Back to Home Page of CD3WD Project or Back to list of CD3WD Publications

CLOSE THIS BOOKLost Crops of Africa: Volume 1 - Grains (BOSTID, 1996, 372 p.)
VIEW THE DOCUMENT(introduction...)
VIEW THE DOCUMENT5. Pearl Millet: Subsistence Types
VIEW THE DOCUMENT6. Pearl Millet: Commercial Types
VIEW THE DOCUMENT8. Sorghum: Subsistence Types
VIEW THE DOCUMENT9. Sorghum: Commercial Types
VIEW THE DOCUMENT10. Sorghum: Specialty Types
VIEW THE DOCUMENT11. Sorghum: Fuel and Utility Types
VIEW THE DOCUMENT13. Other Cultivated Grains

Appendix A

Potential Breakthroughs for Grain Farmers

This book was intended to be solely a survey of Africa's promising grains. However, in drafting it the staff became aware of certain nonbotanical developments that could bring enormous benefit to the use and productivity of Africa's indigenous grains. Some of these promising developments that deal with farming methods are presented here; others dealing with food preparation are given in Appendixes B. C, and D. It should be understood that the innovations described are not the only ones. Indeed, there may be dozens of alternatives for helping to solve the problems described. Nor is it our intention to suggest that these are panaceas. It should be understood further that the novel subjects described here are largely unproved or even undeveloped. Each incorporates a sound and seemingly powerful concept, but whether any will become truly practical in the harsh reality of rural practice and poverty is uncertain. We present them to encourage scientists and administrators to explore these unappreciated topics that just might become vital to Africa's future.


A tiny bird is perhaps the greatest biological limit to African cereal production. The most numerous and most destructive bird on earth, the seed-eating quelea (Quelea quelea) can descend on a farm in such numbers as to consume the entire grain crop in a matter of hours.

Quelea occurs only in Africa, but there its population is estimated to be at least 1.5 billion. Although it holds much of the continent's agriculture hostage, its worst outbreaks are in parts of the eastern and southern regions, where its plagues are worse than those of any locust.

The fields of ripe grain Iying in the path of quelea migrations are essentially doomed. And it is unlikely that the consequences will diminish. Indeed, marginal lands are increasingly employed to grow grains, and future destruction is likely to be even greater.

The quelea's influence is insidious. This bird not only eats enough farm grain to feed millions of people, it destroys the farmers' morale and drains all interest in planting more land.

Where quelea occurs, family members must patrol the ripening fields for weeks, disrupting their lives and restricting all outside activities such as jobs or schooling. Its preferences even dictate what is planted - millions of families now grow dark-seeded, tannin-rich, poorly digestible sorghums, at least in part because the birds, quite naturally, dislike them (see Chapter 10).

Trying to scare away hordes of ravenous birds is clearly futile in all but the smallest plots.

Efforts to control quelea with poisons, napalm, dynamite, pathogens, and electronic devices have failed. Dynamiting the densest concentrations can achieve temporary local control, but a single flock may contain more than two million pairs and spread over an area far too wide for an explosion to have much effect. However, one line of research is now showing some promise.

At sunset each day queleas congregate in patches of tall grasses or trees. As the sky darkens they crowd together, until thousands are packed side by side in a small space.

Researchers at the Zimbabwe Department of National Parks and Wildlife Management have observed that (provided the night is dark and the roost is isolated and fairly homogeneous, such as a patch of bulrushes) the birds are loath to leave. When disturbed, the chattering flock flutters forward a meter or two and only reluctantly decamps into the soundless darkness beyond. Indeed the scientists found that, once the flock had settled in' they could "herd" it around in the roost on moonless nights. By blowing whistles, beating on metal, or making some other disturbance, they could hustle the birds from one end to the other at will.

