1. Labat, Jean-Baptiste, Nouvelle Relation de l'Afrique Occidentale (5 vols.: Paris, 1728), 3: 255, 356.Google Scholar This work is one of several important sources on the history of Africa cited in the present paper that still lack a scholarly edition. It has an inadequate index.

2. Froidevaux, Henri, “La découverte de la chute du Félou (1687), Bulletin de Géographie historique et descriptive, 13 (1898), 304–06, 310–13.Google Scholar

3. Cultru, Prosper, Histoire du Sénégal du XVe siècle à 1870 (Paris, 1910), 13,123–24, 145–47, 151Google Scholar; idem., ed., Premier voyage du Sieur de la Courbe fait à la coste d'Afrique en 1685 (Paris, 1913), 16, 31–32, 66. The latter work, though a primary text, has no index. Cultru listed Labat's main borrowings from the La Courbe account but did not indicate them in detail in the text, and to date no complete analysis of Labat's sources has appeared. La Courbe wrote: “je feray voir, cy-après, dans le relation du voyage que j'ay fait dans la riviere du Senegal, au plus haut ou Ion puisse monter, qu'elles n'ont aucune communication navigable les unes avec les autres” (Cultru, , Histoire, 7Google Scholar, and, for a slightly different reading, Cultru, , Premier voyage, 16Google Scholar). The “relation” mentioned is not extant. Evidence that it was actually written (La Courbe's reference being ambiguous) is provided by Labat, who, in an unguarded moment, remarked, with reference to a specific item of natural history, that “Mr. de la Courbe s'est lourdement trompé dans son Journal” (Labat, , Nouvelle Relation, 3: 177).Google Scholar The item is not in the account by La Courbe that Cultru published; therefore it was in the “relation,” that Labat terms “son journal.” Presumably all of the “relation” was borrowed by Labat, so its description of La Courbe's upriver visit may well be the basis for Labat's description of one of the upriver visits attributed to Brue (cf. Cultru, , Histoire, 128Google Scholar). What was the date of La Courbe's visit? It was not during his 1685–1687 period in Africa, according to his own account. (The date of the visit is wrongly given as ‘1687’ in Broc, Numa, La géographie des philosophes: géographes et voyageurs français au XVIIIe siècle, Paris, [1975], 67n25Google Scholar, despite this author claiming to have seen a 1689 manuscript, “Memoire sur le Senegal” by La Courbe, not cited by earlier historians.) La Courbe spent three further periods in Africa, during 1689-?1692, 1694, and 1706–1710. We know that during the final period he was active in Galam and visited the falls (Cultru, , Histoire, 7Google Scholar, dates corrected in Cultru, , Premier Voyage, lvi–lviiGoogle Scholar, and corrected again in Ly, Abdoulaye, La Compagnie du Sénégal [Paris, 1958], 207, 224, 267Google Scholar, another work without an index). As we have it, La Courbe's account covering only his first period (1685–1687), was written not earlier than late 1690, for in it (Cultru, , Premier Voyage, 135Google Scholar) La Courbe referred to the “late” marquis de Seignelay, whose death occurred on 3 November 1690. If the “relation” was a continuation of this account and was written at the same time, it is plausible that it covered the second period in Africa (1689–1692) and that in 1692 or 1693 LaCourbe wrote a single account of both periods. In the latter year he wrote a mémoire in which he said that Galam had not been visited for some time since his own visit, which was the third French visit there (Cultru, , Premier Voyage, liiGoogle Scholar). The two previous visits were presumably those of Chambonneau in 1686 and 1687, and it would seem that La Courbe's visit was in 1690, as stated by Cultru, , Histoire, 75.Google Scholar

4. Ly, , Compagnie, 155, 255–56.Google Scholar

5. Ritchie, Carson I.A., “Deux textes sur le Sénégal (1673–1677),” Bulletin de l'Institut Fondamental d'Afrique Noire, 30B (1968), 310.Google Scholar For a shortened version, see “Impressions of Senegal in the Seventeenth Century: Excerpts from Louis Chambonneau's Treatise,” African Studies, 26 (1967), 59–93.CrossRefGoogle Scholar

