¹ Curtin, Philip D., The Atlantic Slave Trade: A Census (Madison, Wis., 1969).Google Scholar

² E.g. Fage, J. D., A History of Africa (London, 1978), 266Google Scholar; Curtin, Philip, Feierman, Steven, Thompson, Leonard and Vansina, Jan, African History (London, 1978), 233–4.Google Scholar

³ Phillips, Thomas, ‘Journal of a Voyage…to Cape Monseradoe…(1693–94)’, in Awnsham and John Churchill, A Collection of Voyages and Travels (6 vols., London, 1704–1732), vi, 190.Google Scholar

⁴ Bosman, Willem, Nauwkeurige Beschryving van de Guinese Goud-, Tand- en Slave-kust… (Utrecht, 1704)Google Scholar, Letters XIV. and XXII. We are indebted to Albert van Dantzig, who allowed us to see his revisions of the translation of Bosman. These will appear shortly in History in Africa.

⁵ Foulché-Delbosc, R., ‘Voyage à la Côte Occidentale d'Afrique, en Portugal et en Espagne (1479–1480)’, Revue Hispanique, iv (1897), 175–201, pp. 181, 184.Google Scholar

⁶ Pereira, Duarte Pacheco, Esmeraldo de Situ Orbis (Côte Occidentale de Sud Marocain au Gabon), ed. Mauny, R. (Bissau, 1956), 104Google Scholar. Cf. the map in Winterbottom, Thomas, An Account of the Native Africans in the Neighbourhood of Sierra Leone (London, 1803), 1.Google Scholar

⁷ The best summary of early material on the ‘Pepper Coast’ is to be found in Dalby, David and Hair, P. E. H., ‘“Le langaige [sic] de Guynée”: a sixteenth century vocabulary from the Pepper Coast’, African Language Studies, v (1964), 174–91, pp. 187–91Google Scholar. See also Davis, Ronald W., Ethnohistorical Studies on the Kru Coast (Newark, Del., 1976)Google Scholar; Blake, John W., West Africa: Quest for God and Gold, 1454–1578 (London, 1977).Google Scholar

⁸ Palmer, Colin A., Slaves of the White God (Cambridge, Mass., 1976), 21.CrossRefGoogle Scholar

⁹ Davies, K. G., The Royal African Company (London, 1957), 233Google Scholar; Donnan, Elizabeth, Documents Illustrative of the Slave Trade to America (4 vols., Washington, 1930–5), ii, 308–9.Google Scholar

¹⁰ Curtin, , Census, 122–3Google Scholar. The limits designated for the ‘Windward Coast’ at this period usually extended to Cape Three Points, and the exports accordingly included a little gold, and occasionally some slaves.

¹¹ Public Record Office, T70/1211, Journal of the James, 1675–6;Google Scholar T70/1212, Accounts of the Sarah Bonadventure, 1676–7.Google Scholar

¹² P.R.O., T70/38, Contract between the South Sea Company and the Royal African Company, 1713.Google Scholar

¹³ None of the ships used under this contract appear to have bought a whole cargo on the Windward Coast. In 1715, for example, the Pearl was permitted to buy half her cargo of 300 slaves on the Windward Coast, and the King Solomon was to buy fifty there: British Museum, Add. Ms. 25496, minutes of 31 August 1715 and 5 October 1715.

¹⁴ Donnan, , Documents, iv, 183–7.Google Scholar

¹⁵ P.R.O., T70/362, memoranda dated 28 February 1722, 20 March 1722, 15 October 1722Google Scholar. A few of these may have come from Galinhas rather than Cape Mount.

