The text and plates of this volume have been prepared with the primary object of identification of the trees of the savannah forests of Northern Nigeria. The volume is, in no sense, a Flora, and no attempt has been made to include all the species found in the region under survey. Trees, being distributed over a large number of Families of the plant kingdom, are a small proportion of the total number of flowering plants, and their systematic arrangement would not, except in a few cases, be an aid to identification. Some 30 Families are represented by the 120 species included, in one of which there are as many as 36 species. The botanical names are arranged, therefore, in their alphabetical order so as to group certain genera and species whose similarity aids identification and to make the volume, in its index form, handier for reference.

The text is arranged in paragraphs, the first giving a general description of the form, height, girth, habitat, locality, &c., successive paragraphs describing in detail the bark and slash, thorns, wood, leaves, flowers, fruits and seeds, with a few notes on uses where these are known. These paragraphs are in uniform order for rapid reference.

The text is supplemented by drawings of the flower, leaf, thorn, fruit and seed, sketched from the living specimens. Wherever possible the drawings are actual size to avoid calculations from enlarged or reduced figures, but a number have been enlarged to show structures, or reduced owing to their actual size. In several cases sections of various parts have been made to illustrate peculiarities of structure, a pocket lens being used for this purpose, though practically all the parts drawn are visible to the unaided eye.

The majority of the botanical names have been verified through the Herbarium of the Royal Botanic Gardens at Kew.

In describing the woods, few of which are in use and many of which are little likely to be of use, an attempt has been made to strike a path midway between the complete technical and botanical description employed with regard to timbers and a vague statement as to colour, quality, hardness and grain which helps little to distinguish the individual from others like it. Most of the woods are described from samples collected by the author and shaped by the Sokoto Arts and Crafts School. I am indebted for these, and for the facilities offered for observing their behaviour under tools, to Mr. W.E. Nicholson. The dry weight of seasoned wood is given in all cases and in this connection it may be noted that a number of species produce heavier wood in the north than they do in the south.

A table of flowering seasons has been appended, with a few notes on its significance.

A second appendix arranges the 120 species under their Families. The Genera Plantarum of Bentham and Hooker has been followed.

Finally, there is an index of Hausa names for reference to the plates and text, whose numbers are given in the index. A blank column has been left in the index for the benefit of those who may desire to collect either additional Hausa names or the nomenclature of other tribes. It is hoped that the scope of the volume will be enlarged thereby, so as to cover other parts of Nigeria where the trees are the same but the language different.

The botanical name, being universal, is always given preference. The native names, variable and unreliable as they often are, should be a secondary consideration, their importance resting in their being a means of communication with the native rather than a short cut to the botanical names. In a country where there are several native names for one tree or one native name for several distinct species, the knowledge gained is either limited to a small number of trees or applicable to a small area of country. But once examined and properly identified, no number of native names need confuse the real identity. In collecting native names a warning should be given against the ignorant or ready-to-please native, and independent corroboration is advisable before accepting a name as worth recording. At the same time it is a fact that names genuinely differ in quite small areas of country where the same language is spoken, and a knowledge of the botanical names or real familiarity with the tree species is essential to reliable work.

The identification of 120 species, though seemingly not a large number, is sufficient to afford a thorough familiarity with most of the savannah forests and makes further species, not included here, stand out all the more clearly from their no longer similar companions. Numbers or economic values of further species are sufficient justification for forwarding these for identification to Kew, if they cannot be named from their resemblance to known species or by means of a Flora.

In selecting the 120 species to be included in this volume, the author has had to exercise his own discretion. The collection has been made between 9° and 14° N., and those who know this area will find a great deal they require, while some, in the more northerly latitudes, will find little of importance that is not included. Others, however, will meet with unfortunate omissions, for the following reason: South of 11° N. certain species which more truly belong to the evergreen and intermediate forest zones have established themselves in considerable numbers by streams, in valleys and in pockets of rich soil with surroundings suited to their propagation. Either all or none must be included if this volume is to be representative of such regions, but since they are not truly savannah species their inclusion here is beyond the scope of this volume, which would have to be enlarged very considerably to cover even the forests in the Benue region and south of the Niger, where savannah abounds. All the species included here occur up to 11° N., and with five exceptions up to 12° N., while nearly all occur below 10° N. and many much further south.

