Coconut lethal yellowing disease in Ghana and planting material tolerance

D. Mariau1, S.K. Dery2, A. Sangaré3, Y.P. N'Cho3, R. Philippe1

1CIRAD-CP, BP 5035, 34032 Montpellier Cedex 1, France; 2Coconut disease project, PO Box 245, Sekondi, Ghana; 3IDEFOR-DPO, station de recherche Marc Delorme, 07 BP 13, Abidjan 07, Côte d'Ivoire


Lethal yellowing is the most damaging coconut disease, although the main producing countries (Indian subcontinent, Indonesia, the Philippines and the pacific islands) have been spared so far. However, many tens of thousands of hectares have been devastated in the Americas (Florida, Mexico and many Caribbean countries) and in both West Africa (Togo, Nigeria, Ghana) and East Africa (Tanzania, Kenya).

Etiological agents

The disease is linked to the existence of phytoplasma-like organisms, first detected by various authors in Jamaica in 1972 (Beakbane et al., 1972; Heinze et al., 1972; Plavsic-Banjac et al., 1972). The existence of these phytoplasmas has since been consistently confirmed in all the countries affected by this type of disease, particularly in Africa (Dabek et al., 1976; Dollet and Giannotti, 1976).


The symptoms are very similar from one country to another. The first sign of the disease is premature nut fall, starting with the oldest nuts (Fig. 1). At the same time, the spikelets of the inflorescences still inside the spathes begin to blacken. Yellowing of the leaves rapidly spreads upwards (Fig. 2), leading to the gradual decay and then the death of the coconut palm, generally within a year of the first symptoms (Figs. 3 and 4).

Figure 1 Figure 3
Figure 2 Figure 4

Host plants

Many species of Palmae other than Cocos nucifera are susceptible to the disease. Phytoplasma-like organisms have been observed in at least 30 different species (Thomas, 1979), including the date palm (Phoenix dactylifera) and ornamental palms (Adonidia merrillii, Pritchardia thurstonii). This wide range of hosts could make it difficult to identify disease-tolerant varieties within the Cocos nucifera species.


Even before the etiological agent of the disease was known (and although it was initially thought to be a virus disease), studies were carried out to identify the vector. The role of insects was first proven in Jamaica and Florida, by protecting young coconut palms from insect attacks in zones severely affected by the disease by placing them in cages (Heinze et al., 1972). A few years later, the role of Myndus crudus (Cixiidae) was strongly suspected (Tsai, 1977), before being clearly demonstrated two years after that (Tsai, 1979). In Africa, and more particularly in Ghana, trials are under way and another Myndus species (M. adiopodoumeensis) is strongly suspected (Dery et al., 1995).

Control methods

Shortly after the identification of the etiological agent, it was demonstrated that the disease could be prevented or even cured by injecting a tetracycline-type antibiotic, either preventively or curatively, once the first symptoms had appeared (MacCoy, 1972). This technique has been used as a control method to save coconut palms for tourists, notably in Miami, but such techniques have obviously never been considered for stemming the disease in plantations.

Primarily due to the wide range of M. crudus larva hosts (Tsai and Kirsch, 1978), controlling the vector, as was done with M. taffini, the vector of coconut foliar decay in Vanuatu (Julia, 1982), will no doubt be difficult. In fact, M. taffini larvae develop solely on Hibiscus tiliocens roots; destroying this plant leads to the disappearance of the insect and hence of the disease.

By eliminating the first diseased palms in a plantation, the spread of the disease can, at the very best, be slowed, probably since this eliminates all the possible sources of contamination (Dery and Philippe, 1995).

The only possible control method is, therefore, to seek out varieties or hybrids that are either resistant to or at least as tolerant as possible of the disease. Given that it has proven extremely difficult to reproduce the disease in Florida using the vector insect, and that in many areas the vector is as yet unknown, the only way of testing planting material is to set up performance trials in zones affected by the disease or in neighbouring areas, with all the risks this involves.

Lethal Yellowing disease in Ghana

The disease was first seen in the Far East of the country in 1932, at Cape Saint Paul, close to the Kaincope region in Togo, where the disease was also seen at the time. Over 100,000 coconut palms were killed at the site. The disease reappeared some thirty years later, but in the west of the country, at Cape Three Points. From its initial focus, the disease spread towards Takoradi and westward towards the Ivorian border. Almost all the coconut plantings on the coast, between the Ankobra River and Takoradi, were devastated in twenty years (Fig. 5). A few isolated islands have been spared so far, protected by a natural barrier such as a lagoon. To the north of this coastal strip, there are many generally small foci, particularly north of Takoradi, where the coconut plantings look likely to die out. To the west, the natural barrier formed by the Ankobra River was probably crossed in 1994, with two foci around a dozen kilometres apart. There are continuous coconut plantations right up to the border with the Ivory Coast some 70 km away.

Figure 5

In 1986, the disease jumped some 60-km east, to Cape Coast region, or more precisely near the village of Ayensudu, where a very large focus developed within a few years. Shortly afterwards, a large number of foci of varying sizes appeared throughout the central region, where coconut plantations are now under serious threat (Fig. 6). The disease has probably killed around a million palms in the central and western regions of southern Ghana.

