¹ J. Afr. Hist., XI, no. 2 (1970). The author wishes to thank Jan Vansina of the University of Wisconsin for his usual incisive and stimulating comments on this article in its early stages and for his encouragement in general, and D. H. Jones for his helpful interest at a later stage.Google Scholar

² For one example of the efficacy of the comparative approach see Introduction to Jack, GoodySuccession to High Office (Cambridge, 1966), 1–56.Google Scholar

³ The literature on this is immense, but see especially Rowton, M. B., ‘The Date of Hammurabi’, J. Near Eastern Studies, XVII (1958), 97–111,CrossRefGoogle Scholar and the latest recapitulation of the problem in Hayes, W. C., Rowton, M. B. and Stubbings, F. H. (eds.), ‘Chronology: Egypt, Western Asia, Aegean Bronze Age’, a fascicle in the revised edition of vols. I and II of the Cambridge Ancient History. This includes an extensive bibliography.Google Scholar

⁴ Simmons, D. R., ‘A New Zealand Myth: Kupe, Toi and the “fleet”’, New Zealand J. Hist. III, no. I (04 1969), 28.Google Scholar

⁵ As noted by Halbert, R. W. in Roberton, J. B. and Halbert, R. W., ‘Chronology of Maori Tradition’, Hist. Rev. (Whakatane), no. II (12 1963), 194.Google Scholar This series of short articles in the Hist. Rev., no. II (12 1963), 193–8; no. 12 (09 1964), 139–43; (12 1964), 182–4,Google Scholar together with Halbert, R. W., ‘Genealogies as a means of Maori prehistory’, idem, no. 10 (12 1962), 105–14, constitute an illuminating debate concerning the validity of relying heavily on genealogical evidence for chronological purposes.Google Scholar

⁶ Cohen, W. D., ‘A survey of interlacustrine chronology’, J. Aft. Hist., XI no. 2 (1970), chart opp. p. 190 and pp. 192–4.Google Scholar

⁷ Cohen, , ‘Survey’, 190–1, shows twenty-two of the twenty-seven dynastic units with father/son succession. Significantly those where this pattern is compromised, e.g. Buganda and Rwanda, have been the states most studied.Google Scholar

⁸ Under the succession practices in force among the Celtic states, Macbeth was not a usurper but, on the contrary, the rightful heir to the Scottish throne. Malcolm, as son of Duncan, would have been considered the least eligible contender for immediate succession to his father. However, it was at this time that the principle of filial succession was beginning to make inroads into the Celtic realms. Macbeth's successor failed to retain the throne, and Malcolm and filial succession as a principle ultimately triumphed. And so Macbeth was consigned to history as a usurper when in fact he was not.

⁹ The Pharaohs of Egypt often attempted to chisel the cartouches of their predecessors from public monuments but unaccountably ignored other records. Alan, Gardiner, Egypt of the Pharaohs (Oxford, 1961), 182.Google Scholar Korean annalists ‘took the position of flatly denying the reigns’ of rulers who had been deposed. Yasukasu, Suematsu, ‘Introduction to the Ri Dynasty Annals’, Memoirs of the Research Department of Toyo Bunko, XVII (1958), 148.Google Scholar For similar practices in early Assyria, see Hildegard, Lewy, ‘On Some Problems of Kassite and Assyrian Chronology’, Mélanges Isidore Lévy Annuaire de l'Institut de Phiologie et d'Histoire Orientales et Slaves, XIII (1953), 265.Google Scholar

¹⁰ John, A. Nyakatura, Abakama ba Bunyoro-Kitara (The Bakama of Bunyoro-Kitara), tr. by John, Rowe, in manuscript form at Midwest Center for Research Libraries, Chicago, Illinois, 55.Google Scholar

¹¹ Ibid.

