LUBILOSA is a French acronym, LUtte BIologique contre les LOcustes et SAuteriaux, biological control of locusts and grasshoppers. The project started work late in 1989 from an original concept developed by Prior and Greathead (1989), with the objective of developing a biological means of controlling locusts and grasshoppers. It soon became apparent that oil formulations of deuteromycete fungal spores offered the most promising option. Such fungi will grow on artificial substrates and so can be mass produced quickly in large quantities. Their spores are lipophilic – they suspend more readily in oils than in water. As most locust control operations take place using oil formulations, developing an oil formulation of fungal spores would enable operators to use the same equipment for applying a biological pesticide.

Fungi in nature are most active under moist conditions; part of the original concept was that formulating spores in oil might overcome this constraining requirement for high humidity.

In Phase 1 we demonstrated the technical feasibility of the idea in laboratory and cage trials, involved a network of collaborators and selected an isolate (IMI 330189) of Metarhizium (flavoviride) anisopliae var. acridum for further development. In particular, trials in cages and arenas confirmed that oil formulations were infective even under very dry conditions.

In phase 2, we demonstrated field efficacy; a difficult issue for such highly mobile insects as grasshoppers and locusts. With locusts, fixed plots, unless several square kilometres in size cannot be used, it is necessary to follow bands. Although there were many locust infestations during phase 2, the treatment teams were also very active, and we made little headway in developing the techniques necessary to measure Metarhizium effect on unenclosed desert locust. Progress was better against brown locust; most of the effort however went into scaling up treatments against variegated grasshopper and Sahelian grasshoppers.

With the sound field trial results of phase 2 in hand, phase 3 of LUBILOSA included socio-economic and ecotoxicological elements for the first time. Results emerging from these two topics shaped the project as it developed. In particular, our options for achieving the project objectives became clearer; established biopesticide producers with expertise in solid-state fermentation would need to be involved in order to make a biological control option for grasshoppers and locusts widely available.

Prior to the start of Phase 3, contacts were made with several internationally renowned biopesticide producers to ascertain whether they would be able and willing to reliably meet LUBILOSA's spore requirements in Phase 3; none were able to make that commitment. Accordingly, the building at IITA Cotonou was refurbished and the human and material capacity enhanced to meet expected spore production requirements. The facility has proved an excellent research unit and has enabled us to refine realistic technical specifications, and to test production, contamination control, spore separation, drying and packaging on a large scale.

Economically, however, the technology is not interesting; one of the original concepts of LUBILOSA was to design a medium-technology plant which could be transferred to national programmes or local start-up companies. While this remains a possibility for more robust, competitive and productive microbial control agents (such as Beauveria bassiana and different varieties of Metarhizium anisopliae), large-scale solid-state fermentation appears to be the only economical production method for the LUBILOSA isolate. Expertise in such techniques is limited to perhaps a dozen companies world-wide, and LUBILOSA has contracted with two of these companies to ensure the availability of the Metarhizium product on a large scale.

The importance of the pilot production plant has therefore been (a) to produce spores for the LUBILOSA programme including collaborators; (b) to determine the particular characteristics of our isolate, to provide technical guidance to the commercial producers; and (c) to develop a product specification.

At the beginning of Phase 3, the Metarhizium product consisted of spores (conidia) of unknown moisture content, with conidia attached to each other in chains of variable length, the mixture containing unknown quantities of contaminating micro-organisms, fungal mycelium and substrate particles; in short, a fairly typical mycopesticide. The improvement of the basic dry powder (Technical concentrate, TC) formulation was an incremental, iterative process. With each up-grading of the production facility we were able to test research findings on the importance of different factors affecting the product quality. The end point is the product Technical Specification for 'Green Muscle'. This specification describes in detail what is in the final product, in terms of maximum amounts of permitted contaminants, substrate, moisture content, particle size etc. The detailed specification forms part of the confidential contract between LUBILOSA and the production companies, and thus enables LUBILOSA to ensure the quality of the final industrial product. A less detailed specification is published, and should be of assistance to anyone wishing to develop a usable biopesticide.

Most end-users, from Sahelian farmers to aerial spray operators, have commented that a ready-to-spray formulation is far preferable to a dry powder which must be mixed with formulation ingredients. In response, we developed two formulations, an ‘oil-flowable’ (OF) formulation, which still needs to be diluted with diesel, and an ‘oil suspension’ (SU) formulation which can be sprayed directly from the bottle.

