Pacific

National coconut breeding programme in Papua New Guinea
National coconut breeding programme in Fiji
Vanuatu national coconut breeding programme

National coconut breeding programme in Papua New Guinea

Mathias Faure and John Moxon
Research Officer and Institute Director, PNG Cocoa and Coconut Research Institute, Rabaul, Papua New Guinea

Introduction

Papua New Guinea (PNG) has a land area of about 460 000 sq km and a population of about 3 780 000. Approximately 70% of the land area are unsuitable for agricultural development owing to topography and soil quality. The bulk of the population is located in rural areas, and about 85% of the people obtain their livelihood directly from farming or as paid agricultural labourers (Yarbro 1992).

The land and population distribution by province are presented in Table 1. Coconut is grown in most of the coastal regions of the mainland and islands. The major coconut producing regions of PNG are located on the northern coast of the country's mainland, stretching from West Sepik on the Indonesian boarder down to Milne Bay, and on the Island Provinces of Bouganville, New Britain, New Ireland and Manus. Population densities in the coconut growing provinces are generally low ranging from about one to sixteen persons per square kilometre (Yarbro 1992). The Gazelle Peninsular of East New Britain, which is a major coconut area, is an exception.

Research and development

Coconut research is entrusted to the PNG Cocoa and Coconut Research Institute (CCRI). The headquarters is based at Keravat near Rabaul and the major coconut research station in Madang. The CCRI is a private company registered under the Companies Act. It is jointly owned by the Cocoa Board and the Copra Marketing Board, both of which are government statutory bodies making CCRI also a national institution. Coconut research focuses on breeding, germplasm collection, embryo culture, agronomy, farming systems and crop protection (beetle control). A recent initiative is planned to include 'downstream processing'. CCRI research is funded from annual levies on export of cocoa and copra, by the PNG government, the European Union STABEX, the international EU INCO project, COGENT and CIRAD-CP.

Some previous coconut researches were funded by the Australian Centre for International Agricultural Research (ACIAR) under the general title "Coconut Improvement in the South Pacific". Several papers and reports on various aspects of coconut research were presented during the workshop held in Taveuni in 1993 to mark the end of the ACIAR sponsored research. The proceedings on Coconut Improvement in the South Pacific was published in ACIAR Proceedings No. 53 (Foale and Lynch 1993).

In 1988, the CCRI purchased a 1000 hectare property in the Bouganville Province for the coconut research and development programme. Bouganville produced about 40% of PNG copra exports. Infrastructural development had been completed and the research programme initiated when civil problems began. All work there has been suspended to date. Copra production fell to zero levels.

In 1993, the CCRI purchased a plantation of 500 hectares near Madang, on the mainland of the country, principally for coconut research and also for servicing the cocoa and coconut industries requirements on the mainland. The new research station offices, laboratories, houses and workshops were funded primarily by the Copra Marketing Board. The major land clearing was completed in December 1993. About one third of the land has been planted to research trials and collections. This plantation is called the Stewart Research Station (SRS).

Madang is the site for the regional germplasm repository for the Pacific Region. The inputs from regional and international bodies such as CIRAD-CP, ACIAR, Production and Dissemination of Improved Coconut Cultivars (PDICC) and COGENT will assist the CCRI in addressing the immediate issues and long term development of the coconut industry.

Coconut production and hectarage

The area under coconut cultivation is estimated at about 265 000 ha, making PNG the main coconut grower in the Pacific Region (Yarbro 1992). Coconut occupied about 6% of the total PNG land area (460 000 sq km). The proportion would be higher if areas either suitable for, or under cultivation in every coconut growing province were considered. Total hectarage under coconut cultivation may have changed in recent years due to introductions and acceptance of new crops or of other development replacing coconuts. Thus, up-to-date survey is needed to keep track of the land use under coconut cultivation to help planners in the coconut industry.

There is very little information on the extent of coconut cultivation in PNG. Nevertheless, according to surveys conducted by the Department of Agriculture and Livestock (DAL) in the early 1980's, smallholders (with less than 10 ha) in the major coconut growing provinces had 60 381 ha of monoculture coconut and 41 008 ha of intercropped coconut with cocoa (Table 2). About 5% of the coconut stand in eight provinces were senile, 60% were mature and bearing and 35% were immature. Yarbro (1993) reported further that in East New Britain, East Sepik, Madang and Oro, there were 2% senile, 73% mature and bearing and 26% immature coconut palms.

Largeholders (with 10 ha or more) maintained a declining coconut area of 94 000 ha in 1981 to about 75 000 ha in 1987 as shown in Table 3 (Yarbro 1993). Approximately, at least half of this area was intercropped with cocoa. The reasons for the reduction in planted area were replacement with sole cocoa, oil palm and other crops due to low copra prices and high production costs. It was also extremely dangerous to establish new plantings in many areas, particularly Islands Region, due to beetle pests.

Farm-level productivity

Papua New Guinea's major coconut products are fresh or dry nuts, copra, coconut oils and copra meal. Erosion control coir matting, soap and refined cosmetic oil are produced in small quantities. The palms also provide building materials, fuel, brooms, baskets and mats. The nuts are mainly for domestic consumption apart from small quantities of fresh nuts exported to Australia from time to time. Up-to-date information on coconut production is scarce or at least has not been published. Hence, this report covers only copra production in smallholders and largeholders.