¹² William, Hallo, ‘Zāriqum’, J. Near Eastern Studies, XV (1956), 219–20;Google ScholarHildegard, Lewy, ‘Assyria, c. 2600–1816 B.C.’, fascicle in revised edition of vols. I and II of Cambridge Ancient History, 20, 24;Google Scholaridem, , ‘On Some Problems’, 265, 282.Google Scholar

¹³ Margaret, Priestley and Ivor, Wilks, ‘The Ashanti kings of the eighteenth century: a revised chronology’, J. Afr. Hist. I, no. I (1960), 89–90.Google Scholar

¹⁴ Under certain circumstances interregna can occupy a significant segment of a society's time-depth. Among the Imbangala of Kasanje, only twenty-seven years of true kingship are recognized as existing between c. 1610 and 1911, because a progression of ceremonies was required before complete installation was effected and in most cases this never occurred. Information provided by Joseph, C. Miller, who conducted research among the Imbangala in 1969/1970.Google Scholar

¹⁵ Herodotus, , The Persian Wars. Bk. I, 108–30.Google Scholar Other examples from this period would include Romulus, ‘grandson’ of the penultimate ruler of Alba Longa and hence a direct descendant of the rulers of Troy by the accepted genealogy. The cases of Oedipus and Theseus may also be cited. The practice was also common in medieval Norway. For one case see Gathorne-Hardy, G. M., A Royal Impostor: King Sverre of Norway (London, 1956).Google Scholar

¹⁶ An example of this motif as a bridge between legend and history is that of Piryu, first historical king of Paekche, who, according to the later chronicles, ‘lived long amongst the people’ before ascending the throne seventy years after the death of his father. Gardiner, K. H.J., ‘Some problems concerning the founding of Paekche’, Archiv Orientélnī, XXXVII (1969), 584 n.Google Scholar

¹⁷ In Africa the lost or exiled heir returning to claim his rightful place and continue the dynasty is found twice in early Buganda and twice in Rwanda. Apolo, Kagwa, Basekabaka ba Buganda (History of the Kabakas of Buganda), tr. by John, Rowe, in typescript at Midwest Center for Research Libraries, Chicago, 10–12;Google ScholarJan, Vansina, L'évolution du royaume Rwanda dès origines à 1900 (Brussels, 1962), 49, 51, 52.Google Scholar

¹⁸ Macgowan, J., A History of China from the earliest times to the present (London, 1897), 4–17.Google Scholar

¹⁹ Boston, J. S., ‘Oral Tradition and the History of the Igala’, J. Afr. Hist., x no. 1 (1969), 29, 35.CrossRefGoogle Scholar Similarly Basham, A. L., ‘Prince Vijaya and the Aryanization of Ceylon’, Ceylon Hist. J., I (1952), 171, regards Vijaya, allegedly first ruler of the island, ‘not as an individual but a type–the bold, ruthless Aryan pioneer who was one of the elements responsible for the spread of Aryan culture all over India and beyond’.Google Scholar

²⁰ Gardiner, , ‘Some problems’, 573–6.Google Scholar See also idem, . ‘The beginnings of Korean history’ J. Orient. Soc. Australia, IV no. I (06 1966), 82, 83, 87, 89.Google Scholar

²¹ O'Rahilly, T. F., Early Irish History and Mythology (Dublin, 1957), 260–8, 409–28,Google Scholar and John, V. Kelleher, ‘Early Irish History and Pseudo-History’, Studia Hibernica, II (1963), 113–27.Google Scholar

²² Marjorie, O. Anderson, ‘The Lists of Kings’, Scot. Hist. Rev., XXVIII (1949), 10914;Google ScholarChadwick, H. M., Early Scotland: the Scots and Welsh of Southern Scotland (Cambridge, 1949), 1–17;Google ScholarElaine, Henderson, The Picts (London, 1967), 162–5.Google Scholar

²³ Shah, A. M., and Shroff, R. G., ‘The Vahīvancā Bārots of Gujarat: a caste of genealogists and mythographers’ in Milton, Singer (ed.), Traditional India: Structure and Change (Philadelphia, 1959), 51.Google Scholar

²⁴ Ibid. 46, 47, 51, 56, 64.

²⁵ Kelleher, , ‘Early Irish History’, 125–6.Google Scholar

²⁶ The rulers of Kanem, to establish a genealogical connexion with pre-Islamic Arabia, adopted the simple, if transparent, device of providing their earliest rulers with reigns up to 250, ‘some say 300’, years. Palmer, H. R., Bornu, Sahara and Sudan (London, 1936), 90.Google Scholar

²⁷ The five ‘Good Emperors’ of China were credited with a total reign period of over 600 years, doubtless reflecting the consciously epochal nature of their rule.

