Striga asiatica
(witch weed)

S. asiatica is a hemiparasitic plant, native to Africa and Asia. In common with most other parasitic weeds, it is not especially invasive in natural vegetation, but is much feared in crop land where infestations can build up to ruinous levels, especially with repeated growing of susceptible cereal crops. For this reason it is included in almost all lists of noxious, prohibited plant species. It has recently been reported in Queensland, Australia. There is also evidence for its continuing spread and intensification within a number of countries in Africa in particular in rice in Tanzania and maize in Malawi. A study by Mohamed et al. (2006) suggests that on the basis of climatic data, there are many territories into which Striga species, including S. asiatica, could be introduced and thrive. Global warming could further increase this potential.

S. asiatica (sensu lato) is probably indigenous to most of the countries in which it now occurs in Africa and Asia, though it has apparently spread and increased in importance within many of them in recent decades. The red-flowered form, most serious as a weed, is native to east and southern Africa but has apparently spread to Egypt and Madagascar (Mohamed et al., 2001) and possibly to some West African localities, while the brick-red-flowered form in Ethiopia also seems likely to be a recent introduction (Parker, 1988).

Less aggressive forms occur through most of southern and south-east Asia.

The one clear example of long-distance introduction has been into the USA, where an infestation already covering about 200,000 hectares was only recognised in 1956. In this case, the red-flowered form of the weed is presumed to have originated by accidental importation from Africa. It is not recorded from any other country of North or South America.

Occurrence in Australia was uncertain and eventually discounted by Carter et al. (1996) but has now been confirmed in Queensland (The Observer, 2013). A record of S. asiatica in New Zealand in an earlier edition of the Compendium was erroneous and has been deleted.

A record of S. asiatica in Japan (Holm et al., 1979; EPPO, 2014) published in previous versions of the Compendium is unreliable. No original source was provided for the record in Holm et al. (1979). According to Spallek et al. (2013), S. asiatica is not present in Japan.

There is considerable scope for further spread of S. asiatica, especially to semi-arid regions of North and South America, Southern Europe, North Africa and Australia (Mohamad et al., 2006). The potential pathways for accidental introduction include crop seed and other agricultural produce and packing materials from infested areas, and also soil where this is not strictly prevented. Striga species are listed as quarantine pests in most countries with any developed regulations but the risks of accidental entry are enhanced by the very small size of the seeds which make inspection and enforcement extremely difficult. Deliberate introduction is unlikely other than by research scientists.

Most of the agriculturally important Striga species favour relatively dry, infertile soil conditions and are typically problems in the semi-arid tropics of Africa and Asia. S. asiatica follows this characteristic, confirmed for the main weedy form by Kabiri et al. (2015) in southern Tanzania, where they showed that it was restricted to drier, free-draining soils and did not overlap with the lower, wetter conditions favoured by Rhamphicarpa fistulosa. However, in some countries, e.g. in Indonesia and other South-East Asian countries, some forms of the species attack wild grasses under much wetter conditions. Apart from spilling into such wetter ecologies, S. asiatica also occurs just south and north of the tropical belt, in South Africa and in the USA, respectively. In many countries, S. asiatica is associated with sandy soil conditions, as in Tanzania (Doggett, 1965) but it can occur on a wide range of soil types (Robinson, 1960).

S. asiatica is an obligate parasite and cannot develop without a suitable host plant. Apart from the major crops including sorghum, maize, pearl millet, finger millet, Panicum millets and rice, a wide range of wild hosts are parasitised. For the USA, these are listed by Nelson (1958), the most common being the weed Digitaria sanguinalis. There are many other wild hosts in Africa and Asia. Those recorded up to 1956 are listed in McGrath et al. (1957). Some further host species almost certainly include Andropogon gayanus, Axonopus compressus, Chrysopogon acicularis [C. aciculatus], Digitaria smutsii [D. eriantha], Eleusine indica, Elionurus elegans,Eragrostis malayana [Eragrostis montana], Ischaemum indicum [Polytrias indica], I. timorense, Microchloa indica, Rottboellia cochinchinensis, Sporobolus festivus and Stenotaphrum dimidiatum (synonym S. secundatum). Further hosts are listed by Cochrane and Press (1997). A number of broad-leaved hosts are recorded in McGrath et al. (1957) but occurrence on anything other than Poaceae is quite rare and some records may be suspect. Due to the very fragile connection between host and parasite it is often difficult, in a mixed grass community, to identify the host with any certainty.

