Details of DPV and References

DPV NO: 355 September 1998

Family: Unallocated ssRNA+ viruses

Genus: Umbravirus

Species: Groundnut rosette virus | Acronym: GRV

Groundnut rosette virus

A. F. Murant Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, Scotland, UK

D. J. Robinson Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, Scotland, UK

M. E. Taliansky Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, Scotland, UK

Contents

Introduction

Disease described first by Zimmermann (1907) and subsequently by Storey & Bottomley (1928) and Storey & Ryland (1955, 1957). Causal agent first characterized by Reddy et al. (1985a, 1985b). For reviews of groundnut rosette disease and its causal agents see Murant (1990b), Murant et al. (1993), and Robinson & Taliansky (1997).

Selected synonyms

Groundnut Kraüselkrankheit virus (Zimmermann, 1907; Rev appl. Mycol. 36: 303)

Peanut rosette virus (Rev appl. Mycol. 19: 229)

Groundnut rosette virus (GRV) is a replicating ssRNA which does not produce a coat protein and therefore has no conventional particles. It depends on groundnut rosette assistor virus (GRAV; Luteoviridae) for encapsidation in GRAV coat protein and for transmission by Aphis craccivora in the persistent (circulative, non-propagative) manner. GRV supports the replication of a satellite RNA which is responsible for the symptoms of rosette disease in groundnut and plays an essential role in mediating the dependence of GRV on GRAV for transmission by aphids. GRV and its satellite RNA are manually transmissible to a small range of hosts. The virus complex occurs only in sub-Saharan Africa.

Main Diseases

Rosette is the most destructive virus disease of groundnut (peanut; Arachis hypogaea) in Africa. In some years losses may be small, but in others rosette epidemics can be responsible for devastating losses. Thus in 1975 rosette affected about 0.7 million ha of groundnut in Nigeria and caused yield losses estimated at over 0.5 million tonnes, with a value estimated at US$ 250 million (Yaycock et al., 1976). In 1995, about 43,000 ha were affected in eastern Zambia, with losses amounting to US$ 5 million (Anon., 1996). The unpredictability of outbreaks is an important feature of the disease: the sudden unexpected loss of an important source of protein and cooking oil, of income, and of the seed for the next crop, may lead farmers to abandon growing groundnuts in subsequent years. Following the epidemic in Malawi in 1994/5, the area under groundnut decreased by 23%, from 89,000 to 69,700 ha (Anon., 1996).

Rosette disease is caused by a complex of groundnut rosette virus (GRV), together with an associated satellite RNA, and a helper virus, groundnut rosette assistor virus (GRAV; family Luteoviridae), on which the other two components depend for transmission by the aphid Aphis craccivora. Neither GRV nor GRAV themselves cause obvious symptoms in groundnut. Murant & Kumar (1990) showed that the disease symptoms are caused by the GRV satellite RNA, different variants of the satellite being responsible for the different major forms of rosette disease: chlorotic rosette (Storey & Bottomley, 1928) and green rosette (Hayes, 1932; Smartt, 1961; Hull & Adams, 1968). ‘Mosaic rosette’ (Storey & Ryland, 1957) is caused by mixed infection with the ‘chlorotic’ satellite variant and a ‘mottle’ variant (Murant & Kumar, 1990).

The satellite RNA plays an essential role, as yet unexplained, in mediating the dependence of GRV on GRAV for transmission by aphids (Murant, 1990a). This presumably explains why no satellite-free isolates of GRV have been found in nature, though they can be obtained experimentally (see below).

Plants with chlorotic rosette (Fig. 1, Fig. 2, Fig. 3) show a bright chlorosis of the leaves, usually with a few green islands. The chlorosis may affect the whole plant, or only some shoots or parts of shoots. Plants that are infected early are stunted, with small, curled and puckered leaflets. In the mosaic form of rosette (Fig. 4) the green parts of leaves are more extensive. In green rosette (Fig. 5), the leaves are very dark green, or show a light green and dark green mosaic, and are much reduced in size, with their margins rolled downwards. Plants that are infected early are severely stunted, with much shortened internodes, and appear as small dark green bushes. In all types of rosette, early infection causes severe or total loss of yield. Late infection can cause a decrease in the number and size of pods.

