Striga gesnerioides
(cowpea witchweed)

S. gesnerioides is widely distributed in Africa, from Morocco and Egypt, south to South Africa, also in Arabia and widely in India (up to 2000 m) and Sri Lanka. Holm et al. (1979) indicate occurrence in Australia, but there is no more recent confirmation of this. It also occurs locally in Florida, USA.

A record of S. gesnerioides 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. gesnerioides is not present in Japan.

Across most of its range it occurs only on wild hosts. It is only in the western African countries of Senegal, Mali, Togo, Benin, Burkina Faso, Ghana, Nigeria, Niger, Cameroon and Chad that it constitutes a serious weed problem on cowpea. In Zimbabwe, South Africa and Ethiopia it occurs only very locally on tobacco and/or sweet potato.

All Striga species are listed as prohibited imports into Australia, Israel, Russia and USA.

S. gesnerioides is a plant of the semi-arid tropics, associated both with agriculture and with undisturbed natural vegetation. In southern Africa, Hepper (1990) indicates occurrence in Mopane woodland, rocky grassland and cultivated ground. In West Africa, it is apparently able to grow under quite wet conditions but fails to attack aquatic species growing in water 10 cm deep (Porteres, 1948). A survey by Cardwell and Lane (1995) suggested an association of S. gesnerioides with sandy soils.

S. gesnerioides, as a species, has a wide host range, including annual, perennial and woody species, mainly in Dicotyledonae but also some grass species. The main economic host is cowpea, with tobacco and sweet potato also affected very locally. The most commonly affected hosts occur in the families Acanthaceae, Convolvulaceae, Euphorbiaceae, Fabaceae and Solanaceae (Parker and Riches, 1993). Other genera providing occasional hosts, as listed by Porteres (1948) and others, include Sansevieria (Agavaceae), Commiphora (Burseraceae), Cleome (Capparaceae), Helianthus (Asteraceae), Bergia (Elatinaceae), Echinochloa, Panicum, Rottboellia, Brachiaria, Paspalum, Andropogon, Cymbopogon, Hyparrhenia, Oryza, Setaria, Hyparrhenia, Eleusine (Poaceae), Dysophylla (Lamiaceae), Pterodiscus (Pedaliaceae), Cissus (Vitaceae).

Individual biotypes of S. gesnerioides have a much narrower range of hosts, however. The types attacking cowpea are rarely detected on any wild host. An exception is the occurrence of a Nigerian biotype on Indigofera spicata and I. tinctoria as well as cowpea in pot experiments by Igbinnosa and Okonkwo (1991); another type with deeper purple flowers and more slender stems, which overlaps in distribution, attacks Tephrosia (Fabaceae), Jacquemontia and Merremia species (Convolvulaceae). Although the host range of individual biotypes is generally very limited, the hosts attacked by a single biotype may come from diverse plant families. The form occurring in Florida, USA, has only five or six known hosts but these include sunflower (Asteraceae), sweet potato (Solanaceae), Jacquemontia tamnifolia (Convolvulaceae) and Alysicarpus vaginalis (Fabaceae) in addition to the main host Indigofera hirsuta (Fabaceae) (Upton, 1979). Other forms are apparently specific (or nearly so) to tobacco, or to Euphorbia species, while individual biotypes can also differ in the varieties of cowpea that they can parasitize, as noted under Control. These specificities apparently depend on factors involved after attachment, rather than at the germination stage, but the precise mechanisms are not yet fully understood.

References from before 1957, cited here and in other sections, are usefully abstracted in the compilation by McGrath et al. (1957).

The biology of S. gesnerioides is generally very similar to that of S. hermonthica and S. asiatica. It is an obligate parasite with minute seeds, unable to establish without the help of a host plant. Germination depends on a period of moist conditioning and exposure to germination stimulants in host root exudates, the most important of which is alectrol (so named because it also stimulates germination of the related parasite Alectra vogelii) (Muller et al., 1992). This is closely related to the lactones, strigol and sorgolactone, which stimulate S. hermonthica and S. asiatica. S. gesnerioides is also stimulated by ethylene but is relatively insensitive to the strigol analogues GR 7 and GR 24 (Igbinnosa and Okonkwo, 1992; Maass, 1999). Prolonged conditioning in the absence of stimulant results in a secondary 'wet dormancy' (Reid and Parker, 1979; Maass, 1999). Germination temperature appears to be uncritical over the range 23-33°C. Germination is slower than in other species, taking 2-3 days.

Attachment and penetration of the host root do not appear to differ from the other main species (see Reiss and Bailey, 1998), but the physiology of the established parasite differs in showing very low rates of photosynthesis. The effects on the host also differ in that there is no change in root:shoot ratio. Host roots beyond the point of attachment tend to abort. Host photosynthesis may be reduced to some degree but the greatest damaging effect is attributed to the removal of metabolites from the host (Graves et al., 1992; Hibberd et al., 1996). Fertilization is autogamous (Musselman et al., 1991).

A survey by Cardwell and Lane (1995) suggested an association of S. gesnerioides with sandy soils.

S. gesnerioides in West Africa is frequently affected by several gall-forming Smicronyx species, one of which, S. dorsomaculatus, has recently been described as a new species (Anderson and Cox, 1997). The galls caused by Smicronyx species occur variously in the capsules and in the base of the stem. Markham (1985) concluded that stem galls and plant pathogens, including Macrophomina phaseolina, often significantly reduced vigour and seed production of S. gesnerioides in Mali. It has also been noted that use of insecticides in cowpea can result in the destruction of Smicronyx and an increase in vigour of the parasite (Parker and Riches, 1993).