This was the key. If a barrier (a sheet of glass or transparent plastic, for example) was placed across the middle of the roost, thousands of queleas could be forced to fly into it each night (at least, for three consecutive nights, after which the birds became more cautious). If a holding cage was placed beneath the barrier, at least some of the halfstunned birds tumbled in. They could then be dispatched humanely, or, even better, could be trucked directly to a slaughtering facility and processed like poultry.

In a second step, the Zimbabwe researchers tested tailor-made roosts. In isolated locations (and on sites quelea should find irresistible), they planted plots of napier grass and shaped them with slightly narrowed waists where the barriers and traps could be easily erected.

This seemed like an excellent way to turn a pest into profit, or at least into food, but it proved to have operational difficulties. The biggest problem was that only a few birds ended up in the cages. Those coming in from the fields flew fast enough to stun themselves on the glass, but most of those herded within the roost recovered too fast to fall.

Actually, because of such disappointing results the Zimbabwean authorities dropped the whole idea. They do, however, still use trap roosts to concentrate the birds so that workers with backpack sprayers can get to them with avicides (bird-killing chemicals).

This is much cheaper than using aircraft.

For most parts of rural Africa, killing birds with chemicals is unlikely to be nearly as practical or as appealing as capturing them for food. Thus, even though not yet perfected, the trap- roost concept seems to have promise. Indeed, it might in the end prove ideal for much of rural Africa because it offers the hungry poor both food and source of income. In principle, the operation is simple, cheap, and easy to understand and replicate. Given a new burst of innovation, today's limitations might well be overcome. Nets might be devised or the cages raised so that the chattering flocks would fly right in during the dark of the night and not have to stun themselves at all. Certainly, there seems to be much scope for improvement.

Of course, at this early stage there are many uncertainties even if the method can be made operational. Will it work in locations where the birds normally roost in trees? Could it be modified for use in trees'? Are there grasses better than napier? Will the birds learn, over time, to avoid the seductive patches of grass?

These issues are of course unresolved. However, if this approach can be made to succeed even partially, its effects could be far-reaching. And if it can be brought to perfection, it might transform the production of cereals throughout the quelea combat zone. Relieved of this feathered scourge, farmers could grow the best-adapted, best-tasting, and most nutritious grains. They could plant more land, their children could stay in school during bird season, and they themselves could keep their outside jobs.

Although the trap-roost technique will never be a panacea, it appears to have advantages over other approaches on several grounds:

· Environmental. The method requires no bird-killing chemicals.
· Economic. Trap roosts need no imported materials and farmers can build them with their own labor and materials so the technique could be employed by subsistence farmers, who have no cash to spare for bird control.
· Conservation. Although the fact that other species roost with quelea is a concern that needs to be evaluated, techniques such as use of chemicals and explosions, for instance - are as indiscriminate or more so.
· Logistical. The method is independent of supplies, government, consultants, or high-level training.
· Adaptability. Catching birds in trap roosts seems infinitely adaptable to various locations and to the differing needs of users from subsistence farmers to large-property owners. For instance, a village farmer might install a small trap roost to get a little "poultry" for a party or a corporate farmer may establish many large ones to maximize a crop worth millions.


A small plant is the second largest biological constraint on Africa's cereal production.

Usually called striga or witchweed, it is a parasite that lives off other plants during its first few weeks of life. Its roots bore into neighboring roots and suck out the fluids, leaving the victims dried out and drained of life. (thus the drying-out effect) and to increase its production of roots (at the expense of its leaf growth). Obviously both are processes that greatly reduce grain yields. Information from J.L. Riopel.

Unfortunately, striga (there are two main species, Striga indica and Striga hermonthica) loves maize, sorghum, millet, cowpeas, and other crops. Millions of hectares of African farmland are continually threatened; hundreds of thousands are annually infested. The traditional defense was long, idle fallow - now impossible because of population pressure.

And today when striga breaks out severely, nothing can be done. Farmers usually abandon their land. Some of the most productive sites now lie idle - victims of this abominable sapsucker.

