Coconut genetic improvement programme in Benin

Introduction

Coconut plantations cover an area of around 15 000 ha in Benin and are mostly located within 25 km of the sea coast. Most of the plantations are made up of the West African Tall (WAT) variety with palms 50 to 70 years old. However, there has been substantial extension over recent years with the planting of the PB 121 hybrid. Expansion rate with this hybrid can be estimated at 100 ha per year, along with a fair amount of PB 111 and WAT. Dwarf varieties are found near dwellings or in public places for decoration.

National production amounts to more than 50 million nuts and are used for fresh or green nut consumption, small scale oil production and informally exported to neighbouring countries.

Palm productivity varied from one region to another depending on the importance and care accorded to coconut. Nevertheless, on a national level, it can be estimated at 40-80 nuts/palm/year for WAT and 90-125 nuts/palm/year for PB 121. These hybrids also produced more than 150 nuts/palm/year in irrigated plantations.

Other farmers even estimated PB 121 productivity at more than 180 nuts per year where livestock are reared under coconut and only potassium chloride fertilizer was applied. However, they complained about the small size of the nuts.

A recent study showed that planting densities ranged from 50 to 120 palms/ha in the old WAT plantations, whereas the new hybrid plantations were planted at normal densities of 160 palms/ha.

The coconut sector provides lucrative activities for both men and women in the coastal zone where coconut is grown. The men produce and sell the nuts to enterprising women for an average of 15 francs (US$ 0.03) per nut. The women process the nuts (oil, sweets, appetizers) and market them. While operating costs in the old plantations are limited to occasional weeding and crop picking, the recent PB 121 plantations are more organized, fertilized and sometimes irrigated. These well-organized plantations can bring in net income of 150 000 – 180 000 francs (US$ 300 – US$ 360) per hectare per year.

National coconut germplasm

The very poor living collection at the Sémé-Podji Coconut Research Station contains just four ecotypes, grouped with their corresponding number of palms, as follows: Malayan Yellow Dwarf (1 939); Cameroon Red Dwarf (344); West African Tall (168), and Equatorial Guinea Green Dwarf (29).

All these ecotypes are kept in an isolated seedgarden currently used for pollen and seed production. The small number of WAT palms is due to the fact that the seedgarden was initially planted with double rows of dwarfs alternating with a row of WAT for assisted open-pollination. However, subsequent analysis of the yield data led to the removal of a large number of these palms because of their poor performance.

Attempts to renew this old seedgarden (planted in 1963 and 1964) were launched in 1989, but were unsuccessful.

Comparative performance of hybrid coconuts and other cultivars

The different hybrids and other varieties grown showed varying yield performances (Table 1). The hybrids were found to be more productive than the local variety. However, yields remained generally low, primarily limited by environmental and cultural factors, namely:

a) Rainfall. Coconut palms need at least 1 500 mm of well-distributed rainfall per year, but the coconut zone in Benin has annual means ranging from only 900 mm in the West to 1 200 mm in the East. In addition to being well-below the requirement, the rainfall distribution is very poor.

b) Poor plantation upkeep. Few growers, apart from better educated farmers, regularly fertilize their plantations and practice ‘crop picking’. This leads some farmers to complain about the short economic life span of hybrids compared to the WAT.

c) Soil characteristics. The poor mineral content of the quaternary sand and strong variations in water table also limit the productivity of palms close to the sea.

d) Animal integration. Livestock rearing under coconut, as an added source of income, comes up against lack of pasture and credit.

National replanting programme

Apart from the noteworthy extension of local coconut plantations, without any official directives, by farmers wishing to prepare for their old age, a national study is currently underway to develop a viable strategy for renewing the old WAT coconut plantations (50-70 years old).

The study has initially revealed that coconut plays an important role with regard to:

· ecology – it is the only tree that can grow and constitute a coastal vegetation;

· socioeconomic – its cultivation is the most lucrative agricultural activity in the coastal zone and employs several social categories;

· food – consumption of fresh green nuts and of coconut oil remains high in Benin; and

· tourism – the use of dwarf varieties and sometimes hybrids for decorative purposes is very attractive in tourist areas.

Productivity problems

As mentioned earlier, productivity is primarily limited by environmental and management factors. Market outlets for yields are not currently an obstacle to coconut cultivation, though exports to neighbouring countries are only on an informal basis. Thus, organization of an extension service for coconut cultivation would enable farmers to benefit more from the fruits of their labour.

Breeding strategies

Genetic improvement is based on exploiting the heterosis between genetically distant ecotypes and the search for the best combining abilities. This principle is used in dwarf × tall crosses, which are compared for yield precocity, slow vertical growth, and copra and oil yields in the juvenile and adult stages.

However, this hybridization strategy is handicapped by the small number of ecotypes in the collection and the limited resources available to the breeding station.

Coconut breeding programme for the next ten years

In the coming years, collection for breeding purposes will be enlarged. In 1989, promising varieties from the Ivory Coast were identified for collection and introduction, namely: Polynesia Tall (PYT), Rennell Tall (RIT), Vanuatu Tall (VTT), and Malayan Red Dwarf (MRD).

However, the initiative has yet to get off the ground. The assistance of AFOPDA is requested to hasten the acquisition of necessary ecotypes for the breeding work and for comparative performance trials with the available hybrids.

Suggestion for international collaboration on coconut genetic improvement

As coconut is a plant whose yields vary depending on ecological conditions and on the ecotypes involved, it is suggested that international collaboration on genetic improvement address the following aspects:

· ease of plant material transfer from one producing country to another, respecting the relevant legislative norms;

· simultaneous performance trials with ecotypes or hybrids in several production zones; and

· an active and dynamic coconut R & D information network among producing countries.

National organization involved in coconut breeding

In Benin, only the Sémé-Podji Coconut Research Station is working on coconut breeding. Although this station is under the authority of Institut National des Recherches Agricoles du Benin (INRAB), a state institution, only its staff salaries are paid from the national budget. Hence, it has to operate using its own resources.

Conclusion and recommendation

For more than ten years, Benin has had an estimated 15 000 ha of coconut plantations. Yields are low as most of the plantations contain ageing WAT palms. Nevertheless, these coconut plantations are financially worthwhile for both owners and farmers alike.

The old plantations need to be replanted and the State is looking for ways of succeeding. Coconut breeding work at the Sémé-Podji Station is handicapped by the small number of ecotypes in its collection and limited financial resources.

Genetic improvement is needed to improve yields, but national research organizations are unable to work on yield improvement without international cooperation among producing countries. Such international cooperation should include:

· easy exchange of plant material, respecting the corresponding legislation;

· multi-site hybrid performance trials;

· the search for and mastery of processing technologies that can ensure reasonable industry promotion of the coconut sector; and

· vegetative propagation of high-yielding hybrids.

Table 1. Yield of different ecotypes (nuts/tree/year)

Ecotype	Yield at farmers’ field	Yield in station	Yield in irrigated farms
PB121
(MYD × WAT)	60 – 90	115-120	+150
PB 111
(CRD × WAT)	50 – 80	100 – 105	–
GOA	25 – 60	60 – 80	–

Coconut breeding programme in Côte d’Ivoire

Roland Bourdeix, N’Cho Yavo Pierre and Allassane Sangare
Plant Breeder; Head, Genetics and Breeding; and Director, IDEFOR/DPO Marc DELORME, Coconut Research Station, Abidjan, Côte d’Ivoire

Introduction

The coconut palm in West Africa has a relative discrete existence besides the oil palm which is the traditional perennial oil crop. However, its importance in Côte d’Ivoire is far from being minor, especially in the coastal region where the coconut is the principal cash crop. The value of multiple uses of the coconut palm is not yet realized.

Production and hectarage

In 1966, the coconut grove in Côte d’Ivoire was composed of family-owned farms covering only 10 000 hectares. Since 1967, with the National Coconut Development Programme, new industrial and smallholding farms have been established, increasing the hectarage up to 50 000 in 1980. These coconut groves, located mainly in the coastal region (95%), are monocrops except the central region of the country where intercropping is practised.

The hectarage (ranks second in Africa after Tanzania) is unchanged since 1990 (Table 1). The industrial farms (Palmindustrie and IDEFOR) represent 41% (20 500 ha); the rest (59%) are managed by 3900 smallholders.

The principal limiting climatic factor is rainfall which must be higher or equal to 1500 mm and well-distributed throughout the year. Such condition exists from the littoral to the central part of the country. Within this ecological zone, clay or gravel type of soils are avoided. Sandy and deep soils, with a good water and air exchange capacity, are found very favourable for coconut.

There are several marginal areas (800 000 ha) where oil palm and other perennial crops cannot be grown. However, they are suitable to coconut palm and thus, could be used for development of new farms.

Planting material used and national farm productivity

Generally an average Ivorian coconut grove is composed of 48% hybrids and 52% West African Tall (WAT). Among the grown hybrids, the PB121 (Malayan Yellow Dwarf × WAT) is predominant. The other cultivated hybrids are the PB111 (Cameroon Red Dwarf × WAT) and PB141 (Guinean Green Dwarf × WAT).

In Côte d’Ivoire, genetic erosion related to the planting of hybrids is not a significant problem. WAT is the only local ecotype in the country. This ecotype, which is phenotypically very homogeneous, represents more than 50% of the coconut groves. Moreover/it has been extensively released world-wide which guarantees its preservation in several genebanks.

In 1993, the Ivorian coconut growers produced 681 million nuts (equivalent to 128 900 t copra). The average productivity of 2.7 t/ha placed Côte d’Ivoire at the top in terms of productivity per hectare (Table 1). The difference between the yield of the hybrids and the West African Tall was very significant. At the Marc DELORME Research Station, the best F₁ hybrids yielded twice (average yield of 4 t copra/ha/year) as much as the local WAT.

The industrial plantations of Palmindustrie offered the opportunity of comparing the productivity of 4000 ha of WAT with the 2500 ha of hybrids (90% PB121). From 1985 to 1990, the hybrids yielded 2.4 t of copra while the WAT produced 1.5 t. For the year 1990, the yield was 3.1 t and 2.1 t (De Taffin et al. 1991) for the hybrids and the WAT, respectively.

Similarly, the hybrids out-yielded the local material in different cultivated areas of Côte d’Ivoire. The replanting of the low yielding WAT areas with hybrid material is therefore necessary. The continuous monoculture in the coastal region markedly depleted the soil. Hence, soil fertility restoration becomes a prerequisite before replanting these areas.

Coconut industry in Côte d’Ivoire

In Côte d’Ivoire, the principal coconut products are the oil, desiccated coconut, nuts for consumption, copra and copra cake. These products are exported (80%) to Europe and within the West African region (Table 2).

Since 1980, the Government Estates Palmindustrie had been involved in the extension service to farmers by supplying them with fertilizers. Since the drop of the oil price in the world market in 1986, this extension service has been discontinued.

Palmindustrie is also the principal buyer of the farmers’ copra and sets the price policy. In June 1994, the price was US$140/t for the high-grade copra, and US$120/t for the second grade. Transport cost to the copra mill, located in Abidjan, is reimbursed to the farmers. The international price for the copra was then US$375/t. On the local market, the fresh nuts are sold for CFA50 (US$500) in terms of copra equivalent per ton.

An average farm (2.7 t copra/ha/year) can gross US$378/ha/year. The management cost is estimated at US$175 (including fertilizers) per hectare with net income estimated at US$200 per hectare. It is difficult to give very accurate values since the new production costs are yet to be established due to the recent devaluation of the national currency.