There are distinct ecotypes of S. asiatica with different host specificity and each form may have quite limited host range. Botanga et al. (2002) describe a range of forms in Benin, including red- and yellow-flowered types with differing virulence on maize and sorghum but do not ascribe sub-specific names to these forms. Hence, there are many regions in Africa and Asia where the species occurs but is restricted to wild hosts and does not affect crops.

Genetics

Chromosome number 2n = 38 or 40. S. asiatica (sensu lato) is a highly variable species with many self-perpetuating forms varying in stature, flower colour and host preference. Autogamy tends to keep these forms quite distinct and hybridization between forms or inter-specifically is apparently rare. Genome size of 615 Mb is smaller than that of S. hermonthica (1425 Mb), or of S. gesnerioides (1330 Mb) (Estep et al., 2012).

Reproductive Biology

S. asiatica reproduces by seed only. Most populations appear to be autogamous, self-pollinating even before the flowers open (Musselman et al., 1981). Large numbers of minute seeds are produced but there is no specialised dispersal mechanism. Longevity of at least 14 years has been recorded in the field in both South Africa (Saunders, 1933) and USA (Bebawi et al., 1984).

Physiology and Phenology

The seeds of S. asiatica are less than 0.5 mm long. Most Striga species, including S. asiatica have insufficient resources to establish seedlings independently and are thus obligate parasites, depending on attachment to a host root within a few days of germination. To improve the chances of such attachment the seeds generally germinate only after stimulation by a substance exuded from the roots of a potential host. The natural stimulants in root exudate have been identified as a group of closely related lactone compounds, known as strigolactones, which appear to act on the seeds via the generation of ethylene. These substances are now known to be naturally exuded by the host with the function of stimulating the branching of arbuscular mycorrhizae which act synergistically with the host, taking carbohydrate but greatly enhancing the host’s uptake of nutrient, especially under conditions of low phosphorus and nitrogen (Akiyama et al., 2005). Before the seeds can respond to such a stimulant they have to be imbibed ('conditioned') for a period of about 7-10 days. Beyond this period of moist incubation, there may be a decline in germination, known as secondary dormancy or 'wet dormancy'. Once germinated, the radicle extends and, on contact with a host root, stops elongating, starts swelling and develops sticky hairs which secure the developing haustorium to the host root surface. Apical cells of the Striga radicle then develop an intrusive organ which penetrates the host epidermis, cortex and endodermis and develops good connections with the host xylem, but not with the phloem (Dorr, 1997). For further information see Parker and Riches (1993).

In the absence of a host the seedling dies within a few days, hence, the value of trap crops and ethylene gas which can be used to stimulate 'suicidal germination'.

Until emergence, the parasite seedling draws all its water, mineral and sugar requirements from the host. After emergence, the parasite photosynthesises some of its own sugars, but this photosynthesis is much less efficient than that in a normal green plant and it continues to be dependent on the host plant for vigorous growth (Harpe et al., 1979). Press et al. (1987) confirm that Striga species have C3 type photosynthesis and show that in the related S. hermonthica, the photosynthesis is relatively inefficient and barely balances the losses from respiration. Hence growth is highly dependent on the additional carbohydrate from the host. The Striga species are also abnormal in having almost permanently open stomata, ensuring a high transpiration rate and enhancing the flow of water and food materials from host to parasite. Humid conditions can reduce this flow and reduce parasite vigour, at least partly explaining the association of the weed with relatively arid conditions.

Apart from sapping the host of water and nutrients, S. asiatica has a profound physiological effect on the host from a very early stage, apparently as a result of some toxic or growth-regulatory influence passing from parasite to host. In S. hermonthica it has been shown that photosynthesis is substantially reduced, at least partly due to the closure of host stomata, associated with raised levels of inhibitors such as abscisic acid (ABA). These various influences, which are believed to be similar in S. asiatica, result in a stunting of the host shoot system and an increase in host root growth, thus causing a change in root: shoot balance which favours parasite growth. In hosts affected by S. asiatica there is also a tendency for infected host plants to show wilting symptoms, even under moist conditions. Heavy infestations cause serious stunting, wilting and 'burning' of the foliage and almost total crop failure.