Copra production is concentrated in the Islands Region of the East New Britain, West New Britain, Bougainville and New Ireland provinces which make up to 73% of PNG production. Madang is the major producer on the mainland with 15% of the country's production. From 1983 to 1993, copra production fluctuated sharply, rising from 138 427 t in 1983 to 175 834 t in 1985, falling to 90 824 t in 1991, then slightly increasing to 123 104 t in 1993 (Table 4). In 1994 to 1996, production has increased markedly from 100 057 t to 148 932 t due to higher world prices. Then in 1997 it dropped to 139 703 mainly due to national election and drought period (Table 5). This illustrates that copra production in PNG is price sensitive - growers respond to price signals. Yarbro (1993) reported that the short-run price elasticity of copra supply in PNG has been estimated at 0.19. This means that a 10% increase in price is matched with a 1.9% increase in copra production and vice versa.

Smallholder copra production

The smallholder sector has increased its share of total copra production in recent years accounting for about 70% of annual production in 1988 (Table 6). Smallholders actively replanted coconut following World War II and this continued well into 1960's and 1970's (Yarbro 1992). Most of the plantings after 1970 were hybrid demonstrations by DAL and hybrid coconut (MAREN) planted by few innovative farmers.

Actual up-to-date field records of smallholder copra yields have not been done or at least have not been published. However, results of the surveys by DAL (Anon 1984; Anon 1985) indicated that yields were low ranging from 0.1 t to 1.5 t/ha with an average of 0.7 t/ha over nine provinces. Yarbro (1988) found yields ranging from 0.1 to 0.2 t/ha in a small sample survey in East Britain. A survey of 65 copra producers in East New Britain, East Sepik and Madang showed yields ranging from 0.5 to 1.2 t/ha with an average of 0.7 t for the entire sample (Yarbro 1992). Perhaps the main reasons for low yields may be attributed to senility of palms and different soil conditions.

Largeholder copra production

Copra production in large holdings had fallen primarily in East, West, Central New Britain, and Madang. Most plantations intercropped cocoa and coconut. Hence, some of their coconut overhead costs were combined with cocoa. Copra prices had fallen in recent years to a level approximately the cost of producing copra (labour and processing). Plantations have reduced their copra output accordingly. The decline in plantation production was also due to redevelopment of old coconut plantations with sole cocoa and the abandonment of some plantations. Oil palm has also replaced large areas of senile coconuts in New Ireland and Milne Bay provinces. One or two plantations that could not continue due to low coconut price and high input costs went into receivership and finally sold the plantations. These plantations were either diverted to other uses by the new owners or left to bush by the land owners.

Copra yields on coconut largeholdings are presented in Table 7. Yields ranged from 0.3 t/ha in Morobe and Oro provinces to 1.1 t/ha in East and West New Britain. Planting densities ranged from 82 to 153 palms/ha with an average of 121 palms. Most coconut plantations established before World War II are now approaching senility and are producing fewer nuts compared with younger coconut stands. Further, areas like East and West New Britain and New Ireland had a high incidence of palm deaths resulting from insect pests (Oryctes rhinoceros, Scapanes australis and Rhincophorus bilineatus) which may have contributed to an overall decline in copra production.

National coconut germplasm

Present status

The Coconut Breeding Section of CCRI conducted a nationwide coconut prospection surveys from 1987 to 1992 (Table 8). Morphological characters of palms were considered including vegetative and fruit composition. Collecting of selected populations (Table 9) started in March 1993. To date a total of 42 tall and 11 dwarf ecotypes are planted in collection blocks at SRS. The overall collection consists of Rennell Tall and 41 local tails, 5 exotics and 6 local dwarfs.

Utilization

A hybridization programme has been formulated to cross selected local tails with three dwarfs: Malayan Yellow Dwarf (MYD), Malayan Red Dwarf (MRD) and PNG Brown Dwarf (PBD). Both pollen and seednuts were collected from original sources at the same time. A number of trials have been formulated to use progenies from the current crossing programme for both general combining ability (GCA) tests and selection of the best hybrids for the farming community. The trials will include both population and single plant improvement in order to select the best parents for future hybridization programmes. Other genetic studies will also utilize the current germplasm collection. There are five progeny trials established to date comprised of different male parents identified during the prospection.

Genetic erosion in the country

In PNG, some of the genetic materials, especially landraces, are in danger of being lost through natural causes. Some of the populations have been wiped-out by cyclone in the islands of the Milne Bay province. To prevent further loss of genetic material, assembly of a PNG germplasm collection is being undertaken. Eventually, all the collections will be established in blocks at the SRS. Little replanting has been done in recent years. At present, farmers are using a wide range of tall ecotypes. A small proportion of planting consist of a hybrid from Rennell Tall × Malayan Red Dwarf.

Future priorities for collection, conservation and utilization

Collecting. The prospection surveys have covered most of the sites in the country. The remaining areas for survey include parts of the mainland and the Islands. Exotic materials may be imported once the 'viroid issue' is clarified and safe exchange of germplasm between countries is guaranteed.