A selection has been made in the case of some genera, e.g., Ficus, and Combretum, to familiarise the characters of trees which occur in large numbers and are typical of savannah forest. Some of these are of little importance beyond their occurrence in numbers over large stretches of country. The value of trees varies largely according to locality. Timber value is of little account in regions where it is inaccessible and unexploitable. Food, medicinal and other values take precedence where there is a large population demanding economic produce, and many species must be accounted valuable merely as a soil covering for the prevention of denudation, fixation of shifting surfaces, protection of new growth, grazing areas, precipitation of moisture, retention of conditions advantageous to agriculture or as a basis for the introduction of new, or the encouragement of existing valuable trees from the forestry standpoint. The advance of agriculture or the condensing of population intensifies these values by the destruction of the better types in favour of the poorer, and by the increased demand for the products of the better types. There are many instances of valuable products being obtained from forests which are not only some distance from habitations but whose soil is of a nature that precludes any possibility of supporting a population, since it will not produce crops. For example, Sclerocarya birrea (Danya) grows to large sizes in some parts of Sokoto Province where the soil is of an unworkable kind and where, over wide areas, there are no people. Yet this area is visited from all over the country and hundreds of trees are felled for the manufacture of mortars, trees in many cases suitable for two or three mortars being exploited for one only.

Though much has been written about the various types of savannah forests, a short account of them will not be superfluous here. First it may be pointed out that savannah forests vary so much in their composition and distribution that a survey is practically impossible, and if made, would have to be revised annually. Savannah is a particularly aggressive form of forest growth. Given the opportunity, it at once occupies more land and on its own ground the inferior types oust the superior wherever there is an opening. Any extensive area of cultivation, if abandoned and allowed to regenerate its tree growth, lets in secondary forests containing species which are less exacting as to soil and moisture conditions, and the depreciation of these conditions to the farming methods stops the competitive growth of the trees which formerly occupied the area. Where the area cleared is sufficiently small and the period short enough, the conditions are retained, and are able to close up the clearing with little loss by degeneration.

The fringing forests, belts of evergreen and deciduous trees on the banks of streams, are continually hard pressed by the broad wedges of open savannah between them. Fires take toll of them every year and cultivation not only pushes them down stream from the source but interrupts and cuts them up into islands exposed to threats of extinction.

Savannah is very topographical, and though, within broad limits, it can be divided into two main types, Tree and Bush savannah, the two are so distributed or blended that the differences are often unrecognisable and not able to be recorded in a survey.

Tree savannah is recognised as a tall type having a light, closed canopy, with a sprinkling of under-shrubs and a sparse growth of grass.

Bush savannah is a more open growth of less height, more spreading and lower branched form, with no canopy, and many stunted shrubs and a dense growth of grasses which are normally burnt out each year. There is every variation of this latter type from Combretum and Guiera scrub of barren soils or stony plateaux to the type which closely approaches that of Tree savannah.

The following are some of the recognised sub-types:—

Park Savannah.—This is a natural or artificial formation, the latter a product of agriculture, and is a two-storeyed growth composed of large type trees scattered about and dominating an undergrowth of smaller trees. Type species are Parkia filicoidea, Butyrospermum Parkii, Afzelia africana, Tamarindus indica, Acacia albida, etc. In the artificial product the large trees often owe their presence to their value, having been allowed to stand when the land was cleared for farms, the lower storey being a subsequent growth whose height depends on the period that has elapsed since the farms were abandoned. Natural Park Savannah is well represented by the superior size of Afzelia africana or Paradaniellia Oliveri over the lower storey of Terminalia and other species which cannot attain the height or proportions of these large trees.

Fringing Forests.—These are the narrow belts of forest along stream banks, and are either intrusion of evergreen and other trees from a lower latitude or the remnants of broader belts which have dwindled to a mere fringe. Typical species are Khaya senegalensis, Eugenia guineensis, Adina microcephala, Diospyros mespiliformis, etc., with a number of evergreen shrubs, typically Jasmines. The narrower the belt the lower branched are the trees.

“Kurimi.”—This is the formation found in depressions, where there may or may not be a stream, not necessarily flowing in the dry season. It is an enlarged edition of the fringing belt and more nearly represents the evergreen forest. Trees of large dimensions with long, clean boles are found amidst a luxuriant undergrowth of shrubs and younger trees, which if undisturbed will attain full maturity. “Kurimi” is the chief source of timber in the Northern Provinces and from it such type species as Khaya grandifolia, Chlorophora excelsa, Albizzia Brownei, A. fastigiata, Diospyros mespiliformis, and frequently Oil Palms are obtained.

“Fadamma.”—This is represented by the broad river valley perennially inundated, or a depression of varying size in flat country where water lies during the rains. Typical species are Mitragyne africana, Paradaniellia Oliveri and Borassus flabellifer, the Fan Palm. There is either a dense growth of grass or, in the case of temporary lakes, an area of cracked mud.

