The sources for pre-Arab trans-Saharan contacts are poor, but at least for the central Sahara a picture can be made out. The alignment of rock paintings and engravings of chariots along two trans-Saharari routes has been supposed to prove regular traffic across the desert. The inference is unjustified, but literary and archaeological sources indicate that the conclusion is correct. Herodotus attests the use of a route running west from Egypt to the Fezzan, then apparently south-west via Tassili and Hoggar to the Niger. This corresponds with the central Saharan ‘chariot-route’.

There was also a route to the Garamantes of the Fezzan from the Punic settlements on the coast of Tripolitania. Carthage imported from the Garamantes the precious stones known as ‘carbuncles’, which were apparently brought to the Fezzan from the south-west. Other possible imports are slaves and gold. Carthage imported gold from West Africa by sea, and it seems likely that her explorations down the coast were inspired by an overland trade in gold. But there is no direct evidence for such a trade.

In the second century B.C. Rome replaced Carthage in control of the coast of Tripolitania. Between 20 B.C. and A.D. 86 she fought a series of wars with the Garamantes. Later friendly relations were established, but further trouble led to the organization of the ‘limes Tripolitanus’ after A.D. 201. Trade is attested by imported Roman material in tombs of the Fezzan dating from the late first to the fourth centuries. There is evidence that the Romans imported ivory from the Garamantes, and slaves are now attested directly.

The commodities exported north by the Garamantes came not from the Fezzan, but from farther south. Literary sources refer to hunting expeditions and raids to the south, and finds of Roman material have been made along the ‘chariot-route’ south-west of the Fezzan as far as Ti-m-Missao.

Trade ended with the collapse of Roman rule in North Africa. It was revived with the Byzantine reconquest after A.D. 533, and Christianity penetrated to the Fezzan. In 666 the Arabs overran the Fezzan.

This paper describes a large area of stone-built ruins in northern Tanzania which has so far only been briefly excavated, but which is likely to prove to be a key site in the study of the Iron Age in East Africa. In addition to numerous massive stone circles, terraces and cairns, there are extensive systems of fields and enclosures defined with lines of stones. Excavations carried out in 1964 and 1966 have shown that the small terrace-platforms on the hillsides and the stone circles on the flatter land in the valley were occupied at different periods and by different peoples whose pottery is readily distinguishable. Radiocarbon dates suggest that the terrace sites on the hillsides were occupied during the first millennium A.D., and that the stone circles on the lower slopes in the valley were occupied during the fifteenth century A.D. The purpose of the numerous large and well-built cairns is not yet known, but it appears that they were not burial monuments. No evidence has been found that any of the stone structures were built or occupied by immigrants from outside Africa.

It has not yet been possible to link the systems of fields and enclosures to the hillside terrace-platforms or to the stone circles. A close examination of the main area of fields and of low-level aerial photographs has not produced any evidence that the fields were irrigated, a fact which raises important agricultural and climatic problems in an area which has an average rainfall of less than 380 mm. (15 inches).

The general picture of Engaruka which emerges is of an area which was occupied by different peoples at different times over a period of at least a thousand years. The stone structures which these different peoples built have accumulated to give the impression that there was once a very large population living in the area; in fact it is possible that this population was always less than 4,000 people at any one time.

As a contribution to an already considerable historiography, it is suggested here that the Mane invaders of Sierra Leone comprised two principal elements —a ruling élite originating in the southern section of the Mande world of the Western Sudan, and numerical forces drawn from the area around Cape Mount. The first stage of movement took place in the first half of the sixteenth century, carrying Mande clans to the Liberian coast from the region around Beyla and perhaps even from the hinterland of modem Ghana. There then followed a number of incursions into Sierra Leone during the third quarter of the sixteenth century.

A new ruling class was established in Sierra Leone and adjacent regions, bringing increased exploitation and causing the destruction of the indigenous ivory-carving skills. On the other hand, the Manes brought improved military techniques and advances in the manufacture of iron and cloth. They also profoundly influenced religious and social patterns, particularly with respect to the secret societies of the area.

An inventory of ethnolinguistic units on the Guinea coast can be drawn from early written sources, that is, from Portuguese and other European records of between 1440 and 1700. When this inventory is compared with the present-day inventory it is found that, in the particulars cited, the units have remained very much the same for three, four, or five centuries. Summary evidence relating to the coast, section by section, from the Senegal River to the Cameroons River, is presented, and this includes reference to the linguistic evidence provided by early vocabularies. Not only do all earlier units correspond to present-day units, but the sequence of units along the coast is the same in the earlier as in the present-day inventory. However, some of the units have expanded or contracted; and one of the modern units (Mende) is not recorded before 1700. It is finally suggested that research into the documented period of continuity, through study of the written records, should precede attempts to evaluate the accounts of Völkerwanderungen supplied generously in oral traditions.