And the problem is worsening. Striga is most damaging when crops are stressed by drought or lack of nutrients - phenomena that are increasingly common. Changes in farming practices are also helping striga to conquer ever more countryside. The continuous cropping of cereals, for example, contributes more and more striga seed to the soil.

At present, the only way to keep this weed in check is by carefully crafted farming practices: crop rotations, fertilization, and skillful use of herbicides, for instance. But this is impractical for the millions of subsistence farmers who have no surplus land for crop rotations and can afford neither fertilizer nor herbicides. Also, it would be nearly impossible to train millions of farmers to modify their farming practices, especially in the impoverished zones where striga is most threatening.

A "technological fix" to take care of the problem easily, universally, and permanently has never been found, but there is a possibility that it might be just around the corner. A crack in the plant's biological armor has been discovered, and through it researchers see exciting new prospects.

The excitement is based on the recognition that striga relies heavily on "chemical signals" to locate its victims. The mechanisms of this signaling have now been defined. In addition, approaches have been designed to cut striga's "lines of communication" or to provide misinformation. And control methods are proving successful in laboratory trials and even early field experiments.

Striga seeds refuse to germinate until they receive a chemical signal from the root of a potential host. The signal telegraphs the fact that a victim is nearby and that moisture is adequate for successful germination. The seed may lie dormant for decades awaiting this chemical confirmation that it is safe to come out.

But striga's elegant adaptation provides a window of opportunity. Farmers could, at least in theory, block the signals. Better still, they could supply false signals and trigger striga seeds into suicidal germination. Striga depends so much on the lifeblood of other plants that unless its seedlings can latch onto a root within four days, they die. Each striga plant produces millions of tiny seeds, but a chemical trigger could perhaps fool all of them into germinating. If the land had been newly plowed, the parasite would find no victims and four days later farmers could safely plant their crops.

Recently, scientists have identified chemical signals that trigger striga's germination as well as others that inhibit it. Apparently, the balance between stimulation and inhibition is what determines whether the seed will germinate. Both chemical types are extremely active. The stimulants, for instance, can be diluted 10,000-fold or more and still cause striga seed to germinate. If compounds like these can be synthesized, mimicked, or economically extracted from plant roots, they could be (at least in humanitarian terms) among the most valuable of all organic chemicals. For example, it may be possible to produce striga-suicide sprays, perhaps even in the regions that require the most help. This approach has been exploited by Robert Eplee of the U.S. Department of Agriculture to dramatically reduce striga attachment in greenhouse tests.

Also, another striga signal has been identified. This compound (2,6-dimethoxybenzoquinone) "tells" the germinating striga seedling to form the organ (haustorium) that pierces the victim's root. This, too, may offer a way to overcome striga. For instance, an antagonist chemical might blunt striga's underground weapon. If the pest can find no host' it never develops a growing shoot (apical meristem), it never becomes photosynthetic, and it dies.

Recently, scientists have found that nature is ahead of them. At least one strain of sorghum can already foil striga by producing watersoluble compounds that are striga inhibitors. This sorghum, SAN-39, both resists the parasite and has desirable agronomic characteristics and good-quality grain. Its striga resistance appears to be simply inherited (only one or two genes). Crosses with other cultivars have already been made and promising progeny obtained. Moreover, an assay has been developed to screen breeding material for this resistant characteristic. These results suggest that sorghum breeders may soon be able to breed for striga resistance rapidly and efficiently. Similar progress has been achieved in maize.

It has also been found that some leguminous plants - Crotolaria species are examples - excrete their own striga-stimulating signals but do not serve as hosts. Although the striga germinates, it immediately dies. Thus, plants like these could be employed to deplete the striga seed bank in the soil. They may prove extremely valuable species for fallow crops or alley crops. Crotolaria species (rattleboxes) are legumes, so they not only knock out the parasitic pest, they also enrich the soil with nitrogen and organic matter.

