Push–pull technology

Print PDF

e Newsletter


In the past 17 years, icipe, in collaboration with the Kenya Agricultural Research Institute (KARI), other national partners, NGOs and Rothamsted Research (UK), has discovered and implemented a phenomenally successful technology known as ‘push–pull’. The technology, which simultaneously addresses the major constraints of cereal-based farming systems such as striga weeds, stemborers and poor soil fertility, has been hailed as “the single most effective and efficient low-cost technology for removing the major constraints faced by the majority of smallholders in eastern Africa resulting in an overall and significant improvement of their food security and livelihoods”.


Cereals — maize, sorghum, millet and rice — are the main staple and cash crops for millions of small-scale farmers in most of sub-Saharan Africa (SSA). However, the continent also has the lowest yield of cereals in the world, averaging less than one tonne per hectare. This low cereal productivity is due to a range of biotic constraints, which include insect pests, (notably stemborers) and the parasitic weed Striga, and abiotic constraints, signified by land degradation and poor soil fertility. By ruining cereal yields, these factors keep the food security and livelihoods of millions of people in the region in constant risk.

In 1993, icipe, in collaboration with the Kenya Agricultural Research Institute (KARI), Rothamsted Research (UK) and other partners in eastern Africa, commenced efforts to simultaneously address these biotic and abiotic problems. The outcome was a novel habitat management approach, known as ‘push–pull’. The strategy involves intercropping cereals with a repellent plant such as desmodium, and planting an attractive trap plant, such as Napier grass, as a border crop around this intercrop. Stemborers are repelled or deterred away from the target food crop (push) while, at the same time, they are attracted to the trap crop (pull), leaving the food crop protected. In addition, desmodium stimulates the germination of Striga seeds and inhibits its growth after it germinates. The technology also provides high quality animal fodder. Furthermore, since both companion plant species are perennial, ‘push–pull’ conserves soil moisture and improves soil health and beneficial biodiversity.

Finding the ‘pull’

Push–pull relies on an in-depth understanding of chemical ecology, agrobiodiversity and plant–plant and insect–plant interactions. From the onset, the researchers already knew that some wild grasses acted as ‘trap plants’, enticing egg-laying borer females but depriving the larvae of a suitable environment. In 1994, with funding from the Gatsby Charitable Trust, the researchers studied more than 400 species of wild grasses for efficacy in these dual roles. The results revealed more than 30 grass species with the potential of being used as trap crops to draw the borers away from the maize while reducing their populations. The scientists then consulted farmers and identified Napier grass (Pennisetum purpureum) and Sudan grass (Sorghum sudanense) as the two species that would be most useful as cattle fodder, and which would therefore be ideal from a socio-economic perspective. From a scientific point of view, Napier grass has a particularly ingenious way of defending itself: When the larvae bore into the stem, the grass secretes a sticky gum, physically trapping the borers and preventing most larvae from completing their life cycle. Sudan grass is an attractive habitat for the parasitic wasp Cotesia sesamiae. These tiny insects inject their eggs into the stemborer larvae and, when the eggs hatch, the wasp larvae eat the stemborers. Both grasses attract additional stemborer predators, such as ants, earwigs and spiders. Because intercropping grasses planted among the maize plants would provide too much competition for the cereals, the researchers opted to plant grass border rows around maize fields. The grasses would provide a ‘pull’ and an effective defence mechanism against stemborer attack. In 1997, the scientists began on-farm trials to evaluate the benefits of Napier grass, a perennial plant that is already grown widely for livestock fodder. Working with farmers, they identified ‘bana’, a smooth, broad-leafed variety as the best option. Besides increasing their maize yields, the farmers planting Napier border rows benefited from a ready supply of grass to feed their livestock or sell to other farmers.

Finding the ‘push’

For the ‘push’ part of the technology, the team focused on screening grasses and legumes that could repel stemborer moths. The molasses grass, Melinis minutiflora, caught the attention of the researchers. After conducting trials on the species, the scientists confirmed that the strong, sweet smell of this grass did indeed have a repellent effect on stemborers. Furthermore, the researchers discovered that the molasses grass also attracts C. sesamiae, parasite of stemborers. Their colleagues at Rothamsted Research helped to piece the puzzle together, by investigating the nature of the semiochemicals that attract or repel stemborer moths, and to find out why molasses grass repels stemborers but attracts their natural enemies. They identified a compound that is produced by maize plants, as a key stimulus, a ‘feeding stress’ chemical, when they come under attack from the stemborer. The scientists concluded that molasses grass has evolved an ingenious defence strategy through the release of volatile chemicals that mimic those of damaged plants. In this way, the grass defends itself against the stemborer attack by constitutively releasing a ‘cry for help’ comprising volatile cues that have dual effects: repelling the moths but also attracting the pest’s natural enemies.

