^{page 22 note 1} Svenska Fornm.-fören:s Tidskr. x, 80.

^{page 22 note 2} Ant. Tidskrift f. Sverige, xiv, no. 2.

^{page 22 note 3} Arch. Cantiana, x, 309–10, fig. on p. 310.

^{page 22 note 4} Ibid., xiii, 553.

^{page 22 note 5} Ibid.

^{page 23 note 1} Ibid. v, 310 ff., pl. 1.

^{page 23 note 2} Ibid. vi, 173.

^{page 23 note 3} Burlington Fine Arts Club, The Art of the Dark Ages in Europe, 86, P. 31, pl. xvii.

^{page 23 note 4} Arch. Cantiana, xli, 120–1; E. T. Leeds, Early Anglo-Saxon Art and Archaeology, 50, pl. xiv.

^{page 23 note 5} Robert Plot, Natural History of Oxfordshire (1705), 359 with figure; Archaeologia, LXI, 491, fig. 6; Annaler for Nordisk Oldkyndighed, 1855, p. 291, no. 32 and Atlas 32, where it is figured in reverse.

^{page 23 note 6} B.M. Guide (Anglo-Saxon), fig. 101; Åberg, Anglo-Saxons in England, fig. 87.

^{page 23 note 7} Archaeologia, lxii, 488, fig. 3.

^{page 24 note 1} Compare Atlas for Nordisk Oldkyndighed, 20, from an unknown site, probably S. Germany, imitations of coins of Mauricius Tiberius, as on the pendant (pl. 1, 4).

^{page 24 note 2} Zeitschrift des historischen Vereins für Niedersachsen, 1860, p. 391 seq., Taf. 1, the pertinent portion of which is here reproduced as pl. VII. The eleven bracteates were distributed among several persons and museums, and two of them must have found their way into the hands of a Paris dealer, from whom they were acquired in 1866 by Sir John Evans, their source being given as Scania (i.e. Skåne) or Denmark. In 1908 they were presented to the Ashmolean Museum (A.M. 1909, 824 and 825) by Sir Arthur Evans. Reference to the original publication in the only easily accessible copy in the University Library, Cambridge (for a transcript of which I am greatly indebted to Dr. F. Heichelheim), places the origin of the two bracteates beyond question. They (figured on pl. vii with black backgrounds) are identical with those figured as nos. 3 and 4 on the original plate.

^{page 25 note 1} A. = Atlas for Nordisk Oldkyndighed.

^{page 25 note 2} von Jenny, Germanischer Schmuck, 50, Taf. 49.

^{page 25 note 3} This and similar abbreviations refer to O. Montelius, Från Jernaldern, Pts. I, II, and III (1869).

^{page 27 note 1} It is noteworthy that the frilled head which appears on the specimen from Sarre, grave 90, is also to be seen on many D bracteates from Denmark, among them examples from the Skovborg find (see remarks on p. 26), and again on a scabbard-chape from Sjöröd, about the date of which Åberg expresses himself in most decided manner (Den Nordiska Folkvandringstidens Kronologi, 17, fig. 45).

^{page 27 note 2} P. C. J. A. Boeles, Friesland tot den eelfde Eeuw, 179, pl. xxxix.

^{page 27 note 3} Ibid., no. 4.

^{page 27 note 4} F. Sehested, Fortidsminder og Oldsager fra Egnen om Broholm, 199, pls. XLI–XLIV.

^{page 28 note 1} Stephens, Handbook of Old Northern Runic Monuments, pp. 168–9, nos. 7–9.

^{page 28 note 2} O. R. Janse, Le travail de l'or en Suède à l'époque mérovingienne, 83, 88, 104, and 110.

^{page 30 note 1} The slightly variant opinions about dating are usefully scheduled by Nils Åberg in Den Folkvandringstidens Kronologi.

^{page 30 note 2} Aarbøger, 1851, pp. 127 ff.

^{page 32 note 1} Op. cit. 14–15.

^{page 32 note 1} e.g. at Naesbjerg and Darum in western Jutland not far from Esbjerg; O. Bremer, Ethnographie der germanischen Stämme, 102.

^{page 33 note 1} Anglo-Saxons in England, 82–3.