Environmental Requirements

Most of the agriculturally important Striga species favour relatively dry, infertile, especially low nitrogen, low phosphorus soil conditions and are typically problems in the semi-arid tropics of Africa and Asia. Conversely, fertile, high nitrogen conditions tend to suppress growth. S. asiatica follows these characteristics (Farina et al., 1985), but in some countries, e.g. in Indonesia and other South-East Asian countries, some forms of S. asiatica will attack wild grasses under much wetter monsoon conditions. Apart from spilling into such wetter ecologies, S. asiatica also occurs just south and north of the tropical belt, in South Africa and the USA, respectively. In many countries, S. asiatica is associated with sandy soil conditions, as in Tanzania (Doggett, 1965) but it can occur on a wide range of soil types (Robinson, 1960).

Relatively high temperatures of 30-35°C are optimal for both conditioning and germination of S. asiatica, but the weed can develop over a wide range of temperatures from 22°C upwards and the dormant seeds can survive prolonged freezing (Robinson, 1960). Day length is not critical; photoperiods up to 16 hours allow normal development (Kust, 1964).

S. asiatica is not often seen to be seriously affected by natural enemies, though the Junonia butterflies may occasionally cause conspicuous defoliation in South-East Asia (Boonitee, 1977) as well as in Africa and the USA. The plume moth, Stenoptilodes taprobanes, is also very widely distributed in Asia and Africa, damaging young shoots and flowers. The other 'principal insect enemies of Striga spp.' listed by Greathead (1984) are the noctuid moths Eulocastra argentisparsa and Eulocastra undulata in India, the agromyzid fly, OphiomyiaStrigalis, in East Africa, and several Smicronyx gall-weevils in both Africa and India. Several of these natural enemies have potential as biocontrol agents.

Among the many fungi recorded, none cause widespread or consistent damage, but a powdery mildew, Sphaerotheca fuliginea [Podosphaera fuliginea], is common in both Africa and Asia. Greathead (1984) comments on several which could be considered for biocontrol; these are listed in the section on Biological Control.

It has been shown that Fusarium semitectum var. majus [Fusarium pallidoroseum] greatly reduced the germination, survival and attachment of S. asiatica on the roots of maize (Abbasher et al., 1996). Work with Fusarium spp. attacking S. hermonthica is also likely to be of relevance (Abbasher et al., 1995; Diarra, 1999; Elzein et al., 2002; Beed et al., 2007; Koltai, 2015). A new species of Fusarium, F. brevicatenulatum has been described from S. asiatica in Madagascar (Nirenburg et al., 1998).

Apart from Greathead (1984), other useful surveys of insects and other organisms which attack S. asiatica include Sankaran and Rao (1966), Nag Raj (1966), Greathead and Milner (1971)Girling et al. (1979) and Kroschel et al. (1999a).

Where infestation is suspected, from previous history or from unexpected wilting symptoms, uprooting the crop can reveal the small white seedlings of S. asiatica on the roots, but the attachments are very fragile and gently washing the roots out of the soil will ensure a better chance of finding them.

A technique for detecting the seeds of S. asiatica as contaminants of crop seed is described by Berner et al. (1994). This involves sampling the bottom of sacks, elutriation of samples in turbulent flowing water and collection of seeds and other particles on a 90 µm mesh sieve. Striga seeds are then separated from heavier particles by suspension in a solution of potassium carbonate of specific gravity 1.4 in a separating column. Sound seeds collected at the interface are then transferred to a 60 µm mesh for counting.

There are a number of species related to S. asiatica with which confusion can occur. In Africa, the most likely confusion is with S. passargei but this species has only five ribs on the calyx. In Asia, especially in India, the predominant white-flowered form of S. asiatica is distinguished by its basic 10 (-14) ribs on the calyx from S. densiflora which has only five and S. angustifolia which has 15. In South-East Asia and Australia, there are three further closely-related species: S. multiflora, S. parviflora and S. curviflora, all distinguished by having only five-ribbed calyces.

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Prevention

Eradication

Regulatory control has been enforced in the USA for many years and by the end of the twentieth century was reported to be finally nearing success with 93% of the originally infested area declared Striga-free by the end of 1994 (English et al., 1995) and 98% by 1999 (Patterson, 1999). Some pockets, however, were persisting still in 2011 (Iverson et al., 2011).