¹⁶ Houstoun, James, Some New and Accurate Observations…of the Coast of Guinea… (London, 1725), 15–16.Google Scholar

¹⁷ Donnan, , Documents, ii, 366–80Google Scholar. There are few sources on English trade in this region in the 1730s and 1740s, but it appears to have been very small: Donnan, , Documents, iii, 124Google Scholar; Higgins, W. Robert, ‘Charleston: terminus and entrepot of the colonial slave trade,’ in Kilson, Martin L. and Rotberg, Robert I., eds., The African Diaspora: Interpretive Essays (Cambridge, Mass., 1976), 114–31, p. 127.Google Scholar

¹⁸ Davies, , African Company, 200.Google Scholar

¹⁹ Smith, William, A New Voyage to Guinea (London, 1744), 102Google Scholar; P.R.O., CO 388/25, Snelgrave, 11 May 1726.Google Scholar

²⁰ Curtin, , Census, 128.Google Scholar

²¹ Data from Williamson's, Liverpool Memorandum Book (Liverpool, 1753)Google Scholar, compiled by Williams, Gomer, History of the Liverpool Privateers… (London, 1897), 675–7.Google Scholar

²² Greenwich, National Maritime Museum, LOG/M/46, Newton, J., Journal of the Duke of Argyle and African, 1750–1754.Google Scholar

²³ P.R.O., BT6/3, ff. 99–129Google Scholar. The number of slaves actually purchased was considerably lower than the number ‘gone for’: Newton in 1750 went for 250, but bought only 174; in 1752 he went for 250 but bought only 207; and in 1753 he went for 300 but bought only 87, making up the rest of his cargo with camwood and ivory.

²⁴ Donnan, , Documents, iii, 151, and iv, 314Google Scholar. Those who sold slaves to Godfrey included Patrick Clow (of the Plantain Islands), ‘Mr Wright’ (who lived either on the Banana Islands or at Sherbro), Richard Hall (of York Island, Sherbro), William Skinner (of the Kamaranka River and Sherbro) and William Norie (of the Banana Islands): see Newton, , ‘Journal’; Owen, Nicholas, Journal of a Slave-Dealer, ed. Eveline Martin (London, 1930)Google Scholar; Nantes, , Archives Départementales, B 5004/5, Journal de bord du navire négrier l'Africain, 1738–40.Google Scholar

²⁵ P.R.O., BT6/3; Donnan, Documents, iv, 375.Google Scholar

²⁶ P.R.O., BT6/3; Lloyds' List No. 3591.Google Scholar

²⁷ Le Page, Robert B., Jamaican Creole (London, 1960), 63Google Scholar, from Donnan, , Documents, ii, 545–6.Google Scholar

²⁸ Curtin, , Census, 129Google Scholar. Curtin reproduces a slight error in Le Page's calculations: the ‘Windward Coast’ vessels in Enfield's list in fact account for 32 per cent of the total, not 31.

²⁹ P.R.O., BT 6/3.

³⁰ Edwards, Bryan, The History, Civil and Commercial, of the British Colonies in the West Indies… (2nd ed., 2 vols., London, 1793), ii, 56Google Scholar. An earlier version of the list quoted by appears, Edwards in Campbell, John, A Political Survey of Britain (2 vols, London, 1774), 11, 629–33Google Scholar. This defines the Windward Coast as extending from Cape Roxo (near the Casamance) to Cape Apollonia (near Assini).

³¹ Curtin, , Census, 129.Google Scholar

³² Anstey, Roger, ‘The volume and profitability of the British slave trade, 1761–1807,’ in Engerman, Stanley L. and Genovese, Eugene D., ed., Race and Slavery in the Western Hemisphere: Quantitative Studies (Princeton, 1975), 3–31,Google Scholar pp. 12–13; Philip D. Curtin, ‘Measuring the Atlantic slave trade,’ in Ibid. 107–28, pp. 110, 112.

³³ British Museum, Eg. Ms. 1162A. ‘Papers relating to the commerce of Africa.’ The list is erroneously dated 1754.

³⁴ Anstey, ‘Volume’, 12–13; Curtin, ‘Measuring’, 110–11. Curtin gives Anstey's figure as ‘12.2 per cent’-presumably an oversight; it should be ‘13.1 per cent’. Norris's estimate was made in 1788: P.R.O., BT6/9, Norris 27 February 1788. Curtin gives the date as 1787, while Anstey gives it as 1789: Curtin, , Census, 147Google Scholar; Anstey, , ‘Volume’, 14.Google Scholar

³⁵ E.g. Accounts & Papers (1792), xxxv, 766Google Scholar; Accounts & Papers (1789), xxiv, 633, p. 51Google Scholar; P.R.O., BT6/10; British Museum, Add. MS. 38416; Lloyds' Lists.

