with similar 

 negative results. On the basis of our examination and excavation, 

 which occupied considerable parts of two days, the site consisted of 

 only the one habitation cave and the rock shelter, or more probably 

 shrine, nearby. The following artifacts were recovered or noted at 

 this place. 



From cave i, or rather from the muck at the foot of its steep 

 approach, one complete pot and a large number of sherds were 

 recovered. This pottery was covered with oily black muck and in 

 many cases by a lime incrustation. All sherds were taken to the lagoon 

 and washed, and an attempt was made to fit as many pieces together 