These formulations are robust, with little or no settling, and can be sprayed with spinning disk sprayers of many types; hand-held, vehicle mounted or from an aeroplane.

One of the principal constraints to the implementation of biopesticides is the common perception that they must be stored at low temperatures and have a short shelf life. We wanted to improve the storage properties of Metarhizium spores up to the biological maximum. We found that there is a lot in common between the storage properties of spores and of seeds, so we were able to tap into a body of existing information and develop a model for storage. This model confirmed that spores must be dried to <5% moisture content and sealed hermetically for optimum storage. In this way, spores will keep for several years at temperatures below 20°C. We carried out field validation of the model along the pesticide distribution chain. In southern Benin average temperatures in stores were ca. 30°C; spores will keep their viability for over a year at this temperature. However, maximum temperatures in pesticide stores in the Sahel can reach 50 or 60°C, and average temperatures vary around 35°C; the hottest time of the year is April, just before the rains (and grasshopper infestations) start. At such temperatures, spore will only survive for a few weeks.

In practice, though, even ordinary chemical pesticides are not stored for long in these stores; the Niger plant protection service usually distributes pesticides only shortly before the start of the control operations. We can also make a few simple arrangements, such as improving store ventilation and using insulated delivery boxes. We propose to attach a visible temperature indicator to each package; these indicators cost only a few cents each and turn colour when exposed to unacceptably high temperatures.

The identification of fungus isolates is essential; to support registration dossiers, to check on the identity in production, and to check that field effects are really due to the applied fungus.

Classical fungus taxonomy is based on spore size and shape; but this has led to confusion in the past, and molecular methods are increasingly being used to characterise micro-organisms. For filamentous fungi, the preferred method is sequencing of the ribosomal DNA (rDNA). rDNA contains both conserved and variable sequences. Using this method, the genus Metarhizium was re-classified by Richard Milner’s group at CSIRO, Australia. The LUBILOSA isolate has been moved from Metarhizium flavoviride to Metarhizium anisopliae, and given a new variety name, var. acridum. This result is consistent with our own work in this area, and most fungal taxonomy experts agree with Milner’s suggested classification scheme. From a practical view-point, DNA sequencing facilities are not available in Africa, so there might still be need to develop a field-oriented characterisation method. However, many laboratories offer sequencing services at reasonable rates ($30 – 40 per sample). Similar methods are used to establish paternity law-suits, so the method is highly sensitive and caution will be necessary in the interpretation of results. For instance, if two isolates have minor differences in their sequence data, this does not necessarily mean that they will have ecological differences in the field.

Always the biggest problem in locust research is to be able to tackle actual desert locusts in the field; either there are none, they are in a war zone, or there is a big infestation and spray teams treat them quickly with chemical pesticides. As well as the difficulty of locating the right scale of infestation, quantifying the impact of a control agent on locusts is difficult; the bands move fast, and vary in size and density. We were finally successful in developing a method, involving photographing hopper bands early in the morning. Afterwards, the photographic image is scanned and calibrated, and the area occupied by the band calculated. Work in Phase 3 continued on this topic in the hope of developing a simpler, but adequately precise technique; the conclusion, however, is that all methods are too time consuming to be practical during control operations. We also collected data on the basic ecology of desert locust hopper bands, including feeding and predation.

LUBILOSA has gained a unique understanding of the impact of Metarhizium treatments on grasshopper and locust populations through an ecological approach. ‘Classical’ biological control agents such as parasitic wasps breed and spread in space and time far beyond their original release, and bring the pest population to a permanent equilibrium level below the economic threshold. Natural incidence of Metarhizium is very low; by applying large amounts of inoculum in the form of mycopesticide we artificially and temporarily displace the equilibrium in favour of the fungus and, artificially and temporarily, reduce the grasshopper population.

We showed that under moist conditions, spores are produced on insects killed by the mycopesticide application and these are able to infect other grasshoppers; the effect is limited because predators and scavengers remove most of the Metarhizium -infected grasshoppers, both sick and dead ones.