The eradication programme employed in the USA involved the application of herbicide, not only to kill the emerged weed in the cereal crop (maize) but also to control alternative grass hosts in the rotational broad-leaved crops such as tobacco and soyabeans. The spraying programme in maize, which was repeated several times per season, was based on an intensive survey procedure and was supplemented by injection of ethylene gas to cause suicidal germination. Fumigation with various fumigants was used to finally clear small infestations but these have since been banned for agricultural use. A strict local quarantine was enforced, including vehicle washing and restriction on the movement of many classes of agricultural produce.

Control

Cultural Control and Sanitary Measures

The complex germination requirements, and specialised physiology of S. asiatica, together with certain ecological preferences, can be exploited to some degree in cultural control methods.

Preference for relatively dry conditions can be countered by using irrigation wherever available. Preference for infertile soil conditions can be countered by improving soil fertility via the use of nitrogenous and phosphate manures, green manuring (see Chibudu, 1998) or rotation with leguminous crops. The use of Crotalaria green manure crops has proved especially effective in suppressing S. asiatica in rice in Tanzania (Riches, 2005; Kayeke et al., 2007; Rodenburg et al., 2010). Treatments providing such longer-term improvement in soil fertility are generally preferable to the use of inorganic fertilisers, but in Angola N at 60 kg/ha in the form of ammonium sulphate markedly reduced S, asiatica emergence and greatly increased maize yield. (Dovala and Monteiro, 2014) and Kabambe et al. (2007) concluded that fertiliser was the most important contributor to increased maize yield in Malawi. Urea proved helpful in rice in Tanzania, (Riches et al., 2005) but the cost discouraged most farmers from adopting it.

Most early studies failed to show benefit from application of phosphate (e.g. Farina et al.,1985) but more recent research has shown that phosphate may be as important as nitrogen in reducing exudation of strigolactones (e.g. Lopez-Raez et al., 2008; Chen et al., 2009). Jamil et al. (2013; 2014a) have now shown that ‘micro-dosing’ – the application of diammonium phosphate at 2-4 g per planting hole - can reduce emergence of Striga hermonthica and increase yield in sorghum and in millet. Subsequently Jamil et al. (2014b) have shown that similar benefit can be gained more economically by soaking crop seed (rice, sorghum and millet) in potassium phosphate. In one corresponding result on S. asiatica modest doses of di-ammonium phosphate have provided good results when applied in the ridge before planting maize in Malawi (Shaxson and Riches, 1998).

Preference for low humidity and high transpiration rates can be countered by dense crop planting, inter-cropping with leafy species or later planting when rains have set in more fully and continuously, provided care is taken not to prejudice the growth of the cereal crop to any great extent. Conversely, early planting may be beneficial in more temperate zones when temperatures are sub-optimal for the weed. Under dry conditions it has been shown in Ethiopia that a combination of tie-ridging with skip-row planting (omitting every 3rd row) provided better tolerance of Striga species. Rotation out of susceptible cereals is desirable wherever possible to allow the seedbank to decline, though the longevity of the seed means that one or two break crops provide only partial control. Some alternative crop species, however, can act as 'trap-crops', stimulating germination of the parasite but failing to allow parasitisation. Germination is stimulated but the parasite seedlings fail to penetrate the root beyond the cortex (Hood et al., 1998). These are ideal crops to include in rotation, but must be planted at a time and a density to ensure optimum germination. For S. asiatica, suitable crops include cotton, cowpea, soyabean, pearl millet (where not normally attacked) but even these may need to be grown for several seasons to achieve a useful reduction of the infestation. The choice of crop variety may also be significant (Iwo and Uwah, 2007; YongqingMa, 2015).

In some regions, the choice of trap crop may be influenced by their susceptibility to parasitisation by Alectra vogelii. In Malawi, Kabambe et al. (2008a) found certain soyabean varieties, pigeon pea and velvet bean (Mucuna pruriens) to be suitable where A. vogelii occurred precluding the use of the susceptible cowpea. Moe recently Alectra- resistant cowpea varieties have become available (Kabambe et al., 2014). 'Catch-crops' are susceptible crop or forage species which are used to encourage germination of the weed but are then destroyed before it has time to set seed. These are rarely popular with the farmer but may be suitable in some ecologies, especially if there is a bi-modal rainfall when the less reliable season can perhaps be devoted to such a crop.