All these approaches to the striga problem should be top research priorities, and not only in Africa. This parasite already affects India and has broken out in a small part of the United States. It could easily come to infect much of the world's farmland. Solving the problem now would lift from African agriculture a burden so big that the result might compare with a "Green Revolution." It would also help insulate the rest of the world from the heartbreak of this herbaceous horror. All countries have a stake in the outcome of this challenging research.


Numerous African countries, but especially those in the Sahel, are victimized by the desert locust (Schistocerca gregaria). Controlling this one pest soaks up vast amounts of money, time, and insecticides - 700,000 liters of concentrate were sprayed over 14.5 million hectares in 1988, for instance. It has generally been effective, but in recent years some of the locust's relatives have risen up to become equally menacing. In 1989, for example, grasshoppers - in particular the Senegal grasshopper (Oedalus senegalesis)-arrived just at harvest time, causing 10 times more damage than the locusts had the previous year.

For nearly 30 years Dieldrin was the pesticide of choice. Applied in strips across the desert terrain where locust larvae hatch, it seemed an ideal way to stop the insects before they reached their damaging migratory stage. It worked, it needed no repeated spraying, it was cheap, and it could be stored without degrading even in the scorching heat of the Sahara.

But in the late 1980s, even while locust swarms were swelling to worrisome levels, people began protesting because of Dieldrin's potential toxicity to humans and animals.

On environmental grounds, organophosphorus chemicals and pyrethroids seemed preferable but they remain effective for a few days only and must be reapplied over and over.

This means higher costs, more work, and the destruction of all insect life - even beneficial species.

Now, a new approach to chemical control seems to offer some hope. Research in Germany has shown that oil from the seed of the neem tree (Azadirachta i' indica stops locust nymphs from clustering.

After exposure to even tiny doses, the juvenile locusts fail to form the massive' moving plagues. They remain alive but solitary and lethargic; they sit on the ground, almost motionless, and are thus very susceptible to insectivorous birds. Grasshopper nymphs are affected in the same way.

This is very different from the earlier applications of neem against locusts. Those first attempts used alcoholic extracts of the seed kernel, and were aimed at disrupting metamorphosis or at stopping the adults from feeding on crops. Although highly promising in experiments, they proved less successful in practice.

The new approach uses neem oil rather than neem-kernel extracts. Experiments have shown that at very low concentrations (2.5 liters per hectare) this oil, like Dieldrin, prevents locusts from developing into their migratory swarms. It doesn't kill them but it keeps them in the harmless, solitary (green) form. It apparently disrupts the formation of hormones necessary for the transformation into the yellow-and-black gregarious stage whose plagues are the bane of arid Africa and Arabia.

The neem tree grows throughout West Africa, and thus the locustcontrol agent could, in principle, be locally produced. To press the oil out of the neem kernels and to spray it over the areas where locusts breed and gather requires neither particularly high-technology equipment nor unexpected expense. The oil itself is neither toxic to mammals nor to birds and is biodegradable.

Another approach that may have some localized merit is to provide nesting sites for insectivorous birds. In western China, where another plague locust occurs, farmers have reportedly met with success by protecting, and even building, nesting sites for the feathered locust eaters of the area.


The effects of soil erosion are well known: it devastates farms and forests: worsens the effects of flooding; shortens the useful lifetimes of dams, canals, harbors, and irrigation projects; and pollutes wetlands and coral reefs where myriad valuable organisms breed. But there could now be a way to slow or even stop it.

Hedges of a strong, coarse grass called vetiver have restrained erodible soils for decades in Fiji and several other tropical locations. The hedges are only one plant wide and the land between them is left free for farming, forestry, or other purposes. This persistent grass has neither spread nor become a nuisance. If current experience is applicable elsewhere, vetiver offers a practical and inexpensive solution to the problem of soil losses in many locations. It could become an exceptionally important component of land use, at least in the hot parts of the world.