The breeding programme

The breeding strategies used in Côte d’Ivoire for coconut are based on the conservation and evaluation of accessions in the coconut genebank, the utilization of the germplasm for detecting the best combinations among ecotypes and the improvement through development of hybrids identified through performance tests among progenies (De Lamothe 1970; Gascon and De Lamothe 1976). This scheme has been slightly restructured into two different axes (tall × tall and dwarf × tall) using the reciprocal recurrent selection method (Bourdeix et al. 1990, 1991a and b).

Breeding objectives

In Côte d’Ivoire, the focus of breeding is on yield which is expressed in terms of copra per hectare, being the end-product sold by farmers. The production precocity is another important criterion that can allow a rapid exploitation of the plantation.

The yield on coconut is often limited by some phytophathological problems. As far as coconut diseases are concerned, Côte d’Ivoire is particularly safe. The only disease with economic significance is Phythophtora katsurae which causes bud rot and premature nut-fall in affected palms.

Some tolerant hybrids to this disease are being released by the Marc DELORME Research Station, in addition to chemical control methods. The Aceria mite also damages fruits, causing a reduction in the copra content of the nut. The tolerance to this mite is indirectly considered in the evaluation of yield components.

The coconut is very often cultivated on poor sandy soils where other crops do not grow very well. Pest control and fertilizer application are sometimes lacking in smallholdings. Therefore, it is very important to select planting materials suitable under such harsh environmental conditions. Other environmental stresses, such as drought, are also considered in breeding for desired ideotypes.

Coconut germplasm

Importance of the genetic collection

The need for establishing a coconut genebank in Côte d’Ivoire was expressed very early since the local cultivars showed limited genetic variability. Since 1953, the Institut de Recherches pour les Huiles et Oléagineux (IRHO) has introduced, from the different tropical areas, 53 ecotypes at the Marc DELORME Research Station. This collection has a broad geographic diversity with large sample size (more than 100 palms per population).

The different ecotypes originated from Latin America and Caribbean, Africa, Indian Ocean, the Pacific and Southeast Asia. Thirty-six tails (20 687 palms) and 17 dwarfs (4275 palms) covering 165 hectares, represent this important field genebank. A photographic descriptive manual of this germplasm collection is being made by the Marc DELORME Research Centre.

In addition to the 53 ecotypes, there are 160 selfed families of West African Tall which are selected genitors, as well as, 25 tall × tall hybrids used both for breeding and germplasm conservation.

Germplasm exchange regulation

The import and export of planting materials in the country are subjected to double permits: one from the Scientific Research Ministry and the other from the quarantine service of the Agriculture Ministry. Import of coconut germplasm is possible provided the country of origin does not have a mycoplasma-like-organism or phytomonas type of disease. De Taffin and Wuidart (1981) described the different phytosanitary precautions to be used when receiving imported seednuts.

In Côte d’Ivoire, before export, the seednuts are treated with insecticide and fungicide, dried for 24 hours and put in bags (IRHO CIRAD-CP 1980). Permeable bags can be used by dipping the closed bag containing the seednuts into a treatment solution.

Most of the germplasm exchange is now done in the form of pollen or in vitro zygotic embryos.

Methods of evaluation

Several studies were carried out on the Marc DELORME collection in order to understand the coconut genetic diversity. Morphological and production characters, electrophoresis of isozymes and leaf polyphenols analysis were used to assess diversity.

Characterization and evaluation of ecotypes in a collection require a large amount of standardized measures at different stages of the plant development. The evaluation methods used at the Marc DELORME Station, have been reported in several papers (Meunier et al. 1977, Wuidart and Rognon 1978; De Lamothe and Rognon 1982). Since 1977, a series of papers describing the different populations of the collection has been published. Eight dwarf ecotypes and 11 tall populations have been described. Recently, a synthetic approach has been used to describe the measured parameters of these ecotypes by multivariate analysis.

Seventeen tall ecotypes were studied this way by using 24 morphological characters (N’Cho et al. 1993). This method has allowed some estimate of genetic distances between ecotypes that could guide the breeder in the choice of parental combinations.

Enzymatic electrophoresis was used to assess genetic variability in coconut. This work was conducted using pollen (Benoit and Ghesquiere 1984). The eight ecotypes analysed revealed very low enzyme polymorphism. The variability within ecotypes was lower with the autogamous dwarf varieties than in the tails (except the West African Tall which was monomorphic for the four enzymatic systems tested).

Leaf polyphenols analysis through gas chromatography was initiated on the Marc DELORME collection material (Jay et al. 1989). This technique provided some images of the variability that fitted perfectly the real origin of the ecotypes. It allowed the detection of relationships between and among populations of different origins. However, another test conducted on control populations cultivated in Indonesia, Vanuatu and Côte d’Ivoire showed some limits of the method. The relative polyphenols content seemed to be affected by environmental factors making comparison among populations planted in different countries difficult.

One of the main concerns of breeders is to have a clear idea of the genetic diversity of the species to serve as guide in hybridization work (for example classification of groups having good combining ability). For this purpose, molecular analysis of the genotype will be very useful. This analysis is independent from the environmental interactions that sometimes limit the accuracy of the other evaluation methods. The genetic materials of the Marc DELORME collection are very suitable for such analysis because a significant number of ecotypes from different origins, with large sample size, are available in this genebank. Furthermore, successive progenies from controlled crosses are conserved under field conditions. For several populations, there are selected genitors, their parents, their crosses with different testers, and their selfed families composed of 100 individuals each.

Detection of the best inter-ecotype hybrids

From 1965 to 1993, 121 inter-ecotype hybrids (2- or 3-way hybrids and reciprocal crosses were not included) were tested in Côte d’Ivoire. These breeding trials covered 132 ha (minimum of 1 ha per entry). The techniques used for setting genetic trials have been reported in several papers dealing with the seedbed (Wuidart 1979a, 1981a and 1981c), polybag nursery (Wuidart 1979b, 1981b; Rognon 1971), the field planting (Pomier 1979; Duhamel 1987) and fertilizer application (Coomans and Ochs 1976; Manciot et al. 1980; Ouvrier 1984; De Taffin and Rognon 1991).

Fig. 1 summarizes the different inter-ecotype combinations made by the Marc DELORME Research Station. The first genetic trials used the WAT (from two cycles of mass selection) as a control. These trials compared 35 hybrid combinations from inter-crossing of 35 parent ecotypes.

Table 3 shows the production performance of the best seven identified inter-ecotype hybrids. These hybrids were significantly superior to the WAT control. On average, out of the total 35 hybrids tested, none of them performed less than the control; four of which have even yielded twice that of the WAT.

The other trials planted between 1976 and 1992 used the PB121 as a control (De Lamothe and Benard 1985). These trials have allowed detection of new outstanding hybrids; very productive during the adult stage (9 to 12 years) and more tolerant to Phytophthora than the PB121. Since 1993, improved PB121 (Bourdeix et al. 1992) and PB213 (WAT × Rennel Tall) are being used as control for the genetic trials.

Fig. 1 shows many empty intersections as it is unrealistic to test all possible hybrid combinations. In fact, the 121 inter-ecotype hybrids (26% of the possible combinations) were made during 30 years of continuous breeding research. A choice of the combinations to be tested is, therefore, very important.

At the beginning of the breeding programme, the choice was more exploratory. Criteria such as the apparent genetic distance between parents, the complementarity of their characters and the production level, were used. Progressively, some well-known populations have been used as testers. For the moment, there are two types of trials: basic and complementary trials.

In the tall × tall hybrids improvement, the basic trials consisted of crossing every new accession with two complementary tall testers with well-known characters. In Côte d’Ivoire, the WAT and Rennel Tall (or selected genitors from these two populations) are being used as testers.

The tester used for the dwarf × tall hybrids improvement was unique. In this case, the basic trials consisted of crossing every new accession with a single tester. The tails were tested with the Malayan Yellow Dwarf while the dwarfs used one tall ecotype as a tester (the WAT was used then but the improved WAT × Rennel Tall hybrid is currently preferred).

The complementary trials have more complicated crossing schemes. These trials are designed to provide genetic information on the type of gene action involved in coconut hybrid vigour. Incomplete or complete factorial designs are very often used for this purpose (Fig. 2).

Improvement of the best hybrids

The second phase of the breeding programme was initiated since 1970 (Gascon and De Lamothe 1976) to individually improve the best detected hybrids.

To illustrate the method, 45 selected TAT (Tahitian Tall) palms were crossed individually with the same Malayan Yellow Dwarf (MYD) population to improve PB122 (MYD × Polynesian Tall). The 45 half-sib progenies were tested in comparative hybrid trials.

In theory, the improvement of an inter-ecotype hybrid have to have two complementary crossing designs: palms from each population are crossed individually onto a set of palms in the other population and reciprocally. However, this strategy can be simplified when the two populations present unequal level of variability. For example, in the improvement of the dwarf × tall hybrids, only the combining ability of the tails was tested since the autogamous dwarfs were very homogeneous.

The experimental designs used were (4 × 4 or 5 × 5) lattices that allowed comparison of 15 or 24 half-sib families in addition to a control. The genitors used as male parents were tested individually. They were later selfed in order to obtain progenies for conservation and multiplication. These self-pollinated progenies will also provide pollen for hybrid seed production.

The first results have shown that, by selecting 7-8% of the best families, a genetic gain ranging between 15 to 30% depending on the trials (Bourdeix et al. 1989), could be achieved. It confirmed the variability that exists within the natural population. The analysis of these trials provided the following guides in formulating effective breeding strategies:

· Phenotypic selection of parents is sometimes efficient but cannot be substituted to the progeny test. Some good yielding genitors can have average combining ability value and inversely. Although the progeny test is costly, it is very necessary for genitors selection.

· The best yielding progenies, in terms of copra per hectare, do not show above average number of bunches and copra content per nut. It seems that the genetic progress is mainly due to the improvement in the number of nuts per bunch. However, in some cases, the percentage of copra in the fruit without water is slightly improved.

Creation of complex (3- and 4-way) hybrids

The complex hybrids were developed in Côte d’Ivoire since 1976. The objectives of this programme are to evaluate the genetic variability of hybrid progenies and to select outstanding individuals for multiplication once the somatic embryo culture technique in coconut is available. The following combinations covering 23 hectares, are being tested:

GYD × (WAT × RIT)	LMT × (WAT × RIT)
GYD × (WAT × TAT)	(WAT × TAT) × RIT
(MYD × WAT) × (EGD × RIT)	(MRD × RIT) × (WAT × TAT)
(CRD × MYD) × (WAT × RIT)	(CRD × RIT) × (EGD × WAT)
(MRD × MYD) × (WAT × TAT)	(MRD × WAT) × (TAT × VNT)

Please refer to Fig. 1 for the complete cultivar names of the above germplasm materials.

Evolution of the breeding strategies

A key problem in a coconut breeding strategy is how to integrate the dwarf character in the breeding scheme. In other words what type of hybrid (dwarf × dwarf, dwarf × tall or tall × tall) should be the focus of the genetic improvement?

Dwarf × Dwarf hybrids

The more the dwarf genotype increases in a hybrid, the more precocious this hybrid is. In addition, the resulting reduced bulkiness allows a higher planting density. In Côte d’Ivoire, the hybrid between the Malayan Yellow and Red Dwarfs yielded 3.8 t of copra per hectare during the adult phase, at a planting density of 170 palms/ha (Le Saint et al. 1987). In the same trial, the control Malayan Yellow Dwarf produced one t of copra less than the hybrid, the yield of which is comparable to the dwarf × tall materials (IRHO-CIRAD 1989).

