Reforestation.me – about reforestation, improving soils, cover crops, mixed improved fallows, green manure, mulch, succession, pioneer plants, colonising plants, soil fertility, nitrogen-fixation, permaculture, agroforestry, Rhizobium, Sesbania, Crotalaria, Tephrosia, Mucuna pruriens, Pueraria phaseoloides, Desmodium, Cajanus cajan, Acacia, Tithonia diversifolia.
“Agriculture is our wisest pursuit,
because it will in the end contribute most
to real wealth, good morals, and happiness”
Letter from Thomas Jefferson to George Washington, 1787.
For more information:
Fallows, green manure
In traditional European and Mediterranean agriculture (and elsewhere), a field which has grown annual crops for many years, and where the soil fertility and structure is declining, would be left to rest (natural fallow) for a few years, even up to seven years, and allowed to grow whatever grasses, nitrogen-fixing clovers or other mostly herbaceous legumes, daisy family plants, “weeds” etc., that establish naturally and randomly of their own accord (in slash-and-burn agriculture in the wet tropics, typically ten or more years are set aside for pioneer rain forest trees to grow and restore soils between crops). The crops would also normally be rotated between grasses (e.g. grains such as wheat) and legumes (e.g. beans). The fallow would also commonly reduce crop pests and diseases which may have built up during the cropping period.
The mostly perennial plants that grow during the fallow period would, year by year, build up soil organic matter levels (from top growth and root growth, especially the extensive fibrous root systems of perennial grasses) which improves soil structure and increases biological activity, such as earthworm, bacterial and fungal populations, and increases nutrient fertility and nutrient availability for the following crop. The field would usually be grazed by livestock at intervals during the fallow period. About 80% of what the livestock eat would be returned to the soil as manure, as well as nutrient-rich urine. If the animals have grazed elsewhere beforehand, the manure they deposit is a net addition of organic matter and nutrients to the field (and possibly deficient nutrients if the livestock have been grazing plants growing on a different soil type). Using the AID treatment (see page) the livestock could also be fed with soil improvers beforehand – see more further down this page.
A long term natural fallow allows nature to take its course, leaving the land long enough to improve through the natural process of succession. However, the human population and the demand for more food from less space is constantly increasing, along with conversion of natural vegetation into agricultural land, and so mixed improved fallows are becoming both more popular and necessary as a way to speed up and increase food production.
An improved fallow consists of selecting a particular plant species, or a superior form or provenance of that species, that is much more effective (in increasing nitrogen and organic matter levels in the soil for example) than a random natural fallow, and preferably does so in a shorter time (months to a few years, compared to a natural fallow in tropical slash-and-burn agriculture of around 10 – 20 years). Mixed improved fallows are potentially a further improvement on this, where a carefully selected, preferably complementary mix of two or more plant species are grown, which have the potential to produce still more organic matter and nitrogen – see also the “Complementary Plants” page, click on the button at the top of this page. For example, research at two sites in Western Kenya found that above-ground biomass production over 15 months ranged from 4.1 to 20.5 Mg/ha for four different single legume species fallows (monocultures), and a better response of 7.8 to 23.3 Mg/ha for two different mixed legume fallows consisting of two species each (Ndufa 2009).
Usually at least one of the species would grow as an over-storey, and at least one species would be nitrogen-fixing, to improve the nitrogen fertility of the soil for the following crop. A mixed improved fallow can also be an “insurance policy”, where if one species does not grow well or fails completely for some reason (such as a drought or an outbreak of pests or disease specific to that plant), the other one or more species should still grow well, so that the fallow is not a complete failure.
Deep-rooted improved fallow species (e.g. Sesbania sesban) may be able to reach underground water and so can be planted part or half way through the wet season (relay planted), and continue to grow through the dry season, improving the soil when there is insufficient water for crops or other fallows to be grown (about 9 months). The Sesbania could continue to grow half way into the following wet season (about 12 months), building up organic matter and nitrogen, and it may then be possible to grow a fast-maturing crop through the second half of the wet season; or allow the Sesbania to grow for the whole wet season, and then through the dry season (about 21 months), followed by a crop started at the beginning of the next wet season.