With this unique property, molasses grass was the first plant to be used by the icipe scientists as a ‘push’ plant. The scientists recognised that one of the reasons why the grass was accepted by farmers as a ‘push’ intercrop, was because it provides fodder for cattle. The scientists were therefore keen to find additional plants that would add a further dimension to the habitat management system. They focused their attention on legumes, which would potentially improve soil fertility because of their nitrogen-fixing qualities. In this regard, they shortlisted Desmodium, as a possible ‘push’ intercrop. After evaluation, the silverleaf and the greenleaf desmodium were found to repel stemborers. The scientists proceeded to test these species on maize on-station at the icipe research station in Mbita Point, around Lake Victoria. As it happened, all the experimental plots were infested with striga. But to the amazement of the scientists, they found that the maize plots with a desmodium intercrop not only had little stemborer damage but also became virtually free of striga after only two seasons. This result brought a new dimension to the push–pull technology, as it meant that desmodium could be used to control stemborers and striga, in addition to fixing nitrogen and providing fodder. The crop was, therefore, incorporated as a ‘push’ plant.

‘Push–pull’ in farmers’ fields

In 1997, icipe and partners integrated ‘push–pull’ in maize-, and later sorghum-based cropping systems in Kenya and in eastern Uganda. Today, ‘push–pull’ technology is widely recognised, and is currently being practised by over 35,000 farmers in eastern Africa. This impressive accomplishment was realised through an effective collaboration between the scientists and the national research and extension systems and non-governmental organisations. Together, the teams developed a range of innovative transfer dissemination methods to create awareness and increase the adoption of the ‘push–pull’ technology.

One of the dissemination strategies used is the ‘farmer-to-farmer’ approach, based on the influence that farmers have on their peers in the adoption of new technologies and practices. Farmers are trained through the national agricultural research institutes (NARIs) and nongovernmental organisations (NGOs) participating in the ‘push–pull’ project. They are provided with information regarding the technology as well as technical backstopping.

The icipe scientists also found the use of ‘farmer teachers’ — farmers who are particularly knowledgeable about the push–pull technology — to be an effective method. These farmers were trained, provided with dissemination materials, bicycles for moving around the farms and technology backstopping support. icipe’s research in western Kenya has endorsed the farmer teachers’ approach, showing that one single farmer teacher is able to teach on average 17 farmers, who successfully adopt the technology. Each of these farmers in turn teaches at least two additional farmers, giving a total of 34 farmers adopting the technology per single farmer teacher. The farmer teachers also help to foster collaborations among farmers, scientists and extension personnel.

The scientists also conduct field days every cropping season. These occasions are hosted by farmers already practising ‘push–pull’. The field days provide an opportunity for farmers to evaluate the performance of the technology based on the host’s experience and to make recommendations on areas of improvement. The participants also acquire new skills and knowledge through sharing information and experiences. The events also create awareness among community members as well as interest and demand for the technology.

The researchers also use print materials such as brochures, manuals and posters in the dissemination of the ‘push–pull’ technology, to provide farmers with accurate and reliable information. ‘Push–pull’ print materials are designed to be effective references of technical details, for instance, planting details and illustrated instructions of how to manage the technology.

Moving ‘push–pull’ forward

In SSA, smallholder farming systems are characterised by intercropping cereals with edible legumes. Therefore, as the ‘push–pull’ project progressed, the need to incorporate edible beans into ‘push–pull’ plots became important. In consultation with the farmers, the icipe scientists, therefore, conducted further laboratory studies and field trials on the effects of integrating beans in the maize–desmodium intercrops. They found that integration of beans in the maize–desmodium intercrops, guaranteed farmers a protein source without compromising the striga and stemborer control efficacy of desmodium. In addition, the researchers and the farmers have been able to work out ways of varying the number of Napier grass rows surrounding the cereal fields according to the fodder demand.

Between 2007 and 2008 icipe scientists evaluated the potential of Desmodium intortum, a drought-tolerant desmodium species, for ‘push–pull‘ application in finger millet in western Kenya. Finger millet has outstanding attributes as a subsistence food and fodder crop, and has superior nutritional qualities in the drier areas of Africa. The results of the study showed that the desmodium species offered an effective control of both striga and stemborers for finger millet.