This deeply rooted perennial can already be found throughout Africa, but in most places the idea of using it as a vegetative barrier to erosion is new and untested. However, it is not farfetched. Strips of vetiver certainly are able to catch and hold back soil. The stiff lower stems act as a filter that slows the movement of water enough that it drops its load of soil.

Equally important, the dense, narrow bands of grass cause the runoff water to spread out and slow down so that much of it can soak into the soil before it can rush down the slopes.

This captured moisture allows crops to flourish when those in unprotected neighboring fields are lost to desiccation.

So far, all the international attention has focused on an Indian vetiver (Vetiveria zizanioides).

This is already widespread in Africa and has shown promise for controlling erosion in Nigeria, Ethiopia, Tanzania, Malawi, and South Africa, and appears to be a blessing for many countries. However, Africa has its own native Vetiveria species. These are entirely untested, but they may confer similar benefits. One (Vetiveria nigritana) has long been used to mark out boundaries of properties in northern Nigeria, for instance, and it has been employed for the same purpose in Malawi and Zambia as well.

Vetiver has many interesting and unexpected uses. Tobacco farmers in Zimbabwe report that putting a vetiver hedge around their fields keeps out creeping-grass weeds, such as kikuyu and couch. It even seems to be a good barrier to ground fires.

In the Sahel? vetiver hedges may prove extremely useful as sand barriers. Winds off the Sahara often blow sand with such power that it scythes across the landscape at ankle level, cutting off young crops before they are barely beyond the seedling stage. Rows of vetiver planted on the windward side of fields could be an answer. The stiff stalks would doubtless halt the scurrying sand, providing both a windbreak and a sand trap.

Rows of vetiver planted across wadis may also make excellent waterharvesting barriers.

Once planted, the barriers would be essentially permanent. The deep-rooted grass is likely to find enough soil moisture to survive even the driest seasons in most arable locations.

Although the upper foliage may die back, the stiff, strong lower stalks that block the sand, soil, and water will remain. These are so coarse that not. even goats will graze them to the ground.


As has been noted several times, a major problem with many of Africa's grains - finger millet, fonio, and tef, for example - is that they have tiny seeds. Size alone is holding these crops back. Small seeds create many difficulties. They are hard to store and hard to handle because they pour uncontrollably through even the smallest holes. They also make the crop difficult to plant because the soil must be very finely textured (clods or clumps can overwhelm the seeds' puny energy reserves), and the seeds must be placed precisely at just the right depth. Moreover, because the emerging seedlings are small and weak, they are easily smothered by weeds.

Many innovations could probably be devised to overcome these problems; here we present several examples of seeding devices newly developed in four Third World countries. These are undoubtedly not the only innovations for planting small-seeded crops, but we present them here as guides to those who wish to help Africa's lost crops.

Cameroon In the late 1980s, the Cameroonian Agricultural Tools Manufacturing Industry (CATMI) in Bamenda produced a seeder that, compared to traditional planting by hand, reduces planting time by 60 percent and seed requirements by 33 percent. It is not specifically for small-seeded crops but includes a simple distributer mechanism that can be adjusted to accept seeds of different sizes. It is said to reliably plant the desired number of seeds at the right depth and distance apart.

It is simple to handle, suitable for planting both on ridges and on flat land, durable, easy to maintain, and cheap. In 1988, 30 prototypes were distributed to farmers and research stations for field testing.

After further improvements, 300 more were produced and sent out. Various agricultural services ran information and demonstration campaigns to promote the planter. A line of credit was set up in the Northwest Province to enable small farmers to purchase one. In addition, other provinces were contacted and provided with demonstrators and seed planters.

A survey after the first planting season (1989) indicated that 97 percent of the farmers who tried the implement bought it. Not only did it make the work easier (no back pain) and speeded up planting, but it also reduced the need for hired labor and helped increase both the area farmed and the yields achieved.