Inter-cropping with e.g. cowpea can result in significant reductions in S. asiatica and increases in yield of maize (Chivinge et al., 2001; Atera et al., 2013). The benefit is mainly shown under wet conditions, when the inter-crop further reduces transpiration of the parasite. Under dry conditions, cereal crop yield may be reduced and economics then depend on the productivity and value of the intercrop. In the case of S. hermonthica, the intercrop is more effective when planted intra-row rather than inter-row and in Mozambique within-row pigeon pea was more effective than N and P in increasing yield and suppressing Striga asiatica in maize (Rusinamhodzi et al., 2012).

Using the perennials Desmodiumuncinatum or D. intortum as the intercrop, as in the so-called ‘push-pull- technique (designed to control stem-borer, the Desmodium intercrop repels the moths while a surrounding stand of grass attracts them) appears to have even greater benefit, suppressing S. hermonthica almost completely in maize and greatly enhancing yield, especially as soil fertility is built up over a period of years (Khan et al., 2000; 2006; 2007). In South Africa Reinhardt and Tesfamichael (2011) confirm that a combination of D. uncinatum with added nitrogen can provide 100% suppression of S. asiatica in sorghum. In Angola, comparable results against S. asiatica in maize have been obtained using intercrops of Desmodium uncinatum (cv. D. Silver leaf), Cajanus cajan, Mucuna pruriens, species of Tephrosia. And species of Crotalaria. Trao-cropping with Tripsacum laxum has also proved useful in those trials.

In certain situations, transplanting may be an effective means of reducing Striga attack (Dawoud et al., 1996).

Mechanical Control

Mechanical methods of destruction are not generally satisfactory. If the weed is hoed at ground level, it is almost certain to regenerate from buds below ground and hence hoeing needs to be repeated at quite frequent (2-3 week) intervals. However, good results have been reported with mid-season ridging (Kasembe and Chivinge, 1997). Hand-pulling has a more durable effect but is not practicable for dense infestations and can only be considered for mopping up scattered populations before they build up or after they have been greatly reduced by other methods. Hoeing or hand-pulling are much easier in row-planted crops than in broadcast plantings.

Biological Control

Organisms considered to have potential as biocontrol agents for S. asiatica and/or other Striga species have included the gall-weevil, Smicronyx spp. the agromyzid fly, Ophiomyia Strigalis, the moths, Eulocastra argentisparsa and Eulocastra undulata, the plume moth, Stenoptilodes taprobanes, the powdery mildew Sphaerotheca fuliginea and other fungi including Drechslera longirostrata, Phoma and Cercospora species (Greathead, 1984). Evans (1987) considered Cercosporastrigae the most promising of the fungi. Several different species of Fusarium have been shown to have potential against both S. hermonthica and S. asiatica (Abbasher et al., 1995; 1996). None of these organisms have yet been successfully utilised, but the possibilities for using Fusarium oxysporum are still considered promising (Beed et al., 2007). The strain Foxy 2, widely tested on S. hermonthica is confirmed also to be active against S. asiatica (Elzein et al., 2002). More virulent strains are now being tested along with a novel means of on-farm culturing and it is hoped there could be commercialisation before too long (Koltai, 2015). The only attempts to control S. hermonthica have been with the introduction of Smicronyx albovariegatus and Eulocastra argentisparsa from India into Ethiopia in 1974 and 1978, but there is no evidence that these organisms ever established.

Chemical Control

S. asiatica is readily killed by a number of post-emergence herbicides including 2,4-D; this herbicide has been one of the most important components of the eradication campaign in the USA. In sole-crop cereal, 2,4-D can be applied to the emerged weed but repeat applications may be needed as more parasites emerge. Disadvantages of the use of this herbicide in the peasant farming situation include the cost of herbicide and sprayer; the fact that it is most effective after weed emergence and may do little to prevent the damage and increase yield; and the risk of damage to non-cereal crops, whether adjacent or inter-cropped. Application before Striga emergence has occasionally been effective but is unreliable. In spite of its limitations, this treatment is one of the most important options for control of dense infestations of the weed. While crop safety is normally adequate in maize, sorghum, pearl millet and sugarcane, where application can be directed between crop rows, selectivity in rice and the minor millets may need to be confirmed before widescale use.

Dicamba may be more effective against Striga before emergence, but is more expensive than 2,4-D and is also damaging to broad-leaved inter-crops. Alternatives for post-emergence use in inter-cropped cereal include ametryne, linuron, cyanazine, oxyfluorfen, fomesafen, acifluorfen and lactofen, but all are more expensive than 2,4-D and none have been widely used for this purpose outside the USA. For sparse infestations, cost and the difficulty of application may be alleviated by using inexpensive 'water-pistol' applicators.