Conservation. The establishment of museum blocks for all the collections is in progress, both landrace and exotic materials. Some of the PNG populations that have not been surveyed and selected will be included in the future.

Utilization. The genetic materials will be utilized in hybridization programmes, general combining ability tests, and in other genetic studies including reproductive biology in order to create and provide a range of planting materials to suit the growers' requirements throughout the country. The demand for tender nuts for drinks is increasing in the urban markets which may require consideration in selecting for appropriate variety.

Types of coconut grown in the country

In PNG, the following types of coconut are grown either commercially or for research:

Tall type	Landraces - commercial/research blocks Exotic - research blocks (RLT)
Dwarf type	Landraces - research blocks (PBD)
	Exotic	- seed garden (MYD, MRD) - collection (MRD, MYD, NRD, NYD, NGD)
Hybrid type	Local races - research blocks
	Exotic	- commercial (MYD × Rennell Tall) - research block (MYD × West African Tall)

Comparative performance of coconut hybrids and cultivars

The superior yield and precocity of the MAREN hybrid (Malayan Yellow Dwarf × Rennell Tall) compared with landraces have been demonstrated in PNG at Bubia and Kerevat (Brook 1985) (Table 10). However, MAREN is particularly susceptible to beetle attack. In one of the variety trials at Kerevat, large portion (>50%) of MAREN palms died from insect attack (Table 11). Those that survived were partly damaged and thus, produced poor yields after recovery. The low yields from MAWA (MYD × WAT) compared with MRD × RT was surprising (Table 10). Trials elsewhere in the world have shown that MAWA is a very high yielding cross. MAWA is also susceptible to beetle attack particularly at Kerevat (Table 11).

Rennell Tall performed well in research trials in the country. From cumulative yield data, it outyielded the local tall varieties (Table 10). Copra per nut (g) at Bubia was highest (193 g) for RT and third highest at Kerevat (249 g). However, it is also susceptible to insect attack (Table 11).

Main productivity problems

Generally, all tall landraces are low producers as indicated in one of the research trials at Bubia and Kerevat (Brook 1985) (Table 10). The main productivity problems include low number of nuts per tree and delayed bearing periods. Most exotic materials and cultivars are susceptible to beetle attack causing palm death. Beetle pests prevent planting of new coconut stands in most areas of the islands region and in many areas of the mainland. In contrast, PNG is probably fortunate to be free of major diseases of coconut, at present time. The common insect pests include two beetles, one weevil and a tree hopper:

1. Scapanes australis (Coleoptera: Deynastinae) New Guinea Rhinoceros Beetle with four sub-species

2. Oryctes rhinoceros (Coleoptera: Dynastinae) - Asiatic Rhinoceros Beetle

3. Rhincophorus bilineatus (Coleoptera: Curculionidae) - Black Palm Weevil

4. Segestes decoratus (Ortohoptera: Tettigonidae) - Sexava tree hoppers.

The National Hybrid Coconut Project is believed to have failed in the late 1970's due to beetle and high cost of seednuts. A national coconut replanting programme is now starting as the newly planted hybrid seed garden is just coming to production. The CCRI, through its Scientific Liaison services and the newly established Cocoa and Coconut Extension Agency (CCEA), intends to introduce a hybrid replanting programme to replace ageing coconut stands in areas where beetle pests are not a limiting factor. Coconut palms are preferred shade for cocoa in PNG since cocoa pest problems are reduced and the financial return from the crop combination is greater than monoculture.

Omuru Seedgarden

Omuru Seedgarden is in Madang, about 40 km from the town. It was initially established around 1974 by the then Department of Primary Industry (DPI) to produce hybrids for distribution to the farmers. In 1979, Omuru hybrid seed production started and during the same year, a Hybrid Coconut Project was launched under the National Public Expenditure Plan (Thompson 1981). After five or six years, the project ceased due to high costs of seed production. In 1993, Omuru was taken over by the Coconut Breeding Programme of the CCRI to produce hybrids initially for research purposes. The Seedgarden was earlier planted to MRD, MYD and PBD by DPI. The same dwarf varieties are being used by the CCRI as mother palms for their hybridization programme. The first lot of pollens came from selected local ecotypes. Replanting was done in 1994 as the old stands were too tall for hand pollination purposes and pollen collection.

Manabo Seedgarden

Manabo Seedgarden was the second to be established in the Cape Rodney area of Central province between 1981 and 1982. When the Hybrid Coconut Project ceased operation few years later, Manabo was run down. Due to increasing demand for planting materials in recent years, the CCRI, CCEA and CMB have combined efforts to rehabilitate the Manabo seedgarden. Arrangements are well underway to establish appropriate infrastructure facilities and to commence the production of hybrid seednuts for distribution to the growers.