Nevertheless, dwarfs are not broadly adapted. They are not very tolerant to drought (Ziller 1962). These non-desirable traits could be transmitted to their hybrid progenies. Besides, in the world coconut production, the importance of dwarf varieties is marginal. The genetic variability of dwarf ecotypes is far below the variability found in tails. Hence, the long term genetic potential of the dwarfs is limited.

Nevertheless, the precocity and the high rate of bunch emission are valuable traits of the dwarf ecotypes. Dwarf × dwarf hybrids should be created for cumulating such desirable genes to be incorporated in a breeding scheme even though the primary objective may not be to release this type of material (Bourdeix et al. 1991a).

Dwarf × Tall or Tall × Tall

The importance of the dwarf × tall hybrids in the world is mainly due to their relative precocity. However, this advantage over the tall × tall hybrids is not sufficient for eliminating the latter. The WAT × RIT improved hybrid can have a cumulative production equivalent to the yield of the widely grown hybrid PB121, at the 9th year: This tall × tall hybrid later outyielded the PB121 control (Bourdeix et al., in press).

The impact of the precocity in the choice of hybrid depends on various economic considerations. When one is concerned with profitability during the initial production years of the plantation, dwarf × tall material may be interesting. However, when the long term income is taken into account, improved tall × tall hybrids are to be considered (IDEFOR/DPO 1992).

The earliness and yield are not the only criteria in the choice of planting material. The price of seednut is another factor which can affect this choice. The tall × tall seednuts may have a higher production cost that could be reduced in the future with the use of in vitro propagation technique. The stature of dwarfs, their precocity and existence of markers which allow a detection of illegitimate seednuts, are practical and economically advantageous.

Restricting a breeding strategy to the dwarf × tall hybrids would be equivalent to opposing two populations with unequal genetic variability. Neither the estimates of breeding potentials nor the characters desired by users are enough to discard the tall × tall hybrids. Furthermore, diversification of the type of hybrids cultivated will probably occur in the next decade.

Management of the breeding axes

It is important to maintain at least two breeding schemes: one for the dwarf × tall, the other for tall × tall material. How to define a breeding strategy that allows an optimum exploitation of these two axes?

Harries (1991) has suggested the production of 3-way hybrids using dwarf × dwarf, dwarf × tall or tall × tall female genitors. Some plantations with composite characters could be exploited for fruit or hybrid seednuts production according to the demand. Such evolution of strategy seems inevitable. Indeed, in the evolution of breeding programmes, overlapping generations will occur and combinations will become more complex. Some varieties in the Year 2000 will probably have genes from more than three or four accessions. The overall problem will then be the proper management of parental genitors (no matter what their origin is) in order to get the best progenies.

Based on the outcome of the programme and genetic trials (Bourdeix et al. 1990, 1991a and 1991b), new orientations of the coconut breeding programme were proposed at the Marc DELORME Station. The proposed scheme was based on the reciprocal recurrent selection (RRS) method (Comstock et al. 1949). Nevertheless, this new approach was not intended to discard the previous recommended methods. It was a generalization of principles already used.

Two main areas (improvement of tall × tall and dwarf × tall hybrids) are described. Fig. 2 gives a schematic illustration of the method.

The conception of the dwarf × tall axe was relatively simple. The dwarfs and the tails are two different groups with some complementary characters. In addition, the combining ability between these two groups is very good. Constraints related to the biology of the palm besides other genetic factors lead to the choice of the half-sib families RRS method.

The conception of the tall × tall axe was more difficult. The tall ecotypes group represented the main component of the genetic variability in coconut. High heterosis was expressed in some combinations. Several works were conducted in order to assess for the genetic diversity. Morphological approaches (Harries 1978; N’Cho et al 1993) gave some idea on the partitioning of the tall population. However, the present knowledge of the coconut genetic diversity is not enough for an efficient exploitation in breeding programmes.

The method used for the tall × tall hybrid improvement takes into account the lack of very accurate technique for assessing genetic distances between and among ecotypes. Two artificial populations, based on two ‘founder’ ecotypes, are created and improved in respect of each other by half-sib RRS.

The choice of the ‘founder’ ecotypes should take into account the different constraints of the specific breeding programme. The phytopathological status and the availability of the material are to be considered. For practical reasons, one of the ecotypes should be a local variety. In Vanuatu, the local tall which is tolerant to the lethal foliar decay disease could be used. Results from different research centres show that the Tagnanan and Sri Lanka Tails are potential ‘founders’.

In Côte d’Ivoire, the analysis of genetic trials led to the choice of the Rennel Island Tall (Pacific) and the West African Tall (Côte d’Ivoire and Benin) as ‘founders’. These two ecotypes showed very good combining ability between themselves as well as with other dwarf and tall ecotypes. Moreover, some genitors with excellent general combining ability were identified from these two populations (Bourdeix et al. 1989 and 1992). The use of these particular genitors as testers, has several advantages.

Breeding action plan and expected output

Due to some financial constraints, the area used for the genetic trials is reduced to 10 hectares per year. The planned experimentations will concern:

· Trials on combinations among different tall accessions with the two improved WAT and RIT testers. The control will be a cross between the two testers.

· Trials on combinations of tall ecotypes with the Malayan Yellow Dwarf tester and the improved PB121 as control.

· Trials on combinations among the dwarfs and with the tester WAT × RIT. The control will be MYD × (WAT × RIT).

Summary and conclusions

Most of the breeding programmes were more or less focused on exploiting the hybrid vigour. However, this option is not restrictive but often requires some precision. Various methodological options are possible – should one compare progenies from populations, individuals, half-sib or full-sib families? How to structure the genetic variability in order to optimise the genetic progress? How to anticipate the future clonal propagation? Answers to these questions are required for successful breeding programmes in the future.

A methodological thinking conducted by IDEFOR and CIRAD has led to new orientations of the coconut breeding programme at the IDEFOR/DPO Marc DELORME Station (Côte d’Ivoire). The first phase of the programme concerns the genetic resources – enrichment of the collection and exchange of germplasm, and assessment of the genetic diversity by different techniques. The second phase involves detecting and then improving the best inter-ecotype combinations. The best inter-population hybrids have been released, extensively at the national and international level. In Côte d’Ivoire, these hybrids represent 48% of the coconuts grown, making the country the world’s best productivity record holder. The recent trials have allowed genetic progress ranging from 15 to 20% compared to the previous released hybrids. New materials could yield as much as 3.5 t per hectare in the industrial plantations of Côte d’Ivoire.

Results of various trials have led to a restructuring of the breeding scheme. Two breeding schemes (dwarf × tall and tall × tall), using the recurrent reciprocal selection method, are adopted. This new strategy which is being applied in Côte d’Ivoire, considers the possibility of exploiting the prospects offered by biotechnology. Some specific combinations were made in anticipation of the in vitro propagation. The creation of complex hybrids and their clonal exploitation are expected to increase yield up to four t of copra per hectare in the 21^st century.

References

Benoit, H., M. Ghesquiere. 1984. Electrophorese, compte rendu cocotier. IV. Determinisme genetique, Rapport interne IRHO-CIRAD, (FRA). 11 p.

Bourdeix, R., J. Meunier., Y.P. N’Cho. 1991a. Une strategie de selection du cocotier Cocos Nucifera L. II. Amelioration des hybrides Grand × Grand. Oléagineux 46:367-382.

Bourdeix, R., J. Meunier., Y.P. N’Cho. 1991b. Une strategie de selection du cocotier Cocos Nucifera L. III. Amelioration des hybrides Nain × Grand. Oléagineux 46:361-374.

Bourdeix, R., Y.P. N’Cho. 1992. Influence du sens du croisement sur la production de huit hybrides entre ecotypes Nains et Grands. Oléagineux 47:113-118.

Bourdeix, R., Y.P. N’Cho, J.P. Le Saint. 1990. Une strategie de selection du cocotier. Synthèse des acquis. Oléagineux 45:359-371.

Bourdeix, R., Y.P. N’Cho, A. Sangare. 1995. Amelioration de l’hybride de cocotier Grand Ouest Africain × Grand Rennell. Comparaison avec un temoin PB121 (Nain Jaune Malaisie × Grand quest Africain), a paritre.

Bourdeix, R., Y.P. N’Cho., A. Sangare., L. Baudoin., Nuce de Lamothe M. de. 1992. L’hybride de cocotier PB121 ameliore, croisement de Nain Jaune Malaisie et de géniteurs Grand Ouest-Africain améliores. Oléagineux 47:619-633.

Bourdeix, R., A. Sangare., J.P. Le Saint. 1989. Efficacite des tests hybrides d’aptitude individuelle a la combinatison chez le cocotier: premiere resultats. Oléagineux 44:209-214.

Comstock, R.E., H.F. Robinson, P.H. Harvey. 1949. A breeding procedure to make maximum use of both general and specific combining ability. Agronom. J. 41:360-367.

Coomans, P., R. Ochs. 1976. Rentabilite des fumures minérales sur cocotier dans les conditions du Sud-Est Ivoirien. Oléagineux 31:375-392.

De Taffin G. de et F. Rognon. 1991. Le diagnostic foliaire du cocotier. Conseil de l’IRHO n° 318. Oléagineux 46:169-173.

De Taffin G. et W. Wuidart. 1981. Precaution a prendre avec des semences de cocotier ayant effectue u long voyage. Oléagineux 36(8-9):429-432.

De Taffin G., N. Zakra, C.P. Bonny. 1991. Performance des hybrides de cocotiers Nain × Grand en conditions industrielles en Côte d’Ivoire. Oléagineux 46:187-195.

Duhamel G. 1987. Piquetage des cocoteraies, Consil de l’IRHO n° 280. Oléagineux 421:325-326.

Gascon, J.P., M. de Nuce de Lamothe. 1976. Amelioration du cocotier. Methode et suggestions pour une cooperation internationale. Oléagineux 31:479-482.

Harries, H.C. 1978. Evolution, dissemination and classification of Cocos nucifera L. Botanical Review 44:265-320.

Harries, H.C. 1991. The promise, performance and problems of F1 hybrid Coconut. Pp.380-390 in Coconut Breeding and Management. Proceedings of the National Symposium held from 23rd to 26th November, 1988. Kerala Agricultural University, Vellanikkara 680 654, Trichur, India.

IDEFOR/DPO. 1992. Rapport d’activites Recherches de la Station de Recherches “Marc Delorme”, annees 1991-1992, 361 p.

IRHO-CIRAD-CP. 1980. Coconut seeds from the Institut de Recherches pour les Huiles et Oléagineux. Unpublished.

IRHO-CIRAD. 1989. Rapport d’activities Recherches de la Station de Recherches “Marc Delorme”, annees 1988, 361 p.

Jay, M., R. Bourdeix, F. Potier., C. Sanlaville. 1989. Premier resultats de l’etude des polyphenols foliaires du cocotier. Oleaguineux 44:151-161.

Le Saint, J.P., M. de Nuce de Lamothe. 1987. Les hybrides de cocotiers nains: performance et interet. Oléagineux 42(10):353-362.

Manciot, R., M. Ollagnier, R. Ochs. 1980. Nutrition minerale et fertilisation du cocotier dans le monde. Oléagineux 35:3-55.

Manthriratne, M.A.P. 1971. In: Annual Report of Coconut Research Institute for 1971.