Deep roots may also retrieve leached fertilizer or existing subsoil nutrients, and recycle them to the surface. They can also provide easier and quicker access to deeper soil by the roots of the following crop, which may reduce the need for fertilizer and watering, and may increase crop yield.
Fleshy or tuberous-rooted species may be able to open up compacted soil, improving it for following crops, resulting in a better yield (e.g. Daikon radish, other Brassicaceae and beets, cassava, sweet potato, possibly Tithonia, and sunflowers).
A strategy of maximising organic matter, nitrogen and other nutrients in the shortest possible time through well-designed and managed fallows can not only shorten the fallow period needed, and/or reduce the number of fallows needed, but also result in greater crop production over time. Maximising soil improvement and crop yield is also important where space is limited, or where both space and time are limited, and where fertilizers are unavailable or not affordable.
Fallow plants can also break the life cycle of pests, such as curl grubs eating the roots of sugar cane or pasture (where the grubs can’t live on the roots of the fallow species), suppress nematodes, and reduce a build up of weeds (for example, Mucuna pruriens can be used to smother most weeds). Fallows can also include plants that attract beneficial insects/fauna such as honeybees (for pollination which may increase vegetable or fruit yield and honey production see more uses below), parasitic wasps, and insect-eating birds (pest control), perhaps increase earthworm, microbial and beneficial fungal populations/activity, and provide at least a short term increase in biodiversity in what may be a largely monocultural agricultural landscape. In a rather bizarre example, some farmers in Kenya grow sunflowers to attract aggressive African bees, to deter elephants which keep raiding their crops – hot chillies are also used to deter elephants.
Fallows can provide usable and saleable products such as beans (e.g. Mucuna pruriens, Canavalia ensiformis) or fuelwood (e.g. Sesbania sesban), or fodder, for example. Some species may be best incorporated into the soil, (where strictly speaking they would be called a green manure crop) while some fallows may be good for building up a layer of organic mulch on top of the soil. A mulch on the soil can moderate soil temperature, increase water infiltration and retention, and promote earthworms and beneficial fungi. In time, earthworms and other soil life should incorporate the mulch into the soil, improving soil structure and fertility, and ideally the mulch needs to be topped up continuously.
Where livestock is available, fodder species could be grown, e.g Sesbania sesban combined with siratro Macroptilium atropurpureum. At the end of the fallow, livestock (e.g. cattle, goats) could be concentrated and temporarily confined in the fallow area, where they will eat leaves and finer twigs, turning them into manure of fine texture, which will be deposited on the soil (as well as nutrient-rich urine) and quickly recycled/converted into humus. They could be followed by pigs and poultry. This should attract dung beetles and may increase the population of earthworms, which should incorporate much of the organic matter into the soil, make tunnels, and improve aeration, water infiltration, overall biological activity and nutrient availabilty, etc. The left over larger diameter wood from Sesbania could then be harvested for fuelwood, and a crop grown. Livestock could also be utilised before and between crops, and fallows (see Prinz 1987 below where a combination of an improved fallow and manure was better than either).
Livestock could be fed with soil improvers such as powdered biochar and ash (preferably from fuel efficient stoves), clay for sandy soils, rock phosphate and other rock dusts, deficient nutrients, topsoil containing beneficial micro-organisms, crop/fallow seeds etc. beforehand, to deposit in their manure, which would be further dispersed and incorporated into the soil by earthworms and dung beetles. Livestock could also be fed seeds to disperse in their manure, to establish an improved fallow. Larger livestock such as cattle, and small sized seeds, are likely to work best (e.g. annual Sesbania species and Macroptilium atropurpureum, both of which provide fodder).