6. Les voyages du sieur Le Maire aux Isles Canaries, Cap-Verd, Sénégal, et Gambie (Paris, 1695), 77, 83–84Google Scholar: cf. Cultru, , Histoire, 77–78.Google Scholar Le Maire's work has an index very good by contemporary standards. (Note that the English translation of 1887 [by Edmund Goldsmid, privately printed, Edinburgh] not only omits the preface and the long supplement but silently cuts a number of passages, for instance, 97–116, 157–166 of the original.) Though it is generally supposed that “Fargots” is a misprint for “Sargots,” i.e. Sarakolé, the supposition is weakened by other instances: “son voyage de Galam ou des Fergaux” (Cultru, , Histoire, 122)Google Scholar; “les Fargots” (Thilmans, G., “La relation de François de Paris (1682–1682),” Bulletin de l'Institut Fondamental d'Afrique Noire, 38B (1976), 8).Google Scholar It is possible that “Fargot” represents the Manding term farako “river running over rocks” (Delafosse, Maurice, La langue mandingue, 2, Dictionnaire mandingue-français (Paris, 1955).Google Scholar This term, said to have been applied to a stream south of Galam (Delafosse, , ed., Chroniques du Fouta Sénégalais, Paris, 1913, 232Google Scholar) seems equally applicable to the upper Senegal in the region of the Falls of Félou and the Falls of Gouina, and “Fargots” may therefore have been used by the French ca. 1680 to designate the Sarakolé. “Enguelland” is unidentified, but it looks like a term transmitted into French via Dutch.

7. Ritchie, , “Deux textes,” 329.Google Scholar If the French penetrated Galam in the 1670s, this must have been followed by a period of inactivity, since La Courbe thought that the penetration had only begun in 1685. The possibility that the French were alerted to the existence of upriver falls by references deriving from Portuguese sources is discussed later in this article.

8. [Gaby, F.J.B.] Relation de la Nigritie: contenant une exacte description de ses Royaumes et de leurs Gouvernemens, la Religion, les Moeurs, Coustumes et raretez de ce Rais, (Paris, 1689), second ed. Paris, 1692Google Scholar; reprinted (without further information) Le Missioni Francescane, 6 (1895), 276–82, 331–37, 473–79Google Scholar; 7(1896), 9–17, 108–17, 463–64. The only date for his visit to Africa supplied by Gaby is 1686 (left Paris 11 March 1686, p. 2), apparently an error, since the year of his voyage was undoubtedly 1687--(a) because he sailed on the Catherine, captain Destombes, arriving Senegal 5 June (3, 17, 21) and these details do not match the 1686 voyage of this ship (Ly, , Compagnie, 199–200Google Scholar; Cultru, , Premier Voyage, 122, 149, 182Google Scholar--the 1686 voyage had Chambonneau aboard), and (b) because he travelled with a Company inspector, one François (17, 21), who arrived in 1687 (Cultru, , Histoire, 74, 114–15Google Scholar).

9. Cultru, , Premier Voyage, 182–83Google Scholar; Froidevaux, , “Découverte,” 303.Google Scholar

10. A single example must suffice: [Chambonneau] “Leur derniere grande feste mobile est celle qu'ils nomment ‘Gamon,’ elle se fait le dixiesme de la lune ou elle eschet, qui est leur premier jour de l'an. Ils s'assemblent tous, durant trois nuits depuis soleil couche, jusque a Soleil leve, font le sala, et dansent autour de leur Marabou, qui fait lecture des loix et au bruit des tambours disent tous plusieurs fois: ‘Hiamamar,’ qui est a peu pres, ‘Dieu nous veille garder jusqu'a l'autre annee.’ [Gaby] “Leur derniere feste solemnelle est celle qu'ils nomment Gomon. Elle se fait le dixième de la Lune de May, qui est le premier jour de leur an. Ils s'assemblent tous durant troit nuits, depuis le Soleil couché jusques au Soleil levé, font leurs prieres, et dansent a l'entour de leur Marabou, qui fait lecture de la Loy, (c'est à dire de quelque chapitre de l'Alcoran) puis ils disent tous au bruit du tambour, Hiamamar, Dieu nous veuille garder jusques à l'autre année.” (Ritchie, , Deux textes, 315–16Google Scholar; Gaby, , Nigritie, 46–47).Google Scholar

11. Ibid., 89.

12. Froidevaux, , “Découverte,” 303–305.Google Scholar

13. Labat, , Nouvelle Relation, 1: 30; 2: 180, 205; 3: 180, 204, 308.Google Scholar

14. Ibid., 3: 293–94.

15. Ibid., 2: 156–60. Although interspersed with other material, mainly on natural history, 125–52 of chapter IX represent 15–24 and 52–53 of Cultru's edition of La Courbe's account. However, the remainder of the chapter is not in La Courbe's account. Cultru suggested that this reconnaissance of the falls may have been under the direction of La Courbe in 1690 (Cultru, , Histoire, 145Google Scholar).