Geographical Distribution

Chlorotic rosette disease occurs throughout Africa south of the Sahara. Green rosette disease is found in West Africa and Uganda, and has recently been found in Angola (P. Subrahmanyam, personal communication), Malawi (Subrahmanyam & Mamba, 1993) and Swaziland (Subrahmanyam & Chiyembekeza, 1995). Mosaic rosette is reported only from East Africa. Groundnut rosette disease does not occur in other parts of the world; reports from Argentina, Australia, Fiji, India, Indonesia, Philippines and Russia have not been confirmed and are now discounted.

Host Range and Symptomatology

Experimentally, GRV, along with GRAV and any associated satellite RNA species is transmissible from groundnut to groundnut by grafting. GRV and its satellite RNA, but not GRAV, can be transmitted by manual inoculation to a limited range of dicotyledonous plants, though manual inoculation to groundnut (Arachis hypogaea) may be difficult, especially under greenhouse conditions in summer. A procedure for efficient manual inoculation of GRV to groundnut was reported by Olorunju et al. (1992, 1995). GRV has been transmitted to several other species of Leguminosae (Glycine max, Indigofera nummularifolia, Macrotyloma uniflorus, Phaseolus vulgaris, Stylosanthes gracilis*, S. guayensis, S. mucronata*, S. juncea, S. sundaica*, Tephrosia purpurea, Trifolium incarnatum*, Trifolium repens* and Vigna gracilis) and to a few species in the Amaranthaceae (Gomphrena globosa*), Chenopodiaceae (Chenopodium amaranticolor, C. murale, C. quinoa, Spinacia oleracea*) and Solanaceae (Nicotiana benthamiana, N. clevelandii, N. debneyi, N.occidentalis, N. rustica, N. tabacum Samsun NN) (Okusanya & Watson, 1966; Adams, 1967; Hull & Adams, 1968; Dubern, 1980; Reddy et al., 1985a; Rajeshwari & Murant, 1988; Kumar et al. 1991; D. B. Dangora, personal communication). The plants marked * are also hosts of GRAV.

Diagnostic species

C. amaranticolor. GRV cultures, with or without the satellite RNA, give minute necrotic lesions on inoculated leaves (Fig. 6) about 4 days after inoculation.

N. benthamiana. This is the most sensitive test plant. On inoculated leaves, most GRV cultures, with or without the satellite RNA, give symptoms ranging from almost none to necrotic spots, rings or target spots (Fig. 7). About 7-10 days after inoculation, veinal chlorosis or necrosis appears on the first systemically infected leaf, accompanied usually by slight down-rolling. Later this leaf often shows necrotic spots or line-patterns (Fig. 7). Leaves produced subsequently usually show no more than a mild chlorotic mottle or they may be symptomlessly infected. GRV cultures containing the yellow blotch satellite RNA (Kumar et al., 1991) induce symptoms similar to those of other isolates in inoculated leaves but systemically infected leaves show a brilliant yellow blotch mosaic affecting the entire plant and persisting for its lifetime (Fig. 8). With all isolates there is stunting of the plant.

N. clevelandii (Fig. 9). Isolates with or without satellite RNA induce local chlorotic or necrotic spots or rings followed by systemic chlorosis or necrosis beginning about 7-10 days after inoculation. Later there is systemic mottle or mosaic, accompanied by crinkling and distortion of leaves. These symptoms eventually become less severe but the plant remains stunted. Cultures containing the yellow blotch satellite do not induce distinctive symptoms in this species.

Assay species

C. amaranticolor is the best local lesion host. Arachis hypogaea (groundnut) is used as a test plant in aphid transmission experiments.

Propagation species

N. benthamiana is the best propagation host for studies on virus properties and purification. For long term maintenance, virus cultures are best kept in Arachis hypogaea (groundnut).

Strains

No strains of GRV have been distinguished, but numerous variants of the satellite RNA have been described that affect the symptoms induced by GRV in plants; for example, different satellite variants are responsible for the chlorotic and green forms of rosette disease (Murant & Kumar, 1990). The mosaic form of rosette is caused by GRV containing a mixture of the ‘chlorotic’ satellite variant and a ‘mottle’ variant (Murant & Kumar, 1990). A ‘yellow blotch’ satellite variant discovered in the laboratory (Kumar et al., 1991) is responsible for a striking bright yellow symptom in N. benthamiana which has been useful as an experimental marker. Several satellite variants have been discovered experimentally that induce few or no symptoms in groundnut. One of these down-regulates the replication of GRV (Taliansky & Robinson, 1997a) and, for this reason, isolates containing it are difficult to maintain in culture.