Meunier, J., F. Rognon, M. de Nuce de Lamothe. 1977. L’analyse des composantes de la noix du cocotier. Etude de l’echantillonage. Oléagineux 32:9-14.

N’Cho, Y.P., A. Sangare, R. Bourdeix, F. Bonnot, L. Baudoin. 1993. Evaluation de quelques ecotypes de cocotier par une approche biometrique. I. Etude des populations de Grands. Oléagineux 48:121-132.

Nuce de Lamothe M. de. 1970. Application du principe des croisements interorigines au cocotier. Première resultate obtenus en Côte d’Ivoire. Oléagineux 25:207-210.

Nuce de Lamothe M. de et F. Rognon. 1982. L’observation des caracteristiques de developpement vegetatif, de floraison et de production chez le cocotier. Oléagineux 37:290-296.

Nuce de Lamothe M. de, G. Benard. 1985. L’hybride de cocotier PB213 (GOA × GRL). Oléagineux 40:491-496.

Nuce de Lamothe M. de, G. Benard. 1985. L’hybride de cocotier PB121 (ou Mawa) (NJM × GOA). Oléagineux 40:261-266.

Ouvrier, M. 1984. Exportation par la recolte du cocotier hybride PB121 en fonction de la fumure potassique et magnesienne. Oléagineux 39:263-271.

Pomier, M. 1979. Plantation des cocotiers eleves en sacs de plastiques. Conseil de l’IRHO n° 189. Oléagineux 34:307-310.

Rognon, F. 1971. Les pepinieres de cocotier en sacs de plastique. Conseil de l’IRHO n°106. Oléagineux 26(5):307-310.

Wuidart, W. 1979a. Production de materiel vegetal cocotier. Selection au stade germoir. Conseil de l’IRHO n° 196. Oléagineux 34(8-9):395-397.

Wuidart, W. 1979b. Production de materiel vegetal cocotier. Selection en pepiniere. Conseil de l’IRHO n° 197, Oléagineux 34(10):453-456.

Wuidart, W. 1981a. Production de materiel vegetal cocotier. Selection en pepiniere. Tenue d’un germoir. Conseil de l’IRHO n° 215. Oléagineux 36(6):305-309.

Wuidart, W. 1981b. Production de materiel vegetal cocotier. Pepiniere en sac plastique. Conseil de l’IRHO n° 216. Oléagineux 36(7):367-376.

Wuidart, W. 1981c. Production de materiel vegetal cocotier. Selection en pepiniere. Conseil de l’IRHO n° 218, Oléagineux 36(10):497-500.

Wuidart, W. et F. Rognon. 1978. L’analyse des composantes de la noix du cocotier. Methode de determination du coprah. Oléagineux 33:225-233.

Ziller, R. 1962. La selection du cocotier dans le mode. Oléagineux 17:837-846.

Fig. 1. Panorama of hybrids tested by the Marc DELORME Station (Côte d’Ivoire)

Fig. 2. Illustration of the two types of trials (basic and complementary) in the case of dwarf × tall hybrids improvement

Table 1. Evolution of the coconut production and hectarage

Year	Hectarage (ha)	Production
		Equivalent copra (t)	Nuts (millions)
1989	49 000	84 200	421
1990	48 000	124 600	623
1991	48 000	110 000	550
1992	48 000	97 200	486
1993	48 000	128 900	681

Table 2. Export of coconut products from Côte d’Ivoire

Year	Fresh nuts (millions)	Coconut oil (t)	Desiccated coconut (t)	Copra cake
1989	42.1	44 900	7 800	11 400
1990	62.3	66 400	8 400	13 400
1991	55.0	58 600	8 400	11 500
1992	48.6	51 800	8 500	11 100
1993	68.1	58 000	9 000	5 800

Table 3. Production performance of the first group of outstanding hybrids identified in Côte d’Ivoire compared with the West African Tall control

Trial	Production and percentage compared to the control
	Trial No. 3	Trial No. 8	Trial No. 5	Trial No. 5	Trial No. 5	Trial No. 11	Trial No. 11
Observation period	9-12 ans	9-13 ans	9-15 ans	9-15 ans	9-15 ans	9-14 ans	9-14 ans
Hybrids	WAT×RIT	WAT×VNT	MYD×WAT	MRD×TAT	MRD×TAT	NJM×RIT	CRD×WAT
tested	PB 213	PB 214	PB121	PB 132	PB 122	PB 123	PB 111
Earliness	51	57	51	50	52	40	44
(months)	(91%)	(90%)	(84%)	(82%)	(85%)	(75%)	(83%)
Number of	12.7	13.5	14.8	14.2	14.2	16.5	15.5
bunches/palm	(109%)	(118%)	(123%)	(118%)	(118%)	(120%)	(113%)
Number of	102	100	109	101	110	123	132
fruits/palm	(134%)	(238%)	(188%)	(174%)	(190%)	(164%)	(176%)
Copra/nut	311	212	247	282	253	289	240
(g)	(138%)	(89%)	(105%)	(120%)	(108%)	(128%)	(107%)
Copra/ha	4.26	3.15	3.67	3.88	3.80	4.80	4.35
(t)	(186%)	(235%)	(197%)	(209%)	(204%)	(207%)	(188%)

Coconut breeding programme in Ghana

Joseph Owusu Nipah¹ and Sylvester K. Dery²

¹ Assistant Research Officer, Oil Palm Research Institute, Kade, Ghana
² Coordinator, Coconut Project, Sekondi, Ghana

Introduction

The coconut palm is the most important cash crop in the four coastal regions of Ghana (Greater Accra, Central, Volta and Western Region). Before 1920, its cultivation, as an estate crop, was confined to the Keta area. Its cultivation then spread to other areas of Ghana particularly along the coast (Wills 1962; Agble 1970).

Arkhurst (1991) estimated the area under coconut cultivation to be 43 000 ha with about 80% restricted to the southwestern coastal belts.

Coconut production in Ghana is mainly in smallholdings (0.5-5.0 ha). Out of the annual national production of 224 million nuts, 179 million (80%) were produced by smallholders from an area of 36 000 hectares (Arkhurst 1991).

Yields in smallholder farms were relatively low. Average yield per palm was estimated at 35 nuts per annum (5000 nuts/ha).

Farms which may be classified as well-managed were about 20% of the total area under coconut cultivation. Average yield in these farms, estimated at 45 nuts/palm/annum (6 500 nuts/ha), was below standard. Under Ghanaian conditions, individual West African tall (WAT) palm yielding 70 or more nuts per year had been recorded (Adansi 1970).

Income derived from farms depended on the state of coconut product marketed. According to Raux (1990), about 6.1% of nuts were marketed fresh (green stage). These were sold at 20 cents per nut, yielding a gross income of about 100 000 cents/ha/year (about US $108) for the small scale farmers. With the farmers themselves providing labour to a large extent, net income may be estimated at 90% of the gross. About 12% of nuts were kept for home consumption and for seeds. The rest (81.4%) were processed in industrial as well as village oil mills (Raux 1990). When processed into oil, the farmers derived a net income of about 123 000 cents/ha (US $133).

National coconut germplasm

Table 1 shows the list of coconut varieties currently available in Ghana. With the exception of the local West African tall, the others were brought in from other countries particularly Côte d’Ivoire.

With funding from the EC STD 3 programme, eight ecotypes and three hybrids were requested from the Marc Delorme Station in Côte d’Ivoire to expand the Lethal Yellowing (LY) trials for resistant varieties in Ghana. Seven cultivars were delivered in early May 1994, namely:

Tacunan Green dwarf
Catigan Green dwarf
Panama tall
Tagnanan tall
Indian Lccadive tall
Indian Andaman ordinary
West African tall (Benin)

The other four types, delivered between May and August 1995 are:

Vanuatu tall
Sri Lanka Green dwarf × Vanuatu tall
Vanuatu tall × Panama tall
Malayan Yellow dwarf × Vanuatu tall

All these new accessions have been planted in the resistance trial plots. However, there are plans to establish a coconut field genebank at the Oil Palm Research Institute which is outside the disease zone.

All the cultivars in the country can be considered to be at risk from the Cape St. Paul Wilt Disease (CSPWD), a lethal yellowing type disease. This is because of the saltatory nature of CSPWD. The disease sprang from Cape St. Paul to Cape Three Points, a distance of 450 km. The disease is thus capable of appearing in any part of the coconut belt and also in Côte d’Ivoire. All the listed varieties are in the Western Region which to date has lost 2000 ha of coconut to the CSPWD. The field genebank will thus be located at Kade in the Eastern Region, and replicated at Bunso in the Eastern Region.

With the programme on introducing new varieties in place, the genetic base of coconut in the country is expected to widen. It is still important to introduce more materials for screening for CSPWD resistance. Due to the importance the government of Ghana attaches to coconut research, it does not have special restrictions nor requirements for entry of new coconut varieties into the country. The Oil Palm Research Institute, in collaboration with the Plant Protection and Regulatory Services, takes care of the quarantine and delivery requirements of all introduced entries. The imported materials will be maintained and used for screening and/or breeding resistant varieties against CSPWD.

Productivity problems

There has been a reduction in national production since the late 1970s. Ghana’s coconut production dropped from 160 000 t in 1979 to 110 000 t in 1987 (FAO 1987). Several factors are responsible for the low farm productivity and the drop in production in the country. The low level of farm management practices was one of the major factors. Many of the smallholders are illiterate farmers who have very little knowledge on modern farming practices. This, coupled with the cultivation of low yielding West African tall (WAT) variety, contributed to low productivity. As such, efforts have been made over the past ten years to produce and introduce improved and high-yielding materials to replace WAT.

The rate of adoption of the improved varieties, as well as that of replanting the devastated fields, was quite low. This stems from the pessimistic view of farmers about ever finding a solution to CSPWD.

The environmental limitation which prevented the growing of coconut in the hinterland is another reason for the low production. To overcome this problem, trials will be set up in the immediate future, in collaboration with the Soil Research Institute, on two possible coconut growing areas:

1. Gentle slopes with high water table terminating into marshy valleys occurring in the forest and transitional areas in Ashanti, Brong Ahafo, Central and Volta Regions; and

2. Gentle slopes with high water table in guinea savanna areas mostly along river basins stretching up to latitude 11°N. These, according to Djokoto and Dzomeku (1990), are possible areas for coconut production.

Efforts will also be made to select/breed progenies that are well-suited for the drier conditions prevailing in the hinterland.

The most pressing problem of the coconut industry in Ghana however, is the CSPWD. This lethal yellowing disease is currently causing widespread death of coconut palms in Ghana. Diseases similar to it are known to occur in the Caribbean, Florida, South America, Togo, Nigeria and Cameroon (Romney 1972). Efforts are being made to devise measures that will control the disease.

In the past, trials were mounted to test the various coconut varieties for resistance/tolerance to CSPWD. All the varieties tested, however, succumbed to the disease at various stages of the trials. The varieties tested included MGD, MYD, MRD and CRD.

While some coconut hybrids, notably Malayan Yellow dwarf × Panama tall are considered resistant to the Caribbean strain of Lethal Yellows, they are, nevertheless, reported to be susceptible to the disease in East Africa. The results of trials conducted in Ghana showed the same trend. This may be due to the occurrence of different strains of Lethal Yellowing Phytoplasma in different parts of the world. It has now been proven by researchers at Rothamstead Research Station in the United Kingdom that the East and West African strain of LY MLO are indeed different (Tymon et al. 1993).