Figure 6

Research on varietal susceptibility

Studies in Jamaica revealed that although the local Tall was particularly susceptible to the disease, the Malayan Yellow Dwarf (MYD) was much more tolerant. Performance trials confirmed this initial observation (Been, 1981). While most of the Dwarfs tested proved highly tolerant, none of the Talls showed any real signs of tolerance, although there were substantial differences in susceptibility from one variety to another. Dwarf x Tall hybrids (MYD x Cambodia Tall or MYD x Panama Tall or Maypan) also seemed to show promising signs of tolerance.

Similar trials have been carried out in Tanzania (Schilling and Mpunami, 1990). Here too, the Dwarfs generally performed well compared to the Talls, but the Cameroon Red Dwarf and the Equatorial Guinea Green Dwarf seemed to perform even better than the MYD. However, susceptibility varied substantially from one site to another (Table 1). Among the Talls, certain local Tall origins, such as the variety from the Tanga region, also seemed to be tolerant. The West African Tall and its hybrid with the MYD were particularly susceptible.

Table 1

In Ghana

From 1981 to 1983, seven performance trials were set up in western Ghana, either in the disease zone or nearby (Sangaré et al., 1992). Three trials have not yet been affected by the disease, despite being planted just a few kilometres from foci: they are probably protected by a small strip of forest. Moreover, there are no cases of the disease in another trial, planted over ten years ago in a zone previously affected by the disease. In another trial, also planted in an affected zone, three cases of lethal yellowing were seen when the planting was seven to nine years old, but only one new case has been seen in the past five years.

Only two trials were therefore suitable for studying planting material susceptibility to the disease. The performance of ten varieties (five Talls and five Dwarfs) and seventeen hybrids has been monitored.

The results obtained have sometimes been difficult to interpret, since the coconut palms in the trials suffered substantial mortality right from an early age, due to a severe drought one to two years after planting and severe Oryctes attacks linked to the existence of large numbers of palms killed by the disease in surrounding plantings, which act as larval refuges.

The calculations do not take account of replacement palms, which in many cases are still small and therefore not as vulnerable to the disease. The first cases of the disease were not seen until seven years after planting at Dixcove and ten years after planting at Cape Three Points. For the reasons mentioned above, the number of palms observed was very small in many cases, and the conclusions reached are therefore sometimes provisional.

Overall, the Dwarfs are more tolerant than the Talls, and the SGD (Sri Lanka Green Dwarf) performs particularly well: no deaths have yet been seen out of nineteen individuals (Table 2). However, the CRD (Cameroon Red Dwarf) is highly susceptible, whilst the MYD performs quite well, although the number of individuals under observation is small, due to this Dwarfs high susceptibility to drought and Oryctes attacks. However, when the west of the country was contaminated, it became clear that the MYD eventually succumbs in the event of very high disease pressure.

Table 2

All the Talls tested seem to be susceptible, particularly the WAT, which has been devastated by the disease. The VTT (Vanuatu Tall) seems to be an exception, since no cases of mortality have been seen, but the number of individuals observed is small.

All the hybrids tested seem to be susceptible to some degree, but the number of individuals being observed is often too small to draw any conclusions. For example, the MYD x VTT should perform well, given its reasonably or highly tolerant parents. The apparent good performance of the MYD x MLT (Malayan Tall) is more difficult to explain, since the MLT is highly susceptible. The SGD x VTT, which should perform well, was unfortunately not one of the hybrids tested in this first series of trials.

Comparisons between Jamaica, Tanzania and Ghana

Not all the varieties, particularly the hybrids, could be planted at every site, and effective comparisons are therefore difficult (Table 1), especially since the performance of a given variety can change within a country (Tanzania), where substantial variations have been seen.

The Talls are susceptible everywhere, except perhaps the VTT, which seems to be very tolerant in Ghana but is as susceptible in Jamaica as the WAT is in Ghana. The Dwarfs are more tolerant everywhere. It is in Jamaica that the MYD seems to be the least vulnerable. The SGD seems to be very tolerant in Ghana, whereas it is much more susceptible in Tanzania, where all the varieties seem to be more susceptible to the disease.

Any selection made in a given country or region is therefore unlikely to be valid for another region, as if the aggressiveness of the etiological agent varied from one site to another. It has in fact been shown that there are differences between the various African phytoplasma strains and the Caribbean (Florida) strain in terms of their nucleic acid (Tymon et al., 1995).

Conclusions and prospects

Lethal yellowing, the most devastating coconut disease spreads in a totally unpredictable way. For example, the disease can take several decades to spread a few kilometres in one direction in Ghana, and then jump several hundred kilometres. In Cameroon, the disease developed rapidly in the Kribi region, then for some unknown reason quite abruptly stopped spreading and has never reached a neighbouring performance trial. This was also the case in Ghana, where there has been very little, if any, sign of the disease in trials that were nevertheless set up in regions previously devastated by the disease.