Transmission by Vectors

The aphid Aphis craccivora is the only vector of any importance, though there is a single report of transmission by A. gossypii (Adams, 1966). However, A. craccivora can transmit GRV and its satellite RNA only from source plants that are also infected with GRAV. This is because GRV RNA does not encode a coat protein, and relies on the GRAV coat protein for encapsidation (A.F. Murant, unpublished data) and therefore for aphid transmission. A further complication is that the satellite too must be present in the source plants for GRAV-dependent transmission of GRV to occur, i.e. aphids transmit only GRAV from plants containing GRAV plus a satellite-free culture of GRV (Murant, 1990a). The explanation for this is not known.

The dependence of GRV on GRAV is for acquisition and inoculation by the aphid, not for infection of the inoculated plant. Therefore, in circumstances in which the frequency of transmission of the two viruses is less than 100%, some inoculated plants may become infected with only one of the viruses. Plants that become infected with GRV alone cannot then serve as sources for aphid transmission.

The GRAV/GRV/satellite complex is transmitted by A. craccivora in the persistent (circulative, non-propagative) manner, the aphids retaining the ability to transmit for at least 15 days, possibly for life (Storey & Ryland, 1955; Watson & Okusanya, 1967; Dubern, 1980; Misari et al., 1988). Dubern (1980) found a minimum acquisition access time of 4.5 h, a latent period of 18 h, and a minimum inoculation access period of 3 min. Misari et al. (1988), working with separate cultures of chlorotic rosette and green rosette, reported minimum acquisition access periods of 4 h and 8 h respectively, and median latent periods of 26.4 h and 38.4 h respectively; the minimum inoculation access period was 10 min for both cultures.

In all studies of vector relations reported so far, transmission has been assessed by the appearance of rosette symptoms in the inoculated plants, so that the data are strictly applicable only to GRV plus its satellite. However, the minimum acquisition access times and latent periods for GRV are probably those of GRAV. The minimum inoculation access time for GRV may well be shorter than that for GRAV because GRV can infect mesophyll cells whereas GRAV needs to be inoculated into the phloem.

Watson & Okusanya (1967) reported that A. craccivora populations from Nigeria and Kenya differed in ability to transmit cultures of the rosette virus complex from East and West Africa, but A. F. Murant (unpublished data) found that a population from Malawi transmitted GRAV isolates from Malawi and Nigeria.

Transmission through Seed

There is no evidence for transmission through seed or pollen.

Serology

GRV does not produce a coat protein, but a polyclonal antiserum has been raised against the non-structural 28K ‘movement’ protein (the product of ORF4) by injecting a rabbit with a fusion protein expressed in E. coli. The antiserum reacts with movement protein transiently expressed in N. benthamiana from a potato virus X (PVX)-based vector, but has not been shown to react with extracts from GRV-infected plants (E.V. Ryabov, unpublished data).

Relationships

GRV is a member of the genus Umbravirus (Murant et al., 1995). This genus comprises several imperfectly characterized ssRNA viruses which lack coat protein genes and depend on unrelated helper viruses, usually members of the family Luteoviridae, for encapsidation and for transmission by aphids in a persistent (circulative, non-propagative) manner. The genus Umbravirus is not assigned to a family but its closest affinities seem to be with the family Tombusviridae.

Comparison of nucleotide sequences of genomic RNA molecules and of amino acid sequences of proteins putatively encoded by the same RNA molecules (Taliansky et al., 1996) revealed close homologies between GRV and two other viruses now classified in the genus Umbravirus, carrot mottle mimic virus (syn. Australian isolate of carrot mottle virus; Gibbs et al., 1996a, 1996b) and pea enation mosaic virus-2 (formerly the RNA- 2 of pea enation mosaic virus; Mayo & D'Arcy, 1998).

The sequences of GRV satellite RNAs have up to 59% homology with the satellite RNA associated with pea enation mosaic virus-2 (Demler et al., 1996).

Stability in Sap

(Reddy et al., 1985b). In 0.01 M Tris buffer, pH 8.0, containing 0.02 M sodium sulphite, much GRV infectivity survived after 1 day at room temperature or after 3 days at 4°C. A trace of infectivity survived even after 15 days at 4°C. Incorporation of 1 g/l Mg-bentonite did not affect these results. Buffer extracts were infective after dilution to 10^-3 but not 10^-4. Infectivity was abolished by treating the extracts with ether (50%), chloroform (10 or 50%), or butan-1-ol (8%).