Breeding strategies

Currently, a research programme is in place in the country, aimed at finding solutions to CSPWD. Under the breeding component, varietal resistance trials are being carried out to evaluate the performance of 36 accessions (dwarfs, tails and hybrids) against the disease (Table 1). These are being carried out on eight test plots located at Dixcove, Akwidaa, Cape Three Points, Agona Junction, Princess Town, Dadwen, Axim and Tumentu – all in the Western Region of Ghana.

Crosses of Malayan Yellow dwarfs with Panama tall variety are known to exhibit high tolerance to LY disease in Jamaica. Two observation plots of MYD × Panama tall have been laid out on farmers’ fields in diseased areas.

The disease has so far been recorded in four fields: Dixcove, Cape Three Points, Akwidaa and Agona Junction. Unlike the first two fields where many palms have been affected by the disease, few palms have been affected at Akwidaa and Agona Junction.

So far, Sri Lanka Green dwarf (SGD), Vanuatu tall (VTT) and MYD × VTT are showing some degree of tolerance. Others which have shown tolerance to a lesser degree are MRD, MRD × PYT and CRD.

Observations in some previous trials call for caution when drawing conclusions on resistance/tolerance to CSPWD. In previous experiments at Princess Town, for example, the disease first killed off all the palms except MYD, but later infected the stands of MYD leaving only a single palm which is still able to bear nuts.

Action plan

The broad objective of the breeding programme in Ghana is to breed/select coconut hybrids that are resistant to the CSPWD. The specific objectives include:

a) Exploring the possibility of introducing new varieties;
b) Establishment of pure and isolated stands for the production and extraction of pollen;
c) Evaluating the progenies from the hybridization work being carried out; and
d) Exploiting the possibility of obtaining tolerant mutants through the use of ionizing radiations.

For effective implementation of the breeding programme, a plant breeder from the Oil Palm Research Institute has been stationed in the Western Region to coordinate activities of the breeding programme. It is expected that within the next ten years, considerable progress will be made towards the production of a high yielding progeny that is resistant to the CSPWD.

Institutions involved in coconut breeding projects

1. Oil Palm Research Institute, CSIR, P.O. Box 74, Kade
2. Department of Crop Services, Ministry of Food and Agriculture, P.O. Box M-37, Accra

Funding agencies for coconut breeding

Presently, the Ghana government is the sole sponsor of coconut breeding in the country. The EC STD 3 funding does not provide for breeding although introduction for disease resistance screening is included. The France – Ghana – Côte d’Ivoire Coconut Project used to provide some funds for breeding work but this has now been terminated.

Conclusion and recommendation

Collaborative Research with other countries

CSPWD is not only a problem of Ghana but of the whole West Africa sub-region. Research findings of any one country in the sub-region is likely to benefit all the others. There is, therefore, the need for more collaboration in research among the countries in the sub-region. The most pressing need is the introduction of more germplasm for screening against CSPWD. As preliminary trial results indicated, VTT and SGD are fairly tolerant, indicating that resistant coconut cultivars could be present in the Asian countries.

Financial support is required to introduce more progenies from Asia and other regions.

Development of cloning techniques

Support by other more advance laboratories is also needed in the production of new clones for screening against CSPWD.

Replanting with promising varieties

As an interim measure, the current varieties that have shown some amount of resistance have been included in a replanting programme to maintain the coconut industry until such time that more resistant types are developed.

References

Adansi, M.A. 1970. Coconut improvement in Ghana: The present position of selection and introduction. Pp. 14-15 in Coconut in Ghana (B.J. Chona and M.A. Adansi eds.). Bul. No. 3 of the Crops Research Institute, C.S.I.R. Kumasi, Ghana.

Agble, W.K. 1970. Coconut improvement in Ghana. Pp. 1-2 in Coconut in Ghana (B.J. Chona and M.A. Adansi eds.). Bul. No. 3 of the Crops Research Institute, C.S.I.R. Kumasi, Ghana.

Arkhust, E.D. 1991. Ghana country statement. P. 27 in Proceedings of the First African Coconut Seminar, Arusha, Tanzania, 4-8 Feb 1991. BUROTROP.

Djokoto, L.K. and I.K. Dzomeku. 1990. Potential coconut growing areas in Ghana. Technical report submitted to the National Coconut Development Committee, Ministry of Agriculture, Accra, Ghana. Unpublished.

FAO. 1987. Production Year Book. Vol. 41. FAO, Rome, Italy.

Raux, J. 1990. Possibilities for industrialization of the coconut sector in the Western Region of Ghana. Report presented to the National Coconut Development Committe. (Ministry of Agriculture, Accra.) by Consultant in Agro Industries. UNIDO Div. of Industrial Investment. Unpublished.

Romney, D.H. 1972. Past studies on the present status of lethal yellowing disease of coconut. PANS. 18:386:395.

Tymon, A., P.G. Jones and S.J. Eden-Green. 1993. Detection and discrimination of mycoplasm-like organism (MLO) associated with coconut. E.E.C. STD 3 project A. 0282 Final Technical report. Submitted to the Coordinator, Lethal Yellows Disease of Coconut Project. N.R.I., Chatham Maritime, U.K. 21pp.

Wills, J.B. (ed.). 1962. Agriculture and land use in Ghana. Oxford University Press, London, Accra, New York.

Table 1. Coconut varieties on trial plots

Type	Code
Local
West African Tall	WAT
Dwarf
Malayan Yellow Dwarf	MYD
Sri Lanka Green Dwarf	SGD
Equatorial Guinea Green Dwarf	EGD
Malayan Red Dwarf	MRD
Cameron Red Dwarf	CRD
Tacunan Green Dwarf	TAC
Catigan Green Dwarf	CAT
Dwarf × Tall
Malayan Red Dwarf × West African Tall	MRD × WAT
Malayan Yellow Dwarf × Polynesian Tall	MYD × PYT
Malayan Yellow Dwarf × Malayan Tall	MYD × MLT
Malayan Yellow Dwarf × Rennel Tall	MYD × RLT
Malayan Yellow Dwarf × Vanuatu Tall	MYD × VTT
Sri Lanka Green Dwarf × West African Tall	SGD × WAT
Sri Lanka Green Dwarf × Vanuatu Tall	SGD × VTT
Equatorial Guinea Green Dwarf × West African Tall	EGD × WAT
Equatorial Guinea Green Dwarf × Vanuatu Tall	EGD × VTT
Malayan Red Dwarf × West African Tall	MRD × WAT
Malayan Red Dwarf × Polynesian Tall	MRD × PYT
Cameron Red Dwarf × Malayan Tall	CRD × MLT
Cameron Red Dwarf × Rennel Tall	CRD × RLT
Cameron Red Dwarf × Vanuatu Tall	CRD × VTT
Cameron Red Dwarf × Polynesian Tall	CRD × PYT
Cameron Red Dwarf × West African Tall	CRD × WAT
Tall × Tall
Rennel Tall × West African Tall	RLT × WAT
Rennel Tall × Rennel Tall	RLT × RLT
Malayan Tall × Malayan Tall	MLT × MLT
Polynesian Tall × Polynesian Tall	PYT × PYT
Vanuatu Tall × Malayan Tall	VTT × MLT
Vanuatu Tall × Vanuatu Tall	VTT × VTT
Vanuatu Tall × Panama Tall	VTT × PNT
Panama Tall × Panama Tall	PNT × PNT
Taganan Tall × Taganan Tall	TAG × TAG
Indian Laccadive Tall × Indian Laccadive Tall	LCT × LCT
Indian Andaman Ordinary × Indian Laccadive Tall	ADO × LCT
West African Tall (Benin) × West African Tall	WAT(B) × WAT

The Nigerian coconut breeding programme

E. E. J. Akpan
Chief Research Officer, Nigerian Institute for Oilpalm Research, Benin City, Nigeria

Introduction

The coconut is not indigenous to the country. Most of the vernacular names, where the crop is grown, point to a foreign origin; for example, Edo (Benin) name for coconut is ‘ivi-ebo’ which, when translated, means “white man’s nut”. The vernacular name for coconut in Igala is ‘omoba’, in Ibo ‘aku oyibo’, in Ezon ‘Bebe-imbi’, and in Efic/Ibibio ‘Isip mbakara’. When translated, they all mean “white man’s nut”. In Itsekiri/Urnobo, the name is ‘Kokodie’ which is the direct borrowing of the word ‘coco’. It is not known for certain when the crop was first introduced to Nigeria. However, the first plantation establishment of coconut palm in the country was in Badagry local government area of Lagos State by the Roman Catholic Missionaries in 1876. The planting material was said to have been introduced from Porto Novo (Benin Republic). More planting was done in 1885 to cover an area of approximately 320 ha (Akpan 1989). This plantation is yet to be rehabilitated.

The coconut palm is currently found in the rainforest belt (latitude 10° to the west and latitude 9° to the east) of Nigeria as a mixed crop. In addition, extensive palm groves are available in Lagos State and estimates of 59 trees/ha have been reported (Akpan 1994). Accurate data are not available in terms of hectarage devoted to coconut cultivation and actual nut production. It is estimated that 20 000 ha might have been planted to coconut mainly in smallholdings. The palms are predominantly of the tall variety. However, dwarf palms can also be found around dwelling houses, primarily, as ornamentals. Yield data are not available for coconut smallholdings and grove palms. Even in the few plantations, proper yield records have not been kept. However, it is estimated that some palms are capable of producing up to 100 nuts per palm per year. However, an estimated annual production of 50 nuts per palm might be more realistic since proper management practices are hardly applied and the planting materials had hardly undergone conscious selection for high yield.

The nuts produced are used mainly for domestic consumption, either directly as food or indirectly as oil. Although cultivation is restricted to certain climatic zones in the country, consumption is nationwide. At the cost price of N 5-N 10 (US$ 0.23 – US$ 0.46) per nut, an estimated net income per ha of N 37 000 – N 75 000 (US$ 1690 – US$ 3426) could be realized. However, data from field establishment of improved coconut palms under optimum management practices are not available. Some tall experimental palms produced up to 170 nuts per palm per year generating a maximum estimated annual income of N 225 000 (US$ 10 279) per ha. Considering a more realistic average of 80 nuts per palm per year, an estimated annual income of N 120 000 (US$ 5482) per ha can be realized. Copra export was 7.13 t in 1960 but ceased since 1975 (Isosla Idris 1978) until today because the local consumption increased and with increase in population with no appreciable improvement in production, there was no surplus of copra for export.

National replanting programme

The Federal Government first indicated its policy for the improvement of coconut cultivation in the 1975-80 National Development Plan (NDP). The targets were raising of 77 000 coconut seedlings in Lagos State for rehabilitation of palm groves and the establishment of 1200 ha of coconut plantations in Rivers State. By 1981, establishment of new plantations had been abandoned but rehabilitation of 2000 ha of coconut groves in Lagos State was planned. Unfortunately, this was not carried out. In the 5th NDP (1986-90), the Nigerian Institute for Oilpalm Research (NIFOR), under the then Ministry of Education, Science and Technology, was mandated to provide appropriate research support to the coconut industry in the country. The government realized that there is a great potential for coconut in the country, particularly in the area of generating employment, increasing food production and farmers’ income, and for providing other sources of foreign exchange. The NIFOR has sought to carry out its mandate through:

a) improvement of the genetic potential of the existing coconut germplasm;
b) improvement of agronomic/cultural practices;
c) development of effective management control for pests and diseases;
d) coconut processing and products utilization research; and
e) collecting, processing and analysis of socioeconomic data.

Apart from NIFOR’s mandate, the Federal Government is not involved in coconut cultivation. Nevertheless some state governments, e.g. Lagos State, provide incentives to farmers to rehabilitate their palm groves.

