Our study of caudatum sorghums in Africa and the ethnography, history, and archaeology of regions where caudatum sorghums are grown has led us to the following conclusions.

The distributions of caudatum sorghums and Chari-Nile-speaking peoples coincide so closely that a causal relationship seems probable.

Caudatum sorghums appear to be a product of selection by Chari-Nile-speaking peoples for a hardy crop which will produce a large amount of grain under adverse conditions with a minimum amount of care. Caudatums can produce grain in spite of either droughts or heavy rains during the growing season and can produce grain in places where no other indigenous African crop can grow.

Caudatum may be a relatively new race of sorghum. Available evidence indicates it was developed some time after about 350 A.D. and before about 900 A.D., the date for caudatum found at Daima mound, south of Lake Chad.

By about 1000 A.D. caudatum was probably being grown in many parts of the savanna belt stretching from the eastern shores of Lake Chad to southwestern Ethiopia. This is assuming that the evidence of caudatum culture at Daima is indicative of a general phenomenon occurring in a region with close intercultural contacts.

The large number of varieties of caudatum in the Republic of the Sudan suggests that caudatum sorghums have been grown in the Sudan for a relatively long time and may have been developed there.

A caudatum and cattle-based economy has probably been an important factor in making possible the present level of occupation of regions now inhabited by Chari-Nile-speaking peoples.

On the basis of oral histories and archaeological remains of people in northern East Africa, we have concluded that caudatum-based agriculture was probably introduced to Uganda and Kenya no later than the period from 1500–1800 A.D., when the Luo occupied the areas in which they live at the present. However, it is not unlikely that caudatums were introduced more than once as various Chari-Nile-speaking groups made their way into East Africa.

The presence of the ‘Kisra legend’ in certain western Sudanic societies has long puzzled historians and anthropologists. Attempts by many to explain the phenomenon have been seen as unsatisfactory. In Section I of this study we noted the fact that the Arabic Kasra or Kesra, having been derived from the title of one or the other of two Persian kings of the sixth and seventh centuries, denotes, in von Grunebaum's phrase, ‘a truly royal style of life’. The profound influences of Perso-Arabic elements on many cultures of the southern and western Sudan, even before the spread of Islam in these areas, strongly suggests the possibility that, rather than by any specific migration, the idea of ‘Kisra’ was borne across the Sahara, to the areas where it took root in the form of the Kisra legends. When the geographical situation of those societies having fully-developed Kisra legends is considered, noting that the most detailed and strongly held legends obtain among societies who were constantly threatened by others who were recognized as technologically, and possibly felt as culturally, superior, and among whom the Kisra idea also existed, the origins and distribution of such legends becomes more plausibly explainable. It has been suggested that, through a selective altering of historical tradition, over time, societies who felt so threatened were able to (1) assert their equality to, if not superiority over, the threatening power; (2) justify their successful maintenance of independence in spite of this threat; and/or (3) thus re-establish a basis for societal unity.

Nzinga of Matamba, the seventeenth-century African monarch known primarily for her enmity to the Portuguese in Angola, also faced hostility from her own Mbundu people and the opposition of neighbouring African rulers throughout her long career. Her sex disqualified her from many Mbundu political offices reserved for males, and her origins in the lineageless community at the Mbundu king's royal court made her an outsider in terms of the lineage politics of most Mbundu states. But she overcame these disadvantages by skilful manipulation of the aliens present on the Mbundu borders, Imbangala warrior bands, the Portuguese, and the Dutch, and dominated Mbundu politics and diplomacy until her death in 1663. The domestic forces arrayed against Nzinga triumphed after her death, expelling her chosen successors from the Matamba royal title and omitting her name from the oral traditions of the state. These hypotheses, while not susceptible to direct proof, seem probable on the basis of a re-reading of documentary sources in the light of ethnographic and oral historical evidence collected in 1969–1970.

A drastic decline of aboriginal populations has been an important by-product of European commercial and imperial expansion in many parts of the world. Europeans have been aware of this phenomenon from the early days of Spanish exploration and conquest when Fray Bartolome de las Casas deplored the rapid collapse of Indian populations in the Caribbean. The problem still seemed acute in the nineteenth century. For example, the young Charles Darwin, describing the decline of the Australian Aborigines, placed most of the blame on alcoholism, introduced diseases, and the loss of hunting grounds. But these factors did not fully explain what appeared to be the impending extinction of a variety of peoples. ‘Besides these several evident causes of destruction,’ he wrote,

there appears to be some more mysterious agency at work. Wherever the European has trod, death seems to pursue the aboriginal. We may look to the wide extent of the Americas, Polynesia, the Cape of Good Hope, and Australia, and we find the same result.