³⁶ House of Lords, 5/J/11/2, Return to an order dated 10 July 1799Google Scholar; Accounts & Papers (1801–1802), iv, 451.Google Scholar

³⁷ E.g. House of Lords, ‘Copies of Log Books and Surgeons’ Journals' (19 June 1799)Google Scholar; House of Lords, 5/J/11/2, Table of extracts from log books; P.R.O., C 107/1–15, Rogers papers; California, Huntingdon Library, Journal of Z. Macaulay; Greenwich, National Maritime Museum, LOG/M/21/ Journal of the Sandown.

³⁸ See the articles by Klein, Lamb and Engerman, , and Drake, , in Anstey, Roger and Hair, P.E.H., ed., Liverpool, the African Slave Trade and Abolition (Historical Society of Lancashire and Cheshire, Occasional Series, Vol. 2, 1976).Google Scholar

³⁹ Martin, Gaston, Nantes au XVIII' Siècle. L'Ére des Négriers (Paris, 1931), 188Google Scholar, 207, 218, 289. Gaston Martin's knowledge of African geography seems to have been based almost exclusively on one rather poor eighteenth century map.

⁴⁰ Mettas, Jean, ‘Pour une histoire de la traite des Noirs française: sources et problèmes’, in Emmer, Pieter, Mettas, Jean and Nardin, Jean-Claude, ed., The Atlantic Slave Trade: New Approaches (Paris, 1976), 19–46, p. 32.Google Scholar

⁴¹ Curtin, , Census, 170.Google Scholar

⁴² Mettas, , ‘Pour une histoire’, 32Google Scholar. In some cases, Gaston Martin's sources only state where a ship began to trade and where it finished, without indicating where most of the slaves were actually purchased. One list, for example, includes several ships which began trading at Sherbro, Galinhas or Cape Mount and finished at Anamabou. (Nantes, Arch. Dép., B 4588). Presumably if such ships went to the trouble of sailing beyond Cape Palmas, it was because they expected to obtain a large proportion of their slaves on the Gold Coast. In other cases, fuller details are available: Nantes, Arch. Dép., B 5004 and B 5006. Cf. Donnan, , Documents, iv, 636.Google Scholar

⁴³ Curtin, , Census, 170Google Scholar. Curtin's latest estimates reduce the Windward Coast's share in the 1750s from 38.9 percent to 0.7 per cent and increase its share for the period 1761–1800 to about 2.5 per cent. He gives no reason for making these sizeable alterations: Curtin, , ‘Measuring’, 112–26.Google Scholar

⁴⁴ Rinchon, Dieudonné, Le Trafic Négrier, d'après les Livres de Commerce du Capitaine Gantois Pierre-Ignace-Liévin van Alstein (Paris, 1938), i, 282–301Google Scholar. Some doubt has been cast on the reliability of Rinchon's data: Stein, Robert, ‘Measuring the French slave trade, 1713–1792/3’. J Afr Hist, xix, (1978), 515–21, P. 519.CrossRefGoogle Scholar

⁴⁵ ‘The Choiseul Papers’, Munger Africana Library Notes, iii (1971), 8–9Google Scholar; Paris, Archives Nationales, C⁶24, ‘Mémoire sur la Côte d'Afrique’, and ‘Mémoire sur le commerce à la côte de Guinée’ (1784)Google Scholar; Paris, Arch. Nat., C⁵27, ‘Observations faites par M. Morel’ (1771)Google Scholar and ‘Mémoire sur les côtes et le commerce d'Afrique’ (1782)Google Scholar; Nantes, , Arch. Dép., C. 881 ff. 202–8Google Scholar, ‘Considérations sur le commerce d'Afrique’ (1762)Google Scholar. See also: Forlacroix, Christian, ‘Origine et formation de la Côte d'Ivoire’, Annales de l'Université d'Abidjan, série I (Histoire), 1 (1972), 63–93, PP. 67–8.Google Scholar

