^{page 196 note 1} Brewis, , Archaeologia, LXXIII (1924)Google Scholar, Pl. XXXIX, figs. 19 and 20.

^{page 198 note 1} Their presence here should be connected rather with the Somme line of approach than with the heavier concentration on the middle Rhine. Cf. Mariën, , Handelingen d. Maatsch. v. Geschiedenis en Oudheidkunde te Gent, N.R., IV (1950), 41–77Google Scholar, demonstrating the existence of a ‘Picardo-Scaldean,’ or Somme-Scheldt, culture-area during Hallstatt B. On p. 72 the argument is extended, though less securely, to Hallstatt A.

^{page 198 note 2} Fundb. aus Schwaben, XIX (1911), 8–9Google Scholar, Taf. II, 5 and Abb. 3.

^{page 198 note 3} Behrens, , Bronzezeit Suddeutschlands (1916), 252Google Scholar; Taf. XXIII, 23–6. The sword from grave 2 at Eschborn, near Höchst, is probably also of this type, but is badly broken and is excluded from the dating evidence for want of certainty. The burial itself is securely dated to Hallstatt A. For references see Appendix A (ii), no. 11.

^{page 198 note 4} Chantre, Bronze, Album, Pls. XXIX–XXXII.

^{page 200 note 1} Kutsch, , Nassauische Annalen, XLVIII (1927), 37–43Google Scholar; Abb. 1 and 2.

^{page 200 note 2} Sprater, , Urgeschichte der Pfalz (1928), 98Google Scholar; Abb. 98. The spiral anklets included, with so much else, in this grave, and formerly thought to be characteristic of the preceding Tumulus Culture (Bronze Age D), are now known to occur quite commonly throughout Hallstatt A, and even into Hallstatt B—as in the Champigny burial (Aube). I am indebted to Miss Nancy Sandars for assistance on this point.

^{page 200 note 3} The splendid double-burial of Gammertingen and the useful grave-group known as Eschborn 1, both containing swords and both dating from Hallstatt A, are omitted from this study on the ground that the sword-types in each case have evolved too far to rank with the early varieties we have been discussing and are, therefore, too late for inclusion here. This view is supported by the associated grave-goods at Gammertingen, which are clearly of ‘developed Hallstatt A’ facies. In any event Kutsch's reconstruction of the Eschborn sword, giving it a pommel-tang, is based on evidence too slight to be acceptable; and the details of the blade make the proposal highly improbable. See Appendix A (i), sub no. 7.

A recently discovered sword of Erbenheim type is stated to come from an inhumation burial of the Urnfield period at Heilbronn. This find should be of considerable interest, but has only come to notice at the time of going to press (November, 1951). The contents have not, so far as I am aware, been published and remain in private possession. I am deeply indebted to Dr Zürn (Stuttgart) for his kindness in bringing the discovery to my attention and in sending a photograph of the sword.

^{page 204 note 1} These swords, it may be noted do not occur in Britain, where hybridisation did not take place till a later stage, and then with quite other results.

^{page 204 note 2} Kraft, , Anz. Schweiz. Altertum., N.F., XXIX (1927), 141–2Google Scholar; Taf. XIV, 2 and Abb. 7. Though Kraft does not say as much, it is this type which constitutes the local facies of Naue's Type II, a broadly conceived group with a distribution from the Skagerrak to the Aegean. Vorröm. Schw. (1903), passim. Naue's typology is, however, too generalised to be relevant here, though in a wider context it is still not without value. It may be more useful to observe that these straight-bladed swords to which we are now drawing attention are also the southern (i.e. central European) counterpart to Sprockhoff's Nordic type 11a, characteristic of Montelius III.

^{page 204 note 3} AuhV, Bd. V (1911)Google Scholar, Taf. 38, 623 and 624.

^{page 204 note 4} Appendix B, no. 4. I am again indebted to Miss Sandars for assistance with the dating of this burial.

^{page 204 note 5} Kimmig, , Urnenfelderkultur in Baden (1940)Google Scholar, Taf. 37A.

^{page 204 note 6} Fundb. aus Schwaben, XVI (1908)Google Scholar, Taf. 11, 1–6; XVIII (1910), Taf. II, 8–17.

^{page 204 note 7} Kraft, loc. cit., Abb. 7.

^{page 206 note 1} The only exception is a single example found in Saarland.