Productivity problems

The bulk of existing coconut stands are of uncertain origin with low productivity because they were derived from low yielding, unselected planting materials. There is also the aspect of declining productivity due to senility.

Rehabilitation of coconut groves is one way of increasing coconut production in the country. Establishment of new plantations by individuals, corporate bodies and farmers’ cooperatives can effectively complement rehabilitation efforts. There has been a steady and upward demand for improved variety of coconut by the different groups of planters. Apart from the commercial cultivars, demand for improved dwarf coconut seedlings, which are used mainly for ornamental purposes, has also increased.

Genetic erosion

The area of concentration of coconut palms in Nigeria is Lagos State. This state, previously host to the Federal Government Capital City, is now the seat of the commercial capital of Nigeria. Consequently, there is a tremendous pressure on land. As industries and other infrastructural facilities are being put in place, a lot of coconut palms are being cut down and the risk of genetic erosion is obvious. In addition, the Lagos State government provides incentives to farmers to replace their palm groves with improved varieties. This exercise although good in improving farmers’ incomes, leads to further depletion in genetic variability. The situation is similar in other parts of the country where increasing urbanization leads to decreasing farm land and a steady loss in coconut production.

Breeding strategies

Germplasm resources

The tall variety has been used for commercial production because of its longevity and superior nut components, particularly the copra content. The main disadvantage of the tall is its long gestation period ranging from 7 to 10 years. This is the reason why it is currently being replaced by the F1 hybrids involving tall × dwarf and dwarf × tall crosses in most coconut growing countries in the world. The hybrids combine early fruiting with economic nut characteristics (Satyabalan et al. 1970; Fremond and de Nuce de Lamothe 1972; Manthriratna 1971; and Liyanage 1972). The first experimental hybrids between the Nigerian Tall (NT) and the Nigerian Green Dwarf (NGD) were planted at NIFOR Main Station near Benin City in 1970. The parent population consisted of few palms – 6 NT, 3 NGD, the crosses of which produced 47 F1 hybrid palms. They were planted in unreplicated progeny rows in a field known to be quite sandy with low organic matter content and high water deficit during the dry season. The evaluation revealed that although optimum yield potential was not realized because of adverse environmental conditions, the hybrids nevertheless out-yielded (Table 1) the tall and the dwarf parents (Akpan 1985).

Two hybrid seedgardens have been established, one in Badagry (8 ha) and in the other in the Main Station (10 ha), near the city. The parental materials for the seedgardens were mainly NIFOR’s tall palms whose average nut production ranged from 80 to 170 nuts per palm per year. However, because of a limited number of local superior dwarf cultivars, seedlings were grown from Malaysian, Cameroon and Nigerian dwarf cultivars for hybrid production. The dwarf and tall seedlings were interplanted in the proportion of two rows of dwarf palms to one row of tall palms. Emasculation of dwarf palms led to the production of dwarf × tall hybrids by natural cross-pollination. Hence, the current NIFOR hybrid seeds could well be described as a composite of the three dwarf cultivars with the Nigerian Tall. Other hybrid combinations could be produced in the future.

The dwarf materials in the germplasm bank, consisting of indigenous and introduced genotypes, have been evaluated in terms of yield and other desirable characteristics (Akpan 1989). There was no distinct advantage of indigenous dwarfs, considered to have acquired agroclimatic adaptation, over the exotic ones suggesting that both indigenous and introduced cultivars could be used for direct cultivation and/or for hybrid seed production. At present, a 3-ha seedgarden has been established for the production of superior seedlings of Malayan Yellow Dwarf for distribution to coconut growers.

Demand for improved tall variety of coconut is often minimal and it is envisaged that this pattern is not likely to change in the near future. Therefore, establishment of a seedgarden for large-scale production of improved tall seeds is not considered necessary. Demand for this variety would continue to be met from selected number of existing tall palms.

Breeding objectives

The primary objective of coconut breeding in Nigeria is to be able to provide early maturing, high yielding, disease resistant planting materials to coconut growers. Specifically it involves:

a) establishment of a germplasm bank;

b) study of the floral biology of some of the materials in the genebank which is relevant to the development of a viable breeding scheme;

c) evaluation of the materials in the genebank with respect to yield and other desirable characteristics;

d) nut component studies which have a bearing on the commercial products;

e) identification of superior genotypes of tails and dwarfs;

f) progeny evaluation of selected materials in a well laid-out replicated trials;

g) production of hybrids from selected parent palms;

h) establishment of hybrid seedgardens with elite selected palms to provide superior planting materials; and

i) evaluation of promising hybrids at different locations.

Nigerian coconut germplasm bank

Although the coconut palms are not indigenous to Nigeria (Akpan 1989), they have been grown for a sufficiently long time in the country to have acquired agroclimatic adaptation. These adapted ecotypes are the ones being referred to as indigenous genotypes. They are indispensable for the breeding of superior and adapted varieties. Nevertheless, the amount of genetic variability available to the breeder from indigenous resources is limited and therefore, introduction of exotic materials becomes necessary. The current germplasm collection is shown in Table 2.

Germplasm utilization

The introduced tails from India have been inter-crossed in order to increase the population size and release possible genetic variability. The same exotic population has been inter-crossed with a selected number of local tall palms. A number of high yielding local tall palms are being used in the establishment of hybrid seedgardens and for the production of different hybrid combinations for evaluation.

Future priorities for collecting, conservation and utilization of germplasm

The Nigerian coconut germplasm is very modest but efforts to enlarge the genebank through local collections will be continued. In future germplasm sourcing, attention will be paid to:

a) Lagos State where there is the largest concentration of coconut in Nigeria;

b) areas where palm populations of earlier collections are so low that realistic assessment is difficult; and

c) relatively drier ecological zones of the country where coconut is known to grow. In Nigeria, there are two distinct seasons: the dry season and the rainy season, each lasting for approximately six months. During the dry season, growth and productivity of coconut is adversely affected because of water stress. Materials from the drier ecological zones will be quite useful in breeding for drought tolerance. It is likely that such ecotypes might have developed some adaptation in such water-conserving structures as the thickness of epidermis and cuticle, stomatal distribution and/or stomatal aperture, root density, etc.

Exotic introductions

Earlier, it has been mentioned that the coconut palm is not indigenous to Nigeria and that the amount of genetic variability available from local resources is limited. It is also known that a germplasm bank representing wide genetic diversity is preferable to that with a narrow one because the former can easily be manipulated to achieve breeding objectives. There is, therefore, a compelling need for the introduction of exotic materials. Introductions will be made from such African countries as Côte d’Ivoire, Ghana and Tanzania, and from Asia and the Pacific countries, such as Philippines, Indonesia, Solomon Island, Papua New Guinea, Fiji and Tonga. However, this will be subject to approval by the Nigerian Plant Quarantine Services which has a comprehensive index of coconut diseases and pests worldwide. The genebank will continue to be maintained in the form of field plantings. Acquisition of new collections with sufficient materials to allow replicated trials for comparison with a known local cultivar will be made.

National institutions involved in coconut breeding

NIFOR collaborates with universities in the general areas of plant breeding for mutual exchange of ideas/information although in the specific area of coconut breeding, no other institution in the country is involved. NIFOR also offers practical training to undergraduates studying agriculture and related subjects in some Nigerian Universities.

Collaborative breeding research with other countries

Coconut breeding in Nigeria is relatively in its infancy. There is a need to actively collaborate with other countries particularly, in the area of germplasm collecting, characterization, conservation and utilization. Training, tour/visits to advance coconut breeding laboratories in other countries will be of an immeasurable advantage. So far, germplasm conservation has been in the form of field plantings (ex situ) and maintenance is inadequate because of lack of funds. Training in cryopreservation and embryo culture will be of immediate interest.

Coconut breeding action plan and expected output for the next ten years

a) germplasm collecting, evaluation and conservation

During the period there will be an effective coverage of all the coconut growing areas of the country and with introductions from some parts of Africa, Asia and the Pacific, the germplasm will be sufficiently enlarged. Germplasm will continue to be preserved by field planting but attempts will be made to acquire the latest techniques in in vitro collecting, characterization and conservation by attending workshops. Yield evaluation will include nut component analysis.

b) production and evaluation of coconut hybrids

It has been mentioned that the current NIFOR hybrid seedlings are produced from a Nigerian tall and a mixture of Nigerian Dwarf, Cameroon Dwarf and Malaysian Dwarf. Being the hybrid material currently available, its performance is to be evaluated under optimum management practices at different locations. Other hybrid combinations whose comparative performance is to be assessed are:

· indigenous tall × indigenous dwarf
· indigenous dwarf × indigenous tall
· indigenous tall × exotic dwarf
· exotic dwarf × indigenous tall

From the evaluation results, it will be possible to produce the best hybrid(s) for distribution to farmers. Steps will also be undertaken to introduce the PB 121 hybrid from Côte d’Ivoire for comparative trials with the locally produced hybrids.

c) combining ability studies

This will involve general and specific combining ability tests for the desired traits of promising parent populations for the following crosses:

· local tall × local tall
· local tall × exotic tall and its reciprocal
· local dwarf × local dwarf
· local dwarf × exotic dwarf and its reciprocal

In addition to yield and yield component and combining ability tests, screening of the potential parent materials for resistance to prevalent diseases/pests and drought will be done.

d) establishment of a 60-ha hybrid seedgarden.

Funding agency for coconut breeding projects

At present, the Federal Government allocates funds annually for coconut breeding projects through NIFOR. The Nigerian Government is the sole source of financial support and with so many competing interests for the limited financial resources of the country, research funding appears to have a relatively low priority. International agencies are yet to fund a coconut breeding research in Nigeria.

Conclusion and recommendation

The coconut palm is an important crop in the Nigerian economy, providing food and income to rural and some urban families who grow the crop around their houses. The current exotic cultivars are well adapted to the local environment, suggesting that superior genotypes of both local and exotic materials could be released for direct cultivation and hybrid seed production. Commercial coconut production could prove to be a profitable venture even at the present low domestic price levels. Moreover, domestic demand for the crop is expanding. There are also opportunities for export of many coconut products, hence, the prospects for coconut cultivation in Nigeria are bright.

Steps are being taken to produce good quality and adapted planting materials for coconut growers. However, there should be sufficient encouragement in breeding research by way of providing appropriate financial and logistical support to meet the increasing demand for improved coconut planting materials.

Acknowledgements

I would like to thank the members of the Steering Committee of the Coconut Genetic Resources Network (COGENT) not only for their invitation to attend the workshop on the Standardization of Coconut Breeding Research Techniques but also for undertaking to bear the financial cost of my attendance.

My thanks also go to the Director of NIFOR for permission to attend the workshop.

References

Akpan, E. E. J. 1985. Coconut breeding in NIFOR-Past, present and future. Seminar paper presented at NIFOR, 17^th May 1985.

Akpan, E. E. J. 1989. Evaluation of dwarf coconut genotypes within the Nigerian coconut germplasm bank. Paper presented at International Conference on Palms and Palm Products, 21-25 Nov. 1989, Benin City, Nigeria.

Akpan, E. E. J. 1994. Evaluation of tall coconut genotypes within the Nigerian coconut germplasm bank. Oléagineux 49(1):13-20.

Fremond, Y. and de Nuce de Lamothe, M. 1972. Characteristics and production of the hybrid coconut palm Malayan Dwarf × West African Tall. Pp.309-319 in Proceedings of a Conference on Cocoa and Coconut, Kuala Lumpur, Malaysia.

