¹ The writer owes much to Dr. S. M. Stern for help with points of reading and interpretation, and for verifying transeripts of the Arabic texts. Malam Haliru Binji of Zaria has provided many essential introductions, and contributed several valuable suggestions to the translation. Malam Audu Mahmud Bauchi, of the Nigerian Antiquities Department, and Malam Abubakar Katsina, of Jos, both rendered much patient assistance in the initial examination and selection of documents, and the study of the present texts.

² It is hoped eventually to replace the glass by sheets of unplasticized Perspex, when this material can be obtained. This would not only give better protection against accidental breakage, but also reduce any risk of condensation.

³ Lagos, Government Printer, 1927.

¹ London, 1936; plate VIII, facing p. 62, with a translation at p. 69. It should perhaps be noticed here that none of the ‘Bornu Mahrams’ edited by Sir Richmond Palmer in the work under reference has the appearance of being an original document, even the interesting ‘Mahram “M”’; and that in particular none seems to bear an official seal, though several have the appearance of being correct copies of early originals.

² In Hausa. Published by the North Regional Literature Agency, Zaria, Northern Nigeria, 1956. The writer greatly regrets that his insufficient acquaintance with the Hausa language has prevented his deriving full benefit from this work.

³ ‘Land tenure during the time of the Fung’, Kush, IV, 1956, 48 ff.

⁴ The Persian aspect of certain Fung calligraphy will be further discussed at a later date; Eastern styles seem to have become fashionable in eighteenth- and nineteenth-century Egypt, and may have reached the Nilotic Sudan in this way.

⁵ VIII, 116: ‘And it was on a square sheet of paper, each line being close to the next (al-satruilā jānibi’l-satri), in Maghribi writing, and there was no margin either at the top, or at the side; the text was concluded on the reverse, starting from the foot of the page’.

⁶ Kāti, Mahmūd, Tarikh al-fettach (ed. Houdas and Delafosse, Paris, 1913), translation, p. 139.Google Scholar

⁷ Ed. C. E. J. Whitting, London, 1951, 124–66. This work will henceforth be quoted as ‘Whitting’. Also important for the study of this material is Arnett, E. J., The rise of the Sokoto Fvlani, Kano, 1922, 102–20, which will be quoted as ‘Arnett’, and which provides a paraphrase of the Infāq which is most vauable, especially since Whitting's edition has no systematic division into chapters, and is far from easy to use.Google Scholar

¹ Tisserant, E. and Wiet, G., ‘Une Iettre de l'Almohade Murtad·'’, Hesperis, VI, 1926, 27–53Google Scholar; M. A. Alarcon y Santon and R. Garcia de Linares, Los documentos árabea diplomáticos del Archivo de la Corona de Aragón, Madrid, 1940; E. Lévi-Provencal, ‘Un recueil de lettres officielles almohades’, Hesperis, XXVIII, 1941, 1Google Scholar; Nehlil, M., Lettres chérifiennes, Paris, 1915.Google Scholar

² Some account of the Waziri al-Bukhārī will be found in The occupation of Hausaland, preface (see p. 324, n. 3). He is said by Arnett, Sokoto gazetteer, p. 9, to be the author of aa historical work entitled Ta'nīs al-ikhwān.

³ W. E. N. Kensdale, A catalogue of the Arabie manuscripts preserved in the University Library, Ibadan, Nigeria, Ibadan, 1955–1958, pp. 19–20, and especially nos. 18, 19, and 20, which may be original copies of the author's time, and possibly autographs.

¹ This watermark is mentioned by M. Hiskett, ‘Material relating to the state of learning among the Fulani before their Jihād’, BSOAS, xix, 3, 1957, 550; cf. Heawood, E., Watermarks, I, no. 870 ff. (Veniee c. 1725); Labarre, Dictionary and eneyclopaedia of paper and papermaking, p. 342, figs. 48 and 130.Google Scholar

² See below, p. 329.

³ Encyclopaedia of Islam, new edition, s.v. 'alāma.

