Fig. 6 shows a tranverse section taken at the sudden bend or "elbow," which indicates the commencement of the "ear." In this region there is a prominent ridge which gradually rises and shades off into the "ear." In the section it is seen that the cane is replaced by hard wood (a2) with a triangular cross section which produces the shape of the ridge. Over the "belly" side lies the horn, very thin at this point; it terminates a little way beyond this point. On the back are seen the continuations of the two sinew layers cl, c2. The external bark layers are the same as before.

In Fig. 7 is seen a transverse section through the centre of the handle or "grip." Here the centre is composed of both cane and hardwood; the cane a1 is in direct continuation from the "arms"; the hardwood serves to pad out the handle in order to fit the grip comfortably. The horn, b, is very convex here, and this section cuts through the point of meeting of the two horn strips, which together cover the belly as far as the "elbows"; so that here the end of one of the pieces is represented. The longitudinal sinews are disposed as before. The bark does not extend over the handle, but, as mentioned above, in its place there are coarse sinews, g, wound transversely round in a slightly spiral manner, the ends of which are seen cut across in the section. In the more elaborate specimens there is a layer of shark skin, covering the grip, with thin cork overlying the whole, and affording a good hand-hold.

Fig. 8 is taken from a longitudinal section through the whole of the grip, and shows on a reduced scale the extent and form of the plug of hardwood, a2, and how it ekes out the shape of the hand-hold; the meeting of the two horn strips, b, b, is also seen.

The specimen from which the above description is taken is by no means a fine specimen of its kind, but may be taken as fairly typical of the "Tatar" variety, as the different examples seem to vary more in external finish than in internal structure.

The figures of the complete Persian bow (Plate VI, Fig. 9), and the anatomy of another specimen (Figs. 10-16) are taken from specimens sent to the Oxford Museum, by Colonel Sir R. Murdoch Smith. The two specimens are exactly similar, so that the description of the structure of the one may be taken as applying to that of the other, which is figured entire.

These specimens are estimated by Colonel Murdoch Smith to be certainly 200 years old, and are very good examples of the highest type of composite bow. It is highly improbable that this weapon will ever improve, with the increasing use of firearms in Asia, and we are justified in regarding this as the culminating point in the series.

In shape this bow (Fig. 9) differs from the "Tatar" bow; the unstrung curve is more regular and resembles that of the Scythian bow as generally described, and the "ears," which are relatively much shorter, continue in the same curve with the "arms"; they are moreover not bare, but overlaid with sinew as far as the nocks. The "arms " also, as compared with the "Tatar" bow, are proportionately flatter, wider at the centre, and more tapered towards the "ears" and "grip"; and they are further more markedly plano-convex in section. The specimen figured does not exhibit the recurving in the unstrung state, to the extent of many examples, in some of which the tips actually cross one another.