ON THE SO-CALLED BOW-PULLER OF ANTIQUITY. 147 



The objecls are sometimes found broken and the manner 

 of breaking is of importance. The outer portion of one or 

 both rings is broken away (Fig. 7, alsoPlate IV, Figs. 5, 

 6, 7), indicating the direction of the strain to have been 

 outward, and probably dovvnward, as if the strap or cord 

 was used in binding the object to some body, and the 

 method of wear would sustain this proposition. The 

 spines also in rare cases are found broken and it is usu- 

 ally the outer spine which is mntilated in this way. The 

 tip of the front spine is in 

 some cases bent inward. 

 In no case have I seen a 

 specimen with the spines 

 bent outward. The object, 

 whether heavy or light, has 

 been designed for strength. 



The bow-puller shown in 

 Fig. 5, Plate II, has an 

 openingin one of the rings ; 

 this is probably the result 

 of an imperfection in Cast- 

 ing and not intentional. 

 The same may also be true 

 of a round protuberance on 

 the front of one of the 

 rings of a bow-puller (Fig. 

 6, Plate II). The curious groove seen in the inner edge of 

 the outer part of the ring in Fig. 3, Plate II, is unique so 

 far as I know. 



In an object varying so much in weight, number and 

 length of the spines, ornamentation, or absence of it, the 

 features which seem to have an importance in considering 

 their probable use are first and foremost the two rings 

 springing from the solid body and their usual upward 



Fig. 7. 