16. Labat, , Nouvelle Relation, 3: 356–359.Google Scholar On the evidence of a 1719 letter that indicates that Brue had no personal knowledge of an island below the Falls of Félou, Cultru concluded that Brue never actually visited the falls (Cultru, , Histoire, 151Google Scholar). The conclusion is forced and it remains an open question whether the account of Brue's visit is genuine or whether it is Labat's version of an earlier visit, presumably by La Courbe.

17. If Chambonneau was party to Gaby using his material, why was his name not mentioned? Given the quarrels between agents of the French Company, and not least the serious rivalry between Chambonneau and La Courbe (who arrived back in France in 1687), but also given that both men survived with the Company and were promoted, it may be that Chambonneau encouraged an unofficial publication of material addressed to the Company, which established a claim for field research and exploration during his period of service without linking his name to the leakage. The alternative explanation is, of course, that Gaby was shown the material and published it without Chambonneau's knowledge or permission.

18. Labat, , Nouvelle Relation, 4: 32, 38.Google Scholar The text does not actually describe another visit to the falls, but the map marks a journey to both falls, and the other journeys marked are those undertaken by Sieur Compaignon, on Brue's orders, apparently mainly during 1716. This map confirms La Courbe's activities upriver ca. 1710 by an inscription indicating a river “par lequel Mr de la Courbe vouloit aller à Bamboucq.”

19. Ibid., 2:1. “Isle de Lontou” is shown upriver from the falls: cf “ils arriverent à Lontou, visiterent l'Isle qui est au milieu de la Riviere…” (160). Also this map does not show the stream allegedly joining the Senegal from the north at the falls, the Joto River, shown on the Chambonneau map.

20. Ibid., 3:290; the relevant area reproduced in Froidevaux, , “Découverte,” 305.Google Scholar

21. Ibid., 304n3, 310.

22. Fernandes, Valentim, Description de la Côte Occidentale d'Afrique (Sénégal au Cap de Monte, Archipels), ed. Monod, T., da Mota, A. Teixeira and Mauny, R. (Bissau, 1951), 150.Google Scholar

23. Thilmans, G., “Les planches sénégalaises et mauritiennes des ‘Atlas Vingboons’ (XVIIe siècle),” Bulletin de l'Institut Fondamental d'Afrique Noire, 38B (1975), 100–111Google Scholar; “Careden” is presumably Kaédi. French knowledge of the lower and perhaps middle Senegal is also proved by the elaborate and fairly correct topography of the waterways shown on a 1631 map of Africa by Jean Guérard of Dieppe (reproduced by hand-copying in Santarem, Visconde de, Essai sur l'histoire de la cosmographie [Paris, 1849], Atlas volume, plate XX).Google Scholar French activities on the middle river “vers le grand Roy Sambalame,” the Fula ruler, were mentioned in Jannequin, Claude, Le voyage de Lybie au Royaume du Senega (Paris, 1643), 67.Google Scholar

24. Donelha, André, An account of Sierra Leone and the rivers of Cape Verde (1625), ed. da Mota, A. Teixeira and Hair, P.E.H., (Lisbon, 1977), 123, 273.Google Scholar

25. Pereira, Duarte Pacheco, Esmeraldo de situ orbis: Côte occidentale d'Afrique du Sud Marocain au Gabon, ed. Mauny, R. (Bissau, 1956), 50, 32. My translation.Google Scholar

26. Fernandes, Valentim, Description de la Côte d'Afrique de Ceuta au Sénégal, ed. de Cenival, P. and Monod, T. (Paris, 1938), 69 (f.69), 149Google Scholar; Description de la Côte Occidentale d'Afrique (Sénégal au Cap d Monte, Archipels), ed. Monod, T., da Mota, A. Teixeira, and Mauny, R. (Bissau, 1951), 7 (f.90v), 150.Google Scholar