Oxyfluorfen, trifluralin and pendimethalin have been used in the USA as pre-emergence treatments to prevent Striga emergence, but damage from the parasite is not necessarily reduced, and the cost and difficulty of application make them unsuitable for use in small-scale farming. In Malawi, metolachlor has been tested with some promising but not fully reliable results (Kabambe, 2004; Kabambe et al., 2008b).

Herbicides of the sulphonylurea group, nicosulfuron and rimsulfuron, have shown promise for selective control of S. hermonthica in maize and deserve further testing (Adu-Tutu and Drennan, 1991). However, practical field use of these latter herbicides is likely to depend on combination with herbicide-tolerant crop varieties. For instance, Abayo et al. (1998) report the useful selectivity of imazapyr and chlorsulfuron and related herbicides against both S. hermonthica and S. asiatica when applied into the planting holes with seed of maize having target-site resistance to acetolactate synthase-inhibiting herbicides (obtained by tissue culture mutation rather than by genetic modification). In a similar study, Berner et al. (1997) applied the herbicides imazaquin and nicosulfuron to the maize seed itself before sowing. Further extensive work on these lines has been done in East Africa by e.g. Kanampiu et al. (2002), confirming that when naturally imidazolinone-resistant maize seed is dressed with imazapyr or pyrithiobac for control of S. hermonthica, high levels of Striga control are obtained and it is safe to interplant legumes so long as they are at least 15 cm from the treated maize seed. The technique is now fully developed and commercialised in East Africa and has been shown to be effective against S. asiatica as well as S. hermonthica (Kanampiu et al., 2004; Kabambe et al., 2008c). The technique has also been confirmed effective against S. asiatica in maize in Angola (Dovala and Monteiro, 2013). It is likely that glyphosate will become a useful tool for post-emergence control of Striga species in glyphosate-tolerant maize. It is less likely, however, that herbicide-tolerant varieties of sorghum or pearl millet will be developed. Gressel (1999) reviews the latest results with this technique and suggests novel approaches to reducing the risk of developing herbicide-tolerant Striga.

Various fumigants have been used in USA to eradicate small infestations of S. asiatica but were expensive and are no longer available. Injection of ethylene into the soil, also used in USA to stimulate suicidal germination could be of potential use in Africa but is no doubt too expensive for serious consideration. Similarly various analogues of the natural strigolactone stimulants have been considered as soil treatments to trigger suicidal germination but most are too short-lived in the soil to be effective. More stable analogues have now been tested by Kgosi et al. (2012) and shown to have potential against both S. hermonthica and S. asiatica. These are related to Nijmegen-1 but several have greater activity, especially one derived from 1-tetralone. There are however, no reports of their further testing.

Host Resistance

Sorghum shows wide variation in susceptibility to S. asiatica, and many varieties have been selected which show partial-to-excellent resistance, based on a low production of germination stimulant, apparently controlled by a single recessive gene (Haussmann et al., 2001), and/or some other physiological or mechanical incompatibility which is not yet well understood. A range of S. asiatica-resistant ('SAR') lines have been developed by the International Crops Research Institute for the Semi-arid Tropics (ICRISAT) in India; several of these lines have shown good resistance, but not full immunity, over a wide range of locations both in India and in southern and eastern Africa (Rao, 1984; Ramaiah, 1987; Obilana et al., 1991; Obilana, 1998). Mabasa (1996), however, shows that the SAR lines are not well adapted to Zimbabwe and are out-yielded by local varieties DC 75, SV 1, SV 2 and MMSH 413, even when heavily infested, perhaps indicating some natural tolerance. Other older varieties which have been widely tested and shown to have moderate to good resistance to S. asiatica include N.13, Framida, Serena, Weijita, SRN-4841, 148, 168, 555, IS-2603, IS-4202, ICSV-1007, SRN-39; these varieties have not necessarily been widely adopted but some, especially Framida, 555, 148 and 168 have been used as parents in the development of S. asiatica-resistant lines. Resistance and/or tolerance of sorghum varieties to S. asiatica are generally better and somewhat more reliable than resistance to S. hermonthica. This is also true for the wild relative, Sorghum arundinaceum which demonstrates a tolerance to S. asiatica that is not apparent to S. hermonthica (Gurney et al., 2002a) and which may be exploitable in future plant breeding. In Tanzania, the resistant lines P9405 and P 9406 have been widely tested and are now released under the varietal names ’Hakika’ and ‘Wahi’ respectively, proving popular with farmers (Mbwaga et al., 2007).

