^{Page 173 note 1} I am most grateful to Mr Martin Biddle and to Mr C. E. Blunt for allowing me to draw on research which was unpublished at the time of writing. In addition I owe much to them and to Professor P. Grierson, Professor H. Loyn, Dr D. M. Metcalf, Mrs J. M. Murray, Mr H. E. Pagan, Mr Ian Stewart and Professor D. Whitelock for helpful and sometimes critical comments on a draft of this article. The responsibility for the views finally expressed must, of course, be my own. On the significance of the metrology my opinions have now developed beyond, and sometimes in conflict with, the arguments I set out in ‘Historical Problems of Anglo-Saxon Coinage-(3): Denominations and Weights’, BNJ 38 (1969), 204–22. I use the following abbreviations: BMC = A Catalogue of English Coins in the British Museum; EHD 1 = English Historical Documents c. 500–1042, ed. Dorothy Whitelock (London, 1955); EHD 11 = English Historical Documents 1042–1189, ed. David C. Douglas and George W. Greenaway (London, 1953); SCBI = Sylloge of Coins of the Brit. Isles (London, 1958–).

^{Page 173 note 2} The hoards from Great Britain and Ireland from c. 795 to c. 1105 are conveniently listed and summarized by Dolley, R. H. M., The Hiberno-Norse Coins in the British Museum, SCBI 8, 15–54Google Scholar. The only significant emendation of the listing is a revision of the date of deposit of the Oakham hoard (p. 49, no. 24) from c. 870 to c. 979. A forthcoming paper in NC by C. E. Blunt and C. S. S. Lyon will show that this major hoard, found in or shortly before 1749, contained pence of Edgar, Edward the Martyr and Æthelred (First Small Cross type) and included the so-called Leicester parcel (Dolley, Hiberno-Norse Coins, p. 54, n. (c)). Some additional hoards are listed in Coin Hoards I (R. Numismatic Soc., London, 1975), 87–8.

^{Page 174 note 1} Grierson, P., ‘La Fonction Sociale de la Monnaie en Angleterre aux VII???-VIII??? Siècles’, Moneta e Scambi nell’ Alto Medioevo (Spoleto, 1961), p. 346.Google Scholar

^{Page 174 note 2} Lyon, ‘Historical Problems (3)’, p. 206; see also below, p. 216.

^{Page 175 note 1} The De Moneta of Nicholas Oresme and English Mint Documents, ed. Charles Johnson (London, 1956). P. 51.

^{Page 175 note 2} Morrison, Karl F. with Grunthal, H., Carolingian Coinage (New York, 1967), p. 36Google Scholar, n. 85.

^{Page 175 note 3} Below, pp. 183–4.

^{Page 176 note 1} Petersson, H. Bertil A., Anglo-Saxon Currency (Lund and London, 1969), p. 172.Google Scholar

^{Page 176 note 2} For a description of the plates, see below, App. II.

^{Page 176 note 3} For a full discussion of the dating and distribution of early Anglo-Saxon coinage, see Metcalf, D. M., ‘Monetary Expansion and Recession: Interpreting the Distribution-patterns of Seventh-and Eighth-Century Coins’, Coins and the Archaeologist, ed. John, Casey and Richard, Reece, Brit. Archaeol. Reports 4 (Oxford, 1974), 206–23.Google Scholar

^{Page 176 note 4} ‘La Fonction Sociale’, p. 350. In any event it is not certain that the term ‘thrymsa’ is derived from ‘tremissis’; Wulfstan expresses wergilds in thrymsas in the nor ??? leoda laga, and since a ceorl's wergild of 266 thrymsas is said to be 200 shillings according to the law of the Mercians it has generally been thought that a thrymsa was valued at only three pence.

^{Page 176 note 5} Grierson, P., ‘The President's Address’, NC 7th ser. 4 (1964), xiii–xvii.Google Scholar

^{Page 176 note 6} Rigold, S. E., ‘Imperial Coinage in Southern Gaul’, NC 6th ser. 14 (1954), 104.Google Scholar

^{Page 177 note 1} ‘La Fonction Sociale’, p. 351.

^{Page 177 note 2} Note that Louis the Pious was able to restore the gold to a weight close to the Byzantine solidus; see below, p. 189.

^{Page 177 note 3} Metcalf, D. M. and Merrick, J. M., ‘Studies in the Composition of Early Medieval Coins’, NC 7th ser. 7 (1967), 167–81Google Scholar, esp. 167–72.

^{Page 177 note 4} Metcalf, ‘Monetary Expansion and Recession’, pp. 210–11.

^{Page 177 note 5} EHD 1, 328.

^{Page 177 note 6} See below, p. 189.

^{Page 178 note 1} Grierson, ‘La Fonction Sociale’, p. 352.

^{Page 178 note 2} In the Lindisfarne Gospels the Latin dragmas decem in Luke xv.8 is glossed in the late tenth century clumsily, and therefore presumably with meaning, as ‘ten times five sceattas’. A ‘drachm’ is traditionally one-eighth of an ounce. However in xv.9 the lost dragma is glossed as ‘that shilling’. It seems, though, that the term ‘shilling’ could be used generally to denote a coin, for in other tenth-century sources it glosses numisma and also obelus, which in the form obulum (i.e. half a denarius) is elsewhere glossed as ‘sceatt’ (Wright, Thomas, Anglo-Saxon and Old English Vocabularies, 2nd ed. (ed.Wülcker) 1Google Scholar, col. 1831 line 21, col. 460 line 17 and col. 462 line 17; I am grateful to Professor Whitelock for these references).