Some facts, figures and findings about fallows
“In…Rwanda…a legume fallow of 10 months using Tephrosia vogelii, Crotalaria spp. and Cajanus cajan increased maize yields more than four times as compared with the control, doubled it as compared with stable manure (15 t/hectare per year) and was not significantly different from an application of NPK with 120 kg N, 100 kg P and 100 kg K per hectare…the highest yield was obtained by combining improved fallow with farmyard manure” (Prinz 1987). Actual figures:
Control: 581 kg/Ha maize yield
Stable manure: 1254
Legume fallow 10 month: 2834
legume fallow plus stable manure: 3312
NPK 120:100:100: 3044.
One of the most interesting articles on research into mixed, improved fallows can be found here: Cadish, Gathumbi, Giller and Kamiri “Mixed species improved fallows for Western Kenya” http://ecoport.org/storedReference/559660.pdf . In this research, Sesbania sesban combined with either Crotalaria grahamiana, Macroptilium atropurpureum or groundnut (peanut), resulted in a maize grain yield increase of 1-2 t/ha after a 6-9 month fallow, and 2-5 t/ha after a 12-15 month fallow. “Recommended minimum duration is about 9 months with larger yield benefits obtained for longer duration fallows. The costs of fallow establishment and loss of maize crop are offset by the increased grain yield after the fallow, reduced labor and potential savings in N fertilizer”.
See also http://www.tropicalforages.info/ , and http://www.proseanet.org/, (PROSEA No. 11 Auxiliary Plants), and http://www.echotech.org/technical/az/aztext/azch6gre.htm for more info on fallow species.
“Experiments with Pueraria phaseoloides (tropical kudzu) produced nutrient gains of 59 kg N, 14 kg P, 66 kg K, 53 kg Ca, 28 kg Mg, 113 g Cu and 283 g Zn after 14 month’s fallow and after burning the plants on the area (NCSU 1980). Preliminary results suggest that a one year kudzu fallow’s soil regeneration processes are equal to or better than a 15 – 20 year old forest fallow’s (NCSU 1980, p.179)” (Prinz 1987).
“Fallows improved with sesbania (Sesbania sesban) and tephrosia (Tephrosia vogelii) produced more biomass and fixed more N than those fallows improved with pigeonpea (Cajanus cajan) or mucuna.” (Wortmann and Kaizzi 2000, Uganda). Annual Sesbania species (see below) and some Madagascan provenances of Tephrosia candida (Mafongoya et al 2006) may produce even more biomass in less time (than S. sesban and T. vogelii, respectively).
From the web site Tropicalforages.info: in a trial in S. E. Queensland (subtropical Australia): “Sesbania sesban was the most productive species in the first 18 months, compared with 71 other trees and shrubs. The variety ‘nubica’ grew better than the variety ‘sesban’.”
From Kwesiga and Baxter, Eastern Zambia: “The ICRAF researchers found that growing Sesbania in depleted fields or on fallow lands for 2 or 3 years and then introducing a hybrid maize crop after the fallow period produced exciting and encouraging results. Without N fertilizers, maize yields were 2.3 t/ha after 1 year of Sesbania fallow; 5.6 t/ha, after 2 years; and 6.0 t/ha after 3 years. Continuous maize crops gave only 1.5 t/ha.
Sesbania fallows also produced 10-35 t of fuelwood per hectare after 1-3 years. Such on-farm production of fuelwood eases the burden of labour carried by women and children, as well as reducing pressure on natural woodlands.
When the Sesbania trees are removed (usually by hand), roots are left in the soil: 1.2 t/ha of root mass after 1 year; 1.7 t/ha, after 2 years. Most (90%) of the roots were in the top 50 cm of soil, and as they slowly decayed they provided nutrients to the crops. A few roots reached the water table, at a depth of 7.5 m, where they could access nutrients and water.
A cost-benefit analysis showed that Sesbania fallows are more profitable than continuous cropping, even allowing for the land taken out of maize production. In a farm setting, on severely depleted soils, maize yields increased from 0.15 t/ha to 4.0 t/ha after 2 years of Sesbania fallow”.