We developed some user-friendly computer models based on this research, but the most useful outcome has been an understanding of the environmental constraints on the effectiveness of Metarhizium in the field. Grasshoppers and locusts are able to raise their body temperature in response to an infection, just as we do when we have a fever. The difference is that grasshoppers need to sun themselves to raise their body temperatures. So the old understanding of pathogens being more effective under wet (cloudy) conditions may actually, in the case of grasshoppers, be because they are unable to sun themselves and thus raise their body temperature (thermo-regulate). We are working to incorporate this knowledge into a geographic information system (GIS) that would allow locust control officers to predict whether conditions would be unfavourable for Metarhizium use.

This ecological understanding has enabled us to draft outline use strategies which will enable us to maximally exploit the biological properties of Metarhizium. These use strategies will be refined as users gain operational experience of Metarhizium.

The rationale for developing biological pest control agents is that they will reduce environmental damage during locust and grasshopper control operations. Environmental damage due to chemical pesticides has been investigated but poorly quantified. With an environmentally benign agent, we have the opportunity to make a realistic comparison. Standard chemical pesticides carry the risk of poisoning humans, livestock, birds, lizards, fish, aquatic invertebrates, bees, other beneficial insects and soil invertebrates. Newer pesticides in general are less dangerous, particularly to vertebrates, but are not necessarily any safer towards the non-target invertebrates. We tested Metarhizium also against this range of organisms, and found no particular risks. A few non-target insects can be killed in the laboratory when they are stressed and the fungus dose is high enough. Some of these cases, such as parasitic hymenoptera and termites, we investigated in more detail and found no transmission or evidence for any risk of infection in the field.

One of the principal issues in ecotoxicology is the vast size of the samples and the huge amount of work sorting the samples and analysing the data; naturally this heavy work load leads to delays in the publication of results. The first year’s study (1996) therefore focussed on the development of rapid assessment techniques based on presence-absence sampling. Additionally in the first year, we conducted baseline studies and selected the indicator species. The technique and base-line data were then used in the 1997 study to compare Metarhizium with fenitrothion, and in 1998 with fipronil. In both trials, we observed an absolutely phenomenal degree of specificity and selectivity – Metarhizium biopesticide really does just pick out grasshoppers, leaving everything else alive and well, while the chemical pesticides had some severe effects on non-target arthropods.

Ecotoxicology is in its infancy in the tropics, particularly in Africa. Many test protocols for registration require testing on organisms which are not even present in the ecosystem; earthworms, Daphnia water fleas, pheasants would be better represented in tropical Africa by termites, Streptocephalus crustaceans and lizards. We developed test protocols for lizards, but African countries are only now rationalising their registration requirements and further work in this area needs to be driven by registration authority requests.

The steps necessary to incorporate the use of Metarhizium into broader locust and grasshopper management strategies are more numerous and complex than envisaged at the start of Phase 3. We can divide the issues into four areas; work with national programmes, NGOs and farmers; socio-economic studies; implementation plan; and Phase 4 activities. Work with CILSS, with national programmes, farmers and NGOs is the grass-roots of LUBILOSA; this is where we test the Metarhizium product with the people who will be using it in the long run. As far as possible, we encouraged our network collaborators to submit a plan for a trial; we just sent the spores and the money, and waited for the eventual report. These reports were assembled at annual meetings, in Cotonou, Bamako and Niamey respectively, and are an indicator for the investment in training activities in Phase 2 of LUBILOSA.

Where we wanted to monitor the response of farmers, we sent observers to farmer participatory trials. These were either in the Mono province of southern Benin, in Niger or in Mali. In general, we found that NGOs were more in tune with this approach than the government plant protection services, although particularly in Niger, there was a lot of enthusiasm for a participatory approach.

The relative roles of NGOs and the government services in Africa are complex and evolving. Essentially, the externally-financed NGOs have manpower and operational budgets, but lack specialised expertise. The government plant protection services (PPS) have manpower and skills, but lack operational budgets. LUBILOSA in its research phase, and as chemical companies do, contracted the PPS to carry out trials. The PPS gained specialised technical skills in the process, and are now in a position to provide those skills to NGOs under contract (cascade effect).

We carried out a series of socio-economic studies in Phase 3. Essentially three tools were used; cost-benefit analysis, willingness to pay and macro-economic or market studies. The socio-economist was also involved in supervising participatory trials. The cost of production was assessed by examining the cost of running the pilot plant at IITA; about $21 per ha. Onto this we must add the cost of distribution and profit; however, it is certain that the commercial producers will be able to produce spores at a lower cost. Dose rate trials in Mali in 1998 indicated that a dose of 50g/ha gives adequate control; this could reduce the cost of the formulation by up to 50%.

