24 SMITHSONIAN M (SCKLLANEOUS COLLECTIONS VOL. 95 



The end scrapers arc more variable and nondescript in form than 

 the types just described. They constitute only 2.4 percent of the 

 series, which might be taken as an indication that they vi^ere not as 

 widely used as the other forms. Such was not necessarily the case, 

 however, as numerous implements included in the side scraper class 

 because their predominant features pertain to that form also have 

 an end scraper. So far as shape is concerned, this tool adheres to no 

 particular j^attern. Any random flake seems to have sufficed for such 

 an implement. Its main feature is a scraping edge at one or both 

 ends of the flake. The sides and lateral surfaces generally remain 

 untouched. The ends differ from those of the " snub-nosed " group 

 in that they are not thick and bulky, but are more chisellike in form. 

 They are either straight across, slightly convex, or have a sweeping 

 curve not unlike present-day blunt-end table knives. The size range 

 varies between an example with a length of 35 mm, a breadth of 

 18 mm, and a thickness of 5 mm, and one with a length of 48 mm, a 

 breadth of 22^ mm, and a thickness of 4 mm. 



Tlie " thumbnail " scrapers are not numerous in the collection. 

 Only 1.6 percent of the scrapers are listed under this classification. 

 They constitute a definite type, however. The name is derived from 

 the close resemblance between their shape and that of the ordinary 

 thumbnail. They are thin, roughly rectangular in outline with a con- 

 vex scraping edge. Fragments from channel flakes seem to have been 

 favored as material from which to make these implements. The 

 size range is not great. Examples in the collection are from 14 to 

 15 mm in width, 15 to 17 mm in length, and 2 to 3 mm in thickness. 



The broken series or scraper edges comprise 13.4 percent of the 

 scraper group. Most of the specimens are probably portions of side 

 scrapers, but as previously mentioned they are not sufficiently clearcut 

 in form to warrant more definite classification. 



A curious implement, the only one of its kind thus far found at the 

 site, is one which can be termed a core scraper (fig. 4). It was made 

 from a small core, not from a flake as were the majority of the tools. 

 The long, slender facets where chips were removed in the shaping 

 process show that the maker was possessed of great skill. Whether 

 the object was the ])r()duct of a bit of experimental work or belongs to 

 a definite, although minor, type is a question which can be answered 

 only by additional digging. If no other* examples are found in a com- 

 parable series of specimens, it unquestionably should be considered 

 unique. Core scrapers have been found in parts of Alaska and in 

 some sections of Siberia. This implement is not correlative to the 

 types from those places, however, and it may be that in the last 



