[967] See Report of ... the Brit. Association, 1896, pp. 930-1, and W. Ridgeway, The Early Age of Greece, i, 407-52, 594-630. Cf. Class. Rev., xvi, 1902, pp. 74-5, 88-90.
[968] W. Greenwell, Brit. Barrows, pp. 208-12. Cf. Guide to the Ant. of the Early Iron Age (Brit. Museum), pp. 109-11.
[969] Mr. Romilly Allen (Archaeol. Cambr., 5th ser., xiii, 1896, p. 223) argues from the fewness of the known Late Celtic burials that the period between the introduction of iron and the Roman conquest ‘cannot have been very long’. So also thought Canon Greenwell (Brit. Barrows, p. 212), apparently forgetting what he had very judiciously said on p. 50. The argument would lead to the conclusion that in Scotland the Late Celtic Period was almost non-existent; for only one interment of the Early Iron Age has been found there (p. 435, infra). Many such interments, unmarked by any tumulus, have doubtless escaped notice; and in many French departments they are unknown (L’Anthr., xiv, 1903, p. 386; Rev. de l’École d’anthr., xv, 1905, pp. 218-26).
[970] Sir J. Evans (Coins of the Anc. Britons, p. 39) remarks that, according to Ptolemy (Geogr., ii, 3, § 13), the territory of the Belgae included Ischalis (Ilchester), Aquae Calidae (Bath), and Venta (Winchester), and must therefore have comprised nearly all the area corresponding with Hampshire, Wiltshire, and Somersetshire; but that in each of these counties we find a distinct coinage. Probably, he argues, when the inscribed coins of Somersetshire were struck, the Belgae only occupied the east of Hampshire and the west of Sussex. Without disputing this conclusion, I would suggest that since no one Gallic tribe was called the Belgae, the British Belgae, in the narrower sense of the term, may have been a loose confederation or aggregate of tribes or of pagi, each of which perhaps had its own coinage.
All scholars are, however, aware that it is generally impossible to determine the frontiers of the British tribes, even for the period of the Roman conquest, with any approach to the comparative accuracy which has been attained in the case of those of Gaul (see my Caesar’s Conquest of Gaul, 1899, pp. 330-2). The delimitation of the tribal areas of independent Gaul depends mainly upon the reasonable assumption that they correspond for the most part exactly or nearly with those of the Gallo-Roman cantons (civitates). But in Britain we are not only baffled by the political changes which took place in the restless century that intervened between the invasions of Caesar and the Claudian conquest: we also find that although a recently discovered inscription (Archaeologia, lix, 1904, pp. 121-2) has shown that at all events in the case of the Ordovices the cantonal organization was preserved or adopted by the Roman Empire, yet, as Mommsen says (Provinces, i, 191 [Röm. Gesch., v, 1885, p. 174]), ‘the Britannic tribes, taken in the strict sense, [apparently] disappear as soon as they fall under Roman rule, and of the individual cantons after their annexation there is virtually no mention at all.’ Moreover, the boundaries of the Gallo-Roman civitates served, in principle, to define the areas of episcopal jurisdiction; and the areas of the Gallic dioceses are known. In Britain this source of information is wanting.
[971] Archaeologia, lii, 1890, pp. 387-8.
[972] Archaeol. Rev., ii, 1889, p. 324.
[973] Archaeol. Oxon., 1892-5 (1895), pp. 159-60.
[974] See Guide to the Ant. of the Early Iron Age (Brit. Museum), pp. 83, 98, and p. 240, infra. It has been pointed out (Archaeologia, lii, 1890, pp. 385-7) that the interments by inhumation which have been discovered at Arras in the East Riding of Yorkshire correspond with the Gallic interments of the fourth century before Christ. Since the Belgae appear to have practised cremation, these interments very likely indicate, what we know already, that there was a pre-Belgic Brythonic invasion of Britain: but it does not follow that they were contemporary with those of Gaul, and belonged to the time that immediately followed the close of the British Bronze Age; for, as we shall see (p. 286), inhumation persisted in Britain long after it had become obsolete on the other side of the Channel.
[975] See my article on the ethnology of Britain (pp. 428-45, infra).