32 BULLETIN UNITED STATES GEOLOGICAL SURVEY. 



again resorted to the house-sites, and especially to those filled up by 

 human hands, which was proven to be a fact by finding human skeletons 

 interred at the bottom of the excavation. The corpses were found with- 

 out exception in the subterranean part of the ruined houses, which were 

 here like those at Eogue Elver in size and wooden linings, but with- 

 out the draft-passages for the smoke to escape. Doubled up, the skele- 

 tons were resting near the wall of the excavation, and faced the fire- 

 place, as indicated in sketches (H and I) [Plate 4], which part was the 

 most deeply covered with earth, whereby the remaining surface-inden- 

 tation of such a house-site was easily discernible by an enlarged em- 

 bankment, in contrast to those which were not shaiped through a burial 

 but had adopted the form of an inverted mole-hill by the natural action 

 of filling-up, caused by time and the elements. In one instance, two 

 skeletons were found buried in one house, where a re-opening seemed to 

 be evident by the flattened and unusually enlarged covering earthwork. 

 Such a singular indentation in which a burial was made will be better 

 understood by comparing the section diagram (H) [Plate 4] with that of 

 a common formation (B) [Plate 4], in which latter no burial had occurred. 

 The earth covering the skeletons was strongly mixed with charcoal, 

 pieces of charred wood, fragments of animal bones, and shells black- 

 ened and partially consumed by fire. On the floor on which the skele- 

 tons rested was found a layer of ashes of several inches in thickness. 

 But the fire had not affected the skeletons, as in no instance was any 

 such damage observed, and even the remains of matting, furs, and 

 other similar perishable material were not injured by it. It seems^ 

 therefore, evident that the hut was demolished by fire, after the owner 

 had expired, and was buried in the ruins, covered with rubbish and 

 €arth surrounding his house. Except some glass beads found with a 

 female skull, and three roughly-cast copper buttons with that of a male, 

 nothing was unearthed that had apparently been deposited with the 

 dead. Of course, in the mass of debris we worked over, divers articles 

 were found, but not in such a position as to indicate an intended de- 

 posit of property of the dead in accordance with a religious or super- 

 stitious rite. 



We find another large shell-mound located on loose sand about four 

 hundred yards northward from the main settlement, where all the char- 

 acteristic indications of a permanent settlement are noticed, excepting 

 the house-sites, which likely had become filled up and obliterated by the 

 sand drifts to which this place is exposed, as well as by the heavy rains 

 during the winter. A stream of water passes the base of the dune, but 

 disappears in the sandy beach. Back of the shell-mound, the ground 

 rises gradually for a distance before it reaches the foot of a steep ridge 

 extending back from the shore, and defining the lower boundary of an 

 almost impenetrable country by its rough topography, its forests, and 

 dense growth of underwood, the safe home of all kinds of game, panther, 

 and bear. A few hundred yards up the coast from the shell-mound, near 