¹ ‘Essaie sur l'écriture magrebine’, Publications de l'École des Langues Orientales Vivantes, Paris, 1886, 85–115.

¹ Since a great number of fine examples of the ‘Andalusian’ script were in fact written at Fez, confusion of terminology needs to be avoided.

² It may be notieed that the angular Nigerian book-hands of the nineteenth-twentieth centuries, to which the name of Ajami is applied by Hiskett (BSOAS, XIX, 3, 1957, 550), seem to derive from the Bornu seript.

³ The earliest mark of authentication in the Maghrib appears to have been the calligraphic 'alāma and the use of such formulae as States of Turkish origin have all used the tughrā. The Persian Safavids used an ink seal at the head, but this seal was flame-shaped on formal documents. Other dynasties employing seals generally placed them below the text.

¹ I translate as the Arabie ‘your Gate’, ‘your court’, but it is possible that we have here some vernacular word, or literal rendering of a vernacular expression.

¹ Gowers, W. F., Gazetteer of Kano Province, p. 25. The site of Tumbi is now ineluded in the French-administered Territoire du Niger.Google Scholar

² II, 165; V, 374; Tumbi, V, 376.

³ For this reason we can probably exclude any connexion with the town of Kakali or Kakala, mentioned by Barth, III, 193 ff., as a venue of slave-raids. Kakali was situated south of Dikwa, in the territory of the Musgus, and presumably far from Dan Tanoma's sphere.

⁴ Sarkin Musulmi is the usual Hausa designation of the Sultans of Sokoto, corresponding to the Arabic usage of amīru ‘l-mu’minīn.

⁵ Whitting, p. 157 (= Arnett, p. 112), Presumably these words are quoted from the Kānemi dispatch, though the division of the text is not clear in Whitting's edition.

¹ See n. 5, p. 331.

¹ Gowers, W. F., Gazetteer of Kano Province, London, 1921, 31. The Sokoto archives contain a number of letters from Muhammad al-Hājj, and it is not unlikely that these, when studied, may give some further information about the incident in question.Google Scholar

² Shehu, the form used in Hausa for the Arabie title shaikh . was a specific title of the Kānemi rulers in Bornu, and that by which they are most generally known to-day. The same title is applied to the Fulani leader Usuman Dan Fodio. It is difficult to be entirely con-sistent in the present context with the use of Arabie and vernaeular forms of proper names, but I have tried to retain Arabie forms in transliteration. The question of transliteration is further complicated by the fact that in Nigerian usage the Arabie inflections are pronounced, though in European practice they are commonly not transeribed.

¹ sic legi.

¹ Owing to the break in the paper which comes aeross this line, the reading is doubtful in one or two places. It seems obvious that the Sultan mentioned must be Tanimu of Damagaram, of whose name there are variant spellings, as there are for other rulers of Damagaram. I am inclined to think that the form used here may be Tanīmī which I have not seen in other documents, but which seems a possible variant. The last letter of the word in question herc can scarcely be mim, of which, elsewhere on this page, the final stroke is nearly vertical, though it could eonceivably be nūn. In any event, no other contemporary name seems to fit the require-ments as well.

² W. F. Gowers, Gazetteer of Kano Province, 32.

¹ It is interesting to notice that Abdallahi bin Fūdī, the brother of 'Uthmàn, whose writing is of the type in question, used the Mashriqi pointing of qāf but not of fā.

¹ It is interesting to notice that Abdallahi bin Fūdī, the brother of 'Uthman, whose writing is of the type in question, used the Mashriqi pointing of qāf but not of fā.

¹ Two letters have been lost, but the text is supplied by analogy (see commentary below).

¹ A word is deleted.

² A line of text is deleted, and in tne margin is added the word ‘explanation’, followed by the words inserted between the obeli.

³ Qur'ān, ix, 92–3. For we might expect .

⁴ Qur'ān, IX, 124.

⁵ Qur'ān, 133.

