Here are to be found the survival of the ninth century form shown in Fig. 17, and a remarkable advance in the form of the one at the bottom, which is doubtless the pattern intended to be shown in the sculptured bow, second from the top. The top one is merely given as an example of the perfunctory work the historian has to examine and yet retain his customary calm exterior. Fig. 19 gives some examples of twelfth century bows as depicted by the artists of that period. The first two are evidently intended to represent the type shown in Fig. 17. The sculptor probably found the straight line of the hair inelegant. The third (which is from a MS. in the Bodleian Library) and last show a return to the ninth century form in Fig. 16. [Illustration: FIG. 19.] This is a form that is found so continually through all the centuries, down to the seventeenth and eighteenth, that I am inclined to the belief that it is fairly accurate. It is very much like the outline of the modern double bass bow. In Fig. 20 are given some thirteenth century bows: the one with the curious sword-hilt is remarkable. In the others we find a return to more primitive lines. [Illustration: FIG. 20.] The fourteenth century bows have very little to distinguish them from those of preceding ages, and I give the most noticeable examples I have found in Fig. 21. The second is a very advanced type. Against these must be set those in Fig. 22. [Illustration: FIG. 21.] [Illustration: FIG. 22.] These appear to me as being most probably conventional representations, or copied from older works as suggested above. Of fifteenth century bows, the pictorial and plastic arts record those shown in Fig. 23, together with the usual atavism or return to earlier types. [Illustration: FIG. 23.] This atavism, if credible, is most marked in the sixteenth century as witness those in Fig. 24. [Illustration: FIG. 24.] Here are bows that take us back to before the Norman Conquest, drawn by artists who were contemporary with Gasparo da Salo and Andreas Amati. It is quite out of the question to suppose that such bows were used at that time. The drawings of seventeenth century bows are more convincing. We then get a more definite idea of the nut, which was in most cases a fixture. Also, the head begins to mould itself into something approaching the form of the modern "hatchet." Although there are cases of bows in drawings as far back as the eleventh century (see Fig. 18, etc.) showing great advances, it is not until reaching the seventeenth century, that one can say with any degree of confidence that the perfect bow is on the horizon. CHAPTER IV. THE BEGINNINGS OF THE MODERN BOW--ORNAMENTATION--A POSSIBLE STRADIVARI BOW--THE MOVABLE NUT--THE CREMAILLERE--THE SCREW NUT.