²⁸ Abstracted from The Ōkagami (London, 1967), 442–5.Google Scholar

²⁹ Winstedt, R. O., ‘Notes on the History of Kedah’, J. Malay Branch, Roy. Asiatic Soc., XIV, no. 3 (12 1936), 186–7;Google ScholarBastin, J. S. and Roolvink, R. (eds.), Malayan and Indonesian Studies (Oxford, 1964), Appendix B.Google Scholar

³⁰ Eva, L. R. Meyerowitz, Akan Traditions of Origin (London, 1952), 32.Google Scholar The anomalies of this list and some statistical inferences are discussed in Cohn, Flight, ‘The chronology of the Kings and Queenmothers of Bono-Manso: a revaluation of the evidence’, J. Afr. Hist., XI, no. 2 (1970), 259–67.Google Scholar

³¹ Gardiner, , ‘Some problems’, 573–4.Google Scholar For similar problems in Koguryo, another Korean kingdom of the period see idem, , ‘The Origin and Rise of the Korean Kingdom of Koguryô from the first century B.C. to A.D. 313’, Ph.D. dissertation, University of London, 1964, pp. 210, 330, 332a, 492a, 430a. For Koguryo the problems are posed by the genealogical matter in the chronicles.Google Scholar

³² On Chattisgarh, where the chronicles show thirty-three consecutive father/son successions between C. 250 and 1770 and then eight more generations to 1948, see Sinha, S., ‘State formation and Rajput myth in tribal central India’, Man in India, XLII (1962), 40. This is an extreme example but hardly a rare one.Google Scholar

³³ Thorkild, Jacobsen, The Sumerian King List (Chicago, 1966), passim, esp. tables I and II.Google Scholar

³⁴ Petrie, as late as 1923, showed a date of 5546 B.C. for the first ruler of united Egypt. Petrie, W. M. Flinders, A History of Egypt from the earliest kings to the XVI Dynasty (London, 1923), 10.Google Scholar

³⁵ Hayes, et al. ‘Chronology’, 15–17.Google Scholar

³⁶ Raybaud, L.-P. ‘L'administration du Sénégal de 1781 à 1784’, Annales Africaines (1968), 168–70.Google Scholar

³⁷ I have excluded from the listing the thirteen Ottoman Sultans who ruled from c. 1290 to 1603. This father/son succession pattern was disturbed by two abdications and reaccessions and a period (1402–1413) of collegial rule. Alderson, A. D., The Structure of the Ottoman Dynasty (Oxford, 1956).Google Scholar

³⁸ Acharyya, N. N., The History of Medieval Assam (Gauhati, 1966), 82–109;Google ScholarGait, E. A., A History of Assam (3rd rev. ed., Calcutta, 1963), 85–125.Google ScholarGait, , 305–7, speaks of the conflicting accounts in the chronicles for part of this period.Google Scholar

³³ However, none of the dynasties of ancient and early medieval India has been included. Many of these, e.g. the Hoysalas, display extended father/son succession patterns. For a convenient listing of these dynasties see Trautmann, T. R., ‘Length of Reign and Generation in Ancient India’, J. Am. Orient. Soc. LXXXIX no. 3 (1969), 574–7.Google Scholar The evidence for most of these dynasties consists of inscriptions showing ascendant genealogies for several generations. Many dynasties have been reconstructed almost solely on the basis of these kinds of inscriptions, and the genealogy has usually been interpreted as a kinglist. Doubtless many such inscriptions remain undiscovered–probably some of hitherto unknown rulers. In any event collaterals who may have had short reigns would be less likely to leave epigraphic evidence of their reigns. I would therefore argue that Trautmann's regnal average (20.5) is too high, although his generation average (27.0) would be less affected. On this problem generally, see Sen, B. C., ‘Early Indian approach to history and some problems of its reconstruction’, J. Bihar Res. Soc., LIV (1968), 13,Google Scholar and Raychaudhuri, G., ‘The Early History of the Kachwāhas of Amber’ in Bhandarkar, D. R. (ed.), B.C. Law Volume, Pt. I (Calcutta, 1945), 684.Google Scholar