^{Page 178 note 3} See Grierson, P., ‘Carolingian Europe and the Arabs: the Myth of the Mancus’, Revue Belge de Philologie et d'histoire 32 (1954), 1059–74.CrossRefGoogle Scholar

^{Page 178 note 4} If so, the West Saxon thegn with a wergild of 1200 shillings of silver would have been worth twice as much as the Kentish nobleman of 300 gold shillings; see Bullough, D. A., Annali della Fondazione Italiana per la Storia Amministrativa 2 (Milan, 1965), 651.Google Scholar

^{Page 178 note 5} Blunt, C. E., ‘The Coinage of Offa’, Anglo-Saxon Coins, ed. Dolley, R. H. M. (London, 1961), pp. 39–62Google Scholar, esp. p. 53.

^{Page 179 note 1} Preparatory to ‘Anglo-Saxon England’, being the Collected Papers of Frank Merry Stenton, ed. D. M. Stenton (Oxford, 1970), pp. 371–82. The problem is discussed by Metcalf, ‘Monetary Expansion and Recession’, esp. pp. 212–13.

^{Page 179 note 2} The definitive study of southern minting in the half-century after Offa is Blunt, C. E., Lyon, C. S. S. and Stewart, B. H. I. H., ‘The Coinage of Southern England, 796–840’, BNJ 32 (1963), 1–74.Google Scholar

^{Page 180 note 1} Hoard evidence suggests that Athelstan may still have been ruling when Egbert died; see Lyon, Stewart, ‘Historical Problems of Anglo-Saxon Coinage-(2): the Ninth Century -Offa to Alfred’, BNJ 37 (1968), 226.Google Scholar

^{Page 180 note 2} Blunt, Lyon and Stewart, ‘Southern England’, p. 34.

^{Page 180 note 3} Dolley, Michael, ‘The Location of the Pre-Ælfredian Mint(s) of Wessex’, Proc. of the Hampshire Field Club and Archaeol. Soc. 27 (1970), 57–61.Google Scholar

^{Page 180 note 4} Lyon, ‘Historical Problems (2)’, reviews at pp. 219–38 the coinage of the mid-ninth century and in some respects qualifies the definitive papers by Dolley, R. H. M. and Skaare, K., ‘The Coinage of Æthelwulf, King of the West Saxons, 839–58’, Anglo-Saxon Coins, ed. Dolley, pp. 63–76Google Scholar; Pagan, H. E., ‘Coinage in the Age of Burgred’, BNJ 34 (1965), 11–27Google Scholar; and Dolley, R. H. M. and Blunt, C. E., ‘The Chronology of the Coins of Ælfred the Great, 871–99’, Anglo-Saxon Coins, PP. 77–95.Google Scholar

^{Page 180 note 5} Some coins of Cenwulf (pl. IXd) and Ceolwulf seem to be neither Kentish nor East Anglian. As one of the moneyers reappears under Wiglaf (pl. IXe) it has been assumed that the coins are from a Mercian mint, presumably London. But no product of this mint has yet been recognized for Beornwulf or Ludeca (Blunt, Lyon and Stewart, ‘Southern England’, p. 34).

^{Page 181 note 1} Dolley and Blunt, ‘Ælfred’, p. 81, on the evidence of a lead trial-piece or mint weight of Alfred's from this period found in St Paul's Churchyard, suggest that London may have been a species of open city at which minting could take place. But note the contemporary references to violence (EHD 1, 183, n. 2).

^{Page 181 note 2} See Dolley, R. H. M., ‘An Unpublished Hoard-Provenance for a Penny of Ceolwulf II of Mercia’, BNJ 32 (1963), 88–90Google Scholar; Pagan, H. E., ‘An Unpublished Fragment of a Coin of Ceolwulf II’, BNJ 41 (1972), 14–20Google Scholar; and Blunt, C. E., ‘The Coinage of Æthelred, Archbishop of Canterbury, 870–89’, BNJ 31 (1962), 43–4.Google Scholar

^{Page 181 note 3} Lyon, Stewart, ‘Historical Problems of Anglo-Saxon Coinage-(4): the Viking Age’, BNJ 39 (1970), 196–7Google Scholar. However, the suggestion had already been made and substantiated by Anscombe, Alfred, ‘The Inscription on the Oxford Pennies of the Ohsnaforda Type’, BNJ 3 (1906), 69–100Google Scholar, esp. 86–90. He failed to appreciate that most of the coins were northern imitations, a factor which led Dolley and Blunt (‘Ælfred’, p. 91) to attribute them all to a second Viking mint in Yorkshire, with the tentative suggestion of Horsforth.

^{Page 182 note 1} Dolley, Michael, ‘Towards a Revision of the Internal Chronology of the Coinages of Edward the Elder and Plegmund’, ASE 3 (1974), 175–7Google Scholar, where he cites M. A. O. Donovan's revision of the date of Plegmund's death from 914 to 923 and shows that this is in keeping with the numismatic evidence.

^{Page 182 note 2} Dolley (Ibid. p. 175, and, with Blunt, ‘Ælfred’, p. 86) seems to have overlooked the fact that at least four of the moneyers of Plegmund's early coinage were at work for the king quite early in Edward's reign, namely Ælfstan, Athelstan, Edmund and Hunferth. It is, of course, possible that they had moved to another mint in the meantime. The coinage of Edward the Elder is still in need of proper classification.

^{Page 182 note 3} The definitive study of this reign is Blunt, C. E., ‘The Coinage of Athelstan, 924–939: a Survey’, BNJ 42 (1974), 35–160Google Scholar. (The volume was a special one to mark Mr Blunt's seventieth birthday.)