Pure forests are rare in savannah and Acacia Seyal and Isoberlinia species are type cases. Gregarious clumps of various sizes are, however, one of the commonest features and a large number of trees is concerned. Examples are the Acacias Seyal, arabica, albida, Senegal and campylacantha; Anogeissus, Isoberlinia, Bauhinia, Anona, Bombax, Terminalia, Combretum, Detarium, Gardenia, Parinarium, Pseudocedrela, Stereospermum, Ximenia, Uapaca, Boswellia, Monotes, etc.

Savannah also shows many examples of a dominant species, the presence of which are indications of the composition of a forest. The presence of this or that species sets a standard of comparative quality to an area by which it may be valued from various standpoints.

Over wide areas of savannah there are to be found many evidences of previous peoples in the shape of walled ruins, foundations of corn stores, grindstones and heaps of hand-picked stones indicative of cultivation. The dating of such remains gives the approximate age of the forest growth, and from this evidence it would appear that certain types of savannah deteriorate considerably with age and that farming will regenerate them to a certain extent. A comparison of 20-30 year old forest with the untouched older forest shows the former to be denser, more healthy and of straighter and more even growth, while the latter contains a large number of very old trees in a state of decay, with great open crowns, liable to be blown down by storms, their branches burned by fires and riddled with fungus. The new growth which is to take their place consists mostly of grasses, stunted shrubs and small trees of species which are incapable of forming a canopy. As mentioned above, provided the area farmed is not too large, the support given it by the surrounding forests is sufficient to ensure effective regeneration, otherwise a poorer type takes hold.

In conclusion a short account of the growth and characteristics of savannah forests is given. As would be expected, the trees, being subject throughout life to extremely hard conditions, counter with defensive measures for self-preservation. Most of them are prolific seeders and many have the habit of retaining their fruits or seeds on the tree for many months, sometimes right round to the next flowering season. Germination of seeds is a matter of chance when the tree is liable to ground fires from October to April or May. The life of a seedling, too, is precarious, since at the age of six months it may be subject to a devastating fire. In later life a fire, occurring in April, may, if the grass is high, burn the bark to ashes on the outside, destroying all fruits or seeds, killing the twigs and small branches and apparently destroying all life. Yet, a month later leaves will spring from all but the burnt tips, the tree losing a year’s height growth, but in a position to put on another year’s stem girth. Wounds on the trunk of a tree heal over with the formation of a hollow or rotten core. Trees 4-5 feet in girth have been felled and their stems found to consist of a mere shell two or three inches thick. The hollow is filled either with the workings of termites or the fermenting sap which is forced up from the ground level and will pour out of the stump. Savannah trees are adapted to overcome most of these adverse conditions. Heavy crops of fruit and seed, rapid ripening of seeds or fire-resisting fruit coats and various means of distribution are means to ensure germination. Deep or thickened tap roots of seedlings are put down in the first year so that though the seedling may be levelled by fire it will appear with renewed vigour soon after. A very short flowering period for the individual tree is a marked characteristic of many species. Fire has a quickening effect on the flowering and leaf bearing of trees to a marked degree. A grass fire in November may produce flowers and leaves in December on trees which, if unburnt, would not have flowered till February. As a rule, no tree will flower again if the first flowers have been burnt off, though it will, of course, bear leaves. One or two exceptions, very rare, have been noticed by the author, but these were due to the peculiarity that some trees exhibit of bearing flowers on some of their branches and not on others, so that the flowerless twigs were probably excited by the fire into bearing flowers after it. Leaves are readily replaced after attacks by locusts which generally take place early in the season. The end of the cold weather and first sign of heat is the spring of growth. Few trees have to wait for rain before they produce all the signs of maturity except the complete growth of their leaves. The bark of young seedlings, subject to fire, is very thick and corky and that of old trees is still more so, while the inner layers are fibrous and full of sap. Almost all species will coppice very well, many throwing up shoots up to 10 feet high in a season, others only a foot or two. Root shoots are also very common and a large number of what are apparently seedlings in the forest are root shoots. Most trees are anchored very firmly to the ground by lateral roots far longer than the height of the tree, a protection against storms. A tree which is completely ringed will often endeavour to join up the cut, meanwhile continuing its season’s foliage, while the smallest connection with the root is sufficient for it to live for many years. Most new leaves, as a protection against the sun, are reddish in colour, many a brilliant crimson, the green being produced through all gradations of colour from the red. Others are hairy, scaly or covered with a bloom, the protective coverings wearing off with maturity. The proportion of sapwood to hard is large and many species show marked differences in this proportion according to their locality, some showing no heartwood in the north where it is produced in the south. The weights of woods are often considerably heavier in the north than is the case for the same species in the south. There are some localities where wide areas are infested with termites to such an extent that, during the dry weather their whole surface, even to the tips of the twigs, is completely covered with the earth carried by the termites. Although there are exceptions, it is observed that apparent damage is negligible and that the old scales are removed from the bark, exposing the fresh surface.