We made the first estimate of the ‘externalities’ of chemical pesticide usage in Africa – we concluded that the medical, veterinary and pesticide disposal costs met by society when farmers use chemical pesticides comes to about $1.75 per ha.

Benefits are harder to estimate; locust and grasshopper attacks are notoriously variable, and the best efforts of GTZ, USAID and NRI have failed to provide convincing data. We attempted a similar approach, but were let down by low levels of grasshopper attack in 1998. We developed alternative approaches using regression analyses and farmer’s perspectives. The regression model needs more data collection, but should provide reliable estimates in the longer run.

Adding the farmer’s perspectives to the crop loss estimates gave substantially higher estimates of the benefits of controlling locusts. We estimated willingness to pay by participatory rural appraisal; such estimates can be unreliable, but the consistency of the estimates from several sites in Niger and Mali gives us some confidence in the figure obtained of $7 per ha. Clearly there is a gap between this figure and our estimate of production cost of $21 per ha ($11 at the lower application rate). Whether we will need to turn to donors to make up the difference, or whether the commercial producers can deliver at a lower cost, remains to be seen. In South Africa, the national parks commission may pay the difference in price between a normal pesticide product and Metarhizium.

Markets were analysed, and the results are collated in a report Assuming only 10% market penetration, there are excellent long-term prospects for the companies.

We often see research projects which produce good results, but fail at the implementation stage. Sometimes the developed technology did not respond to farmer’s needs; more often, the process of technology transfer was not given proper attention. Mindful of this problem, we drew up an implementation plan. The plan includes a careful analysis of the technology, pulling together the economic analysis of production, efficacy data, markets and considering several different scenarios. All scenarios considered that commercialisation is the only feasible implementation route. We didn’t dismiss the option of some sort of low-tech production system in Africa were this to emerge as competitive. But, we concluded that the only option with a chance of success in the near future would be to license the LUBILOSA technology to companies already having expertise in large scale solid state fermentation.

We also addressed Intellectual Property Rights (IPR) within LUBILOSA – a highly complex and sensitive issue in a multi-donor, multi-institutional project. Various documents were drawn up in the course of the transfer process; technical specifications, ecotoxicological profile, mammalian toxicological profile, multi-party IPR agreement, trust fund proposal, confidentiality agreements, licensing agreements. These are atypical for scientists, some of whom were dealing with confidential information for the first time; but were necessary to gain the confidence of, and enable the interaction with, the commercial production companies.

According to criteria laid out in the implementation plan, LUBILOSA proposed two small to medium-sized companies (SME) to produce Metarhizium, and, following discussions with donors and partners, licensing agreements were signed in November 1998 with Biological Control Products (BCP) of South Africa and in January 1999 with Natural Plant Protection of France. BCP submitted a registration dossier to the South African authorities in October 1998, and this was accepted. NPP are expected to submit their dossier to CILSS in June 1999. Both companies expect to scale up production to meet demand in 1999. For a simple product with established markets, that should be the end of the story. However, biological control of locusts is a novel area, and there will be much work to establish a demand for Green Muscle (BCP have purchased this name from LUBILOSA; NPP have yet to decide whether to use the same name or come up with their own.).

So the third phase of LUBILOSA has seen the successful culmination of the process of developing a robust and effective biological pesticide for locust and grasshopper control world-wide and transferring it to the private sector. A product is now registered and in production in South Africa, with a second producer for the West Africa market poised to start production in France, and eventually transfer to Senegal or Cote d’Ivoire. Nevertheless, the consensus opinion is that just launching the product will not necessarily lead to the scale of adoption which the project sponsors, participants, clients and spectators feel justifies the investment to date; indeed, without further support the products may fail. A modest fourth phase will therefore go ahead to ensure the large-scale uptake Metarhizium. This will involve activities to develop both the supply side and the demand side of the economic equation. On the supply side, we have product stewardship; helping out the production companies in case of problems and helping to advertise the availability of Green Muscle. On the demand side, stimulating demand by information, publicity and advocacy, and developing integrated use strategies which capitalise on the particular properties of Metarhizium.

[ Main Page | Project Summary | Field Trials | More Information ]

[ Download Area | Publication List | Links | Staff and Collaborators | FAQs ]

[ Acronyms | Contact Us | Press Release ]

[ IITA Project B - Developing Biological Control Options | IITA news | IITA Home ]