^{page 207 note 1} Sprockhoff, , Griffzungenschwerter (1931), 22CrossRefGoogle Scholar, Taf. 6.

^{page 207 note 2} Ibid., 22, Taf. 8, 15–18. It is many years since Sophus Miiller assigned the comparable sword from Boeslunde also to Montelius III (strictly to Müller's Period 5, i.e. Montelius IIIa). But that, apparently, was on typological grounds only, which he made no effort to expound. Mém. Soc. Roy. Ant. du Nord, N.S., 1908–1909, 126–7Google Scholar, fig. 115.

^{page 207 note 3} Ibid., 21–3.

^{page 207 note 4} Ibid., 96 and 108, Taf. 18, 4.

^{page 207 note 5} Ibid., 108, Taf. 18, 1 and 25, 15. Having seen the Bremen sword, I am confident it once had a pommeltang which has since been broken off. It was certainly an imported piece.

^{page 208 note 1} Sprockhoff, , Chronologische Skizze, in Reinecke Festschrift (1950), pp. 133–49Google Scholar.

^{page 208 note 2} The current view that the beginning of our Late Bronze Age was marked by the simultaneous appearance of the leaf-shaped sword and the socketed axe, though still lacking formal proof, is here retained as a sound working concept and one which, in any event, is not likely to prove far wrong.

^{page 208 note 3} P.P.S., 1948, pp. 215–17Google Scholar.

^{page 208 note 4} Ibid., pp. 233–4.

^{page 208 note 5} Childe still wishes, it seems—Prehistoric Migrations in Europe (1950), p. 203Google Scholar—to equate Montelius II and III with Reinecke's Bronze Age D and Hallstatt A and B; but on present evidence it is difficult to go beyond Sprockhoff's position. In any event I cannot agree that the analogy between the solid-hilted sword from Spandau and those in the Huelva hoard is sufficiently close to be significant.

^{page 209 note 1} As detailed reference is being made to my former note, it may be observed that there is an error on p. 234 (loc. cit.); in line 5 for ‘D to early E’ read ‘C and D.’

^{page 210 note 1} On the other hand his reconstruction of the sword from Eschborn I as a representative of this type, complete with pommel-tang (ibid. Abb. 1, 12), is unacceptable, and is not adopted here.

^{page 210 note 2} The only other sword with a pommel-tang found in Britain is the magnificent and well-known example from the Thames at Millwall. (Brewis, , Archaeologia, LXXIII (1924)Google Scholar, fig. 21). This is one of the longest swords ever recovered in this country (L 804, or 31⅝ inches), and stands very close to the Erbenheim type. It has, however, a slot in the tang, and for that reason (in the writer's view) must represent a development from the original form. How very little later that development was, may be surmised from the general character of this piece alone.

^{page 211 note 1} A very late derivative of this type is illustrated by Sprockhoff in an interesting setting (ibid. 41 and 109, no. 14; Taf. 20, showing all the objects found). It comes from a tumulus-grave with stone cist found at Latdorf, Pohlsberg (Anhalt), which must fall within Montelius V. This is a very late survival indeed—the south German equivalent being the second half of Ha B—and it can only be said that the sword is not to be accounted a true representative of the Hemigkofen type, though the influence upon it of developed examples of the type is clear to see. (Not on map).

Other related variants excluded from the list comprise a small dagger-like piece found near Aegerten, Kt. Bern (L 393. 3:5. M: Zürich, 16575), and a fine sword which may possibly belong here from Schwenningen, O/A Rottweil, Württemberg; the drawing in Kraft is rough and I have not seen the original. (L c. 575. 5:4. M: Rottweil. Kraft, , Kultur d. Bronzezeit in SD., 132Google Scholar, Taf. XV, 5).

Besides these I have noted good typical examples, without recorded locality, at St. Germain (two, both casts), and in R-GZM, Mainz. Another in the Mayer collection at Liverpool is attributed, but unconvincingly, to ‘Hungary.’

^{page 212 note 1} A fine example, without locality, formerly in the Lipperheide Collection, is in the R-GZM at Mainz, O28536. There is a contemporary repair at the end of the tang. L 628. 3:4. Mainz. Zeitschr. 31 (1936), 68Google Scholar, Abb. 1.

^{page 213 note 1} A fine example without locality is in the R-GZM at Mainz, O 25056. L 642. 3:4.