The savannah species are, then, well equipped to withstand fire, drought, insects, wounds, storms, and damage by man and it can be imagined how readily they will supplant better types, since the harder the conditions the better adapted are the trees to meet them. The wide distribution of most species renders them more aggressive, and an individual species that will grow into a fine tree under the best conditions can still grow under the worst, though it may differ in form and feature so as to be hardly recognisable as the same species.

If, as it is supposed, savannah forests are depreciating of their own accord, some factor other than climate, which has not been demonstrated by records, must be at work. It is sufficient to point to the increase and spread of the population from the towns to the forests to find a reason for the displacement of good forest by lower grades. Shifting cultivation encourages a worse type each time it is practiced and late fires are very destructive.

For valuable assistance given in identifying the trees and shrubs in this volume, the author is indebted to Dr. A. W. Hill, Director of the Royal Botanic Gardens, Kew, and Mr. J. Hutchinson of the Herbarium. To Dr. J. M. Dalziel he is indebted for the “Hausa Botanical Vocabulary,” a valuable short-cut to further research. Other books referred to are the “Flora of Tropical Africa” and Kew Bulletin, Additional Series IX., neither of which has been used other than for purposes of classification, it being the author’s aim to describe and illustrate the material as it appears to be, and not only as it actually is from the scientific point of view, avoiding botanical terms except where they have no alternatives.

H. V. L.

March, 1925.

ACACIA ALBIDA Delile.—Gawo. LEGUMINOSAE.

This is the largest of the Acacias in the north, and attains a height of over 60 feet with girths over 12 feet. It is particularly common in Sokoto, but even there is locally distributed, and gregarious clumps are commonly met with. It prefers dry, sandy soils. The form of the young and the old tree is very different. The former has a straight bole with acutely ascending branches forming a high, flat-topped crown very similar to that of Paradaniellia in shape. The bole may be clean for over 20 feet, but in young trees there are usually clusters of thorny twigs at no great height. As age increases the crown widens and the limbs get heavier and more spreading until the form is like that of Parkia. The bole is then quite clean and considerably thickened at the base, the roots spreading above ground and forming flanges, thick but of little height. Its distinguishing peculiarity is the habit of shedding its leaves at the approach of the rains and putting them forth at the first sign of the dry season, in September. It grows fast up to full height and then slowly.

The Bark is uniform dull grey. Long wide fissures and prominent ridges of hard bark ascend the bole and the bark appears as if it was stretched without being cast off. Light brown patches, quickly turning grey, are left by the falling scales. The slash is pale brown and fibrous.

The Thorns are in pairs, under ½ inch long and slightly recurved. They are pale brown with white bases. The young tree is armed all over, but as the tree ages the thorns leave the stem and finally the higher branches are free as if protection seemed to be unnecessary at a height.

The Wood.—The heartwood is a light, clean yellow. The sapwood is a dirty white. Frequently the whole wood is a dirty white or grey, but this is due to discolourations from mould and the wood should always be seasoned in dry air and not allowed to get wet. In transverse section the rings are indistinct and wide apart, the pores are small and evenly distributed, mostly in festoons with twin pores here and there, the soft tissue plainly seen as concentric lines with the hard tissue alternating. The rays are extremely fine, very regular and closely spaced and quite invisible to the naked eye. In vertical section there are faint bands of colour, with light reflecting flecks in tangential section. The wood is soft, very easy to work with all tools, seasons well if looked after, and finishes well under the plane. The weight is 35 lbs. a cubic foot.

The Leaves are bipinnate, 4 inches long with 6-7 pairs of pinnae bearing some 10-15 pairs of dusty, grey-green leaflets, covered with tiny hairs. The foliage in the mass appears bluish-green. It appears at the end of September and is susceptible to insect, caterpillar and locust.

The Flowers, from October to December, are in dense spikes 3-4 inches long, scented. The lower inch of the spike is flowerless. Several spikes are borne at the twig ends. Each flower has a 5-lobed cream-coloured calyx with pink centre, 40-50 stamens, and a short pistil with slightly clubbed stigma.

The Fruits are pods, at first green and sickle shaped, ripening to a bright orange and twisting into strange shapes. They are 4-6 inches long, 1 inch wide, concave on one side, convex on the other. They contain dark brown shiny seeds ⅜ inch long. The pods ripen in January and February and fall entire, rotting on the ground.

Uses.—Inferior canoes, of the stitched together type are made. The ripe pods are collected and fed to cattle, sheep and goats. The common brown kite nests frequently in its branches in March and April.