⁴⁰ An example of this may be seen in a description of the succession in the Punjab state of Bashahr: ‘It is said that for 61 generations the Raja never had more than one son. For the succeeding 60 generations the Raja has had two or even three sons but as a rule only the eldest son had sons of his own. But, the direct line is likely to become extinct, for the present Raja is seventy years of age and his only son died leaving an infant son who died eleven months later.’ Punjab State Gazetteers: Simla Hill States, sub Bashahr, , 5.Google Scholar Thus we are to believe that after 121 generations of unfailing father/son succession, two successive generations became extinct within eleven months without ruling. For an almost identical description of the succession in an early Thai state, delivered with the same lack of incredulity, see Kachorn, Sukhabanij, ‘Proposed dating of the Yonok Chiengsaen Dynasty’, J. Burma Res. Soc. XLIII (1910), 61–2.Google Scholar

⁴¹ How stereotypical may be seen from a comparison of the succession in the Mossi state of Bussuma in Georges, Chéron, ‘Contribution à l'histoire du Mossi. Traditions relatives au cercle de Kaya (Haute-Volta)’, Bulletin de la Comité d'Etudes Historiques et Scientifiques de l'Afrique Occidentale Française (1924), 662–79,Google Scholar and in Junzo, Kawada, ‘Chronologie des Bussuma Naba’, Notes et Documents Voltaīques, II, no. 2 (1969), 42–54.Google Scholar Chéron shows twenty-six rulers for twenty-three generations over 400 years. Kawada shows only eight generations and only five father/son successions. Michel, Izard, Introduction à l'histoire des royaumes Mossi (2 vols., Paris, 1970), II, 240–3, accepts neither, feeling Kawada shows too few generations to accommodate a foundation date of c. 1600 for the state, but a forty year generational average where fraternal succession is common does not seem excessive.Google Scholar

⁴² Kagamé, , La notion de la génération appliquée a la généalogie dynastique et de l'histoire du Rwanda des Xe-XIe siècles à nos jours (Brussels, 1959), passim, esp. p. 45 (emphasis added).Google Scholar

⁴³ See Vansina, J., ‘The Use of Oral Tradition in African Cultural History’ in Gabel, C. and Bennett, N. R. (eds.), Reconstructing African Cultural History (Boston, 1967), 62.Google Scholar

⁴⁴ For the chronicle version see Rajputana, and Ajmer, , List of Ruling Princes, Notables and Other Personages (Delhi, 1938), 78–87.Google Scholar For the collateral successions see Raychaudhuri, , ‘Early History’, 683–94Google Scholar and Bhargava, V. S., ‘The Successor of Raja Bharmall of Amber (Jaipur)’, J. Andhra Hist. Res. Soc. XXXI (1965–1966), 51–4.Google Scholar For similar practices in the Puranic genealogies, see Basham, A. L., ‘The Average Length of the Generation and the Reign in Ancient India’ in idem, , Studies in Indian History and Culture (Calcutta, 1964), 81. Basham speculates that the Puranic kinglists omit one in seven rulers ‘owing to their being younger sons or mere upstarts’.Google Scholar

⁴⁵ Buck, , Ethnology of Mangareva (Honolulu, 1938), 357.Google Scholar For a discussion of the concept of ariki as first-born see Aarno, A. Koskinen, Ariki the First-Born: an analysis of a Polynesian chiefly title (Helsinki, 1960),Google Scholar See also Bell, F. L. S., ‘A Functional Interpretation of Inheritance and Succession in Central Polynesia’, Oceania, III (1932), 185–94.Google Scholar

