^^HoS^^^'S CHARACTER OF THE REJECTS. 17 



the well known types of European paleolithic implements. Such a 

 piece is shown in a, plate vi. It represents an abortive attempt at 

 blade-making, the heavy end having been much battered in attempts 

 to reduce the thickness. This specimen was so discolored by lichens 

 that a good photograph could not be made. An additional example of 

 rejection from too great thickness is given in b, on the same plate. In 

 this case a pronounced hump has developed on one side, and repeated 

 blows on the edge of the specimen employed to remove the excrescence 

 have only tended to increase the difficulty. Plate vii is intended to 

 show still more fully this important class of rejects. One variety recur- 

 ring somewhat rarely is shown in plate viii. The sides are obscurely 

 notched, giving an ax-like outline, but it can not be determined whether 

 or not this was intentional, representing the beginning of some peculiar 

 specialization,,or whether it is the result of repeated attempts to reduce 

 the great thickness of the middle part of the specimen by strokes first 

 on one edge and then on the other. These specimens are nearly all 

 highly convex on both sides. 



Besides the varieties of shop refuse, rejects and broken incipient 

 implements, referred to and illustrated above, there are among the 

 rejectage many interesting fortuitous shapes — shapes produced in shap- 

 ing implements but not themselves the subject of the shaping opera- 

 tions. There are flakes and fragments in great diversity of shape. 

 Fracture is often eccentric and unique forms are produced, some of 

 which are so suggestive as to lead the operator to the fashioning of 

 new and unheard of forms. The long slender flakes are often excellent 

 knife blades, and many must have been utilized in the arts without 

 modification. Others are slender and dagger-like, making effective 

 perforators or piercing tools or weapons. 



In shaping the quarry blades the most marked tendency toward 

 abortive fracture is in the direction of too great thickness. Fracture 

 by blows delivered on or near the edge of the specimen does not carry 

 across the face of the specimen, but rises quickly, resulting in high 

 backs or peaks with facets recalling those of a turtle's back. Common 

 forms have already been presented. Very often these forms are pro- 

 nounced pyramids, as shown in a, plate ix. Eccentric shapes occur, such 

 as that shown in c, where a curved spawl has been worked on one side 

 only with the view of reducing the convexity. These shapes grade 

 imperceptibly into other conical or pyramidal forms, which are cores 

 resulting from the removal of flakes for some unknown use — perhaps 

 as knives — or to be carried away for the manufacture of small arrow 

 points, scrapers, and the like. They resemble the well known cores of 

 obsidian, so common in Mexico, from which thin blade-like flakes were 

 removed for knives and razors. Two specimens of these cores are 

 shown in plate x. It is very hard to draw the line between such cores 

 and the high-backed failures previously mentioned, and we class them 

 as cores only because it seems unlikely that the flaking could have 

 BULL. u=21 2 