In recent years, Napier stunt disease (NSD) started to threaten Napier grass in the ‘push–pull’ regions and beyond. The icipe scientists immediately embarked on studies, which confirmed that the leafhopper Recilia banda is the vector of the Napier stunt phytoplasma in western Kenya. The studies also indicated that the species might be the key vector of NSD in the region. The scientists are advancing studies on Napier grass cultivars that might be resistant to NSD, as well as integrated pest management strategies for its control.

The effect of climate change is likely to affect the effectiveness of the push–pull strategy especially on some desmodium species, which require sufficient moisture to establish, and during planting and after harvesting of the cereal crops. In response to this challenge, icipe initiated a study in Sudan to identify drought-tolerant species of desmodium.

Making desmodium seed available

One of the limiting factors for the ‘push–pull’ technology has been the lack of sufficient quantities of desmodium seeds. icipe and partners have undertaken a three-pronged approach to resolve this issue. The first is through commercialisation, whereby, in collaboration with Western Seed Company Ltd, one of the main private seed producers in East Africa, commercial quantities of desmodium seed are produced through a network of farmers and farmer groups. The company sub-contracts farmers to produce quality seeds on its behalf. It then buys, cleans, controls seed quality and obtains certification from the Kenya Plant Health Inspectorate Service (KEPHIS) for germination and viability. The company then packs and distributes certified seed through its network of agro-stockists.

The second approach to ensure availability of desmodium seed is through farmer groups, who have been trained in its production, processing and marketing, using phytosanitary guidelines. This approach is aimed at boosting the production of desmodium seeds, while improving its access and affordability.

Third, farmers have been trained to establish desmodium through vegetative propagation of the vines.

Impact assessment

In 2009, icipe commissioned an independent impact assessment of ‘push–pull’ to establish its effect on the livelihoods of smallholder farmers and their perceptions towards the technology.

The study found that 19 percent of the farmers in the villages under assessment had adopted ‘push–pull’, citing the technology’s ability to address the major cereal production constraints concurrently as the main attraction. The farmers also mentioned the low cost of implementing ‘push–pull’, and the use of Napier grass and desmodium as fodder as other motivating factors for adopting the technology.

Based on the impact assessment, the ‘push–pull’ technology has contributed significantly to reducing the vulnerability of farm families by ensuring higher and better yields. Of the assessed farmers, 75 percent indicated maize yield increases of between three- to fourfold. For instance, farmers using ‘push–pull’ were able to harvest more than five tonnes of maize per hectare from plots that previously yielded below one tonne per hectare. In addition, ‘push–pull’ has become a ‘springboard’ for diversifying the farming system, especially incorporating dairy operations using Napier and desmodium as fodder.

These benefits have contributed to the increased well-being at household and village levels. By selling their surplus grain, milk and fodder, ‘push–pull’ farmers earn extra income, which they use to pay school fees for children, purchase household items, and improve their housing, overall nutrition and health. The study thus suggests ‘push–pull’ as “probably the single most effective and efficient low-cost technology for removing major constraints faced by the majority of smallholder farmers in the region, resulting in an overall and significant improvement of their food security and livelihoods”.

On a national scale, the economic benefit of ‘push–pull’ is estimated at US$ 2–3 million annually. In addition, the technology contributes to national food security, rural employment, better education and increased farming knowledge. Furthermore, ‘push–pull’ is an environmentally friendly technology that is likely to increase agrobiodiversity and contribute to provision of ecosystem services.

Pushing forward

‘Push–pull’ responds directly to the rising uncertainties in Africa’s rain-fed agriculture due to the continent’s vulnerability to climate change as it addresses both abiotic and biotic constraints, resulting in improved farm productivity, incomes and food security.

icipe and partners hope to build on their phenomenal 17-year ‘push–pull’ experience, aiming to extend the benefits to 1 million smallholder farmers in SSA by 2020. The researchers also hope to spread the technology to regions with harsher climate conditions. There is a great deal of scope for further expansion, as the technology has not been extended to millions of farmers within and beyond East Africa, who could benefit from it. Moreover, with the rising uncertainties in the region’s rain-fed agriculture due to the continent’s vulnerability to climate change, there is a demand for the more robust cropping system provided by the technology and for its further adaptation to withstand the increasingly adverse and changeable conditions.

In sum, the full potential of ‘push–pull’ to enhance the food security and prosperity of millions of smallholders in Africa could be realised by:

  • Extending the technology and integrating it with livestock development enterprises to areas of similar agroecologies to those where it is currently practised.
  • Ensuring accessibility and availability of companion planting materials in the target areas.
  • Adapting the technology to the increasingly dry and hot conditions with less predictable rainfall associated with climate change.