^{Page 182 note 4} Dolley, R. H. M. and Pirie, E. J. E., ‘The Repercussions on Chester's Prosperity of the Viking Descent on Cheshire in 980’, BNJ 33 (1964), 39–44Google Scholar. Derby's decline can be seen from Hildebrand, B. E., Anglosachsiska Mynt i Svenska Kongliga Myntkabinettet funna i Sveriges Jord, 2nd ed. (Stockholm, 1881)Google Scholar, where only sixteen out of some 4500 coins of Æthelred II are from the Derby mint.

^{Page 182 note 5} This reform is discussed below, p. 192. London provides only five per cent of the moneyers and seven per cent of the coins known for the Reform – First Small Cross type – figures which cannot wholly be explained by hoard distortion.

^{Page 182 note 6} Blunt, C. E., ‘Tenth-Century Half-Pennies and C. Roach Smith's Plate of Coins Found in London’, BNJ 31 (1962), 44–8Google Scholar. My own claim to have discovered an authentic halfpenny of Edward the Confessor (BNJ 34 (1965), 42–5) has not gained general acceptance.

^{Page 183 note 1} The extant gold coins are discussed in this context by Lyon, ‘Historical Problems (3)’, pp. 207–8. Documentary references to mancuses frequently relate to multiples of thirty silver pence, but it seems certain that Eadred was referring to the minting of mancuses in gold. Whether minted mancuses fulfilled any regular economic function in Anglo-Saxon England must be doubtful.

^{Page 183 note 2} Lyon, ‘Historical Problems (4)’, pp. 208 and 213, briefly summarizes these weights on the basis of the coins listed in BMC, Anglo-Saxon Series 1 and 11.

^{Page 183 note 3} I owe to Mr Martin Biddle the suggestion of a link with town-founding.

^{Page 183 note 4} See below, p. 185.

^{Page 184 note 1} The antiquity of the Tower pound can only be conjectured. Mint documents of the thirteenth century divide what appears to be its pennyweight into 24 grains (De Moneta, ed. Johnson, p. 67), each of which would have been about six-and-a-quarter per cent lighter than the Troy (barley) grain. Other sources, in both England and Scotland, say the sterling penny should weigh 32 grains of wheat, which would not be dissimilar (see Grierson, ‘La Fonction Sociale’, p. 368).

^{Page 184 note 2} The weights of 128 coins listed by Morrison, Carolingian Coinage, pp. 74–87, are: 1·50 g and over, 1; 1·45–1·49 g, 2; 1·40–1·44 g, 4; 1·35–1·39 g, 5;1·30–1·34 g, 19; 1·25–1·29 g, 18; 1·20–1·24 g, 11; 1·15–1·19 g, 23; 1·10–1·14 g, 13; 1·05–1·09 g, 8; 1·00–1·04 g, 2; and under 1·00 g, 22.

^{Page 184 note 3} Derived from Blunt, ‘Offa’, p. 54. The median is the middle coin when all the coins are ranked in order of weight. The weight of the median is useful as a statistical measure because it is less affected than the average weight by the inclusion of worn or slightly damaged coins. (The median weight of Pepin's coins is 1·18 g, or about 18 grains, which is marginally less than for Offa's early issues.)

^{Page 184 note 4} For a table giving the weight in Troy grains and the weight in grams equivalent to each number of pence per Tower ounce from sixteen to forty inclusive, see below, App. I.

^{Page 185 note 1} For a brief discussion of the date, and for other references, see Morrison, , Carolingian Coinage, p. 4Google Scholar; see also Ibid. pp. 38–58 for the weight of the reformed denier.

^{Page 185 note 2} ‘Et trecenti tales nummi antiquam per viginti et quinque solidos efficiunt libram’ (Prou, M., Monnaies Carolingiennes (Paris, 1896), p. xxxviiGoogle Scholar, quoting Mabillon, Vetera Analecta). Prou dates this document c. 845.

^{Page 185 note 3} The interpretation of these scales has been the subject of much controversy; see, e.g., Morrison, , Carolingian Coinage, pp. 35–6Google Scholar, and Lyon, ‘Historical Problems (3)’, pp. 220–1. Essentially, three classes of clergy (described relatively as fortiores, mediocres and minores) are common to two distinct scales which run: (i) pound of silver, half-pound, five solidi; (ii) 3 ounces, 1½ ounces (a Beauvais manuscript: thirty deniers), one solidus. The ratio between the first two classes is clearly two to one in each scale, and it therefore seems reasonable to assume that there is also a common ratio between the first and third classes. If so, and regardless of what that ratio may be, it follows directly that the first scale must be five times the second and that the pound of silver must contain 15 ounces. The solidi mentioned in the Aquitaine source must be the traditional Frankish accounting solidi of twelve deniers, and if it can be assumed that the pound of silver in the Capitulary of Bishops was also reckoned at twenty-five solidi the two scales are fully reconciled by an internal ratio of one to one-half to one-fifth, the solidus emerging with a rating of three-fifths of an ounce. (This rating is, in fact, in line with the ninth-century comment about the Gauls which Morrison queries on p. 36.) The whole system appears to have an exact parallel in the twelfth-century Scottish pound of David I, which was stated a century later to have contained 15 ounces and weighed twenty-five shillings of sterlings (The Acts of the Parliaments of Scotland, ed. T. Thomson (Edinburgh, 1844) 1, 309Google Scholar). Although a ratio of one to one-half to one-quarter might seem more natural for the scales in the capitulary, and was assumed by myself and also, by implication, by Morrison in the references cited, it could be sustained only if there were a different solidus, expressed in silver weight and equivalent to fifteen old deniers. Prou also arrived at this conclusion (Monnaies Carolingiennes, p. xl). Twenty of these solidi would comprise a pound of silver and each would weigh three-quarters of an ounce. The idea has its attractions, because a West Saxon shilling rated at five pence could then be seen as equivalent to one-third of a solidus rated at fifteen deniers, just as a Kentish gold shilling appears to have been one-third of a solidus mancus. But there seems to be no other evidence to support it.