⁴⁶ Paris, Arch. Nat., C⁶27. British traders alleged in 1777 that the French carried on ‘by much the greatest part of the trade’ between Sierra Leone and Cape Apollonia, paying ‘high prices for Negroes and other African produce’: British Museum, Eg. MS. 1162A, fol. 178, Germain et al., 1777Google Scholar. But there is no evidence to support this claim. In the same year, a Frenchman complained that trade at Bassam was dominated by English ships: Paris, Arch. Nat., C⁶26, ‘Mémoire pour servir à faire de Nouveaux Établissemens’ (1777).Google Scholar

⁴⁷ It may turn out that the proportion of slaves bought on the Windward Coast by Nantes ships was higher than for other French ports: see, for example, Mettas, Jean, ‘Honfleur et la traite des Noirs au XVIII^e siècle‘, Revue française d'Histoire d'Outre-Mer, lx (1973). 5–26, p. 10.Google Scholar

⁴⁸ Debien, G., Houdaille, J., Massio, R. and Richard, R., ‘Les origines des esclaves des Antilles’, Bulletin de l'I.F.A.N., série B, xxiii–xxix (1961–7)Google Scholar. Curtin identifies Debien's ‘Kissi’ with the Kisi of upper Guinea-Conakry and suggests that they were probably sold on the Windward Coast (Curtin, , Census, 185, 196Google Scholar). This interpretation is supported by Rodney, Walter, ‘The Guinea Coast’, in Gray, Richard, ed., The Cambridge History of Africa, iv (Cambridge, 1975), 223–324, p. 291Google Scholar. A more plausible theory, however, is that they were people from the Kise Kise River, perhaps speaking the Kisekise dialect of Susu: such slaves were sold at Sierra Leone and the Iles de Los (Mettas, , ‘Honfleur’, 23)Google Scholar. The ‘Canga’ or ‘Kanga’ slaves whom Curtin attributes to the Windward Coast may have come from north of Cape Mount too: Hair, who has examined the literature on this subject, concludes that ‘though Kanga was sometimes considered to extend down the Liberian coast, it tended to be more often associated with the coast between Sherbro Island and Cape Mount and the hinterland of this coast’: Hair, P. E. H., ‘An ethnolinguistic inventory of the Lower Guinea Coast before 1700: Part I’, African Language Review, vii (1968), 47–73,Google Scholar p. 71. The slaves called ‘Bobo’, as Curtin himself admits, may have been shipped on the Gold Coast (Curtin, , Census, 185Google Scholar). Thus the Windward Coast's share of slaves who reached the French West Indies in the second half of the eighteenth century may have been as low as I per cent.

⁴⁹ Brun, Samuel, Schiffarten… (Basel, 1624), 48–56;Google Scholar The Hague, Archief van de Nederlandse Bezittingen ter Kuste van Guinea, Rijksarchief, Oude W.I.C. II, instructions dated 3 February 1634 (Nassau); Ratelband, K., Vijf Dagregisters van het Kasteel Sao Jorge da Mina 1645–7 ('S-Gravenhage, 1953), ixxix, 80Google Scholar, 138, 155; Van Brakel, S., ‘Eene Memorie over den Handel der West-Indische Compagnie omstreeks 1670’, Bijdragen en Mededeelingen van het Historisch Genootschap (Utrecht), xxxv (1914), 87–104,Google Scholar p. 101.