Purification

Preparations of the infective ssRNA of GRV have not been obtained free from host RNA, and no infection-specific band can be detected by electrophoresis of whole RNA extracts from infected plants. However, unlike healthy plants, infected plants yield abundant double-stranded RNA (dsRNA) which gives a characteristic pattern of electrophoretic bands (Reddy et al., 1985b; Murant et al., 1988; Murant & Kumar, 1990), with three characteristic dsRNA species (Fig. 10). Two of these species, dsRNA-1 (4.0 kbp) and dsRNA-2 (1.3 kbp), appear to be double-stranded forms of the genomic and a sub-genomic RNA respectively. The third and much the most abundant species, dsRNA-3 (0.9 kbp), is a double-stranded form of the satellite RNA, from which GRV cultures can be freed experimentally (Fig. 10) (see below). The characteristic dsRNA band pattern is a useful diagnostic aid and Breyel et al. (1988) described a quick procedure for analysing multiple samples electrophoretically. However, somewhat similar dsRNA band patterns are given by other putative umbraviruses (A.F. Murant, unpublished data), so that such analyses should be interpreted with caution.

Preparation of infective ssRNA
The following procedure was described by Murant et al. (1988):
1. Grind fresh leaf tissue (10 g) to a fine powder in liquid nitrogen with a pestle and mortar. Stir the powder for 30 min at room temperature with 20 ml TSE buffer (50 mM Tris-HCl, 0.1 M NaCl, 1 mM disodium EDTA, pH 7.0), 3.0 ml 10% SDS, 40 mg bentonite and 30 ml ‘phenol reagent’ (9 vol. water-saturated phenol + 1 vol. m-cresol, containing 0.1% 8-hydroxy-quinoline).
2. Precipitate total nucleic acids from the aqueous phase with 70% ethanol and place at -20°C for at least 16 h, then wash three times with cold 70% ethanol. For infectivity assay, resuspend the pellets in 0.1 M phosphate buffer pH 7.4 containing 1 mg/ml bentonite, and inoculate immediately to test plants.

Preparation of dsRNA
The following procedure was described by Murant et al. (1988):
1. Make preparations of total nucleic acid as described above but after the phenol extraction step adjust the aqueous phase to 20% (v/v) ethanol. Apply to columns of Whatman CF-11 cellulose as described by Dodds & Bar-Joseph (1983), and elute the dsRNA fraction with ethanol-free TSE buffer. Recover nucleic acid from the eluate by ethanol precipitation.
2. To remove any contaminating DNA, resuspend the sample in 50 mM Tris-HCl, 50 mM NaCl, 10 mM MgCl₂, pH 7.5, and treat with 10 µg/ml DNase I for 30 min at 30°C. Then, to remove contaminating ssRNA, adjust to 0.3 M NaCl and treat with 10 ng/ml RNase A for 1 h at 30°C.
3. Adjust the preparation to 2% SDS and re-extract with ‘phenol reagent’.
4. A procedure for electrophoretic analysis of dsRNA preparations is given by Murant et al. (1988).

Preparation of satellite-free cultures of GRV
The method used by Murant & Kumar (1990) was as follows:
1. Electrophorese dsRNA preparations in 0.9% low-gelling-temperature agarose gels and excise the dsRNA-1 bands from several tracks.
2. Boil the gel slices for 5 min in 0.5 ml 10 mM Tris-HCl, 1 mM EDTA, pH 7.4, then cool rapidly on ice, add 4 mg/ml bentonite, and inoculate the preparation both undiluted and at a dilution of 1/10 to Nicotiana benthamiana. Isolates obtained from the highest infective dilution usually lack dsRNA-3 but possess dsRNA-1 and dsRNA-2.
3. Biologically active preparations of dsRNA-3 are prepared and inoculated in the same way, but of course the satellite does not replicate in the plants unless they are also inoculated with melted dsRNA-1. Unlike dsRNA-1, dsRNA-3 does not need to be melted to yield biological activity (Kumar et al., 1991).

Particle Structure

No virus-like particles have been seen in preparations from plants infected only with GRV, and no candidate for the role of a particle protein has been identified among the proteins putatively encoded by the genomic RNA. The virus seems to exist in plants as an infective ssRNA. However, its relative stability in crude sap suggests that it is protected in some way against degradation by ribonucleases.