^{Page 186 note 1} There is some evidence that an accounting ounce may have been created at the weight of twenty new deniers, for Morrison (Ibid. pp. 61–2) discusses a number of Carolingian mint weights most of which approximate to multiples of 35 g.

^{Page 186 note 2} Both references are given by Professor Miskimin, Harry A. in a controversial article, ‘Two Reforms of Charlemagne? Weights and Measures in the Middle Ages’, EconHR 20 (1967), 35–52, at p. 51Google Scholar. One reference points to nine new deniers having the value of a solidus of old deniers (i.e. twelve). The other equates six mancuses (i.e. 180 new deniers) with a Venetian pound, which would therefore contain 12 ounces if the reformed Carolingian pound contained 16. Miskimin, however, starts from a division (more correctly, a redivision) of the reformed pound into 12 heavy ounces, and therefore obtains a Venetian pound of 9 such ounces. He also considers the Capitulare Episcoporum but concentrates on a single subsidiary scale in which vassi dominici are charged half a pound if they have 200 casati, five solidi if they have a hundred and 1 ounce et faciant biduanas if they have only fifty or thirty. He deduces from this, by assuming a proportional series of one-half to one-quarter to one-eighth, that the pound of c. 780 contained twenty solidi, or 8 ounces. This leads him to postulate a two-stage reform of the pound, from 8 to 9 and then to 12 (heavy) ounces. But if the pound of c. 780 really consisted of twenty-five solidi, or 15 (light) ounces, as the main scales seem to suggest, the subsidiary scale must be in the less obvious series one-half to one-fifth to one-fifteenth and the 8-ounce pound cannot have existed.

^{Page 186 note 3} A pound of 16 Tower ounces would equal 15 English Troy ounces; medieval French Troy weight was nearly two per cent lighter, and 15 ounces would have weighed 459 g. However, Miskimin seeks to identify the reformed pound with 16 French Troy ounces (489·6 g, or almost exactly 1½ Roman pounds), and tries to bridge the gap with the coinage by adding up to ten per cent to the weight of the surviving deniers on account of wear. In contrast, Grierson (‘Money and Coinage under Charlemagne’, Karl der Grosse 1 (Düsseldorf, 1965), 501–46, at p. 530), sees the commercial pound as containing 16 ounces on a scale in which 15 ounces would be the correct minted weight of 240 deniers. By assuming a weight for the denier of 1·7 g, or 32 Paris grains of ·053 g (576 of which constituted the medieval French Troy ounce), he arrives at a commercial pound of 437 g (i.e. 256 deniers), which is identical with 15 Tower ounces and therefore seems too light. Other views are discussed by Morrison, , Carolingian Coinage, pp. 34–58.Google Scholar

^{Page 187 note 1} See above, p. 185, n. 3.

^{Page 187 note 2} Chadwick, H. M., Studies on Anglo-Saxon Institutions (Cambridge, 1905), pp. 43–4.Google Scholar

^{Page 187 note 3} See, e.g., Thompson, J. D. A., Inventory of British Coin Hoards AD 600–1500, R. Numismatic Soc. Special Publ. 1 (London, 1956)Google Scholar, nos. 117, 123, 176 and 366 (the site for which is London, Middle Temple).

^{Page 188 note 1} EHD 1, 381.

^{Page 188 note 2} Both the treaty and the reform seem likely to have a terminus post quem of 886, the treaty because it includes London in English territory (EHD 1, 580) and the reform because the London Monogram issue appears to be among the earliest on the heavy standard, if not the earliest. If the Two-Line types had begun much before Archbishop Æthelred's death in 888 one might have expected more than a single example of his to have survived. This coin provides a terminus ante quem for the currency reform, as does Guthrum Athelstan's death in 890 for the treaty. (See EHD 1, 183–4, for these obits.)

^{Page 188 note 3} The members of a peace guild in the London area agreed to contribute four pence towards it, ‘and each man was to pay his shilling who had property that was worth thirty pence’ (EHD 1 388). Perhaps shilling is being used here in a general sense. In a Wessex charter (Sawyer, P. H.Anglo-Saxon Charters (London, 1968)Google Scholar, no. 1417) the clergy of the New Minster at Winchester lease some land at Chiseldon, Wiltshire, to a thegn of Athelstan's: for non-payment of rent there is a fine of sixty denarii for the first offence, thirty solidi for the second, and the property is to be forfeited if the rent is still not paid. A shilling of four pence would imply that the fine was doubled after the first offence, but a shilling of five pence would imply an increase of two-and-a-half times.

^{Page 188 note 4} EHD 11, 399.

^{Page 189 note 1} In the Edict of Pitres, 864. The subject is discussed by Lyon, ‘Historical Problems (3)’, p. 215, and also pp. 221–2, where the observed relationship is demonstrated.

^{Page 189 note 2} See above, p. 183.

^{Page 189 note 3} For an example, see EHD 11, 573.

^{Page 189 note 4} Lyon, ‘Historical Problems (3)’, p. 221.

^{Page 189 note 5} Ibid. p. 208.

^{Page 190 note 1} See above, p. 187, and p. 188, n. 3.