⁵⁰ Postma, Johannes, ‘The origin of African slaves: the Dutch activities on the Guinea Coast’, in Engerman, and Genovese, , Race and Slavery, 33–49, p. 38Google Scholar. However, when merchants from the Austrian Netherlands sent two frigates to buy slaves at River Sestos and Cape Lahou in 1718, they came into conflict with the Dutch West India Company: Everaert, John G., ‘Commerce d'Afrique et traite négrière dans les Pays-Bas autrichiens’, in Emmer et al., Atlantic Slave Trade, 177–85, p. 179.Google Scholar

⁵¹ Postma, , ‘Origin’, 44–8Google Scholar. Curtin explicitly includes Postma's figures for the stretch of coast between Assini and Axim in his estimates for the Windward Coast, although according to his definition slaves obtained there ought to belong to the Gold Coast: Curtin, , ‘Measuring’, 113Google Scholar, 118. This makes his estimates for the Dutch trade on the Windward Coast about 2 per cent higher than Postma's figures warrant.

⁵² Paris, Arch. Nat., C⁶18, ‘Remarques. État en apperçu des Esclaves que peuvent retirer les Nations de l'Europe de la Côte Occidentale d'Afrique.’

⁵³ Curtin, , Census, 226.Google Scholar

⁵⁴ Curtin, , ‘Measuring’, 112.Google Scholar

⁵⁵ Feierman, Curtinet al., African History, 233Google Scholar. Cf. Curtin, , ‘Measuring’, 121.Google Scholar

⁵⁶ The British naval vessel Guernsey, which visited Cape Mesurado in 1722 to fetch wood and water, did not encounter anyone buying slaves. Several members of the crew were killed by the people of Cape Mesurado, and others were returned only after the payment of a ransom: Gwyn, Julian, ‘An incident on the Grain Coast, 1722’, Mariner's Mirror, ivi (1970), 313–25;Google Scholar Greenwich, Nat. Mar. Mus., ADM/L/G/179, F. Percy, ‘Journal’. This incident confirms the picture given by Von der Groeben in 1681, Snoek in 1701, Atkins in 1721 and Smith in 1726: all four reported that the people of Cape Mount and Cape Mesurado were afraid of being ‘panyarred’ by Europeans and that the relationship between them and Europeans was extremely sensitive: Von der Groeben, Otto F., Guineische Reisebeschreibung (Marienwerder, 1964), 35Google Scholar; Bosman, , Beschryving, Letter XXIIGoogle Scholar; Atkins, John, A Voyage to Guinea, Brasil and the West-Indies (London, 1735), 58Google Scholar; Smith, , New Voyage, 104.Google Scholar

⁵⁷ Labat, J. B., Voyages du Chevalier des Marchais en Guinée (4 vols., Paris, 1730), 1, 105.Google Scholar

⁵⁸ Villault de Bellefond, N., Relation des Costes d'Afrique (Paris, 1669), 98–113.Google Scholar

⁵⁹ Paris, Bibliothèque Nationale, Manuscrits, Français 24, 223Google Scholar, ‘Journal du voyage de Guinée…par le Chevalier des Marchais…’ (1724–1726).Google Scholar

⁶⁰ Add, British Museum. Ms. 19560, ‘Journal de navigation… par le Sieur des Marchais’ (1704).Google Scholar

⁶¹ ‘Journal du voyage de Guinée’, ff. 28–9. The parts of this document relating to the Gold coast and Whydah are highly suspect; but Des Marchais may well have revisited Cape Mesurado in 1724.

⁶² Arch, Nantes. Dép., B 5004/4, ‘Journal de bord du navire La Diligente’ (1750).Google Scholar

⁶³ Person, Yves, ‘Le Soudan nigérien et la Guinée occidentale’, in Deschamps, Hubert, ed., Histoire Générale de l'Afrique Noire de Madagascar et des Archipels (2 vols., Paris, 1970), 1, 283–303, p. 284.Google Scholar

⁶⁴ Idem, ‘Les Kissi et leurs statuettes de pierre dans le cadre de l'histoire ouestafricaine’, Bulletin de l'I.F.A.N., série B, 23 (1961), 1–57, pp. 13–14Google Scholar, 53.