⁴⁶ Caillot, A-C-E, Mythes, légendes et traditions des Polynesiens (Paris, 1914), 306–7;Google ScholarCollocott, E. E. V., ‘An Experiment in Tongan History’, J. Polynesian Soc. XXXIII (1924), 166–70;Google ScholarWood, A. H., A History and Geography of Tonga (Nuku'alofa, 1932), 65et passim;Google ScholarClaessen, H. J. M., ‘A survey of the history of Tonga: some new views’, Bijdragen tot de Taal-, Land-, en Volkenkunde, CXXIV (1968), 506;Google ScholarWilliamson, R. W., The Social and Political System of Central Polynesia (3 vols., Cambridge, 1924), I, 135–51.Google Scholar

⁴⁷ Claessen, , ‘Survey’, 516.Google Scholar

⁴⁸ Koskinen, , Ariki, 12–13.Google Scholar

⁴⁹ Of course, in this regard it is unclear how the traditions may have handled any usurpations. They may have omitted them, or they may have retroactively clothed any usurpers with the raiment of lineal legitimacy.

⁵⁰ Raymond, Firth, ‘Succession to Chieftainship in Tikopia’, Oceania, XXX (1960), 168. Significantly, most of these successions fall within the historical period.Google Scholar

⁵¹ Hans, Cory, A History of Bukoba District (Dar-es-Salaam, 1962), 53.Google Scholar

⁵² Wood, , History, 8, with emphasis added.Google Scholar

⁵³ All data abstracted from Wilhelm, Karl, Prinz, von Isenburg, Stammtafeln zur Geschichte der Europaischen Staaten (2 vols., Marburg, 1960), 1, passim.Google Scholar

⁵⁴ The Native States should provide a suitable laboratory for testing this hypothesis. During he nineteenth and twentieth centuries there were a large number of adoptions to the gadi among these principalities–a number so large that it needs more extended analysis to determine whether specific local factors accounted for it.

⁵⁵ Martin, Southwold, ‘Succession to the Throne in Buganda’ in Goody, J. R. (ed.), Succession, 104–6;Google Scholaridem, , ‘The History of a History: Royal Succession in Buganda’ in Lewis, I. M. (ed.), History and Social Anthropology (London, 1968), 140–7.Google Scholar

⁵⁶ ‘The Nipani ruler, though he had six wives, had no son’, wrote the compiler of an Indian Gazetteer. Gazetteer of Bombay Presidency, I, no. 2, History of Konkan, Dekhan and the Southern Maratha Country (Bombay, 1880), 670. Similar statements have been made elsewhere regarding lack of male heirs to polygynous rulers, and it is apparent that polygyny per se is no guarantor of male progeny.Google Scholar

⁵⁷ Eóin, Macneil, ‘Pre-Christian Kings of Tara’, J. Roy. Soc. Antiquaries of Ireland. LVII (1924), 154.Google Scholar

⁵⁸ That of Kigala, fifth and seventh kabaka of Buganda. Even here it is the result of abdication and reaccession. Kagwa, , Basekabaka, 23.Google Scholar

⁵⁹ Two analogous cases would be the Russian principalities under Mongol rule and the Shan states in Burma during the nineteenth century. Nikolas, Serge de Baumgarten, Généalogies des branches régnantes des Rurikides du XIIIe au XVe siècles (Paris, 1927),Google Scholarpassim; James, G. Scott and Hardiman, J. P. (comps.), Gazetteer of Upper Burma and the Shan States (2 vols., Rangoon, 1900–1901), s.v. individual states.Google Scholar

⁶⁰ In contrast the founders of Chinese dynasties had the following reigns: Chou 19 years; Ts'in, 12 years; Han, 9 years; E. Tam, 5 years; Sui, 23 years; T'ang, 8 years; Sung, 16 years; Yüan, 34 years; Ming, 30 years; Ts'ing, 17 years.