^{Page 190 note 2} It seems clear that the ora was adapted to fit into the English system of weights and measures. Brøndsted, Johannes, The Vikings (London, 1960), p. 181Google Scholar, derives its etymology from the Latin aureus and gives its weight as c. 26·4 g (407 grains) at the time of the migrations. He states that it fell to 24·5 g (378 grains) as the Viking period developed. However, if used to denote 16 nominal pennyweights on Alfred's (or Edgar's) reformed standard it must have been assimilated at more than 400 grains and probably nearer to the Tower ounce of 450 grains. Its use as a weight is apparent from IV Æthelred 9·2, where those in charge of towns are instructed so to arrange matters ‘ut omne pondus ad mercatum sit pondus quo pecunia mea recipitur; et eorum singulum signetur ita, cur XV ore libram faciant’. From this, the Brompton version, it is clear that market weights were to be based on a pound of 15 oras, this pound apparently being identical with that at which the king's money would be taken back whenever it was due by weight. (This code has survived only in the Latin of the twelfth-century compilation, Quadripartitus. Liebermann, F. (Die Gesetze der Angelsacbsen (Halle, 1903), 1, 236–7)Google Scholar inexplicably prefers an apparently corrupt version which reads ‘ut omne pondus sit marcatum ad pondus pecunia mea recipitur …’ and is virtually incapable of translation, although the Brompton text is quoted.)

^{Page 190 note 3} See above, p. 185.

^{Page 190 note 4} See above, p. 183.

^{Page 190 note 5} The definitive studies of this series are Lyon, C. S. S., ‘A Reappraisal of the Sceatta and Styca Coinage of Northumbria’, BNJ 28 (1956), 227–42Google Scholar, and Pagan, H. E., ‘Northumbrian Numismatic Chronology in the Ninth Century’, BNJ 38 (1969), 1–15Google Scholar. Pagan reinterprets Lyon's numismatic chronology and proposes a radical revision of the historical dates, both of the kings and of the archbishops.

^{Page 191 note 1} Although the term styca describes the widow's mite in the Old English gloss to the Lindisfarne Gospels it seems to be yet another word of general connotation, being cognate with the modern German Stück, ‘bit’ or ‘piece’.

^{Page 191 note 2} The hoard, and the Northumbrian Viking issues included in it, are discussed by C. S. S. Lyon and B. H. I. H. Stewart, ‘The Northumbrian Viking Coins in the Cuerdale Hoard’, Anglo-Saxon Coins, ed. Dolley, pp. 96–121. For the East Anglian issues, see Blunt, C. E., ‘The St Edmund Memorial Coinage’, Proc. of tbe Suffolk Inst. of Archaeology 31 (1967–1969), 234–54.Google Scholar

^{Page 192 note 1} Stewart, Ian, ‘The St Martin Coins of Lincoln’, BNJ 36 (1967), 46–54.Google Scholar

^{Page 192 note 2} The definitive study of these issues is Dolley, R. H. M., ‘The Post-Brunanburh Viking Coinage of York’, Nordisk Namismatisk Årsskrift 1957–1958, 13–88Google Scholar. My synopsis here modifies some of Dolley's conclusions in the light of recent unpublished research by Mr Blunt into the coinage of the Danelaw south of the Humber.

^{Page 192 note 3} Space does not permit these to be illustrated here. A much more extensive range of illustrations of the Anglo-Saxon coinage is to be found in North, J. J., English Hammered Coinage I, c. 650–1272 (London, 1963)Google Scholar. This contains a comprehensive catalogue of Anglo-Saxon coinage in accordance with classifications published up to the early 1960s. It includes extensive lists of types, varieties, mints and moneyers.

^{Page 193 note 1} No definitive account yet exists of the coinages of Athelstan's four successors. Edgar's pre-reform coinage is sketched out by Dolley, R. H. M. and Metcalf, D. M., ‘The Reform of the English Coinage under Eadgar’, Anglo-Saxon Coins, ed. Dolley, pp. 136–68Google Scholar, but their paper is mainly concerned with the reform itself, for which it is a basic source.

^{Page 193 note 2} Blunt, ‘Offa’, p. 51, says the hoard evidence for Offa is disappointing in the extreme, with only seven of over 250 coins known to him having a recorded hoard provenance.

^{Page 193 note 3} E.g., two pence of Archbishop Plegmund, from the same pair of dies, have been recorded as having been struck using a London Monogram penny and a St Edmund Memorial penny as flans; see Banks, F. and Purvey, Frank, ‘Two Overstruck Pennies of Archbishop Plegmund’, BNJ 36 (1967), 189–90Google Scholar. It is odd that a St Edmund coin should have been used, for the weight would surely have been too light. (The overstruck coin is a fragment, so its full weight is conjectural.) Perhaps there is a connection between such irregularities and the closing of Plegmund's mint; see above, p. 181.

^{Page 193 note 4} Thompson, Inventory, no. 123.

^{Page 194 note 1} For a discussion of the various issues, see Lyon, ‘Historical Problems (2)’, pp. 229–37.

^{Page 194 note 2} The date of Edgar's reform, set by Dolley and Metcalf, ‘Eadgar’, at 973, has been the subject of much controversy. Petersson, Anglo-Saxon Currency, p. 74, reviews the evidence but uses special pleading to place the date nearer to 975, the annalistic year under which Roger of Wendover records it. However, coins in Edgar's name account for nearly half of the total recorded by Petersson for the First Small Cross issue, which ended early in Æthelred's reign. While admitting that there must have been a massive recoinage of previous issues, my own opinion is that the reform must be earlier than 975, since, pace Petersson, it is unlikely that there was a significant volume of posthumous coinage; see below, p. 212.

^{Page 194 note 3} Dolley and Metcalf, ‘Eadgar’, pp. 144–58, give a graphic description of the reform and its consequences. Although in this article I qualify some of their findings in the light of the research of another decade, their paper is essential reading for any serious student of the late Anglo-Saxon monetary system.