⁶⁵ Curtin, , ‘Measuring’, 121.Google Scholar

⁶⁶ Feierman, Curtinet al., African History, 234.Google Scholar

⁶⁷ Mercier, P., ‘Guinée centrale et orientale’, in Deschamps, Histoire Générale, 1, 305–27, p. 326.Google Scholar

⁶⁸ Person, Yves, ‘En quête d'une chronologie ivoirienne’, in Vansina, J., Mauny, R. and Thomas, L. V., ed., The Historian in Tropical Africa (London, 1964), 322–38, p. 327.Google Scholar

⁶⁹ Ibid. 332; idem, ‘Soudan nigérien’, 301. Person's chronology with regard to Kong has been challenged in a short but convincing article by Levtzion, Nehemia, ‘Note surles états dyula de Kong et de Bobo’, Bulletin de Liaison (Linguistique, Ethnologie), Centre Universitaire de Recherches de Développement, Université d'Abidjan, i (1971), 61–2.Google Scholar

⁷⁰ Person, , ‘Soudan nigérien’, 285.Google Scholar

⁷¹ Calculations based on Postma, ‘Origin’, 44–7Google Scholar. According to a French document dated 1784, the people of Cape Lahou refused to accept a Dutch establishment; but in peacetime the Dutch always kept a ship stationed there. After staying seven or eight months, it would exchange its cargo for gold and ivory: Paris, Arch. Nat., C⁶24, ‘Mémoire sur le commerce à la Côte de Guinée’ (1784)Google Scholar. In 1838–9, Bouët referred to a former Dutch establishment at Cape Lahou: Paris, Bibliothèque Sainte-Généviève, Réserve, MS. 1813, Bouët, and Broquant, , ‘Esquisse commerciale de la côte occidentale d'Afrique, depuis Gallinas jusqu'au Gabon’ (1838–1839)Google Scholar. But Postma (‘Origin’, 39) says that the Dutch never established trading stations on the Windward Coast.

⁷² P.R.O., BT 6/3. Cf. P.R.O., CO 267/9, Norris 29 May 1790.Google Scholar

⁷³ See, for example, Curtin, Philip D., ed., Africa Remembered: Narratives by West Africans from the Era of the Slave Trade (Madison, 1967)Google Scholar; Barber, John, A History of the Amistad Captives (New York, 1840).Google Scholar

⁷⁴ Clarkson, Thomas, An Essay on the Slavery and Commerce of the Human Species in Three Parts (London, 1786), 29.Google Scholar

⁷⁵ Newton, , ‘Journal’; P.R.O., C 107/14, Walker 14 October 1787Google Scholar to Rogers.

⁷⁶ Owen, , Journal, 24Google Scholar; Newton, John, Thoughts upon the African Slave Trade (London, 1788), 5–6Google Scholar. Towards the end of the century, however, a few white men did settle at Cape Mount and Cape Mesurado: see, for example, Sierra Leone Gazette, May 1808Google Scholar; British Museum, Add. Ms. 12131, Mr Parfitt's Information (1796).Google Scholar

⁷⁷ P.R.O., BT 6/9, Newton 15.2.1788Google Scholar. As Curtin says, ‘The supply of slaves was not drawn from the distant hinterland, but from scattered sources close to the coast’: Curtin, , Census, 226.Google Scholar

⁷⁸ Newton, ‘Journal’.

⁷⁹ Accounts & Papers (1790), XXIX, 698Google Scholar, Fraser, 1 February 1790.Google Scholar

⁸⁰ Klein, Herbert S. and Engerman, Stanley L., ‘Slave mortality on British ships, 1791–1797’Google Scholar, in Anstey, and Hair, , Liverpool, 113–25,Google Scholar p. 116; Drake, B. K., ‘The Liverpool-African voyage c. 1790–1807:Google Scholar commercial problems’, in Ibid. 126–56, p. 146. Anstey, , however, calculates that ships which went to the Sierra Leone region took even longer: Roger Anstey, The Atlantic Slave Trade and British Abolition (London, 1975), 24.Google Scholar

⁸¹ Curtin, , Feierman, et al., African History, 234Google Scholar; Curtin, , Census, 247, 258.Google Scholar