27. The first vernacular term is Fula mayo “river” (Gaden, Henri, Le Poular, 2, Lexique poular-français, [Paris, 1914]).Google Scholar The second vernacular term is shared by the closely-related languages of the three ethnic groups found today in Galam and neighboring Bambouk--the Sarakolé, Manding, and Bambara--as these languages are recorded in modern or historical dictionaries, thus: ko “stream”+le variant of de, a diminutive, hence kole “small stream” (Bazin, H., Dictionnaire bambara-français [Paris, 1906])Google Scholar: ko “river” + ni/li/ti, a diminutive, hence ko-li “stream” (Delafosse, Langue Mandingue); collé “marigot” (see below) (“Vocabulaires…receuillis à la côte d'Afrique pour le service de l'ancienne Compagnie Royale du Sénégal…,” Mémoires de la Société Ethnologique, 2 (1845), 2:205–67.Google Scholar This not altogether accurate transcription of MS Fonds africain 4 of the Bibliothèque Nationale, Paris, has been checked against the original, the item cited being under the head “François-Saracolé”); Kollé/Khollé “fleuve” (in Sarakolé, Delafosse, , Haut-Sénégal-Niger [3 vols.: Paris, 1912], 1:60n1.Google Scholar) The same term was commented on by Labat in 1728: “toutes les Rivieres de ce pays sont appellées Marigots par les François et colez par les Négres Mandingues, Saracolez, Foules de Cassou et autres des environs” (Labat, , Nouvelle Relation, 3: 336Google Scholar).

28. de Barros, João, Asia, Primeira decoda, ed. Cidade, H. and Murias, M. (Lisbon, 1945), 104–05Google Scholar, my translation. Cf. Crone, G.R., ed., The Voyages of Cadamosto and Other Documents on Western Africa in the Second Half of the Fifteenth Century (London, 1937), 136–37.Google Scholar The African sections of Barros have not been adequately edited to date. The African terms in the Barros extract are almost certainly from the Mande languages. “Huaba” most probably represents Manding wa “cavité, alvéole, creux de rocher” + ba “grand” (Delafosse, Langue Mandingue). “Burto” is more difficult, mainly because it is not clear what metaphor the Africans were invoking. It is unlikely that the reference was to the waters forming a rainbow, because the local term invokes, not a bow, but a sword: e.g. Manding sa-nkala or alla-ta-fa, Bambara alla-Ka-mourou “God's sword, the rainbow” (ibid; Bazin, Dictionnaire). Local terms for “bow” do not resemble “burto.” However, if an African did compare the spurting of the waters with the firing of arrows from a bow, “burto” may relate to Manding bu “point of arrow,” or Bambara byento and Serawoké boto “quiver” (Delafosse, Langue Mandingue; Bazin, Dictionnaire; Koelle, S.W., Polyglotta Africana [London, 1854]).Google Scholar An alternative explanation is that the terms derives from Manding butu “écroulement, avalanche, tomber de haut” (Delafosse, Langue Mandingue). If the terms in Barros' extract are actually from Manding, they could have been supplied by Manding informants who knew both the Senegal and the Gambia rivers, which might explain Barros' apparent confusion of the two. But his description of the falls can apply only to the Senegal.

29. Fernandes, , Description, 149.Google Scholar

30. Barros, Asia, livro 2, caps. 5–8, 12. See the discussion of this material in da Mota, Avelino Teixeira, “D. João Bemoim e a expedição portuguesa ao Senegal em 1489,” Boletim Cultural da Guiné Portuguesa, 26 (1971), 63–111Google Scholar (also no. 93, série separatas, Agrupamento de Estudos de Cartografia Antiga, Junta de Investigacões do Ultramar, Lisbon, 1971). It cannot be said that exploration of the Senegal River was adequately represented on pre-1550 maps, but then neither was the more clearly evidenced exploration of the Gambia, both rivers being generally represented by straight lines heading into the interior. However, an anonymous Italian map of West Africa of ca. 1470 does show a feature on the Senegal River which resembles the “Isle du Morphil” of eighteenth-century maps (da Mota, Teixeira, Mar, além mar; estudos e ensaios de história e geografia (Lisbon, 1972), plate 10.Google Scholar It is perhaps significant that only twenty years after the discovery of the Senegal River, it was reported that “the tide ascends the river more than 60 miles, according to information …from Portuguese Christian sailors who have been up it.” Viagens de Luis de Cadamosto, ed. Peres, D. (Lisbon, 1948), 26–27Google Scholar, my translation. Though the term ‘Galam’ is not recorded before ca. 1600, the Portuguese had apparently heard the term “Sarakolé” by the 1490s and perhaps by the 1450s (da Mota, Teixeira, “Un document nouveau sur l'histoire des Peuls au Sénégal pendant les XVème et XVIème siècles,” Boletim Cultural da Guiné Portuguesa, 24 (1969), 781–860.Google Scholar (also no. 56 of Série separatas, Agrupamento de Estudos de Cartografia Antiga, Lisbon, 1969, reference on p. 20 of latter); Gomes, Diogo; De la première découverte de la Guinée, ed. Monod, T., Mauny, R., and Duval, G. (Bissau, 1959), 41, 65–66 “Cereculle”).Google Scholar