The only way of controlling the disease is to breed tolerant varieties or hybrids. Very few varieties have shown real tolerance, and the level varies depending on disease "pressure". In Ghana, the Malayan Yellow Dwarf performs well, but in the Princess Town region, where the disease has now disappeared after wiping out all the coconut plantations in the are a, this Dwarf was unable to resist destruction, except for a single individual.

In Ghana, the tolerance of the Vanuatu Tall, the hybrid between this Tall and the Malayan Yellow Dwarf, and even better still, between the Sri Lanka Green Dwarf, needs to be tested on a large scale. However, a highly susceptible control (WAT) should also be set up, given the often-surprising epidemiology of the disease. Other varieties will also have to be tested particularly those from the Philippines Indonesia and India.

A hybrid tolerant of the disease in one region will not necessarily be tolerant in another, which has been confirmed by research on the pathogen genome.


BEAKBANE A.B., SLATER C.H.W., POSNETTE A.F., 1972. Mycoplasmas in the phloem of coconut, Cocos nucifera L., with lethal yellowing disease. J. Hortic. Sci. 47:265.

BEEN B.O., 1981. Observations on field resistance to lethal yellowing in coconut varieties and hybrids in Jamaica. Oleagineux 36 (1): 9-12.

DABEK A.J., JOHNSON C.G., HARRIES H.C., 1976. Mycoplasma-like organisms associated with Kaincope and Cape St-Paul wilt diseases of coconut palms in West Africa. PANS 22 (3): 354-358.

DERY S.K., PH]LIPPE R., 1995. Results of preliminary study of the epidemiology of the Cape St Paul Wilt disease of coconut in Ghana. International Workshop on lethal yellowing-like diseases on coconut, Elmina, Ghana, Nov. 1995, 14 pp.

DERY S.K., PHILIPPE R., MARIAU D., 1995. Results of preliminary observations on Auchenorrhynca (Homoptera) suspected vectors of lethal yellowing disease in Ghana. International Workshop on lethal yellowing-like diseases on coconut, Elmina, Ghana, Nov. 1995, 19 pp.

DOLLET M., G]ANNOTTI J., 1976. Maladie de Kaincopé : présence de mycoplasmes dans le phloème des cocotiers malades. Oléagineux 31 (4) :169-171.

HEINZE K., PETZOLD H., MARW]TZ R., 1972. Beitrug zur Atiologie der Todlichen Vergiltung der Kokospalme. Phytopathol. Z. 74:230 237.

HEINZE K.G., SCHUILING M., ROMNEY D.H., 1972. The possible cause of lethal yellowing disease of coconut. Plant Prot. Bull. 20: 58-68.

JULIA J.F., 1982. Myndus taffini (Homoptera Cixiidae), vecteur du dépérissement foliaire des cocotiers au Vanuatu. Oléagineux 37 (8-9): 409414.

McCoY R.E., 1972. Remission of lethal yellowing in coconut palm treated with tetracycline antibiotics. Plant Dis. Rep. 56:1019-1021.

PLAVSIC-BANJAC G., HUNT P., MABAMOROSCH K., 1972. Mycoplasma-like bodies associated with lethal yellowing disease of coconut palms Phytopathology 62: 298-299.

SANGARÉ A., TAFFIN G.DE, FRANQUEVILLE H.DE, ARKHUST E.D., POMIER M., 1992. Le jaunissement mortel du cocotier au Ghana. Premiers résultats sur le comportement au champ du matériel végétal. Oléagineux 47 (12): 699-704.

SCHUILING M., MPUNAMI A., 1990. Lethal disease of coconut palm in Tanzania: review of research up to date and preliminary result of resistance trials. In: La problematica del amarillamiento letal del cocotero en Mexico, Merida, Mexique, 14-16 janv. 1989. Yucatan, Mexique, Centro de Investigacion Cientifica de Yucatan, p. 171-183.

THOMAS D.L., 1979. Mycoplasma-like bodies associated with lethal declines of palms in Florida. Phytopathology 69: 928-934.

TSAI J.H., 1977. Attempts to transmit lethal yellowing of coconut palm by the planthopper Haplaxius crudus. Plant Dis. Rep. 61: 304-307.

TSAI J.H., 19,9. Vector transmission of mycoplasmal agents of plant diseases. In: The mycoplasmas, Vol. 111, R.F. Whitcomb et J.G. Tully éd., New York, Etats-Unis, p. 266-307.

TSA; J.H., K[RSCH O.H, 1978. Bionomics of Haplaxius crudus (Homoptera: Cixiidae). Environ. Entomol. 7: 305-308.

TYMON A., JONES P., EDEN GREEN S., 1995. Detection and discrimination of mycoplasma-like organisms (MLO) associated with coconut lethal yellowing and other diseases. July 5 1992 -February 4 1993. Final technical report. Project No. A0282. Kent, Grande-Bretagne, Natural Resources Institute, 21 p. (document interne).

PACSOF Home Page

VPE Home Page

VPE Table of Contents

VPE Photo gallery

Virtual Palm Encyclopedia Site Map
Powered by FreeFind.

This site is copyrighted © 1998, 1999, 2000, Palm & Cycad Societies of Florida (PACSOF)
For questions or comments, e-mail the webmaster.
Internet hosting provided by Zone 10, Inc.