⁸² P.R.O., CO 267/40, Hyde 1 July 1815Google Scholar to SoS Enc, Saunders 28 June 1815Google Scholar; P.R.O., FO 84/19, MacCarthy 26 September 1822Google Scholar to F.O. Enc, Gordon 6 September 1822Google Scholar; P.R.O., FO 84/332, Stevenson 10 November 1840Google Scholar to F.O. Enc, Bell 28 July 1840Google Scholar; P.R.O., CO 267/167, Admiralty 25 July 1841Google Scholar to C.O. Enc, Canot 12 November 1840Google Scholar; Paris, Arch. Nat., Section Outre-Mer, Généralités 177/1417 No. 47ee; Parliamentary Papers 1820, XII, 90Google Scholar and 1821, XXIII, 366, 20, Reports of Sir Collier, George; 16th Report of the African Institution (London, 1822) 69–71Google Scholar; African Repository, November 1830 and December 1837Google Scholar; Bridge, Horatio, Journal of an African Cruiser (London, 1845), 52Google Scholar; Mayer, Brantz, Captain Canot: Or, Twenty years of an African Slaver (New York, 1854), chapters 53–65Google Scholar; Holsoe, Svend E., ‘A study of relations between settlers and indigenous peoples in Western Liberia, 1821–1847’, African Historical Studies, IV, ii (1971), 331–62, pp. 342–6CrossRefGoogle Scholar; Daget, Serge, ‘Long cours et négriers nantais du trafic illégal (1814–1831)’, in Mettas, Emmer and Nardin, , Atlantic Slave Trade, 90–134,Google Scholar p. 112; Peterson, John, Province of Freedom (London, 1969), 178Google Scholar. Regional contributions to the slave trade between 1821 and 1843 are discussed in Eltis, David, ‘The export of slaves from Africa, 1821–1843’, J Econ Hist, xxxvii (1977), 409–33CrossRefGoogle Scholar. Unfortunately, he includes the Windward Coast in the category ‘Other Known Points’, so that it is impossible to assess the precise share it occupied. It should also be remembered that a few slaves from the northern part of this coast were probably transferred to Galinhas and shipped from there: African Repository, November 1830.Google Scholar

⁸³ Bouët and Broquant, ‘Esquisse’. Bouët later published a picture of a ‘slave factory on the Grain Coast’, as well as a map on which the importance of the slave trade was indicated by a thick black line extending from Sherbro to New Sestos: Bouët-Willaumez, Edouard, Description Nautique des Côtes de l'Afrique Occidentale (Paris, 1846), 179Google Scholar; idem, Commerce et Traite des Noirs aux Côtes occidentales d'Afrique (Paris, 1848).Google Scholar

⁸⁴ Parliamentary Papers, 1821, XXIII, 366Google Scholar, 20, Collier 16 September 1820 to Admiralty.

⁸⁵ Curtin, , Census, 258Google Scholar. One possible exception is the ‘Gbe’ (perhaps Bakwe, Sikon, or Gere) recaptive mentioned by Koelle: Koelle, Sigismund, Polyglotta Africana (London, 1854), 4Google Scholar. Many eastern ‘Kru’ wordlists are given in Clarke, John, Specimens of Dialects (Berwick-upon-Tweed, 1848)Google Scholar. Clarke, who collected his material in Jamaica and at Fernando Po, may have obtained these particular wordlists not from recaptive slaves but from sailors, just as Koelle obtained western ‘Kru’ wordlists from sailors at Sierra Leone. But the possibility that there were recaptives at Fernando Po who spoke eastern ‘Kru’ languages cannot be excluded.

⁸⁶ Koelle, , Polyglotta, 3–4.Google Scholar

⁸⁷ Curtin, , Census, 150.Google Scholar

⁸⁸ Curtin, , ‘Measuring’, 112.Google Scholar

⁸⁹ Ibid. 112.

⁹⁰ This is the highest of the alternatives given in Curtin, Census, Chapter 8. It should be noted that the totals of which these are percentages do not refer to the same stretch of the African coast.