Genome Properties

The GRV genome is a single segment of ssRNA. The complete nucleotide sequence has been determined (Taliansky et al., 1996); its mol. wt is 1.3 × 10⁶ (4019 nucleotides). The sequences of ten variants of the satellite RNA were determined by Blok et al. (1994); their nucleotide numbers ranged from 895 to 903. Radioactive and non-radioactive cDNA probes complementary to the satellite RNA are useful for detection of GRV infection (Blok et al., 1995). Probes complementary to the GRV genomic RNA are also available but seem likely to be less useful because of the lower molar concentration of genomic RNA than of the satellite RNA in infected plants.

Neither the genomic RNA nor the satellite RNA are polyadenylated, nor do either of them possess a 3'-terminal tRNA-like sequence. It is not known whether they have 5'-terminal genome-linked proteins (VPg) or m⁷G RNA cap structures. However, the biological activity of RNA transcripts from full-length cDNA clones of the satellite RNA is unaffected by addition of a cap (Taliansky & Robinson, 1997a).

The GRV genomic RNA sequence (Taliansky et al., 1996) contains four large open reading frames (ORFs) (Fig. 11). ORF2 includes sequences that encode motifs characteristic of viral RNA-dependent RNA polymerases and is probably expressed by a -1 frameshift mechanism as a fusion with the ORF1 product. ORFs 3 and 4 almost completely overlap in different reading frames and are probably expressed from subgenomic RNA (Taliansky et al., 1996). ORF4 codes for a 28 K protein which is involved in virus movement (Ryabov et al., 1998); its amino acid sequence has significant similarity with those of several other viral movement proteins, especially with the 3a protein of cucumber mosaic virus (CMV). The function of the ORF3 product is not known; it has significant sequence similarity with the corresponding products of the umbraviruses carrot mottle mimic virus (Gibbs et al., 1996a, 1996b) and pea enation mosaic virus-2 (Taliansky et al., 1996), but not with any other viral or non-viral proteins in existing databases. None of the putative protein products of GRV RNA seems to be a particle protein.

The ten GRV satellite RNA sequences determined by Blok et al. (1994) contained from two to five short ORFs, but mutational analysis (Taliansky & Robinson,1997b) has shown that no satellite-encoded ORFs are required, either for satellite replication or for symptom induction in N. benthamiana. However, three functional untranslated RNA elements have been identified. One (designated R) is essential for RNA replication and appears to act in cis. Symptom induction in N. benthamiana and groundnut involves two further untranslated RNA elements acting in trans: (A) located in the left half, and (B) located in the right half of the satellite RNA molecule. Element A contains the determinant that is unique to the yellow blotch satellite.

Relations with Cells and Tissues

The localization of the proteins encoded by GRV ORFs 3 and 4 was studied in Nicotiana spp. by expressing the proteins as fusions with the jellyfish green fluorescent protein (GFP) from modified PVX and tobacco mosaic virus (TMV) vectors (Ryabov et al., 1998). The GFP-GRV ORF3 fusion protein was found in granules associated with membranes of the endoplasmic reticulum, and accumulated mainly in large cytoplasmic inclusion bodies lying close to the nuclei. This protein was also detected in nucleoli. The GRV ORF4-GFP fusion product was associated with the plasmodesmata of epidermal and mesophyll cells, as is found with several other viral movement proteins, including the 3a movement protein of CMV, which has sequence similarity with the GRV ORF4 protein. In addition, the GRV ORF4-GFP fusion product was detected in the cytoplasmic inclusion bodies of mesophyll cells.

Notes

Groundnuts, which came originally from South America, are cultivated worldwide throughout the tropics and sub-tropics, but only in Africa are they affected by rosette disease. This suggests that the causal agents of rosette are endemic to Africa and are pathogens of some wild African plant species. However, groundnut is the only plant in which GRV, GRAV or the GRV satellite have been found occurring naturally. Experimentally, GRV and GRAV have each been transmitted to a few other plant species but the only known hosts of both viruses, apart from groundnut, are Stylosanthes gracilis, S. mucronata, S. sundaica, Trifolium incarnatum, T. repens, Gomphrena globosa and Spinacia oleracea, and none of these is likely to act as a dry season host of Aphis craccivora. However, since neither of the viruses is seed-borne it seems likely that at least one species that is a host of the virus complex and of the aphid vector must exist. There is disagreement (Evans, 1954; Booker, 1963; Hildebrand et al., 1991) about the extent to which persistence of the viruses through the dry season may now depend on survival of infected groundnut plants (‘groundkeepers’) in favoured areas, such as around watercourses. In regions where this does not happen, initial infection may depend on the influx of viruliferous aphids from other parts of Africa on prevailing winds (Bunting, 1950; Adams, 1967; Rossel, 1977).