^{Page 195 note 1} Grierson, P., ‘The President's Address: Numismatics and the Historian’, NC 7th ser. 2 (1962), ix–xiv.Google Scholar

^{Page 195 note 2} An analysis of Æthelred's moneyers, which I hope to publish when it has been completed, indicates that the continuity of moneyers from one issue to another varies considerably when expressed as a proportion of the moneyers active in the earlier issue. This might prove useful as an indicator of the relative durations of different issues.

^{Page 195 note 3} Dolley and Metcalf, ‘Eadgar’, p. 166, n. 2, seek to use the coins of Edgar in the 1950 Chester hoard to prove that he did not drop the standard weight below about 21 grains. But that hoard excluded the latest pre-reform issues, e.g. BMC, Anglo-Saxon Series, Type ii, from north-west Mercia (pl. Xlh), and Type iii, without mint name, from York. Of thirty-six undamaged coins of the latter type in SCBI Edinburgh 1, half weigh under 19 grains and a quarter weigh less than 17 grains.

^{Page 196 note 1} ‘Deinde per totam Angliam novam fieri praecepit monetam, quia vetus vitio tonsorum adeo erat corrupta, ut vix nummus obolum appenderet in statera’ (Flores Historiarum, ed. H. R. Luard, Rolls Ser. (1890), s.a. 975).

^{Page 196 note 2} See the tables, Morrison, , Carolingian Coinage, pp. 40–55.Google Scholar

^{Page 196 note 3} See below, p. 200.

^{Page 196 note 4} Anglo-Saxon Currency, pp. 195–253. All the weights quoted in this discussion of the late Anglo-Saxon coinage are derived from Petersson's tables.

^{Page 196 note 5} See, e.g., Lyon, C. S. S., Meer, G. van der and Dolley, R. H. M., ‘Some Scandinavian Coins in the Names of Æthelrsed, Cnut, and Harthacnut attributed by Hildebrand to English Mints’, BNJ 30 (1961), 235–51.Google Scholar

^{Page 196 note 6} Not listed in Thompson, Inventory; see above, p. 173, n. 2.

^{Page 196 note 7} Thompson, Inventory, no. 85.

^{Page 197 note 1} Dolley and Pirie, ‘Repercussions’, p. 44.

^{Page 197 note 2} Petersson, , Anglo-Saxon Currency, pp. 81–4Google Scholar, proposes a septennial cycle which counts the Hand types as one issue; see also Dolley, Michael, ‘Some Irish Evidence for the Date of the Crux Coins of Æthelred II’, ASE 2 (1975), 145–54Google Scholar, esp. 153, for a recent restatement of the sexennial hypothesis.

^{Page 197 note 3} Dolley, ‘Some Irish Evidence’, p. 152.

^{Page 197 note 4} I have so far noted no fewer than fifty-six moneyers for the Crux type with mint signatures of Colchester, Cambridge, Hertford, Southwark (Suð gewe[orc]) and Thetford, compared with only six in Hand and eighteen in Long Cross. The number could be reduced by regarding moneyers with the same name at different mints as one, but the increase during the Crux issue would still be remarkable. Since Sudbury is within this belt, it may be necessary to reconsider the assignation to Southwark of most of the coins of this type bearing the mint signature Suðby[rig]: however, one of the fourteen moneyers who use that signature employs the same obverse die with reverses reading SVÐB and SVÐG; see Dolley, R. H. M., ‘A Note on the Mints of Sudbury and Southwark at the End of the Reign of Æthelraed II’, BNJ 28 (1956), 264–9Google Scholar, Meer, G. van der, ‘Some Corrections to and Comments on B. E. Hildebrand's Catalogue of the Anglo-Saxon Coins in the Swedish Royal Coin Cabinet’, Anglo-Saxon Coins, ed. Dolley, pp. 169–87Google Scholar, esp. 176. (Miss van der Meer's promised study of the two mints has not been published.)

^{Page 197 note 5} Lyon, Stewart, ‘Analysis of the Material’, Mossop, H. R., The Lincoln Mint c. 890–1279 (Newcastle upon Tyne, 1970), p. 18.Google Scholar

^{Page 200 note 1} Dolley, ‘Some Irish Evidence’, p. 154, suggests that Long Cross coins were popular because of their heavy weight.

^{Page 200 note 2} Ibid. pp. 147–54. However, a forthcoming paper by Mark Blackburn on the mint of Watchet will demonstrate that the evidence of the Watchet-Dublin die link discussed by Dolley cannot be used numismatically to confirm (or reject) the date of 997.

^{Page 200 note 3} Stewart, B. H. I. H., ‘The Small Crux Issue of Æthelræd’, BNJ 28 (1957), 509–17Google Scholar, and Smart, Veronia J., ‘A Subsidiary Issue of Æthelred II's Long Cross’, BNJ 34 (1965), 37–41.Google Scholar

^{Page 200 note 4} Dolley and Metcalf, ‘Eadgar’, p. 153.

^{Page 200 note 5} Dolley, ‘Some Irish Evidence’, p. 153, argues from hoard evidence that First Hand coins had been demonetized before the inception of Crux. But the two hoards which he cites (n. 2) were small and the absence of First Hand coins may not be of any significance. There are Irish hoards which mix First Hand and Second Hand coins in quantity, once when Benediction Hand is present; see Dolley, R. H. M., ‘New Light on the Mullingar Find of Hand Pence of Æthelræd II’, BNJ 35 (1966), 12–21Google Scholar, esp. the listings, pp. 17 and 21.

^{Page 201 note 1} Stewart, Ian, ‘Notes on the Intermediate Small Cross and Transitional Crux Types of Ethelred II’, BNJ 37 (1968), 16–24Google Scholar, updates an earlier paper by Dolley but does not give a weight distribution.