Department of Agriculture and Livestock (DAL) Demo Blocks

During the implementation of the 'Hybrid Coconut Project', a number of hybrid coconut demonstration blocks were established throughout PNG. No accurate data were collected during the course of the project because of funding constraints. When the project stopped around 1985, the demo blocks were run down, making monitoring even more difficult to carry out. However, the recent demand for hybrid planting material prompted CCRI to visit several DAL demo blocks to collect and analyze data, and provide interim recommendation on the type of hybrid coconuts for the growers. Mr Ovasuru has been assigned to visit these sites to collect the necessary data. The locations covered to date include:

Hawain	East Sepik Province
Mushu	East Sepik Province
Amele	Madang Province
Popondette (PATI)	Oro Province
Palmalmal	East New Britain
Talasea	West New Britain

The hybrids included in the demo blocks are:

1) MRD × KKT (Karkar Tall)
2) MRD × MVT (Markham Tall)
3) MRD × GLT (Gazelle Tall)
4) MRD × RLT (Rennell Tall)
5) MRD × WAT (West African Tall)

Staff training and development

With financial aid from AusAID (Australian scholarship), Mr Tore Ovasuru (Coconut Breeder) successfully completed his MSc at Sydney University in 1994. Analysis of the prospection data was done by Mr. Tore Ovasuru as part of his MSc degree. Since Mr Ovasuru's degree course was for ten months (Master of Agriculture by course work), he was not able to conduct molecular studies using leaf samples of coconut germplasm to establish genetic distance between populations. There are plans for a PhD course work in the near future which may involve molecular studies on PNG germplasm.

In a separate PhD study by R. Ashburner, RAPD was used to characterize the germplasm of the South Pacific including one or two ecotypes from PNG. The other components of Ashburner's study were to collect and conserve coconut germplasm in the South Pacific.

Mr Mathias Faure (Research Officer), also with the assistance of AusAID scholarship, completed his MSc at the Sydney University in 1993. In 1995, he enrolled at La Trobe University to pursue PhD studies with a thesis on 'Reproductive biology of coconut in PNG'. The expected time of completion will be early 1999.

Various short courses and workshops organized by COGENT for both Messrs Ovasuru and Faure were useful for the continuity and appropriate conduct of the coconut breeding research in PNG. Other collaborating organisations such as PRAP (Pacific Regional Agricultural Program), PDICC, VCTC (Vanuatu Coconut Training Centre) are also acknowledged for their support.

Besides training, COGENT's major role in the campaign for coconut germplasm exchange and selection of the regional genebanks was a major achievement in coconut breeding programme. PNG has been nominated as the host for an international genebank for the Pacific region. Infrastructure planning and establishment are well underway at the Stewart Research Station (SRS). Currently, only previously introduced exotic materials (MRD, MYD, NGD, NRD, NYD, RIT & WAT) could be found in the proposed genebank at SRS.

Breeding programme

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

National coconut research and development is the responsibility of the CCRI. A coconut breeding programme has been formulated and will be finalized pending the characterization of the PNG germplasm. The objectives of the current coconut breeding programme are:

1. Breeding for resistance against major insect pests;
2. Develop hybrids and tails with high yielding abilities and improved precocity; and
3. Improve and develop better parents for hybridization.

Suggestions for collaborative breeding research with other countries

· Exchange germplasm materials at the institute, regional and global level. At present, it is only possible to exchange pollen due to quarantine restrictions;

· Exchange information and upgrade information network;

· Design similar experiments with standardized designs for comparisons; and

· Standardize breeding techniques.

The PNG CCRI is the official and only research body responsible for coconut research and development. All correspondence should be addressed to:

The Institute Director
PNG Cocoa and Coconut Research Institute
P. O. Box 1846
Rabaul, ENBP
Papua New Guinea
Tel: (675) 983 9131/983 9108
Fax: (675) 983 9115
E Mail: ccri@datec.com.pg

Funding agencies for coconut research projects

Annual financial contributions are received from the Cocoa Board and Copra Marketing Board of PNG based on the levy of the exported crops. At present, this research levy is approximately K 20.00/t (US$ 15/t) for cocoa and K 2.00/t (US$ 1.5/t) for copra. The Copra Marketing Board has further contributed substantial funds for the establishment of the CCRI's SRS and for research activities.

The PNG National Government makes an annual contribution to CCRI for general research operations.

The European Union under STABEX (a price stabilization scheme) has funded the entire beetle control research programme for the last four years which included construction of a new entomology laboratory and equipment at SRS.

ACIAR funded the Coconut Improvement Project, covering germplasm prospection, characterization, embryo culture and conservation, from 1987 to 1993. Roger Ashburner's RFLP research and the analyses of the germplasm prospection data were supported by ACIAR until July 1994.

CIRAD-CP funds the salary and some operating costs of a CIRAD-CP Agronomist (Dr Jean Ollivier), based at the SRS in Madang. The Copra Marketing Board provides additional funds for purchase and development of the research laboratories. Most recently, the CCRI has participated and received funding from an international European Union INCO for a coconut beetle project.

ADB financial support was obtained through COGENT for specific coconut research projects including the establishment of a regional genebank at SRS.

Conclusion and recommendation

Research and development of the coconut industry is of high priority to the nation. Coconut is a major daily source of food and income for the lowland people and feed source for their animals. It is the only source of income for many farmers on the islands and on mainland areas with poor soils. Coconut is the third most important agricultural export commodity of the lowlands. Coconut is extremely important for cocoa cultivation. There was a significant decline in coconut production and area planted to coconut over the years. The reasons for the decline included low world market prices and replacement of coconut stands by other cash crops. However, the recent high world price of copra increased production and export levels. In 1997, copra exports rose to around 140 000 t.