Isola Idris, A.D. 1978. Copra marketing – Ministry of Trade Industry And Cooperatives, Ikeja, Lagos State, Nigeria.

Liyanage, D. V. 1972. Production of improved coconut by hybridization. Oléagineux 27(12):597-599.

Manthriratna, M. A. P. 1971. Some results of field experimentation with typica × Nana F₁ hybrid. Ceylon Coconut Quarterly 22(3/4):107-113.

National Development Plan. 1975-80. Federal Republic of Nigeria. Pp. 76-79.

National Development Plan. 1981-85. Federal Republic of Nigeria. P. 92.

National Development Plan. 1986-90. Federal Republic of Nigeria. P. 86.

Satyabalan, K., T. C. Ratnam, and P. V. Kunjan. 1970. Hybrid vigour in nut and copra characters of coconut hybrids. Ind. Journ. Agric. Sci. 40:12.

Table 1. Nigerian Tall × Nigerian Green Dwarf (F₁ experimental hybrids)

Type	Flowering time (months)	No. of nut/palm		Weight of copra per nut (gm)
		Range	Mean
Tall	78.0	1-29	8.2	168
Dwarf	40.2	3-94	37.3	89
Hybrid	52.3	5-105	44.9	102

Table 2. NIFOR germplasm collection

Collection	Origin	No. of trees
FOREIGN
Cameroon Green Dwarf	Cameroon	8+
Malayan Green Dwarf	Malaysia	25+
Malayan Yellow Dwarf	Malaysia	19+
Malayan Red Dwarf	Malaysia	6+
Nylor Gading	India	2+
Bengal Tall	India	3+
Gon Thembili	India	3+
Mysore Tall	India	3+
Gangabondam	India	3+
Spicata	India	3+
Navasi	India	2
Bombay Tall	India	3+
West Coast Tall	India	3+
Regia	Sri Lanka	16
Pumila	Sri Lanka	16
Eburnea	Sri Lanka	16
TOTAL -16		131
LOCAL
Evboneka Tall		95
NIFOR Tall		175*+
Iguedaikan Tall		82*
Obazuwa Tall		37*
Okunuvbe Tall		43*
Ekiadolor Tall		94*
Ugbogiobo Tall		63*
Uwan Tall		90*
Moor Plantation Tall		14*+
lyowa Tall		156
Utekon Tall		117*
Eko-Abetu Tall		27*
Aihuobabekun Tall		120*
Koko Tall		130*
Opobo Tall		69*
Badgry Tall		232*+
Ekenobore Tall		98
Uyo Tall		17
Abak Tall		36
Ikot Ekpene Tall		26
Degema Tall		23
Eket Tall		72
Igwikhinmwin Tall		42
Oyun Tall		23
llorin Tall		20
Irepodun Tall		18
Kogi Tall		16
Okene Tall		175
Umuahia Green Dwarf		18
Abakaliki Red Dwarf		6
Abak Green Dwarf		33
Uwan Green Dwarf		21
Ughelli Red Dwarf		19
TOTAL – 33		2397

* collections already evaluated
+ collections utilized

Note: These materials were mostly established in the field as unreplicated progeny rows.

Coconut breeding research in Tanzania

K.E. Mkumbo and A. Kullaya
Coconut Breeders, Mikocheni Agricultural Research Institute, Dar es Salaam, Tanzania

Introduction

The need to promote the coconut industry in Tanzania has been more felt during recent years, when diminishing production trends became evident. Before this, Tanzania used to export copra to the international market, but the business was curtailed by certain production constraints. In addition, domestic demand for this commodity has been high and could not be met by the prevailing production levels. Coconut remains the most important source of vegetable oil, accounting for about 40% of the total vegetable oil consumption in Tanzania (Balingasa 1990).

Coconut production and farm productivity

In Tanzania, coconut is cultivated by more than 300 000 small-scale farmers, accounting for about 95% of the production. According to recent estimates there are more than 25 million palms occupying an estimated area of over 252 000 hectares giving an average of 0.8 ha per farming family.

According to a recent study, the coconut yields are estimated at an average of 38-40 nuts/palm/year (Ashimogo et al. 1996) compared with 24 nuts/palm/year estimated in 1984 (Krain and Kabonge 1992; Balingasa 1990). This increase is largely attributed to the fact that farmers have started to adopt improved crop husbandry practices like weeding and proper intercropping. The availability of water appears to have a significant influence on yield, given the varied nature of rainfall intensity and distribution along the coastal belt. A recent survey conducted (Bellin 1994) shows that, under rainfed conditions, palms bear 50-60 nuts/palm/year; with some individual palms having as much as 80 nuts per year. In valleys and in areas where ground water is available throughout the year, palms bear over 100 nuts/palm/year; with some individual palms in this group producing up to 200 nuts per annum.

Productivity problems

Growing conditions for the coconut palms in Tanzania is characterized by sub-optimal rainfall and erratic distribution pattern. The mean yearly precipitation ranges from 600-1900 mm. The presence of lethal disease (LD) and serious pests like Pseudotheraptus wayi, all imposed tremendous production constraint. The lack of quality planting materials, inadequate extension services and low level of crop management standards further aggravated the problem (Balingasa 1990). Aware of the various constraints on coconut production, the National Coconut Development Programme (NCDP), which is now being implemented by the Mikocheni Agricultural Research Institute (MARI). NCDP was established in 1979/80 with the overall responsibility to promote the productivity of coconut industry in Tanzania through various research and development activities. A Selection, Multiplication and Breeding Section within NCDP was charged with the responsibility of breeding for high yielding varieties, adapted to the local environmental conditions and resistant to prevailing diseases.

National coconut germplasm and conservation status

The success of a crop improvement programme is ensured by its ready access to maximum genetic variability. In view of this, a range of genetic material has been introduced in the country to establish seed farms and test for disease resistance and agronomic requirements (Table 1). The list includes 12 local dwarfs, 19 local tails and 41 introduced germplasm. Aside from imported germplasm, 30 EAT sub-populations and three dwarfs were collected locally. Both the introduced and locally collected genetic materials are in different sites and are being evaluated with respect to vegetative growth, yield, fruit component characteristics and LD resistance.

Types of coconut grown

Although the list of coconut germplasm under MARI is long, it is not the case in farmers’ fields. The predominant and in some cases exclusively tall variety grown in Tanzania is local East African Tall (EAT). It is widely cultivated along the whole coastal belt, the islands and even inland where the climate and edaphic factors allow.

The common dwarf variety in farmers’ field is the Pemba Red Dwarf (PRD). It is more prevalent in the Isles (Unguja and Pemba). On the mainland, this dwarf is mainly cultivated for its sweet juice from immature fruits and for ornamental purposes. It is occasionally used as a source of vegetable oil.

Hybrids known to be grown by the coconut farmers are the CAMWA (Cameroon Red Dwarf × West African Tall) and MAWA (Malayan Yellow Dwarf × West African Tall). They were widely accepted by farmers in their first years of bearing, but due to disappointing low yields, and susceptibility to LD and drought, seed production and planting of these hybrids was stopped in 1990/91.

Performance of coconut hybrids and cultivars

Under optimum cultural management and high soil water table, the performance of MAWA and CAMWA hybrids is substantially superior to the local tall (EAT) and dwarf (PRD). However under average and poor management (a case with Tanzanian small holders) and adverse soil and climatic conditions, EAT is a better choice.

Genetic erosion

Coconut genetic resources may be lost through genetic erosion if steps are not taken to conserve the available material, either in situ or ex situ. The major factors contributing to genetic erosion are:

· Drought: There are palm losses in several areas of the coconut growing belt, because of wilting, as a result of prolonged dry spells.

· Lethal disease (LD): Presence of LD caused by phytoplasma has been recorded in Tanzania, Kenya and Mozambique. LD is the major contributor to genetic erosion in Tanzania. From 1960s to 1991, the overall palm losses due to LD in Tanzania is approximately 8 million palms which is 38% of the total palm population (Schuilling 1992).

· Urbanization and old age of palms: The high rate of urbanization in places like Zanzibar, Tanga, Bagamoyo to mention a few, and the old age of original palms (most of which could be more than 120 years) which constitute the genetic base, has accelerated the genetic erosion.

Future priorities for germplasm collecting, conservation and utilization

The high susceptibility of the imported germplasm to LD and the disappointing agronomic performance of imported hybrids makes the search for suitable genes in the local EAT a first priority. It should be mentioned here, that at the onset of the project in the late 70s much emphasis was put on the introduced germplasm, and this delusion has been noted with great concern by the public. In response to this, efforts are being concentrated on the improvement of the local germplasm, through selection and cross breeding programme (NCDP Ann Rep 1989/90).

Most of the germplasm will be collected from LD infected farms and only few from the disease-free areas. The collected materials will then be screened under LD conditions and the promising populations will be utilized and conserved as base/active collection in the field genebanks.

Breeding Objectives

The breeding objectives are problem-specific, prioritized into primary and secondary objectives in relation to the more pressing needs. They are summarized as follows:

Primary:

· Resistance to lethal disease (LD)
· Tolerance to drought stress
· Productivity under minimum input cultivation
· Responsiveness to improved cultural practices

Secondary:

· Precocity
· Good root/shoot ratio
· High harvest index
· High oil content
· Long economic life span
· Resistance to insect pests
· Resistance to minor diseases

The Breeding Strategy

The long-term strategy to achieve the aforementioned objectives has been to screen different genotypes against lethal disease, to use the survivor palms in cross breeding and to test the combining ability of the resulting progenies in terms of yield, resistance to LD and tolerance to drought stress. The occurrence of LD makes it necessary to screen all the germplasm materials before using them for breeding. This is an expensive and time consuming exercise (Kullaya et al. 1993). In view of this, MARI is collaborating with other institutions in:

· Developing molecular biology techniques to identify genetic markers for LD resistance and drought tolerance.

· Using DNA probes and primers to detect at an early age the presence of phytoplasmas (formerly referred to as mycoplasma-like organisms) in in vitro plantlets.

The success of these techniques is expected to complement significantly the coconut breeding programme.

Suggestions for collaborative breeding research with other countries

The MARI will continue to acquire coconut genetic materials from other countries. However, emphasis shall be on germplasm likely to be resistant and/or tolerant to LD and drought. For instance, it would be interesting to import germplasm that have been reported to resist similar MLO-diseases elsewhere, but the Plant Protection Department regulations which have put the coconuts materials under prohibited import restriction, makes it difficult to acquire coconut germplasm from other countries (Annex 1). The problem is further exacerbated by restrictions imposed by the exporting countries. However, this problem can be overcome if the FAO/IPGRI guidelines on safe movement of coconut germplasm can be implemented at international level. In anticipation to future germplasm exchange with other coconut growing countries, the MARI decided to establish an embryo culture laboratory.

In the light of these constraints, the following collaborative research with other countries/organizations are suggested:

1. In Tanzania the more pressing problems are lethal disease, drought stress, serious pest incidence and poor farm management practices. However, it is most probable that these are also common problems in other coconut growing countries. It is also true that other countries have problems which affect the productivity of their coconut groves different from those that occur in Tanzania. Nevertheless, they do not occur in isolation. Collaborative research at regional and international level should be able to cut short time and resource investment by avoiding duplication of research activities.

2. Use of molecular techniques (biotechnology) in the following areas:

· Screening for drought tolerance
· Diagnosis of phytoplasmas
· Development of molecular markers for useful agronomic traits.