⁶¹ Roberton, J. B. W., ‘Genealogies as a basis for Maori chronology’, J. Polynesian Soc. LXV (1956), 50,Google Scholar accepts eighteen years as ‘a very reasonable generation length, Robert, C. Suggs, ‘Historical Traditions and Archaeology in Polynesia’, Am. Anthropologist, LXII (1960), 767,Google Scholar postulates a twenty-year average. Stokes, J. F. G., ‘New bases for Hawaiian chronology’, Ann. Report Hawaiian Hist. Soc. XLI (1933), 52–62, begins with a minimum generation length and adds ‘corrections’ for such factors as sterility, unrecorded death of eldest son, interregna, homosexuality, etc. But his minimum figure (15) is too low, and the allowance made for these factors is so small that his final figure, too, is twenty years.Google Scholar

⁶² Among the Southern Hiung-nu, six brothers ruled consecutively from 31 B.C. to A.D. 46. Wilhelm, Eberhard, ‘Chronologische Uebersicht über die Geschichte der Hunnen in der späteren Han Zeit (25 n. Chr.-220 n. Chr.)’, Türk Tarih Bulleten, IV (1940), 433.Google Scholar Among the ^cAbd-al-Wadids of Tilimsan, eight brothers ruled, but not consecutively, from 1389 to 1461. de Zambaur, E., Manuel de généalogie et de chronologie pour l'histoire de l'Islam (Hannover, 1927), 77. Among the Mossi kingdoms there are several examples of six to eight brothers ruling, if one accepts the genealogies offered. See Izard, Introduction, passim. There is an example in the German principalities of three brothers ruling over a period of eighty-eight years. The extreme possibilities here are probably too great to provide reasonably limiting bases.Google Scholar

⁶³ For African rotational systems see Goody, J. R., ‘Circulating Succession among the Gonja’ in idem, , Succession, 157–69,Google Scholar and Smith, M. G., ‘The Dynastic Chronology of the Fulani Zaria’, J. Hist. Soc. Nigeria, II, no. 2 (1961), 277–85.Google Scholar

⁶⁴ Goody, , Succession, 47,Google Scholar correctly observed that the version of the succession in Laigin that he used was ‘suspiciously perfect in form’. It should be replaced by that in John, Ryan, ‘The ancestry of St. Laurence O'Toole’, Reportorium Novum, I (1955), table opp. p. 75.Google Scholar Some other studies of Irish succession systems include Paul, Walsh, ‘Leinster states and kings’, Irish Ecclesiastical Record, LIII (1939), 57–61;Google ScholarByrne, F. J., ‘“Clann Ollaman Uaisle Emna”’, Studia Hibernica, IV (1964), 55–61;Google ScholarDonncha, O. Corráin, ‘Studies in Vest Munster History, I, The Regnal Succession in Ciarraighe Luachra, 741–1165’, J. Kerry Archaeological and Hist. Soc., 1 (1968), 46–55;Google ScholarJames, Hogan, ‘The Irish Law of Kingship, with special reference to Ailech and Cenél Eoghain’, Proc. Roy. Irish Acad. XL., sect. C, no. 3 (1932), 188–93, 247–9.Google Scholar

⁶⁵ E.g. Fraser, P. M., ‘The Tribal-Cycles of Eponymous Priests at Lindos and Kamiros’, Eranos, LI (1953), 325.Google Scholar

⁶⁶ Amadou, Wade, ‘Chronique du Wälo sénégalais (1186?–1855)’, BIFAN, XXVI (1964), 465–6;Google ScholarSmith, M. G., Government in Zazzau (London, 1960), charts at end.Google Scholar

⁶⁷ Ronald, Cohen, ‘The Bornu King Lists’ in Butler, J. (ed.), Boston University Papers on Africa. II. African History (Boston, 1966), 47;Google ScholarRuth, Berger, ‘Oral Traditions in Karagive’, paper given at East African Institute of Social Research (06 1963), 6. Fieldwork currently being conducted in the Fanti and Alem states has shown the alarming propensity of tradition to absorb printed sources, of whatever quality, as soon as they become available. The traditions are then remoulded to fit this ‘new’ information.Google Scholar

⁶⁸ Simmons, , ‘A New Zealand Myth’, 28, 30.Google Scholar

⁶⁹ James, Tod, Annals and Antiquities of Rajasthān (2 vols. in I: London, 1950),Google Scholar and Harold, Wilberforce-Bell, The History of Kathiawad from the earliest times (London, 1916), passim, esp. genealogical tables at end.Google Scholar