A national coconut development programme failed in the 1970's due to debilitating beetle pests, high costs, and ineffective promotion and distribution of seednuts. Redevelopment of the coconut industry is not possible in many areas until a solution to the beetle pest is found. Substantial funds have been successfully secured for this work for the last four years from the European Union STABEX. Recently, the CCRI has joined an EU INCO International Project on beetle control.

Internal downstream processing of coconut is confined mainly to oil production. A new oil mill was recently established in Madang by the Copra Marketing Board and others are planned for future establishment.

Coconut research and development is the responsibility of the CCRI. This programme was initiated a few years ago and is now a major component of the CCRI's overall programme. The main component projects of the programme are: comprehensive germplasm collecting, hybrids and tall palm improvement/breeding, agronomy, plant protection (beetle control), farming systems research and downstream processing.

It is felt that the programme could be enhanced by:

a) Collaborative work and cooperation among institutions, at regional and global levels;
b) Intensifying research on beetle control;
c) A standard method of germplasm characterization;
d) Providing a solution to the 'viroid issue' to enable global and regional exchange of germplasm;
e) Information exchange of new research findings;
f) Promotion of downstream processing;
g) Socioeconomic studies of smallholder coconut farming systems; and
h) Intensifying coconut farming systems research with particular regard to cash crop diversification.

References

Anon. 1984. Hand Book on Agriculture Statistics. Papua New Guinea Department of Primary Industry (Department of Agriculture and Livestock). Port Moresby.

Anon. 1985. Commodity Summary: Copra. Papua New Guinea Department of Agriculture and Livestock. Port Moresby.

Brook, R. M. 1985. Early yields from dwarf × tall coconut experiment. Harvest Vol. 11, No. 2. Papua New Guinea.

Foale, M. A. and P. W. Lynch. 1993. Coconut Improvement in the South Pacific, ACIAR Proceedings No. 53.

Papua New Guinea National Statistical Office. 1991. 1990 National Population Census, Preliminary Figures. Port Moresby.

Thompson, P. 1981. Hybrid coconuts in Papua New Guinea. Harvest Vol. 7 No. 3. Papua New Guinea.

Yarbro, S. 1988. Farmers' receptivity to new technologies in coconut - Papua New Guinea. Asian and Pacific Coconut Community, Jakarta, Indonesia.

Yarbro, S. 1992. Domestic marketing of coconut products in Papua New Guinea. Asian and Pacific Coconut Community, Jakarta, Indonesia.

Yarbro, S. 1993. Processing of coconut products in Papua New Guinea. Asian and Pacific Coconut Community, Jakarta, Indonesia.

Table 1. Provincial areas and populations, 1990

Source: 1990 Census and 1979 National Roads Inventory

* The 1980 census figure
** Excludes the population of North Solomons

Source: Department of Primary, Handbook of Agricultural Statistics, 1984

Source: National Statistical Office, Rural Industries

Year	Area (ha)
1981	93 734
1982	92 104
1983	92 063
1984	89 410
1985	84 998
1986	81 034
1987	75 245

Table 4. Copra production (t) in Papua New Guinea by province

Source: Copra Marketing Board of Papua New Guinea.

Province	1983	1984	1985	1986	1987	1988	1989	1990	1991	1992	1993
East New Britain	51 904	58 597	58 923	55 543	53 362	50 362	50 172	51 558	42 068	44 253	45 110
North Solomons	19 462	26 658	30 782	25 834	24 790	27 081	26 756	1 801	1 984	10 212	10 898
Madang	22 927	22 525	24 568	25 618	24 390	18 735	20 294	25 997	20 811	23 896	14 825
New Ireland	15 455	16 589	21 700	17 765	16 566	15 002	12 438	11 426	8 874	14 316	15 797
West New Britain	11 975	13 421	16 400	14 615	15 454	12 531	14 150	15 349	10 839	14 876	15 777
Milne Bay	5 448	7 155	8 424	7 714	5 546	4 636	3 695	4 257	2 811	4 082	3 856
Central & Gulf	4 256	4 204	5 224	4 000	2 499	2 119	1 471	1 409	625	832	701
East & West Sepik	2 856	3 717	3 179	3 059	2 885	2 205	2 202	2 669	1 892	2 402	3 004
Morobe	2 662	3 470	4 450	3 587	3 230	2 514	1 972	1 806	773	1 355	1 535
Manus	1 482	1 888	2 184	757	595	667	588	380	147	378	601
Total	138 427	158 224	175 834	158 492	149 317	135 852	133 738	116 652	90 824	116 602	123 104

Table 5. Copra production (t)

Source: Copra Marketing Board of Papua New Guinea

Source: Copra Marketing Board of Papua New Guinea

Source: Yarbro 1992

Province	Number of sites
East New Britain	23
West New Britain	4
New Ireland	11
North Solomons	6
Manus	27
Morobe	3
Madang	9
East Sepik	28
Sandaun	7
Central	6
Milne Bay	7
Oro	6
Gulf	11
Western	5
TOTAL	153