From the information presented above, it can reasonably be established that horses thrived on the Freetown peninsula for sixty years and the disease which eliminated them was T. brucei. This disease was prevalent for at least 15 years although it eventually disappeared, probably before 1900. In addition, clearance of the forest, in three overlapping phases, was carried out prior to and during the outbreak.

From the ecology of the three species of tsetse fly possibly implicated in the outbreak of the disease in Sierra Leone, it can be deduced that deforestation and the consequent regeneration and cultivation led to favourable conditions for the expansion of at least G. palpalis and G. longipalpis, and possibly also of G. morsitans. Taking into account the information currently available on the inter-relationships between tsetse fly and trypanosome species, it can also be suggested that G. longipalpis was probably the primary source of the disease but that the disease was sustained, particularly around Freetown itself, by G. palpalis and possibly by other biting flies. That G. longipalpis or even G. morsitans probably initiated the outbreak makes the eventual disappearance of the disease more understandable since both of these species relied upon regenerated forest and an abundance of game, resources which diminished as agriculture progressed. It is likely that these primary carriers infected horses and cattle moving between the provinces and Freetown, thereby providing the link in spreading the disease to the Colony.

It appears therefore that the removal of the forest led to the outbreak of trypanosomiasis in horses. A different and more rational approach to forestry in Sierra Leone would have avoided this altogether. In fact it is fortunate that the consequences of this indiscriminate deforestation were not more serious since all three tsetse flies in question are carriers of the human sleeping sickness, T. gambiense. For some reason, however, this species of trypanosome, whilst common in the interior, is not prevalent in the coastal areas of West Africa, though there was in fact, an epidemic of this disease in the 1930s in the Kaiahun area, associated with G. palpalis.

Since the horse is relatively sensitive to tsetse fly and its associated trypanosomes, this animal is a good indicator of the presence of trypanosomiasis. Dr McCoy in his memorandum mentions that horses thrive in the Gambia and at Lagos and that ‘loin disease’ was unknown in these places. He makes no reference, however, to horses in Ghana. The history of the horse in other parts of West Africa may therefore be worth pursuing in this respect.

The history of the London-based Natal Land and Colonisation Company is explored against the background of the evolving political economy of rural Natal. In the early years of the colony, white-controlled farming operations consistently failed. The landholdings of bankrupt colonists passed into the hands of a small group of men with capital. In 1861 this group activated its links with financiers in Britain to float the Natal Land and Colonisation Company. The Company ‘bought’ 250,000 acres of surplus lands from them in return for an injection of metropolitan capital into productive operations to be carried out on the remaining mainly coastal lands, or into further speculative activity. In fact, white-controlled farming activity in the interior continued to stagnate. Money which the Company loaned to white farmers in the 1860s, secured as mortgages on their farms, was not repaid, and the Company took over the lands of the bankrupt until in 1874 it controlled 657,000 acres in Natal. Anxious for a sizeable and more reliable source of income, the Company, in common with some colonists, concentrated on extracting rent from Africans, as yet the only successful farming population of the Natal interior. The increasing importance of this source of income to the Company was rudely interrupted in the 1890s by a fundamental shift in the Natal political economy. New mining centres in South Africa looked to Natal to furnish some of their needs for raw material and labour. The balance of economic and political forces favoured those who demanded labour, not rent, from Natal Africans. The Company switched its capital in good time out of renting land to African farmers and into renting and property development in the growing urban areas of white South Africa. Its properties were brought within the empire of the Eagle Star Insurance Company in 1948.

This paper argues that interpretations which would view pre-colonial Akan political life as ‘normative’ and structured may be incorrect, at least in so far as stool succession is concerned. Contemporaneous evidence for this early period is at best sparse and at worst simply non-existent and seldom allows even tentative hypotheses. Rather, it is necessary to infer past practices from more recent data, whether this be observation of present behaviour or recent testimony about the past. In this case I have used the testimony presented at various stool and jurisdictional disputes during the colonial period for which records survive. These are generally, of course, ex parte statements and can be used only with caution. However, there is a surprising consensus throughout these records that both the principles and the patterns of stool succession and paramountcy in the pre-colonial period were variegated and even extemporaneous although, not surprisingly, there is much dispute about the reasons for this. On balance, this testimony suggests that a re-interpretation of early Akan political culture using a wider range of evidence is desirable.

Although this implies that the impact of colonial ‘indirect’ rule was not as profound as has often been supposed, I have not discussed this problem directly except as it bears on the quality of the data. Nor have I attempted to analyse the day-to-day dynamics of political life, either for the earlier period (which would be impossible on the evidence) or for the colonial period (which would be irrelevant for comparison). Nevertheless, within the restricted compass of stool succession and paramountcy the argument here is that colonial rule involved little fundamental change from earlier practices. If anything, it probably served to delimit a greater range of previous options by seeking to codify them.