Peru Cultivos Andinos, Universidad Nacional Tecnica del Altiplano, Avenida de la Infancia N° 440, Huanchac, Cuzco, Peru. In the Andean city of Cuzco, Luis Sumar Kalinowski has created a seeder capable of handling kiwicha, whose seeds are as small as sand grains. It is a simple, almost cost-free device that can sow large areas evenly and in uniform rows. It may also work well with Africa's small seeds.

One version of the Sumar seeder uses a scrap piece of plastic pipe with a foam-plastic cup taped to the end. A nail is pushed gently through the bottom of the cup to leave a hole of known diameter.

Another version employs a commercially available plastic end piece, which is drilled to provide the hole. In either case, seed placed in the pipe trickles out at a constant rate, and the farmer can vary the seeding density by walking faster or slower.

Indeed, by measuring the flow of seed through the hole, it is easy to calculate how fast to walk (in paces per minute, for example) to sow the desired density of seed. With a little practice, the farmer can attain an accuracy rivaling that of mechanical drills. For the method to work, however it is important that the seeds be clean and free of straw, small stones, or other debris that could block the hole. Engineers at Morogoro have designed and developed a low-cost, hand-operated device known as the Magulu hand planter. It includes an attachment that can be fastened to a hand hoe and can be used to plant both maize and beans in a straight row. It is said that to plant a hectare of land using the Magulu hand planter takes between 18 and 27 man-hours as compared with 80 man-hours using the conventional method of planting by hand hoe.

Thailand The Asian Institute of Technology (AIT), which is located near Bangkok, has developed a mechanical seeder that is now being popularized in many Asian countries. In one stroke, this so-called "jab seeder" makes a hole, drops a seed, and covers the site, without the operator ever having to bend over.

The seeder weighs only about 1.5 kg and costs about US$10.00 (including labor, materials, and mark-up). In Thailand, a farmer can recover the cost. in terms of labor saved, in only 5 days and on an area as little as one-fifth of a hectare. Mass production is expected to reduce the cost even further.

In Thailand's northern province of Chiang Mai, the idea has already caught on: a number of local manufacturers are producing mechanical seeders based on the AIT model.

At present, this machine is not intended specifically for small seeds. It is used mainly with soybean, rice, maize, and mungbean. But even with these crops, it brings big advantages in labor saving and yield.

In Nepal, field tests have found that - at wages of 25 rupees (US$1) a day - a farmer can recover the cost of a jab seeder by planting maize or soybean in just I hectare of land. Fifty seeders made locally by the Agricultural Tools Factory in Birganj cost US$13.50 each.

By making a less onerous and more systematic operation, the jab seeder could well increase grain-crop productivity and thereby benefit millions of Africa's grain farmers.


Seed planters are probably the main need for small-seeded crops, but they are not the only need. Various appropriate technologies are required also for harvesting, storing, shipping, and handling tiny cereal grains. Some of these might come from techniques devised to produce ornamentals, forages, and vegetable crops, many of which also have minute seeds.

Southern Great Plains with precision for half a century. I worked with native grass seeds myself for 25 years. Some of the seeds are smaller than tef, fonio, or finger millet. We had equipment that would mete out seed at low seeding rates very accurately and plant them with precision. Our planters, processors, and cleaners are, perhaps, too sophisticated for subsistence farmers. but modified versions are well within the capabilities of most village mechanics and blacksmiths. The technology has been available for a long time. Suggest you contact Chet Dewald, Southern Great Plains Range Research Station, 2000 18th Street' Woodward, Oklahoma 73801."

Also, it is not impossible that the size of the seeds could be increased through selection and breeding. Luis Sumar has already created a simple machine for doing this in the case of kiwicha. The Sumar sorter uses a small blower and a sloping plastic pipe. The seeds are blown up the pipe and drop into different containers, depending on their weight. With it, Sumar has increased the grain size in kiwicha. He keeps only the heaviest for planting, so that over the years the crops produce seeds that are ever larger, on average. The use of such a simple, inexpensive device in Africa might dramatically benefit fonio, finger millet, and tef, to mention just three cereals.