Table 9. Selected coconut germplasm in Papua New Guinea

1 Markham Tall (MVT) 2 Karkar Tall (KKT) 3 Gazelle Tall - Ulatava (ULA) - Pellevarua (GLT 1) - Raulawat (GLT 2) - Natava (GLT 3) - New Massava (GLT 4) 4 Gazelle Yellow Fruited Tall (GYT) 5 Gazelle Red Fruited Tall (GRT) 6 Gazelle Many Fruited Tall (GMT) 7 Milne Bay Tall (MBT) 8 Oro Tall (OLT) 9 Vailala Tall (VLT) 10 Namatanai Tall (NLT) 11 Manus Tall (MLT) 12 East Sepik Tall (EST 13 Bouganville Large Fruited Tall (BLT) 14 Nuguria Tall (FIT) 15 West New Britain Tall (WNT) 16 Talasea Red Fruited Tall (TRT) 17 Kiwai Tall (KWT) 18 PNG Yellow Dwarf (PYD) 19 PNG Red Dwarf (PRD) 20 Rabaul Red Dwarf (RRD) 21 Lokea Red Dwarf (IRD) 22 PNG Brown Dwarf (PBD) 23 PNG Green Dwarf (PGD)

Table 10. Cumulative yields of dry copra at Bubia up to the end of 1982 and nut sizes (dry copra basis) at Bubia and Kerevat in 1982

Source: Brook 1985

* Not represented

Source: Brook 1985

National coconut breeding programme in Fiji

Joeli Kauvere
Agriculture Officer, Taveuni Coconut Centre, Waiyevo, Taveuni Island, Fiji

Introduction

Although the coconut industry in Fiji has been on the decline for the last three decades, coconut remains a major crop in Fiji. Apart from its nutritional importance, it is estimated that about 40 000 households rely on coconut as a source of cash income. Approximately 65 000 hectares are under coconut with 60% senility.

Coconut usage is split between subsistence consumption, and local and export market of whole nuts and copra. Copra processed by local mills varies between 10 000 and 19 000 t/year against 35 000 to 41 000 t in the early 1960s with copra meal as a by-product. Most of the copra is exported while meal is used locally as a feed for poultry and livestock.

The domestic consumption of coconut can be roughly estimated at 12 000 t of copra equivalent. Hence, copra productivity in Fiji remains around 350 kg ha^-1, which is very low compared to the world average of 700 kg ha^-1. While good arable land is needed, most land under senile coconut is improperly used, if not wasted.

Although soil and climatic conditions are favourable, coconut is losing ground in the country's economy. The farmers' income is decreasing and self-sufficiency in coconut products in the future is not guaranteed, unless the industry is rehabilitated soon. Senility of a large proportion of palms along with plummeting prices of coconut oil in the world market are the two major reasons for the industry's constant declining pattern, resulting in poor maintenance of plantations by farmers.

To address the industry's poor performance and to reverse the trend, the Fiji Government decided to rehabilitate part of the existing coconut areas (approximately 35 000 ha) using high yielding hybrids in 1986. Along with intercropping and livestock integration, the scheme was expected to raise the productivity per hectare. Land not replanted to coconut was made available for other agricultural purposes.

National coconut germplasm

Coconut germplasm, collection and multiplication

The establishment of coconut germplasm blocks was constrained by difficulties in obtaining pure seednuts in the Pacific.

The Taveuni Coconut Centre (TCC) has 17.3 ha germplasm collection and 14.8 ha multiplication blocks. It holds four dwarf varieties (MRD, MYD, MD and NLD) along with three tall varieties (RLT, ROT and FJI). No further addition of coconut germplasm has been done due to quarantine restrictions.

Fiji tall progenies

The TCC maintains and monitors progenies of Fiji Tall palms, which were selected from two populations of Taveuni.

It provides the basis for germplasm selection for the breeding programme, particularly on pure Fiji Tall or on hybrids derived from Fiji Tall.

Detailed observation and identification of different types of Fiji Tall and Dwarf for breeding purposes will be done in the near future.

Comparative performance of coconut hybrids and local cultivars

Table 1 shows the comparative precocity for both Malayan Tall and Dwarf parents, and for their hybrids. The remarkable precocity of the hybrids is inherited from Malayan Dwarfs. The Niu Leka Dwarf confirmed its peculiarity of being a late bloomer (only 33% of palms flower after 50 months).

The average number of nuts per palm of the Fiji Tall and hybrids are shown in Table 2.

National replanting programme

The initial plan is to plant 34 000 hectares within 20 years, concentrating on the main coconut growing provinces which are contributing a substantial amount to national copra production. Production from replanted hybrids commenced in 1994 with gradual increase to 30 000 t by the year 2000. Meanwhile, efforts will be made to rehabilitate 22 000 ha coconut groves, which are less than 40 years old, through mixedcropping and animal integration.

Breeding strategies

Coconut breeding action plan

The first phase of the Coconut Research and Development Program (CRDP) was the establishment of the Taveuni Coconut Centre (TCC) with the following triple mandate:

1. production and release of hybrid seednuts;

2. applied research on best hybrids adapted to Fiji condition along with formulation of planting maintenance techniques; and

3. providing extension support and training services for extension staff and farmers.