From the book PROSEA no. 11 Auxiliary Plants: “Under favourable conditions, (the annual) Sesbania rostrata grows very fast, reaching a height of 2 m in 60 days, accumulating 8-11 t above-ground dry matter per ha”(page 241). “(The annual) Sesbania sericea can produce large amounts of organic matter in a short period. In Hawaii, several selections produced over 15 t dry matter per ha in 14 weeks, about half of it in stems. In India, sesbania grown as a sole crop produced 20-30 t/ha fresh above-ground biomass (4-6 t dry matter, containing 60-100 kg N) in 60 days. In the Philippines, the reported yields of sesbania in 60 days were even higher (7.8-9.9 t/ha of dry matter containing 170-225 kg N)”. “Intercropping S. sericea with maize to provide green manure for a subsequent wheat crop has been tried in India. When sown simultaneously with maize, S. sericea smothered the maize; delaying sowing S. sericea by 6 weeks did not depress maize yield and boosted the following wheat crop by about 20-40%” (page 239). Also: http://www.cabdirect.org/abstracts/19876762984.html/
Applying 4 t/hectare of Sesbania rostrata mulch “was able to increase the grain yield from wet rice cultivation from 3.1 t/hectare to 4.7 t/hectare” (Prinz 1987).
Some photos and examples of fallow plant species, and succession
Tephrosia candida. This plant is native to India, and is used in tropical Africa as a cover crop. In a trial in Zambia, Madagascan provenances of this plant produced twice as much biomass as the native Tephrosia vogelii (Mafongoya et al 2006).
Coconut fibre (coir) pots (or a similar homemade equivalent) could be useful for growing larger fallow species such as Sesbania sesban and Tephrosia candida, and then planted without having to remove the pot, which means no root disturbance and probably better and quicker establishment. Plants could be sown in the pots (a few seeds to a pot, and cut down the least vigorous seedlings later) before the beginning of the wet/growing season, so that they get off to an early start. In the case of fast-growing annual Sesbania species (or Desmodium tortuosum) for example, an early start may leave enough time to grow a fast-maturing grain crop in the second half of the wet season. Smaller plant species with smaller seeds (e.g. Crotalaria spp., Macroptilium spp.) could have perhaps three or more plants growing in a single pot.
Vigorous nitrogen-fixing groundcovers/climbers can be used to smother weeds and improve soil fertility by adding nitrogen, organic matter and leafy mulch. Mucuna pruriens would be one of the most popular plants for these tasks. The photo above is the related Vandasina retusa, formerly Hardenbergia retusa, a perennial tropical climber native to Australia, which may also have potential for roadside cuttings. Photo: Cairns, Australia.
Mucuna pruriens used as a cover crop in Central Africa, and commonly used throughout the wet tropics. In Mediterranean climates, Dipogon lignosus, syn. Dolichos lignosus may do a similar job (only grown where it is not a weed), and in colder climates, Coronilla varia, which has also been used successfully in Europe and the USA as a nitrogen-fixing groundcover beneath maize. For more possible cover crops/green manure crops in colder climates, see http://plants.usda.gov/java/covercrops/
Improved fallows contd…
Desmodium tortuosum, Florida beggar-weed. Two to three metres tall, vacant lot, North Cairns, sandy soil, 50 metres from the sea, approx. 2500mm rainfall p.a. This plant is an invasive weed in North Queensland, (the seed pods stick to fur and feathers) but could be useful in the West Indies where it is native, as a nitrogen-fixing reforestation plant, where the seeds could be fed to livestock for dispersal (or dispersed in seed balls).