^{page 33 note 2} Åberg, Den Nordiska Folkvandringstiden Kronologi, 12, fig. 35.

^{page 33 note 3} Sehested, op. cit. 214, pl. XLVI, 19 a, and 214, fig. in text. Some ambiguity in regard to the date of these two brooches is introduced into Åberg's Kronologi (p. 38) by his reference to them in his description of a brooch from Nørre Tranders, Aalborg Amt, N. Jutland. This brooch he assigns to the seventh century. No actual date is given for the other two, but the natural inference is that he wishes to rank all three together, since he also draws attention to the trefoil-shaped knobs which they all possess. For his dating of the Nørre Tranders brooch Åberg relies particularly on one detail of external form, the use on each side of the foot of heads with curved beaks instead of gaping jaws. Apart from this feature there is nothing to prevent its being compared with brooches assigned by him to the late fourth and early fifth centuries, among which is one from Möllebakken, Bornholm (fig. 47), with the same beaked heads. Insistence on this feature alone furnishes a doubtful criterion, so long as nothing can be judged of what must have been the chief elements in the decoration owing to the entire loss of the cloisons or filigree ornament with which the surface of the brooch was originally filled. If I have not misinterpreted Åberg, the feature on which he relies ranks poorly in comparison with all the other known qualities of the Elsehoved and Skodborg brooches, particularly the former on which there is among much filigree-work a little animal at each side of the head-plate. These can be closely paralleled with those on the great gold collar from Mone, Vestergötland (e.g. Saim. Altgerm. Thierom., figs. 501 and 502 a), which in turn retains in decadent form the splayed animal figure, also to be seen in similar guise on the Öland collar (ibid., fig. 500), in purer form on that from Olleberg, Vestergötland, as also on the large silver brooch from Meilby, Aalborg Amt, N. Jutland (Kronologi, fig. 32). The collars are all deemed to be earlier than 500, antedating the full flowering of Salin'ls style I. With the Meilby brooch was associated a small cruciform brooch, which, if found in this country, would on Åberg's own reckoning be assigned to the middle or latter half of the fifth century. Included also in the Elsehoved find was a gold tore stamped with spot-filled half-moons, accepted by most Scandinavian archaeologists as not current beyond 550, and by some as generally earlier. It would be curious to deny to the owners of the wealth of gold represented in these hoards the possession of contemporary brooches of equal magnificence. There is, moreover, no evidence for the use of the distinctive trefoil knobs in any later period: they are the forerunners of the zoomorphically ornamented knobs of a large series of brooches both in Scandinavia and in England, Why, after all, should the Elsehoved brooch be divorced from the obvious date of the remainder of the hoard, when it contains seven solidi, looped for suspension, the latest a coin of Anastasius (A.D. 491–518) and oval beads of spirally wound gold wire (cp. the loop on the Honorius coin-pendant, pl. 1, 1) such as have been found in other similar hoards? The Stodmarsh brooch, it should be noted, has lateral heads (or pseudo-heads) of exactly the same shape as those on the Nørre Tranders example.

^{page 35 note 1} Archaeologia, xci, 1 ff.

^{page 35 note 2} J. M. Kemble, Horae Ferales, 212, pl. xviii, 4; von Ledebur, Königl. Museum, p. 7, tab. 1; for one closely similar from East Prussia see Salin, fig. 133.

^{page 36 note 1} Franken und Westgoten, 99, n. 1.

^{page 36 note 2} Åberg's ‘mehrere Englische Fibeln’ (loc. cit.) in this connexion is an understatement. Outside Kent those with the median bar as compared with those without number at least two to one.

^{page 36 note 3} Altgermanische Thierornamentik, 145.

^{page 36 note 4} Ashmolean Museum, 1909. 154. Among archaeological papers bequeathed by Sir Arthur Evansis a letter confirming the accuracy of the provenience of this brooch, which appeared to have gone astray from a large group otherwise confined to East Anglia and adjacent areas.

^{page 37 note 1} Atlas, fig. 165; Herje Öberg, Guldbrakteaterne från Nordens folksvandringstid, 179, figs. 113 and 122.

^{page 37 note 2} Particularly on bracteates from Kragelund, Stenholt, Viborg Amt (S. Müller, Ordning, 557).