suppose that the angle of junction has been worn away, but in that case one would expect the two converging lines to thin off and become gradually weak as they approach, and this is not the case. Or, on the other hand, one might admit that the angle was never closed, but argue that this is only a piece of carelessness on the part of the engraver; which is not impossible, though hardly probable. But the real difficulty of Professor Sayce's interpretation lies in the left hand line of the supposed W. A first glance at the weight or cast does indeed show something which looks like a fragment of the upper part of the desired line. But on more careful examination under a powerful lens this fragment resolves itself into two elements (1) a clearly defined but very short cut, which has not the direction required for the left limb of a W, but rather runs parallel to the main or upright line of the ל; a splintered break proceeding from the lower part of the right hand edge of this cut, and trending downwards to the right. The distinction between the true cut and the break is perfectly clear to me in the original, but of course not so clear in the cast, which does not render the toolmarks quite sharply, and does not show at all the difference of surface between a saw cut and a splintered break. After having made out the composite character of this little stroke on the original, I persuade myself that with great care and strong magnifying power I can see even on the cast that the line is partly sawn, and partly due to splintering; but the study of the cast alone would hardly suggest this distinction, and so would leave it a very open question whether the whole stroke is cut (in which case it can hardly be anything else than a fragment of the fourth arm of a W) or the whole due to a superficial fracture (in which case a W is impossible). My own opinion as to the nature of the stroke is hardly more favourable to the reading W than the view that it is wholly due to a fracture; for it is the break alone which, by trending to the right as it descends, gives the line as a whole the appearance of running in the proper direction for the fourth limb of a shin.
I am afraid that these observations on the difficulties attending both the rival interpretations leave the matter more puzzling than ever; but there is one point not hitherto noticed on which I think that I can throw some light. I am convinced that the inscriptions on the two sides of the weight are not of the same date. The whole רבע של inscription (to name it after Professor Sayce's reading) is much more worn than the רבע נצג How can this have happened if the two inscriptions are contemporaneous? Not by weathering, one side being protected and the other not; for then there would be a difference in the surface texture of the two sides. But that is not the case, as can be seen by taking the points of the weight between the thumb and forefinger and gently rotating it, at the same time observing the reflections of the light that falls on the surface. The whole surface has been worn by similar agencies, producing a uniform texture and polish. At the same time, the weight has no tendency to roll over upon the more worn side; so that there is no