Ways of protecting groundnut crops against rosette disease include the removal of groundkeepers, early sowing at high plant density (which reduces disease incidence because the landing response of the aphid vector is inhibited as the ground becomes covered by plant growth; Booker, 1963; A'Brook, 1964; Farrell, 1976) and, if affordable, the use of insecticides (Davies, 1972). However, the best approach lies in the development of resistant cultivars. Resistance is currently available in late-maturing cultivars derived from groundnut material found in the border region between Côte d'Ivoire and Burkina Faso (Sauger & Catharinet, 1954; De Berchoux, 1958). The resistance, which does not amount to absolute immunity, is controlled by two independent recessive genes and is effective against both chlorotic and green forms of rosette (Nigam & Bock, 1990; Olorunju et al., 1992). The resistance is directed against GRV (and therefore the GRV satellite RNA) but is not effective against GRAV (Bock et al., 1990). Recently, progress has been made in transferring this form of resistance into early-maturing cultivars, which are the type needed in most parts of Africa. Another form of resistance may exist in the wild diploid species Arachis chacoense, which seems immune to both GRV and GRAV (Murant et al., 1991): a high degree of resistance to rosette was found in a hybrid derivative from an interspecific cross of A. hypogaea x A. chacoense (Moss et al., 1993).

A possibility for the future is the deployment of transgenic forms of resistance. Because of the relative difficulty in transforming groundnut, the use of constructs directed against GRV has been evaluated in Nicotiana benthamiana (Taliansky et al., 1998). Resistance to GRV infection was detected in plants transformed with constructs derived from a mild variant of the satellite RNA. However, this strategy has not yet been tested for protection against rosette disease in groundnut.

Resistance in groundnut to Aphis craccivora was reported by Padgham et al. (1990) and this could prove a useful additional approach, especially because the aphid is an important pest in its own right.

Acknowledgements

All photographs copyright Scottish Crop Research Institute.

Figures

References list for DPV: Groundnut rosette virus (355)