^{Page 201 note 2} Lyon, C. S. S., ‘Variations in Currency in Late Anglo-Saxon England’, Mints, Dies and Currency: Essays in Memory of Albert Baldwin, ed. Carson, R. A. G. (London, 1971), pp. 101–20Google Scholar, esp. 107.

^{Page 202 note 1} Brand, J. D., ‘Meretricious Metrology’, Numismatic Circular 75 (1967), 63–6.Google Scholar

^{Page 202 note 2} Lyon, C. S. S., ‘The Significance of the Sack of Oxford in 1009/1010 for the Chronology of the Coinage of Æthelred II’, BNJ 35 (1966), 34–7, at p. 37.Google Scholar

^{Page 202 note 3} Dolley and Metcalf, ‘Eadgar’, p. 153. One of the objections raised by Petersson, Anglo-Saxon Currency, pp. 76–8, namely that because no light Long Cross coins of Wilton are known there could have been an interval of some years between the last minting at Wilton and the first minting at Salisbury, is answered by the analysis of weights by county given by Lyon, ‘Variations in Currency’, p. 113. The mints in Hampshire and Wiltshire, inter alia, consistently maintained a heavy standard, even when they were using dies produced late in the issue.

^{Page 202 note 4} Lyon, ‘The Sack of Oxford’, pp. 34–7.

^{Page 203 note 1} Dolley, Michael, ‘The Nummular Brooch from Sulgrave’, England Before the Conquest: Studies in Primary Sources presented to Dorothy Whitelock, ed. Peter, Clemoes and Kathleen, Hughes (Cambridge, 1971), p. 340.Google Scholar

^{Page 203 note 2} Lyon, ‘Variations in Currency’, p. 103 (referring to Hildebrand's Type Ae).

^{Page 203 note 3} Lyon, van der Meer and Dolley, ‘Some Scandinavian Coins’, p. 249.

^{Page 203 note 4} Lyon, ‘Variations in Currency’, p. 104.

^{Page 203 note 5} The sole exception is the Short Cross type of Edward the Confessor (c. 1048–50), which was struck on tiny flans more suited to halfpence. The average weight (c. 15¾ grains) corresponds to twenty-nine per ounce, but many were minted very much lighter.

^{Page 204 note 1} Thompson, Inventory, no. 37.

^{Page 204 note 2} See the table of weights, BMC, The Norman Kings 1, cliii.

^{Page 205 note 1} De Moneta, ed. Johnson, pp. 57 and 76, where it is stated that the pound must contain 20s 3d.

^{Page 205 note 2} ‘Royal Revenue and Domesday Terminology’, EconHR 2nd ser. 20 (1967), 221–8.

^{Page 205 note 3} Pace Finn, R. Welldon, Domesday Book: a Guide (London and Chichester, 1973), p. 81.Google Scholar

^{Page 206 note 1} Anglo-Saxon Currency, pp. 29–56.

^{Page 206 note 2} Ibid. pp. 49–53.

^{Page 206 note 3} Ibid. pp. 29–36.

^{Page 206 note 4} This is how I interpret the complex arguments of Petersson, Ibid. pp. 40–2.

^{Page 207 note 1} See, e.g., II Athelstan 12 (EHD 1, 384) and II Cnut 24 (EHD 1, 422).

^{Page 207 note 2} Anglo-Saxon Currency, p. 49.

^{Page 209 note 1} EHD 11, 573.

^{Page 209 note 2} EHD 11, 579.

^{Page 209 note 3} I.e. Troy weight.

^{Page 209 note 4} From Worcester; see Brooke, G. C., BMC, The Norman Kings 1, cxxxiii–cxl.Google Scholar

^{Page 209 note 5} De Moneta, ed. Johnson, pp. 57 and 76. The 24 Tower grains would have a weight close to 22½ Troy grains (1·46 g).

^{Page 210 note 1} For an account of a simulation of the Anglo-Saxon die cutting and minting processes, see Sellwood, D., ‘Medieval Minting Techniques’, BNJ 31 (1962), 57–65Google Scholar. Sellwood concludes that one engraver working eight hours a day for ten days could have produced a pair of dies for each of 150 moneyers, and that a team of three workmen at a mint could probably turn out a coin every five seconds from blanks already prepared.

^{Page 210 note 2} Jones, F. Elmore and Blunt, C. E., ‘The Tenth-Century Mint Æt Weardbyrig’, BNJ 28 (1957), 494–8Google Scholar, and Dolley, R. H. M. and Jones, F. Elmore, ‘The Mints Æt Gothabyrig and Æt Sith(m)estebyrig’, BNJ 28 (1956), 270–82Google Scholar. The third mint is identified as Cissbury, Sussex.

^{Page 210 note 3} Dolley and Metcalf, ‘Eadgar’, pp. 146–7.

^{Page 210 note 4} Biddle, M., ‘Organization and Society: iii. The Mint’, Winchester in the Early Middle Ages, ed. Biddle, M., Winchester Stud. 1 (Oxford, 1976), 396–422.Google Scholar

^{Page 210 note 5} II Athelstan (Grateley) 14 (EHD 1, 384).

^{Page 210 note 6} IV Æthelred 9, which contains the passage quoted above, p. 190, n. 2, begins by saying (in A. J. Robertson's translation) that ‘moneyers shall be fewer in number than they have been in the past. In every principal town (there shall be) three, and in every other town (there shall be) one.’ But the date of this code is uncertain and the instruction does not appear to have been carried out.