The NCDP is part of the research network titled ‘Improvement of coconut by biotechnology: Application to breeding and crop protection’ that collaborates three other international institutions: Max Planck Institute for Plant Breeding (MPIZ), Cologne, Germany, the Agricultural Research Centre, CIMA, in Vitoria, Spain and Albay Research Centre in the Philippines. The objectives of this network are to develop molecular biology techniques for:

· Genome analysis of the coconut palm with the aim of studying and assessing the genetic diversity within and between different coconut varieties/populations using polymorphic markers [Restriction Fragment Length Polymorphism (RFLP), Randomly Amplified Polymorphic DNA (RAPD) and Inverse Sequence Tagged Repeats (ISTR)].

· Construction of a genetic linkage map of coconut using RFLP, RAPD and ISTR markers

· Development of non-radioactive and highly sensitive techniques for indexing of germplasm for important disease pathogens like phytoplasmas, formerly referred to as MLO and Cadang Cadang Coconut Viroid (CCCVd).

3. Research on culture and cryopreservation of zygotic coconut embryos and clonal propagation of coconut palms.

The NCDP (MARI), virtually has all the basic and necessary facilities required for tissue culture and molecular biology. What is needed is the identification of specific areas of research that can be carried out jointly with other countries. The Tanzanian government, through the NCDP (MARI), will be in a position to provide land, facilities, materials and staff within limits for collaboration or joint projects with other interested countries.

Coconut breeding action plan and expected output in the next 10 years

In view of the results so far obtained from the introduced germplasm, it is apparent that the local EAT is, and will continue to be, the predominant coconut population in Tanzania. Despite its low yield potential, EAT is better adapted to the prevailing conditions than most of the introduced genetic materials. Research efforts will then be directed on improving the local EAT through selection and cross-breeding with promising parent materials or genitors. Selection will be done simultaneously with the performance evaluation of progenies from the cross-breeding programme. The selection within existing EAT seed sources or seed gardens, will be open pollinated, selfed or sibbed. Selected hybrids from tall × tall or tall × dwarf crosses will be compared with the EAT selections. The ultimate goal is to release an improved variety in terms of LD resistance and drought tolerance by the year 2003. The long term coconut breeding plan for Tanzania is summarized in Table 2.

Conclusion and recommendation

Morphological and vegetative traits used in the coconut characterization and evaluation schemes do not necessarily express the genetic diversity in the field genebanks. Similarly, large field collections do not necessarily maximize variability in the populations. It is therefore, imperative to use complementary molecular techniques which provide direct genetic information (regardless of the environment) of the variables in question. Use of these techniques could involve RFLP, RAPD, AFLP, ISTR, microsatelites, etc. These techniques should be emphasized in determining and understanding the variation existing in the country’s coconut genetic resources.

Coconut germplasm conservation must be accomplished by multi-disciplinary and systematic evaluation, full documentation and effective information exchange among different disciplines, so as to enhance the usefulness of large collections and to justify long term investment.

References

Ashimogo G., Evelyne L., Henry M. and Dismas M. 1996. National Coconut Development Programme. Coconut Impact Assesment Survey. Final Report. June 1996.

Balingasa E. N. 1990. Coconut Seed Multiplication and Breeding Report. National Coconut Development Programme, Tanzania.

Bellin H. 1994. Yield and Yield Potential of the EAT Coconut Palm in Different Agro-ecological Zones in the Coastal Belt of Tanzania. Pp. 159-172 in National Coconut Development Programme, Annual Report, July 1993 – June 1994.

Krain E and Kabonge P.M.D. 1992. How to Grow Coconuts. Dar-es-Salaam University Press, Dar-es-Salaam, Tanzania.

Kullaya A., Harries H.C., Mkumbo K.E. and Masumbuko L.I. 1993. Coconut Breeding Research Programme in Tanzania.

National Coconut Development Programme, Annual Report, 1989. 90 pp.

Schuiling M. 1992. Final Report of GTZ of Plant Pathologist. Research on Lethal Disease from January 1980 – June 1992.

Table 1. List of local (bold) and introduced (unbolded) coconut genetic materials of MARI

No	International cultivar name	International code		Local code		First accession no.	Planting date	Number of alive and legitimate trees
1	Cambodia Tuk Sap Tall	KATO2		CBT		NCDP-T8	1987	230
2	Cameroon Red Dwarf	CRD		CRD		NCDP-D6	1981	4000
3	Cameroon Red Dwarf					NCDP-D6 R1	1992	125
4	Catigan Green Dwarf	CATD		CAT		NCDP-D10	1990	140
5	East African Tall Bagamoyo	EAT18		EATBAG		NCDP-T1N BAG	1988	110
6	East African Tall Boma	EAT25		EATBOM		NCDP-T1U BOM	1988	150
7	East African Tall Boza	EAT14		EATBOZ		NCDP-TU BOZ	1988	128
8	East African Tall Chambezi	EAT02		EATCHA		NCDP-T1B CHA	1981	277
9	East African Tall Faza	EAT13		EATFAZ		NCDP-T1I FAZ	1992	90
10	East African Tall Jambiani	EAT04		EATJAM		NCDP-T1D JAM	1985	125
11	East African Tall Kiembesamaki	EAT05		EATKIE		NCDP-T1D KIE	1986	77
12	East African Tall Kifumangao	EAT03		EATKIF		NCDP-T1C KIF	1985	100
13	East African Tall Kimanga	EAT15		EATKIM		NCDP-T1K KIM	1988	122
14	East African Tall Kinowe	EAT08		EATKIN		NCDP-T1E KIN	1986	48
15	East African Tall Kiombamvua	EAT06		EATKIO		NCDP-T1D KIO	1985	125
16	East African Tall LBS	EAT12		EATLBS		NCDP-T1H LBS	1980	70
17	East African Tall LBS			EATLBS			1993	70
18	East African Tall Lamu	EAT29		EATLAM		NCDP-T1Y LAM	1990	150
19	East African Tall Mafia	EAT01		EATMAF		NCDP-T1A MFA	1984	2000
20	East African Tall Mazizini	EAT07		EATMAZ		NCDP-T1D MAZ	1986	360
21	East African Tall Mchukwi	EAT17		EATMCH		NCDP-T1M MCH	1988	100
22	East African Tall Mikindani	EAT21		EATMIK		NCDP-T1Q MIK	1988	150
23	East African Tall Msambweni	EAT27		EATBWE		NCDP-T1W BWE	1992	158
24	East African Tall Msanga Mkuu	EAT22		EATMSA		NCDP-T1R MSA	1989	120
25	East African Tall Msuka	EAT09		EATMSU		NCDP-T1E MSU	1986	61
26	East African Tall Mtoni	EAT28		EATMTO		NCDP-T1X MTO	1989	150
27	East African Tall Mwambani	EAT23		EATMWA		NCDP-T1S MWA	1988	200
28	East African Tall Ng’apa	EAT20		EATNGP		NCDP-T1P NGP	1988	100
29	East African Tall Pehehe	EAT11		EATPEH		NCDP-T1G PEH	1987	54
30	East African Tall Singino	EAT19		EATKLW		NCDP-T1O KLW	1988	80
31	East African Tall Songomnara	EAT30		EATSMR		NCDP-T1Z SMR	1990	150
32	East African Tall Songosongo	EAT16		EATSON		NCDP-T1L SON	1992	169
33	East African Tall Tumaini	EAT24		EATTUM		NCDP-T1T TUM	1989	120
34	East African Tall Vuo	EAT26		EATVUO		NCDP-T1V VUO	1990	100
35	East African Tall Yellow	EAT10		EATYEL		NCDP-T1F YEL	1986	66
36	Equatorial Guinea Green Dwarf	EGD		BGD		NCDP-D7	1981	900
37	Equatorial Guinea Green Dwarf	EGD		BGD		NCDP-D7 R1	1988
38	Karkar Tall	KKT		KKT		NCDP-T16	1989	180
39	King Coconut	RTB		KC		NCDP-T11	1988	43
40	Laccadives Micro Tall	LMT		LACO		NCDP-T14	1990	150
41	Laccadive Ordinary Tall	LCT		LACM		NCDP-T13	1990	150
42	Madang Brown Dwarf	MBD		PGBD		NCDP-D11	1990	150
43	Malayan Green Dwarf	MGD		MGD MLI		NDP-D4 MLI	1972	45
44	Malayan Green Dwarf	MGD				NCDP-D4 M R1	1991	156
45	Malayan Green Dwarf	MGD		MGD ZNZ		NCDP-D4 ZNZ	1985	30
46	Malayan Red Dwarf	MRD				NCDP-D5	1981	4000
47	Malayan Red Dwarf	MRD				NCDP-D5 R1	1991	118
48	Malayan Tall	MLT				NCDP-T7	1983	250
49	Malayan Tall	MLT				NCDP-T7 R1	1995
50	Malayan Yellow Dwarf	MYD		MYD MLI		NCDP-D2	1972	9
51	Malayan Yellow Dwarf	MYD				NCDP-D3	1981	6400
52	Malayan Yellow Dwarf	MYD			NCDP-D3 R1		1992
53	Niu Leka Dwarf	NLAD	NLGD		NCDP-D12		1990	150
54	Panama Aguadulce Tall	PNT01	PNT		NCDP-T10 PN1		1989	78
55	Panama Monagre Tall	PNT02	PNT		NCDP-T10 PN2		1989	78
56	Pemba Red Dwarf (Tanga)	PRD01	PRDTAN		NCDP-D1 M R1		1992
57	Pemba Red Dwarf		PRD MLI		NCDP-D1 MLI		1972	54
58	Pemba Red Dwarf (Zanzibar)	PRD02	PRD ZNZ		NCDP-D1 Z R1		1992
59	Pemba Red Dwarf		PRD ZNZ		NCDP-D1 ZNZ		1983	390
60	Pumilla Green Dwarf	PGD	SGD		NCDP-D8		1983	200
61	Pumilla Green Dwarf		SGD		NCDP-D8 R1		1988
62	Rennell Island Tall	RIT	RLT		NCDP-T6		1983	300
63	Rennell Island Tall		RLT		NCDP-T6 R1		1995
64	Rotuman Tall	RTMT	ROT		NCDP-T15		1989	185
65	Tacunan Green Dwarf	TACD	TCN		NCDP-D9		1990	140
66	Tagnanan Tall	TAGT	TGT		NCDP-T12		1989	270
67	Tahitian Tall	TAT	PYT		NCDP-T5		1983	1100
68	Thailand Tall	THT	NCDP-T9		1984		240
69	Vanuatu Tall	VTT	NHT		NCDP-T4		1982	142
70	Vanuatu Tall		NHT		NCDP-T4 R1		1995
71	West African Tall	WAT	WAT		NCDP-T2 OP		1981	1250
72	West African Tall	WAT	WAT		NCDP-T3 HP		1981	250

Table 2. Activity schedule for long-term EAT improvement in Tanzania

Annex 1. Requirements for entry, quarantine and use of coconut materials in Tanzania

The following are the quarantine regulations with regard to entry of coconut genetic materials into Tanzania:

· Prohibited planting materials from all countries including vegetative materials as cutting, flowers, foliage and all tissues.

· Ungerminated nuts in husk are not allowed to enter Tanzania from the following countries: Indonesia, Madagascar, New World, Seychelles, West Indies, Pacific Islands, the Philippines and all countries where the following occurs: Melitoma insularie (Fair), Diocalandra taitensis (Green) and Rhadinaphelenchus cocophilus (Cobb).