⁷⁰ Jeifreys, M. D. W., ‘Some Historical Notes on the Ntem’, J. Hist. Soc. Nigeria, II, no. 2 (1961), 260–76; [?], ‘The Central Corridor Area [of Tanzania]’, paper given at SOAS Conference on African Chronology, p. 6.Google Scholar

⁷¹ Dorothy, B. Barrère, ‘Cosmogenic Genealogies from Hawaii’, J. Polynesian Soc. LXX (1961), 419–28;Google Scholaridem, , ‘Revisions and Adulterations in Polynesian Creation Myths’ in Genevieve, A. Highland et al. (eds.), Polynesian Culture History (Honolulu, 1967), 103–17.Google Scholar

⁷² The Salum kinglist, based on rainy seasons, is supported in every instance where European records are available. Jean, Boulègue, ‘Contribution a la chronologie du royaume de Saloum’, BIFAN, XXVIII (1966), 657–62.Google Scholar The Mandan Indians of North America counted by winters and made drawings on buffalo hides each year for mnemonic purposes. A forty-three-year sample of their recorded history has shown this method to be basically accurate. Howard, J. H., ‘Butterfly's Mandan Winter Count, 1833–1876’, Ethnohistory, VII (1960), 28–43;CrossRefGoogle ScholarSmith, M. W., ‘Mandan “history” as reflected in Butterfly's winter count’, idem. 199–205.Google Scholar

⁷³ See above, pp. 312–3.

⁷⁴ Katate, A. G., Abagabe wa Ankole (Abagabe of Ankole), tr. by John, Rowe, in typescript at Midwest Center for Research Libraries, Chicago, p. 80, with emphasis added.Google Scholar

⁷⁵ See Richard, Gray, ‘Eclipse Maps’, J. Afr. Hist., VI (1965), 251–62;Google Scholaridem, . ‘Annular Eclipse Maps’, J. Aft. Hist., IX (1968), 547–57.Google Scholar

⁷⁶ For references to the ‘Budomel’, that is, the Damel of Kayor, between 1455 and 1529, well before the traditional date for the creation of the title, see Alvise, Ca da Mosto, Viagens (Lisbon, 1950), 34–9,Google Scholar and Jean, Boulègue and Benjamin, Pinto-Bull, ‘Les relations de Cayor avec la Portugal dans la première moitié du XVIe siècle d' après deux documents nouveaux’, BIFAN, XXVII (1966), 663–4.Google Scholar Compare with Félix, Brigaud, Histoire traditio, mejle du Sénégal (Saint-Louis, 1962), 85, 93 n.Google Scholar For Muslim references to the dissolution of the Wolof empire, see ^cAbd-al-Rahmān, aṣ-ṣa^cdi (tr. Ch., Houdas), Tarikh es-Soudan (Paris, 1964), 127.Google Scholar

⁷⁷ Sékéné, Mody Cissoko, ‘Civilisation Wolof-Serère’, Présence Africaine (1967), 121–44.Google Scholar

⁷⁸ For the Kongo, Jan, Vansina, Kingdoms of the Savanna (Madison, 1966);Google Scholar for Ankole, , Oberg, K., ‘The Kingdom of Ankole in Uganda’ in Fortes, M. and Evans-Pritchard, E. E. (eds.), African Political Systems (London, 1940), 47.Google Scholar

⁷⁹ E.g. Priestley, and Wilks, , ‘Ashanti kings’, 87.Google Scholar

⁸⁰ Quoted in James, Murdoch, A History of Japan (3 vols. London, 1903–1926), 11,56.Google Scholar

⁸¹ For examples see Donald, Ferguson (ed.), ‘The History of Ceylon from the earliest times to A.D. 1600 as related by João de Barros and Diogo do Couto’, J. Roy. Asiatic Soc. Ceylon Branch, XX (1908), 24, 29, 38, 43.Google Scholar

⁸² Paranavitana, S., ‘The Emperor of Ceylon at the time of the arrival of the Portuguese in 1505’, Univ. Ceylon Rev. XIX (1961), 28–30.Google Scholar For another interpretation see Somaratna, G. P. V., ‘Political History of the Kingdom of Kottē (c. 1400–1521)’, Ph.D. thesis, University of London, 1969, p. 461.Google Scholar