Where infestation is suspected, from previous history or from symptoms of chlorosis, uprooting the crop can reveal the nodules of young Striga seedlings. These can vary in size from a few millimetres to over 2 cm in diameter and are somewhat irregular in shape.

A technique for detecting the seeds of Striga spp. 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. However, the seeds of S. gesnerioides are not readily distinguished from those of other Striga species such as S. asiatica or S. hermonthica (see Musselman and Parker, 1981).

S. gesnerioides is not readily confused with other Striga species, due to the lack of expanded leaves (see Parker and Riches, 1993, for a key to agriculturally important species).

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.

Cultural Control

No cultural control methods have been widely developed or adopted. Crop rotation should be effective over the long term but is rarely practicable and there has been little research on the potential for trap crops, although pigeon pea, velvet bean (Mucuna species), sorghum and soyabean have been suggested (Wild, 1948; Igbinnosa and Okonkwo, 1991; Berner and Williams, 1998). Cardwell and Lane (1995) also note an apparent absence of S. gesnerioides where cotton is grown. The use of manures or fertilizer to increase soil fertility has apparently much less impact on this species than on those attacking cereal crops (Parker and Riches, 1993). Hand-pulling is difficult and liable to uproot the crop itself, though it should be used where new sporadic infestations are discovered.

Chemical Control

Some herbicides have shown moderate promise for conventional pre-emergence application (see Parker and Riches, 1993), but the farmers affected by S. gesnerioides are not generally in a position to use these and there has been no field use. A more recent development has been the demonstration that S. gesenerioides may be controlled by application of the herbicide imazaquin to the crop seed itself (Berner et al., 1994) but it is uncertain whether this has yet been used in practice.

Biological Control

S. gesnerioides is often very heavily affected by Smicronyx gall-forming weevils and while there has been no attempt to exploit these for biological control it has been noted that this natural control can be adversely affected by insecticide use. The only effort towards biological control has been the testing of the ethylene-generating bacterium Pseudomonas syringae as a means of inducing suicidal germination (Berner et al., 1999). Promising progress in the development of mycoherbicides for control of S. hermonthica, based on Fusarium species, has recently been reported (Watson et al., 2000), and could have relevance for control of S. gesnerioides in the future.

Host-Plant Resistance

The main control measure available for cowpea is varietal resistance. The most important source of resistance is the landrace B301, originally selected for its partial resistance to Alectra vogelii in Botswana (Parker and Riches, 1993). B301 fortunately shows high-level resistance to A. vogelii in West Africa (based on two dominant genes) as well as to S. gesnerioides (based on a single dominant gene) (Singh et al., 1993; Atokple et al., 1995). The resistance, or virtual immunity, of this line has been effective against all biotypes of the parasite in West Africa except that occurring locally in southern Benin. Lane et al. (1996) describe the existence of five known parasite biotypes, varying in their virulence on different 'resistant' varieties of cowpea. Two other sources of resistance, Suvita-2 and IT82D-849, have different single dominant genes for resistance to the Mali biotype, and a different pattern of response to the five parasite biotypes (Atokple et al., 1995). While B301 and IT82D-849 resist four of the known biotypes, Suvita-2 is resistant only in Mali. Fortunately, although each of these three sources is susceptible in southern Benin, other resistant lines, 58-57 and IT81D-994, are resistant to this Benin biotype (Lane et al., 1993) even though they show susceptibility to biotypes in Niger and Nigeria. Further lines with resistance to some biotypes include APL-1 and 87-2 (Moore et al., 1995) but none of these, other than B301, and to some extent IT81D-994, has cross-resistance to Alectra. The mechanisms of resistance are not fully understood but involve a failure of the parasite to develop normally following penetration of the parasite haustorium into the host root (Reiss et al., 1995).

Although B301 is an agronomically poor line, the simple dominance of the resistance character has allowed its ready transfer into more desirable varieties (Singh and Emechebe, 1991). IITA (International Institute for Tropical Agriculture) has now developed lines with resistance to Striga and Alectra, as well as to various other pests and diseases. Singh (1999) lists the most promising of these and indicates that IT90K-76 and IT90K-59, both with resistance from B301, have already been released in Nigeria and South Africa, respectively. Progress is also being made in the development of varieties with combined resistance to Alectra and all five biotypes of S. gesnerioides, using crosses between 58-57 and the B301-derived IT90K-76 (Singh and Emechebe, 1997). In Senegal, Cisse et al. (1995) report a useful degree of resistance in the variety Mouride, whereas the variety KN-1 demonstrates a degree of tolerance, being less damaged in spite of parasite development (Gworgwor, 1991).

Although there has been no evidence as yet for the breakdown of resistance after repeated trials of the varieties based on B301 and other lines, vigilance will be needed to detect any such breakdown and minimize the risks of a build-up of more virulent biotypes. Shawe and Ingrouille (1993) used isoenzyme techniques to demonstrate differences between the populations of S. gesnerioides (from Niger) which emerged on a susceptible variety and on Suvita-2 which is partially resistant to the Niger biotype, emphasizing the risk of selection for virulence in any but totally immune varieties.