31. de Figueiredo, Manuel, Hydrografia…com os roteiros (Lisbon, 1608 and later editions)Google Scholar; Carneiro, Antonio de Mariz, Regimento de pilotos (Lisbon, 1642 and later editions).Google Scholar For later Portuguese or Portuguese-inspired maps that fail to show Félou (or map the upper Senegal), see Cortesão, A. and da Mota, Avelino Teixeira, Portugaliae Monumenta Cartographica, (6 vols.: Lisbon, 1960), 5: plate 556/CGoogle Scholar, Albernaz atlas of 1665; Mortier, P., “Carte particuliere des costes de l'Afrique qui comprend le royaume de Cacheo, le Province de Gelofo, etc.,” Suite de Neptune François (Amsterdam, 1700).Google Scholar

32. Between the early Portuguese exploration and the later French, the Dutch also visited the middle river. Dutch maps record names along the river, but only up to Kaédi, and they mark them along a misplaced and imaginary river running northward from the main east-west stream. This curious displacement of the Senegal apparaently began with a Dutch manuscript map of 1628 (Thilmans, “Planches Sénégalaises,”) and was often repeated, for instance, on the map of Africa in Olfert Dapper, , Naukeurige Beschrijvinge der Afvikaensohe Gewesten (Amsterdam, 1668) and on the map of ca. 1700Google Scholar, “Genehoa, Jaloffi et Sierraleone regna” by P. Schenk of Amsterdam.

33. David, Pierre, Journal d'un voiage fait en Bambouo en 1744, ed. Delcourt, André (Paris, 1974), fig. 4.Google Scholar The Labat map is very clearly derived from this one, with only the most minor modifications by Labat's engraver.

34. Guillaume Delisle (1675–1726), “L'Afrique…1700…chez l'Auteur Rue des Canettes;” “L'Afrique…1700…chez l'Auteur sur le Quai des Horloges” [i.e. 1707 revision]; “Carte de la Barbarie, de la Nigritie et de la Guinée…1707;” “Carte d'Afrique…1722.” Reproductions of the first and last can be conveniently found in Tooley, R. V., Collectors' Guide to Maps of the African Continent and Southern Africa (London, 1969), plates 51–52Google Scholar; and of the third in Klemp, E., Africa in Maps (Leipzig, 1968), plate 25.Google Scholar

35. It is just possible that an earlier naming of “Félou” may yet be found in the rich store of manuscript maps in the Bibliothèque Nationale. From Fonds D'Anville in this library I have seen and checked two maps of Senegambia, both illustrating French activities in the 1690s, both said to be based on information from La Courbe (Ge. DD. 2987, portfolio 119, nos. 8099 and 8138). One of these maps has many resemblances to another manuscript map in the same library (Service Hydrographique, portfolio 111/2/10) published in Ly, Compagnie, opp. 284. But unfortunately none of these three maps extends to the upper Senegal.

36. Labat, , Nouvelle Relation, 2: 151–52.Google Scholar

37. Cultru, , Premier Voyage, xxix–xxxi, xlvii.Google Scholar

38. This paper represents research undertaken in the course of editing, with other scholars, the writings of John Barbot for the Hakluyt Society. Barbot visited Senegal in 1681, and the information on French activities he then collected may have been supplemented by later information from former colleagues in the French Senegal Company before it was in corporated into the manuscript text he completed in 1688. It was therefore necessary to study the course of French activities on the Senegal River in the 1680s in order to determine whether or not Barbot supplies original and additional information. In fact his 1688 text does supply a little original information on the lower and middle Senegal but almost none on the upper river, and it has no direct reference to the Falls of Félou. Barbot's later printed text of 1732 adds only a little on the upper river and most of this is clearly derived from printed sources published between 1688 and 1713, the date of his death. An informant in 1711 told Barbot about explorations far beyond “the falls that are about Galama” but nowhere in the printed text are the falls named. However, a map in the manuscript text of 1688 marks the town of “Leyde” on the frontiers of Galam, a town otherwise first mentioned by Labat. The name appears to have been inserted on the map after the manuscript was completed, but the insertion was probably made before 1700, since in the 1700s Barbot abandoned his French manuscript text in order to prepare an English version. Thus the inserted name apparently represents information about French activities in the mid-1680s that reached Barbot a decade or so later.