The East African ivory trade is an ancient one: East African ivory is soft ivory and is ideal for carving, and was always in great demand. It figures prominently in the earliest reference to trading activities on the East African Coast. But the great development came in the nineteenth century when an increased demand for ivory in America and Europe coincided with the opening up of East Africa by Arab traders and European explorers. The onslaught on the ivory resources of the interior took the form of a two-way thrust—from the north by the Egyptians who penetrated into the Sudan and Equatoria, and by the Arabs from the east coast of Africa. The establishment of European protectorates and a settled administration in the 1890s ended this exploitation.

During the nineteenth century ivory over-topped all rivals in trade value— even slaves. The uses of ivory were wide and novel—it played the same part in the nineteenth century as do plastics in the mid-twentieth—but it was always a much more expensive article.

The Sanford Exploring Expedition was Henry Shelton Sanford's main financial interest in Africa. Sanford established the Expedition in 1886 to foster trade development along the upper Congo. Possessing special State concessions granted by Leopold II of Belgium, Sanford's company founded settlements at Kinshasa, Matadi, South Mancagues, Luebe and Equator, and brought the first commercial steamer over the cataracts to Stanley Pool. This steamer, the Florida, proved of little value, yet a second steamer, the New York, was purchased in 1887. Lt. Emory H. Taunt was the principal company agent, the administrator in Africa; William S. Parminter was second in authority. Delays caused by damages to the Florida and hardships in establishing trading settlements on the Congo drained the company's capital. In 1888 Sanford tried to interest various American businessmen—A. H. Alden, George H. Alden, Charles B. Flint, John D. Rockefeller, H. M. Flagler and others—in investing in the expedition. His attempt failed, and after brief negotiations with an English financier, M. Rilneff, the Sanford board made agreements with a Belgian firm. In December 1889 the Sanford Exploring Expedition merged with a Belgian organization, becoming the Société Anonyme Beige pour i'Industrie et Commerce du Haut Congo. Sanford retained a minor interest in the company, but his stock was sold by his widow shortly after his death.

The French government, in the early part of this period, from 1854 to the turn of the century, did not have a consistent or systematic ‘Islamic policy’ for ts colonial possessions. There were, however, certain patterns of administration which, quite unintentionally, gave a new impetus to the spread of Islam in Vest Africa. The first section of the article deals with this period, when the basis for later policy was laid but when policy was not yet systematically articulated.

The creation of the Service des Affaires Musulmanes et Sahariennes in Paris in 1900, and of the Service des Affaires Musulmanes in Dakar in 1906, together with the works of scholar-administrators such as Le Chatelier, Arnaud and Marty, marks the definition of a general policy towards Islam in colonial territories. This policy was aimed, in particular, to secure the loyalty of the Muslim notables, and to use them as intermediaries and tools of administration.

In a final section, the article deals with the renewed fear of Islam which affected France, with the intensification of pan-Islamic propaganda from Turkey immediately before the First World War, and with the change in policy which resulted. The outbreak of war, which enabled the Muslim élite to demonstrate its real loyalty to France, provided, however, a final reassurance.

The term ‘balkanization’, as applied to colonial policy in Africa, frequently suggests a European ‘divide and rule’ policy, intended to fragment pre-existing African unity. An examination of the policy of France towards the former federation of French West Africa indicates that the term is inaccurate to describe French policy.

The federation was created by the French for the French. It served to streamline French administration during the period of expansion, to help develop the component territories' economies, and to guarantee the security of French private investments. France conceived of the territories as the basic political units and hence modern political life was implanted there rather than in the federation. Post World War II French public investments flowed chiefly to the territories, common federal services were decentralized as territories acquired expertise and funds to run their own, and the quasi-federal Senate called the Grand Conseil was allowed to expire. Other aspects of French policy had unintended centrifugal effects, such as the metropolitan party structure and consequent dispersal of African representatives in Paris, and the influence of some French parties in Africa. Since 1956, France logically responded to African leaders' demands for more power and eventual autonomy in the territories, and left it to the Africans to decide whether or not to continue their federal relationship.

By 1956, the federation had outlived its usefulness to France. It is improbable that the metropole could have ‘saved’ it, because political territorial roots were too strong, the Africans were not agreed, and relations between France and Africa had already become mainly bilateral with the territories directly. Consequently it appears that the federation, a purely French creation, was simply permitted to fade away once its functions were no longer relevant to French needs. It is true that France preferred eight small, powerless states to a strong federation when national independence approached, but ‘powerless’ and ‘strong’ in this context are but relative terms. The story of the federation of French West Africa suggests that such political structures can survive the colonial period only if they are anchored in fundamental African needs; this was not the case with that federation.