Suggestion for collaborative breeding research with other countries

The emphasis should be in breeding for total copra because of the wide variation that exists in copra/nut ratio between populations. The high heritability of this trait ensures its rapid improvement by selection. More attention could be focused on improving the fruit characteristics, i.e. less husk and more albumen. Scientific organizations should work together in developing coconut breeding research techniques at regional basis specifically on the following aspects:

a) Breeding research techniques with emphasis on local cultivars; comparative trials with introduced cultivars;

b) Germplasm collecting, conservation and exchange; and

c) Use of tissue culture and embryo culture for collecting of germplasm from local and foreign sources.

The success of coconut breeding in meeting its national objectives depends on the following requirements:

a) Establishment of a centre for applied research on coconut and seed production aiming at improving coconut farmers' income especially by intercropping, mixed farming and diversification of products. The centre should also strengthen coconut timber utilization aspect associated with coconut replanting development; and

b) Coconut rehabilitation programmes to manage replanting projects and isolated nurseries.

Variety	TCC site	Age at 50% flowering palms (months)
Malayan Red Dwarf (MRD)	Collection	41
	Multi. Blocks	37
	Mean	39
Malayan Yellow Dwarf (MYD)	Collection	38
	Multi. Blocks	39
	Mean	38.5
Rennel Tall (RLT)	Collection	56
	Multi. Blocks	47
	Mean	51.5
Rotuma Tall (ROT)	Collection	59
	Multi. Blocks	49
	Mean	54
Fiji Tall (FJT)	Collection	54
	Tuvumaca	51
	Ura	46
	Hybrid Expt.	46
	Mean	49.3
Hybrid:
MRD × RLT	expt.	39
MRD × ROT	expt.	39
MRD × FJT	expt.	39

Table 2. Average number of nuts per palm

Population/hybrid	1	2	3	4	5	6	Total	Mean
FJT	0.7	2.2	2.0	2.6	0.0	1.7	9.2	1.5
MRD × FJT	19.5	24.3	12.4	7.8	15.7	27.1	106.8	17.8
MRD × ROT	22.3	18.7	13.5	5.5	18.0	8.4	86.4	14.4
MRD × RLT	10.6	10.6	10.7	7.2	0.3	0.0	39.4	6.6
MRD × NLD	7.2	12.1	7.5	15.4	0.0	0.0	42.2	7.0
Total	60.3	67.9	46.1	38.5	34.0	37.2	284.0
Mean	12.1	13.6	9.2	7.7	6.8	7.4		9.5

Legend:

MRD: Malayan Red Dwarf
FJT: Fiji Tall
ROT: Rotuma Tall
RLT: Rennel Tall (Solomon Is.)
NLD: Niu Leka Dwarf (Fiji)

Period	Number of seednuts	Year	Number
			Seedling	Hectare
May 1993 - April 1994	99 000	1994	59 400	383
May 1994 - April 1995	206 000	1995	123 600	797
May 1995 - April 1996	258 000	1996	154 800	999
May 1996 - April 1997	258 000	1997	154 800	999

Vanuatu national coconut breeding programme

Gérard Duhamel
Project Coordinator, PDICC-PRAP, VARTC, Espiritu Santo, Vanuatu

Introduction

Present coconut production and hectarage

The estimated area planted to coconut in Vanuatu is 30 000 ha which provides 50% of the national export. The estimated local consumption is equivalent to 17 000 t of copra. The smallholders ensure about 90% of the export whereas, the share of large plantations is decreasing constantly.

Present farm level productivity

Coconut cultivation in Vanuatu involves two contrasting sectors:

a) large plantations, with an average of 133 ha, often ageing (60% > 50 years old), largely involved in other activities (livestock), and have low yields (0.46 t ha^-1 average); and

b) smallholders, who own 80% of the coconut grove, cultivate small farms (3.38 ha average) with younger palms and higher yield (0.63 t ha^-1).

National coconut germplasm

Present status

Vanuatu germplasm collection is entirely maintained in the field. It is made up of local populations and imported ecotypes. It can be summarized as follows:

Local	8 local tall populations (Vanuatu tall or VTT)	5.5 ha
	1 dwarf ecotype (Vanuatu Red dwarf or VRD)	0.5 ha
South Pacific	10 tall accessions (7 ecotypes)	4.6 ha
	5 dwarf accessions (4 ecotypes)	2.0 ha
Southeast Asia	4 tall ecotypes	2.9 ha
	9 dwarf accessions (7 ecotypes)	4.4 ha
Others (Africa, America)	1 tall ecotype	0.7 ha
	2 dwarf ecotypes	1.0 ha
TOTAL	23 tall/17 dwarf accessions	21.0 ha

Utilization

The germplasm is characterized according to CIRAD (Centre de Coopération Internationale en Recherche Agronomique pour le Développement) standards and used in breeding programmes. Their susceptibility to an endemic disease, CFD (coconut foliar decay), is evaluated in the field and by artificial inoculation under cage.

Genetic erosion

The presence of CFD prompts growers to plant almost exclusively the local tall which is tolerant. Replanting is actively done by smallholders at the rate of 2 000 ha/year and as such the local population can hardly be considered as endangered, at least in the main cultivation areas. In the more marginal areas, (south of the country) the reconversion of some plantations, along with cyclones, reduced the local populations. However, the genetic studies made on population coming from different islands showed a great homogeneity which might be related to the selection pressure imposed by the disease. In total, the risks of genetic erosion on the short term are limited.