⁶ A hadiih which has been frequently attested with minor variants, cf. Wensinck, Concordance de la tradition musulman s.v..

¹ This title is no. 64 in the list of Bello's works published by W. E. N. Kensdale, ‘Arabio literatures of the Western Sudan’, JRAS 1958, 56. See also the same author's Catalogue of the Arabie manuscripts preserved in the University Library, Ibadan, Nigeria, p. 20, no. 18.

¹ The references to the Infāq are as follows: (a) Whitting, Infāq al-maisūr, p. 128 (= Arnett, Rise of the Sokoto Fulani, p. 103); (b) Whitting, p. 135 (= Arnett, p. 108); (c) Whitting, p. 139 (= Arnett, p. 109); (d) Whitting, p. 142 (= Arnett, p. 110); (e) Whitting, p. 161 (= Arnett, p. 113). These concordances are given because the contexts are not easy to find, but it should be noted that the texts of the two editions do not entirely correspond.

² Whitting, p. 180 (= Arnett, p. 122).

¹ Gazetteer of place-names on the map of Nigeria, Zaria, 1949, p. 18, s.v. Banaga.

² Whitting, p. 98, who, however, reads (Yānāgha). We should probably prefer to follow Arnett, p. 80, in reading ‘Banaga’, though it must be noted that there are other discrepancies between the two editions at this point.

³ Arnett, E. J., Gazetteer of Sokoto Province, London, 1920, p. 31: ‘In 1820, Gunki, Sarkin Gobir, was killed in battle at Kadaye. Bello unsuccessfully attacked Bakura, and the Dakakerri Country’.Google Scholar

⁴ idem, p. 8. I first learnt of this interesting circumstance in conversation with Mr. Adri·an Webb and Malam 'Umaru Jabka at Maru Teacher Training Centre.

¹ e.g. Bibliothėque Nationale No. 2334, a Persian firmān of Ahmad Jalair dated Tabriz, 773/1372, of which the entire heading is lost, and which commences only with the matter, as is clear from photographs in my possession; cf. H. Massé, ‘Ordonnance rendu par le Prince Ilkhanien Ahmad Jalair’, JA, CCXXX, 1938, 465. Also BM Or. 4934 (iii).

¹ The Pasha does not adopt the Nigerian spelling . a It is assumed here that a word has fallen out of the text before

³ Read

¹ The text suggests a corruption at this point. The expression is not strictly parallel with and this unnaturalness suggests that there may be need for an emendation.

² Read .

³ Dr. Stern tells me that this unclassical form is a typical Maghribi colloquial usage for .

⁴ The copyist has deleted the definite article of as the grammar requires.

⁵ The translator has some doubt whether the syntax is correctly followed at this point.

¹ The unfamiliar word . is not Arabic. Its meaning is not clear, but a suggestion (originating with Malara Audu Mahmud) is that it may be a Kanuri phrase, mai gana ‘small king’, Arabized as a single word. Alternatively, Mr. D. Cowan has suggested a connexion with the Moroccan Arabic word qināwī ‘Guinean’.

² ‘Brother-in-law’ is also an admissible translation. The relationship of this personage to Sultan Bello will no doubt be precisely established as more documents become available.

³ i.e. ‘Thank you for the gift’.

⁴ Strictly, the text would signify that the firearms had arrived, presumably from Sokoto, in the same convoy as the slaves and the emissary Mādī. Yet the sense seems to require reference rather to a return present from the Pasha to Sultan Bello, since there is an implication that the firearms are to reach Bello. It is intrinsically improbable that Bello would send a present of firearms to Tripoli, where such products were more easily obtainable, and less valued. I therefore understand the text as translated above.

⁶ The Ottoman title of rank, ‘Commander of a Standard’ ‘Sanjakbay’.

⁶ i.e. of God.

¹ Rossi, E., ‘Per la storia della penetrazione turca nell' interno della Libia’ Oriente Moderno IX, 1929, 154.Google Scholar

² Encyc. of Islam s.v. Karamanli.