Some possible strategies for mixed improved fallows and cropping
Desmodium tortuosum could also be used as an improved fallow species, and/or for fodder. More biomass is likely to be produced if the plants are cut back by about two thirds when they are approximately one metre tall, to induce branching. This plant, or similar fast-growing N-fixing plants (such as annual Sesbania spp), could be combined with a tall, fast-growing grass such as annual forage sorghum to produce even more organic matter (see the “Complementary Plants” page – photos of sugarcane and Sesbania cannabina growing together). However, if a grass is used, it could potentially carry over diseases or pests to a following grass crop (eg. maize, sorghum, millet) and sudex has been shown to leave allelopathic compounds in the soil for about 4-8 weeks. Crop rotation usually reduces the problems of pest and disease carry-over, so if a grass is used in a mixed improved fallow, it may be better to grow an unrelated crop after the fallow, e.g. cassava, okra, sweet potato, cow peas, pigeon pea. Because of its weediness, this species of Desmodium should only be used where it is native, or perhaps where it is already a weed, but managed so that it does not set seed.
In this same vacant lot, Crotalaria pallida showed slower growth in the early part of the wet season, but grew faster later and for longer into the dry season. The Desmodium grew fast early in the wet season, and petered out towards the end of the wet season. Thus, a fast-maturing grain variety could perhaps be grown early in the wet, with Crotalaria or Sesbania sesban sown/planted later, to grow on into the dry season (and possibly into the first half of the next wet season); and in another plot, Desmodium (or possibly a fast-growing annual Sesbania species) could be sown early in the wet season, (or even grown in pots or a raised nursery bed well before the wet season and then planted at the beginning of the wet season, to get an early start), then cut down after a good amount of growth, and followed by a fast-maturing grain variety, which would hopefully mature before the dry season stopped growth. This strategy could potentially provide two staggered grain crops in one wet season, and a short improved fallow for each field in one wet season.
Another possibility: a fallow could perhaps consist of Desmodium, or a fast-growing annual Sesbania species – annual Sesbania species (and Desmodium) should grow faster and produce more biomass than Sesbania sesban, but will not grow through the dry season. Once the plants grow to a height of approximately 30-60 cm tall, the top approx. 10 cm could be cut off to induce branching and therefore more biomass production, and finally cut at or below ground level (after about 3-4 months). Then Sesbania sesban could be planted to grow through the remainder of the wet season, and through the dry season, (total fallow about 12 months) and then crops could be grown at the beginning of the following wet season. Alternatively, Sesbania sesban could continue to grow through the first half of the following wet season, and then grow a crop (total fallow about 15 months); or Sesbania could grow through the whole following wet season and through the following dry season (total fallow about 24 months), and then grow a crop started at the beginning of the wet season. Since annual Sesbania species and Desmodium tortuosum grow faster initially than Sesbania sesban, the total biomass production should be greater than if just Sesbania sesban was grown throughout the fallow period.
Annual sesbania species generally grow best in moist to wet soils. If the field can be formed into ridges and furrows, the annual sesbania should grow better in the furrows where water collects and infiltrates. In a mixed improved fallow, many other fallow species prefer well drained soils, and so they could be established on the ridges (e.g. Mucuna). Alternatively, the field could be formed into a grid pattern of ridges and basins, with sesbania in the basins, where water collects. Forming the fields may also make it easier to add fertilizers such as human urine, for fallows or crops.
Another possibility: annual Sesbania or Desmodium tortuosum planted at the beginning of the wet season, or preferably started earlier in pots, (assuming a wet season of around 5-6 months), then grow Mucuna pruriens about half way through (perhaps 2-3 months after the beginning of the wet season). The Mucuna should then have about 3 months to grow initially as a groundcover, and then up and over the top of the Sesbania or Desmodium (which would continue growing, but may eventually be swamped), and the Mucuna should therefore produce more biomass than if it was just covering the ground. The Sesbania or Desmodium would have a total of around 4-6 months or more to grow, and, in spite of competition, the combined biomass production and nitrogen fixation should be perhaps 10% or more than growing just one species, in a relatively short term fallow. This strategy should also be reasonably affective at controlling weeds.
Another possibility: relay plant Sesbania sesban mixed in with the crop part way through the wet/growing season, so it grows through the latter part of the wet season, and then through the dry season. Then grow Mucuna pruriens at the start of the following wet season, so that it grows up and over the Sesbania (total fallow about 12 months). Cut down and start a fast-maturing crop about halfway through the wet season. This strategy should also be good for weed control (see also the section below on ridges, furrows and black plastic mulch).