1. A’Brook, Ann. appl. Biol. 54: 199, 1964.
2. Adams, Rhodesia, Zambia Malawi J. agric. Res. 4: 125, 1966.
3. Adams, Rhodesia, Zambia Malawi J. agric. Res. 5: 145, 1967.
4. Anon., Ann. Progr. Rep., SADC/ICRISAT Groundnut Project, 1996, Chitedze Research Station, Malawi, 1996.
5. Blok, Ziegler, Robinson & Murant, Virology 202: 25, 1994.
6. Blok, Ziegler, Scott, Dangora, Robinson & Murant, Ann. appl. Biol. 127: 321, 1995.
7. Bock, Murant, & Rajeshwari, Ann. appl. Biol. 117: 379, 1990.
8. Booker, Ann. appl. Biol. 52: 125, 1963.
9. Breyel, Casper, Ansa, Kuhn, Misari & Demski, J. Phytopath. 121: 118, 1988.
10. Bunting, Ann. appl. Biol. 37: 699, 1950.
11. Davies, Bull. ent. Res. 62: 169, 1972.
12. De Berchoux, Oléagineux 13: 237, 1958.
13. Demler, Rucker, de Zoeten, Ziegler, Robinson & Murant, J. gen. Virol. 77: 2847, 1996.
14. Dodds & Bar-Joseph, Phytopathology 73: 419, 1983.
15. Dubern, Phytopath. Z. 99: 318, 1980.
16. Evans, Ann. appl. Biol. 41: 189, 1954.
17. Farrell, Bull. ent. Res. 66: 317, 1976.
18. Gibbs, Cooper & Waterhouse, Virology 224: 310, 1996a.
19. Gibbs, Ziegler, Robinson, Waterhouse & Cooper, Molec. Pl. Path. On-Line 1996/1111 gibbs, 1996b.
20. Hayes, Trop. Agric., Trinidad 9: 211, 1932.
21. Hildebrand, Bock & Nigam, p.8, in: Groundnut Virus Diseases in Africa, Proc. 4th Meeting Consult. Gp Collab. Res. Groundnut Rosette Virus Disease, Montpellier, France 1990, ICRISAT, 1991.
22. Hull & Adams, Ann. appl. Biol. 62: 139, 1968.
23. Kumar, Murant & Robinson, Ann. appl. Biol. 118: 555, 1991.
24. Mayo & D’Arcy, in: The Luteoviridae, Smith & Barker, eds, Wallingford: CAB International, in press, 1998.
25. Misari, Abraham, Demski, Ansa, Kuhn, Casper & Breyel, Pl. Dis. 72: 250, 1988.
26. Moss, Singh, Subrahmanyam, Hildebrand & Murant, Int. Arachis Newsl. No. 13, 22, 1993.
27. Murant, J. gen. Virol. 71: 2163, 1990a.
28. Murant, Rep. Scott. Crop Res. Inst., 1989, 138, 1990b.
29. Murant & Kumar, Ann. appl. Biol. 117: 85, 1990.
30. Murant, Rajeshwari, Robinson & Raschke, J. gen. Virol. 69: 1479, 1988.
31. Murant, Kumar & Robinson, p.7, in: Groundnut Virus Diseases in Africa, Proc. 4th Meeting Consult. Gp Collab. Res. Groundnut Rosette Virus Disease, Montpellier, France 1990, ICRISAT, 1991.
32. Murant, Robinson, Blok, Scott, Torrance & Farmer, Ann. Rep. Scott. Crop Res. Inst., 1992, 85, 1993.
33. Murant, Robinson & Gibbs, pp. 388-391, in: Virus Taxonomy - Classification and Nomenclature of Viruses. 6th Rep. Int. Comm. Taxonomy of Viruses, Murphy et al., eds, Vienna: Springer-Verlag, 1995.
34. Nigam & Bock, Ann. appl. Biol. 117: 553, 1990.
35. Okusanya & Watson, Ann. appl. Biol. 58: 377, 1966.
36. Olorunju, Kuhn, Demski, Misari & Ansa, Pl. Dis. 76: 95, 1992.
37. Padgham, Kimmins & Ranga Rao, Ann. appl. Biol. 117: 285, 1990.
38. Rajeshwari & Murant, Ann. appl. Biol. 112: 403, 1988.
39. Reddy, Murant, Duncan, Ansa, Demski & Kuhn, Ann. appl. Biol. 107: 57, 1985a.
40. Reddy, Murant, Raschke, Mayo & Ansa, Ann. appl. Biol. 107: 65, 1985b.
41. Robinson & Taliansky, Ann. Rep. Scott. Crop Res. Inst., 1996/7, 170, 1997.
42. Rossel, Trop. Grain Legume Bull. 8: 41, 1977.
43. Ryabov, Oparka, Santa Cruz, Robinson & Taliansky, Virology 242: 303, 1998.
44. Sauger & Catharinet, L'agronomie tropicale 9: 28, 1954.
45. Smartt, Emp. J. exp. Agric. 29: 79, 1961.
46. Storey & Bottomley, Ann. appl. Biol. 15: 26, 1928.
47. Storey & Ryland, Ann. appl. Biol. 43: 423, 1955.
48. Storey & Ryland, Ann. appl. Biol. 45: 318, 1957.
49. Subrahmanyam & Chiyembekeza, Int. Arachis Newsl. No.15, 22, 1995.
50. Subrahmanyam & Mamba, Int. Arachis Newsl. No.13, 7, 1993.
51. Taliansky & Robinson, Virology 230: 228, 1997a.
52. Taliansky & Robinson, J. gen. Virol. 78: 1277, 1997b.
53. Taliansky, Robinson & Murant, J. gen. Virol. 77, 2335, 1996.
54. Taliansky, Ryabov & Robinson, Mol. Pl. Microbe Interact. 11: 367, 1998.
55. Watson & Okusanya, Ann. appl. Biol. 60: 199, 1967.
56. Yaycock, Rossel & Harkness, Samaru Conference Paper No. 9, Institute of Agricultural Research, Samaru, Nigeria, 1976.
57. Zimmermann, Der Pflanzer 3: 129, 1907.
58. Olorunju, Kuhn, Ansa, Misari & Demski, Peanut Sci. 22: 56, 1995.