Future priorities for collection, conservation and utilization

Considering the above situation, collection of local populations is not among the first priorities for Vanuatu. However, attention should be given to the marginal areas where large plantations could disappear.

CFD severely limits the use of imported materials for varietal improvement. Priority will be given to the ecotypes which are less sensitive to CFD like Renelle Is. Tall (RIT), which can improve the local cultivars. Selected ecotypes will be combined with the local tolerant ones. In this domain, the enrichment of the collection will be limited to some South Pacific ecotypes, in connection with the Production and Dissemination of Improved Coconut Cultivars (PDICC) programme.

The main priority for conserving the current genetic resources is the rejuvenation of the existing collection. This programme is partially realized for dwarfs.

Types of coconut in the country

There are three major types of coconut grown in Vanuatu: tails, tall × tall hybrids and dwarf × tall hybrids. The specific cultivars under each group type are shown in Table 1.

National replanting programme

From 1982 to 1983, the Kokonas Development Project (KDP) promoted more productive coconut plantations, by providing improved seeds and seedlings produced at the Vanuatu Agricultural Research and Training Centre (VARTC). During its life-span, the project established about 20 nurseries spread across the country and provided planting materials for about 2700 ha. As part of this project, 500 ha were planted with a hybrid which was issued from 1989 onward. The project covered only 10 to 15% of Vanuatu's replanted area. The rest, about 2000 ha/year, are replanted by smallholders, allowing the renewal of the national coconut groves in a 45-year period.

Since KDP stopped its activity, coconut development is managed by the Department of Agriculture using the 2% tax levied on copra.

Main productivity problems

Factors limiting productivity, in addition to CFD, include insufficient maintenance of the plantation (manuring) and in some cases, the age of the plantation.

Breeding strategies

The breeding programme is being implemented at the VARTC located at Saraoutou, Santo Island. The main objectives of the programme, in order of importance, are as follows:

a) tolerance to coconut foliar decay;
b) a nut size at least equivalent to that of Vanuatu tall (VTT) to be broadly accepted;
c) high yield;
d) germination curve similar to that of the local tall (VTT); and
e) ability to adapt under the traditional production scheme, particularly: affordability of planting material, response to local cultural management and tolerance to low levels of fertilizer. Related to this, improvement of the local tall and of tall × tall hybrids remain interesting even though dwarf × tall would give a higher yield.

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

Presently, the coconut germplasm resources consist of three types of cultivars, namely:

· a tall cultivar (VTT) tolerant to CFD, precocious, with small nuts and average productivity;

· a local hybrid cultivar (VRD × VTT), tolerant to CFD but with a very slow germination and very small nuts; and

· hybrid cultivars involving introduced ecotypes, productive but more or less susceptible to CFD.

a) In the short term, the exploitation of the variability of the local tall leads to rapid progress in terms of cultivar improvement. Selected progenies can be used to establish improved seedgardens. This will allow the production of 'tall' seeds through natural pollination as well as hybrid seeds through assisted pollination of the local dwarf using the 'tall' pollen.

b) In the medium term, Renelle Is. Tall (RIT) is among the least susceptible exotic cultivar and its hybrid with Vanuatu tall (VTT) is almost free of CFD. Two operations have to be realized:

i) reproduction of VIT × RIT crosses with individuals screened and selected for improved CFD tolerance, and

ii) build up of a recombined population involving (RIT × VTT) × (RIT × VTT) and (RIT × VTT) × VTT crosses aiming at accumulating the qualities of both ecotypes. In the next 10 years, the diffusion of RIT × VTT hybrids and the set up of hybrid trials involving the recombined population and the local dwarf may be expected.

c) In the long term, the improvement of the dwarf × tall hybrid implies the recombination of the local dwarf with good combining ability, i.e. the Brazilian Green dwarf (BGD). Selection for tolerance, nut weight and germination speed shall be done up to the F₂ generation. Progeny test will begin with the F₃s.

As a complement to hybridization, setting up of tests for tolerance to CFD, in the field and through artificial inoculation under cages, will be necessary.

Moreover, the VARTC shares regional responsibilities, through the PDICC programme, to test several tall × dwarf hybrids involving tall ecotypes from South Pacific. Four trials have already been planted and three others are in preparation.

Suggestions for collaborative breeding research with other countries

As a member of the PDICC, the VARTC contributes to improvement of the knowledge on South Pacific germplasm through training and exchange of information among coconut breeders in the region. Although the presence of CFD prevents the Centre from providing breeding material for the moment, the experimental results will assist other research centres in the region to determine the most interesting hybrid material for their own programme. The centre shall increase its collection of ecotypes from the region.

The second phase of a regional collaboration work may involve a series of multi-location tests, comparing a limited range of hybrids in contrasting environments.

The main obstacle to exchanges between coconut research centres comes from the fears concerning the possible transmission of cadang-cadang viroid (CCVd) related sequences, and in the case of Vanuatu, of CFD. Continued research aiming at establishing the exact nature of their transmission and at developing and using a safe methodology for exchanging breeding material is a must to allow rapid collecting and appreciably contribute to improving the coconut germplasm resources.

Natio