^{Page 210 note 7} II Athelstan 14.1 (EHD 1, 384), III Æthelred (Wantage) 8 (EHD 1, 404) and II Cnut 8 (EHD 1, 420). For a general discussion, see Kinsey, R. S., ‘Anglo-Saxon Law and Practice Relating to Mints and Moneyers’, BNJ 29 (1958), 12–24.Google Scholar

^{Page 211 note 1} Most are listed in Smart, Veronica J., ‘Moneyers of the Late Anglo-Saxon Coinage 973–1016’, Commentationes de Nummis Saecukrum IX-XI in Suecia Repertis 11 (Stockholm, 1968), 193–276.Google Scholar

^{Page 211 note 2} Stewart, Ian, ‘Reflections on some Wessex Mints and their Moneyers’, NC 7th ser. 15 (1975), 219–29Google Scholar; also Biddle, ‘Organization and Society’, p. 401, n. 4. Later, Henry I's decree concerning the coinage (Gesetze, ed. Liebermann 1, 523) prohibits a moneyer from reminting money except in his own county.

^{Page 211 note 3} Brooke, , BMC, The Norman Kings 1, cxxxvGoogle Scholar, n. 1.

^{Page 211 note 4} An important recent paper on this theme is Archibald, Marion M., ‘English Medieval Coins as Dating Evidence’, Coins and the Archaeologist, ed. Casey and Reece, pp. 234–71Google Scholar. In terms of a critical appraisal of the classification of the coinage it covers much the same ground as this article but from an archaeological standpoint, and is strongly recommended for further reading.

^{Page 212 note 1} Morrison, , Carolingian Coinage, p. 21.Google Scholar

^{Page 213 note 1} Dolley, R. H. M., ‘Three Late Anglo-Saxon Notes’, BNJ 28 (1955), 99–105Google Scholar, discusses the Cadbury mint. For Cissbury, see above, p. 210, n. 2.

^{Page 213 note 2} For a discussion of reverse inscriptions, see Stewart, B. H. I. H., ‘Montla and Mot on Anglo-Saxon Coins’, BNJ 31 (1962), 27–42Google Scholar. However, although moneta is unlikely to be a contraction for monetarius, because it can occur after a moneyer's name in the genitive case, it may at times denote ‘workshop’ rather than ‘die’. It is clear from Biddle, ‘Organization and Society’, pp. 397–400, that in twelfth-century Winchester the term monete was used in that context.

^{Page 213 note 3} Illustrations are given by Lyon, ‘Historical Problems (4)’, pp. 202–3.

^{Page 214 note 1} The classic study of style is Dolley, R. H. M., Some Reflections on Hildebrand Type A of Æhelræd II, Antikvarist Arkiv 9 (Stockholm, 1958)Google Scholar, in which a clear separation is made between First, Intermediate and Last Small Cross, and the latter is regionally divided. But undoubtedly it can be further subdivided, and when this is done many of the anomalies of style which are apparent in Dolley's listings by mint (pp. 18–27) are likely to be removed.

^{Page 214 note 2} For a salutary discussion of this subject, see Kent, J. P. C., ‘Interpreting Coin-Finds’, Coins and the Archaeologist, ed. Casey and Reece, pp. 184–200.Google Scholar

^{Page 214 note 3} See above, p. 200, n. 5.

^{Page 214 note 4} A possible example is given by Dolley, R. H. M., ‘The Shaftesbury Hoard of Pence of Æthelræd II’ NC 6th ser. 16 (1956), 267–80Google Scholar, esp. 273. Of ninety-two Long Cross coins listed, forty-one are from the mints of Lincoln and York but only two are from Winchester, with a single coin from Shaftesbury itself.

^{Page 214 note 5} The Morley St Peter (Norfolk) hoard of 1958 (Dolley, Hiberno-Norse Coins, no. 69), deposited early in Athelstan's reign, contained a single coin of Athelstan, 665 imitative portrait coins in the name of Edward the Elder from a local mint (?Norwich), ninety-two late non-portrait coins of Edward but no mid-period coins of his reign, and a parcel of eighty pence of Alfred, one penny of Ceolwulf II and six early pence of Edward which would not have been out of place in the Cuerdale hoard deposited twenty years earlier. (There was also a handful of Viking coins.) Full publication of this hoard is still awaited, but a summary list appears in Dolley, R. H. M., ‘The Morley St Peter Treasure Trove’, Numismatic Circular 66 (1958), 113–14.Google Scholar

^{Page 215 note 1} See above, p. 212.

^{Page 215 note 2} See above, p. 190, n. 5.

^{Page 215 note 3} For an early but important discussion of this subject, see Dolley, R. H. M., ‘The Relevance of Obverse Die-Links to some Problems of the Later Anglo-Saxon Coinage’, Commentationes de Nummis Saeculorum IX-XI in Suecia Reperth 1 (Stockholm, 1961), 155–72.Google Scholar

^{Page 215 note 4} See above, p. 213, n. 3.

^{Page 216 note 1} See above, p. 196, n. 5. There are many other examples.

^{Page 216 note 2} Lyon, ‘Historical Problems (3)’, p. 206.

^{Page 217 note 1} Anglo-Saxon Currency, p. 254. The figures I quote are for the columns in his table 49a, headed ‘same dies’.

^{Page 217 note 2} See above, p. 209.

^{Page 217 note 3} For a discussion of the various techniques available, see Grierson, Philip, Numismatics (Oxford, 1975), PP. 149–55.Google Scholar

^{Page 217 note 4} See above, p. 177.

^{Page 217 note 5} Petersson, , Anglo-Saxon Currency, pp. 156–61.Google Scholar

^{Page 217 note 6} Metcalf and Merrick, ‘Composition’, p. 175.

^{Page 218 note 1} Lyon, C. Stewart S., ‘Consultation in Research’, BNJ 35 (1966), 226–30.Google Scholar

^{Page 218 note 2} Mossop, , The Lincoln Mint, pp. 15–17.Google Scholar

^{Page 218 note 3} For a discussion of this, see Grierson, , Numismatics, pp. 15–57.Google Scholar