Phosphorus deficiency: fallows typically accumulate nitrogen and organic matter, but many soils are also either deficient in phosphorus, or the P is unavailable, limiting plant growth. Tithonia diversifolia has a root association with Glomus spp., which makes it particularly efficient at accumulating P and N (typically the leaf content is about 4% for both). Human manure contains phosphorus and other nutrients, and so Tithonia grown on Arborloos (see the “Arborloo” page), should efficiently recycle these nutrients into its stems and leaves. The prunings from Tithonia could be composted or steeped in water to make a “fertilizer tea”, which could then be applied to crops (and perhaps fallows). However, prunings applied fresh as mulch on the soil surface may take root and then grow as a weed in the crop.
During the cropping phase, the nitrogen-fixing Desmodium heterocarpon ssp. ovalifolium could be grown as a groundcover, and may help to suppress the weed Striga (see http://www.push-pull.net).
In tropical semi-arid regions such as the Sahel, a fallow of Acacia holosericea and/or Acacia torulosa (perhaps grown on mini swales or demi-lunes, or Zai holes), plus a drought tolerant groundcover of Canavalia sp. or similar, could build up mulch, soil organic matter and nitrogen, but would probably need 3 – 5 years to significantly improve the soil.
Possible fallow strategies continued… black plastic mulch, ridges and furrows.
Black plastic mulch can greatly increase the productivity of crops, but potentially also increase the growth of fallow species. An improved fallow of annual Sesbania or S. sesban should grow particularly well in furrows, with the ridges covered in plastic. The increased moisture and temperature in this system should suit most Sesbania spp. very well, and result in more growth.
If a quick maturing maize (or other cereal crop) is used, it may be possible to get worthwhile growth out of an annual Sesbania fallow in the first half of the growing season, which is then removed, and then still harvest a quick maturing maize or other cereal crop in the second half of the growing season (Sesbania grown in the furrows, and maize/cereal on the ridges).
Also with ridges and furrows, and plastic – plant/sow Sesbania in the furrows at the beginning of the wet season, and about halfway through, grow Mucuna pruriens or Lablab purpureus or similar on the ridges, to grow up and over the Sesbania. If it works, this combination could provide a great deal of organic matter and nitrogen. It would also produce usable and possibly saleable products in the form of beans and fuel wood.
The exact same spot as the photo further up the page – the Desmodium has died off, leaving just twiggy stems, and the Guinea grass Panicum maximum, started to take over through and to the end of the dry season, as the Desmodium died off (first) and then the Crotalaria.
Nitrogen-fixing nodules on the roots of Crotalaria pallida.
Crotalaria species add nitrogen and organic matter to improve soils, and the yield of the crops following them. They can be effectively combined with Sesbania sesban, for example, for still more nitrogen-fixation and organic matter. In East africa, Crotalaria grahamiana is commonly used.
In the photo below, Crotalaria pallida is growing with Panicum maximum. Combining a grass with a nitrogen-fixing plant has been a tried-and-tested method (for thousands of years) for increasing organic matter and nitrogen in soils, such as Berseem clover and rice in ancient Egypt, Azolla and rice in China/SE Asia, and for at least hundreds of years, white clover and ryegrass in Northern Europe. The nitrogen-fixing plants can also be trees, for example the tree Faidherbia albida could be grown as an overstorey to the grass Pennisetum purpureum, for cut-and-carry fodder for stall-fed livestock. For more information on mixed, improved fallows visit, for example, http://www.cglrc.cgiar.org/icraf/improvedfallow.pdf/ .
Mixtures of grasses and nitrogen-fixing plants add organic matter and nitrogen to the soil, Crotalaria pallida and Panicum maximum growing together, self seeded on a vacant lot.
