164 BULLETIN OF THE ESSEX INSTITUTE. 



3. As a Caltrop or Tribulus. The spines are too short 

 and blunt in many of them and the long-spined ones would 

 not remain in position ; they would show no signs of wear ; 

 furthermore the Tribulus is known and has no resemblance 

 to this object. 



4. As a Screw-Driver. The idea is unthinkable. 



5. As a Spear-Thrower. The varying character of 

 the spines and signs of wear are against the idea ; further- 

 more the amentum used by the ancients for spear-throwing 

 is well known as a leathern strap attached to the middle 

 of the spear. 



6. As a Snaffle or Curb. Again the Variation in the 

 length of the spines, and the fact that in no case has any 

 device of this nature been represented on a horse's head 

 in ancient sculpture, are sufficient to disprove the idea. 



7. As a Bit inside the mouth. The jointed bit was 

 co-existent with it, and the ponderous character of some 

 of the bow-pullers, and the lightness of others, would mil- 

 itate against this conception of its use. 



8. As a Lamp Wick Holder. The signs of great wear 

 and its manner of breakage renders this supposition val- 

 ueless. 



9. As an object to prevent a load from slipping. The 

 small tubercles which take the place of long spines in 

 some of them would render it useless for that purpose. 



10. As a Curb to hold in the hand for grasping reins 

 or anything eise. The great length of the spines in some 

 specimens would preclude its use in that way. 



EXPLAN ATION OF PLATES. 



PL ATE I. 



Natural size. 



Figs. 1, 2. Bow-puller. Plain type, front and top view. Collection, 

 author. 