A large number of maize varieties have been screened against S. asiatica in the USA and resistance has been shown in the inbred TZi-30 and the hybrid Pioneer 3181 (Ransom et al., 1990). TZi-30 is one of the parents involved in the development of hybrids such as 8322-13 with resistance and/or tolerance to S. hermonthica in Nigeria (Kim and Winslow, 1992), and there is some evidence that the hybrids do have some corresponding resistance or tolerance to S. asiatica. Pierce et al. (2003) have demonstrated resistance to S.asiatica in line IWD STR Co, associated with low stimulant exudation, and tolerance in line 98 Syn WEC, showing no significant damage despite high stimulant exudation and normal development of the parasite. Tolerance has also been demonstrated in the East African maize variety Staha (Gurney et al., 2002b). Tolerant varieties, which allow growth of the parasite but suffer less damage, need to be used in conjunction with other methods such as herbicides to avoid increasing the infestation, but may be popular with farmers wishing to obtain an economic return from infested land.

In India the pearl millet inbred TF-23A, and hybrids derived from it, have shown resistance to S. asiatica (Mathur and Bhargava, 1971) but there is no recent confirmation of their continued use. In general, work on resistance to S. hermonthica in Africa has been generally unsuccessful. There are no reports of resistance in any of the minor millets.

Some varieties of rice show good resistance to S. hermonthica (Harahap et al., 1992) and some of these lines, including the breeding line IR-49255-BB-52, and some hybrids between O. sativa and O. glaberrima, are also resistant to S. asiatica (Riches et al., 1996; Johnson et al., 1997; Rodenburg et al., 2010). More recently a series of 18 NERICA lines, developed by crossing O. sativa with the African species O. glaberrima, have been released and have now been tested for resistance to S. hermonthica and S. asiatica (Rodenburg et al., 2015b). NERICA-1, NERICA-9 and NERICA-17 show tolerance to S. asiatica in the field in Tanzania while NERICA-2 and NERICA-10 show resistance. This resistance is shown to be due to a failure of the parasite to penetrate the endodermis (Cissoko et al., 2011).

Sugarcane varieties differ widely in response to S. asiatica, and varieties which have shown useful resistance or tolerance include Co-290, Co-281, CP-33-224, CP-50-28, CP-53-19 and F-31-762 in the USA (Robinson and Stokes, 1960), and N-17, 52/219 and 75-F-2573 in South Africa (Visser, 1987). In Kenya, some varieties show resistance to S. hermonthica (Mbogo and Osoro, 1991).

IPM Programmes

Integrated programmes have been devised for the control of S. hermonthica in countries such as the Gambia, but have not been widely proven or adopted. A programme for control of this relatively large species in a peasant-farming situation almost inevitably relies on hand-pulling as a final means of removing plants after the infestation has been reduced by the use of other methods such as crop rotation, growing less susceptible varieties, manuring, intercropping and delayed planting. In the case of S. asiatica, effective hand-pulling is much less feasible as the plant is so much smaller and the seed matures and sheds rather faster. Hence, there has tended to be greater reliance on single control approaches such as the use of resistant varieties (of sorghum) or herbicides.

IPM programmes must ensure that S. asiatica is not allowed to set seed, because of the very large numbers of seed produced and their longevity in the soil. The range of possible measures to be incorporated into systems of control, according to infestation level, are discussed by Parker and Riches (1993). Other discussions of integrated control and the problems of educating farmers and designing and introducing control methods appear in the volume edited by Kroschel et al. (1999). Also Mbwaga et al. (2007) in Tanzania suggest resistant sorghum varieties in conjunction with animal manure and tied ridges to conserve soil moisture, for integrated control of both S. hermonthica and S. asiatica. In Nigeria, it is recommended that the use of S. asiatica-resistant varieties needs to be combined with high nitrogen fertilization (Anjorin et al., 2013).

19/01/2016 Updated by:

Chris Parker, Consultant, UK

20/11/2009 Updated by:

Chris Parker, Consultant, UK