The same spot as the photo above, but close to the end of the dry season – the Crotalaria has died off, and the Guinea grass Panicum maximum has taken over. The two plants combined would have added significant nitrogen and organic matter to the soil. If a crop were to be grown at the start of the wet season in this example, the grass could be grazed, and then killed by ploughing or using glyphosate.
Pioneer plants/colonising plants and succession
The first stage of succession. A Chenopod (left), a grass (foreground), and Desmodioum tortuosum (right) growing in a car park in Cairns – pioneer plants doing their thing, building organic matter and nitrogen fertility, creating an improved soil and atmospheric micro-climate for later stage successional plants.
Acacia flavescens, Eucalyptus tesselaris and Ageratum conyzoides (introduced) colonising the bare subsoil of an eroded gully, Clifton Beach, Cairns.
Acacia simsii direct seeded on a rocky roadside cutting, Campbell’s lookout, Cairns.
Melastoma malabathricum or Native Lasiandra is a common pioneer or colonising plant in North Queensland rain forests, spread by birds and other animals eating the fruit and dispersing the seeds. The name means “black mouth”, since the edible fruits stain the mouth.
The photo below is the same area, about two years later. The ferns and Ageratum are gone, and the trees are becoming established.
This magnificent Acmena graveolens tree was rotten on the inside and blew over during cyclone Yasi, Feb 2011. Photo taken standing on the fallen trunk. Near Hutchinson creek, Daintree.
A bleeding heart tree, Homalanthus novoguineensis, has germinated and is growing beside the fallen trunk. Photo taken one year later.
Cosmos sulphureus. Asteraceae species are commonly pioneer or colonising plants in the early stages of succession after soil disturbance. Asteraceae species can be used as part of a mixed fallow, or intercropped and the cut foliage used as a mulch on the surrounding crop. Tithonia diversifolia has a root association with mycorrhizal fungi which improves nutrient uptake, so its leaves are high in nitrogen and phosphorus. This may also be true of many other Asteraceae. Some Asteraceae species also suppress nematodes, and most attract beneficial insects. Cosmos sulphureus tends to self-seed, and could be useful as part of a mixed, improved fallow, but only in places where it is native, or has already been introduced.
Update: with more recent experience growing this plant, it appears to be allelopathic and so is therefore not such a good choice for a mixed fallow. It may be best grown as a garden plant, pruned occasionally and the prunings composted, or used to make a fertilizer tea (like Tithonia).
Some other plants in the daisy family that could be promising include Montanoa spp., (similar to Tithonia),Vernonia galamensis and other Vernonia species, Tagetes palmeri, T. lemmonii, Ageratum spp., etc. Some of these are, or could be, invasive weeds. An example of an Asteraceae plant used as a fallow is Chromalaeana odorata, which although it is an introduced weed in West Africa, is nevertheless valued for its ability to accumulate nutrients, add organic matter and suppress weeds in fallow fields.
Bunch, Roland. Green Manures and Cover Crops. http://www.echotech.org/technical/az/aztext/azch6gre.htm/
Kwesiga, F and Baxter, J.”Sesbania fallows for increased maize production in Zambia”.http://www.idrc.ca/en/ev-31960-201-1-DO_TOPIC.html
Mafongoya et al. 2006. “Tephrosia species and provenances for improved fallows in southern Africa”. Agroforestry Systems. Volume 59, Number 3, 279-288. http://www.springerlink.com/content/k07746536g37021m/
Ndufa, J. K. (2009). “Above-ground biomass productivity and interaction in mixed legume species fallows on two sites in Western Kenya.”
Prinz, Dieter. Improved fallow. Increasing the productivity of smallholder farming systems. Ileia – Mei 1987 Vol 3. No. 1. http://www.agriculturesnetwork.org/
Wortmann, C. S. and Kaizzi, C. K. (2000). Tree legumes in medium-term fallows: N fixation, nitrate recovery and effects on subsequent crops”. African Crop Science Journal. Vol 8, No 3. http://ajol.info/index.php/acsj/article/view/27691