On-farm Reproduction of Coconut Palms for Breeding, Seednut Production and In-situ Conservation: A guide for coconut farmers to better conserve, breed, use and market their coconut seednuts

Roland Bourdeix, Carmel Pilotti and Vincent Johnson Suva, Fiji, 2025 © Pacific Community (SPC) 2025 All rights for commercial/for profit reproduction or translation, in any form, reserved. SPC authorises the partial reproduction or translation of this material for scientific, educational or research purposes, provided that SPC and the source document are properly acknowledged. Permission to reproduce the document and/or translate in whole, in any form, whether for commercial/for profit or non-profit purposes, must be requested in writing. Original SPC artwork may not be altered or separately published without permission. Original text: English Pacific Community Cataloguing-in-publication data Bourdeix, R. (Roland) On-farm reproduction of coconut palms for breeding, seednut production and in-situ conservation: a guide for coconut farmers to better conserve, breed, use and market their coconut seednuts – first edition / by Roland Bourdeix, Carmel Pilotti and Vincent Johnson 1. Coconut – Oceania. 2. Coconut palm – Oceania. 3. Plant breeding – Oceania. 4. Seeds – Collection and preservation — Oceania. 5. Germplasm resources conservation – Oceania. 6. Coconut farmers – Oceania. I. Bourdeix, R. (Roland) II. Pilotti, Carmel A. III. Johnson, Vincent IV. Title V. Pacific Community 634.6180995 AACR2 ISBN: 978-982-00-1647-7 Authors Roland Bourdeix, CIRAD, UMR AGAP Institut, Montpellier, France. UMR AGAP Institut, Univ Montpellier, CIRAD, INRAE, Institut Agro, Montpellier, France. Email: [email protected] V. Johnson. GreenQuills International, Prades-le-Lez, France; Alliance of Bioversity International and CIAT, Montpellier, France. Email: vinbjohnso[email protected]; [email protected]rg C. Pilotti. Correspondent author. Associate Scientist Coconut Genetic Resources. Pacific Community, Land Resources Division, Fiji. Email: [email protected]t Cover page photo: Farmer and extensionist trainees in Taveuni nursery, Fiji/V. Johnson. Prepared for publication at SPC’s Suva Regional Office, Private Mail Bag, Suva, Fiji, 2025 www.spc.int | [email protected] iii Contents List of figures .............................................................................................................................. v Acronyms ...................................................................................................................................vii Glossary of key technical terms ............................................................................................... viii Acknowledgements ....................................................................................................................xi Preface ....................................................................................................................................... xii Executive summary .................................................................................................................... 1 1. INTRODUCTION ................................................................................................................... 2 2. BIOLOGY AND DIVERSITY OF COCONUT VARIETIES ............................................................ 5 2.1. Four kinds of coconut palms and their possible hybrids ............................................. 5 2.2. Differentiating between coconut palms ...................................................................... 8 2.3. Naming of coconut varieties and populations ............................................................ 9 2.4. Coconut inflorescences and flowers.......................................................................... 13 2.5. Different ways coconuts are produced in nature ...................................................... 14 2.6. Colours of coconut fruits and sprouts ....................................................................... 18 3. WAYS FARMERS CAN PRODUCE BETTER, MORE DIVERSE SEEDNUTS .............................. 23 3.1. Brief overview of genetics and plant breeding ......................................................... 23 3.2. Brief overview of coconut seednut production methods ......................................... 25 3.3. Dealing with open-pollinated seednuts .................................................................... 27 3.4. Controlled natural (open) pollination ........................................................................ 29 3.5. Assisted hand pollination .......................................................................................... 33 3.6. Controlled pollination with bagging .......................................................................... 37 3.7. Toward seednut production methods tailored to farmers ....................................... 38 3.7.1. Natural (open) pollination ......................................................................................... 39 3.7.2. Controlled natural (open) pollination ........................................................................ 39 3.7.3. Assisted hand pollination. .......................................................................................... 42 3.7.4. Controlled pollination with bagging .......................................................................... 44 4. IMPROVING SEEDNUT QUALITY ........................................................................................ 48 4.1. Coconut palm sensitivity to environmental, seasonal and annual variations .......... 48 4.2. Selection of parent palms .......................................................................................... 52 4.3. Efficiency of pollination techniques .......................................................................... 54 4.4. Selection at seedbed and nursery stages .................................................................. 55 4.5. Traceability of seednut production, use and sales .................................................... 56 iv 5. OTHER INITIATIVES TO PROMOTE THE CONSERVATION OF FARMERS’ VARIETIES .......... 58 5.1. The Polymotu concept ............................................................................................... 59 5.2. Community or private coconut genebanks ............................................................... 62 5.3. Strengthening relationships with national genebank ............................................... 65 5.4. Importance of compiling and publishing varietal catalogues ................................... 66 5.5. Communicating about varieties and seednuts on social media ................................ 68 5.6. Creating coconut ecomuseums and learning centres ............................................... 69 6. CONCLUSION ..................................................................................................................... 71 References ................................................................................................................................ 73 Glossary of additional technical terms ..................................................................................... 75 Annex 1. Updated guidelines for producing technical coconut photos................................... 82 Annex 2. Forms for parent palm selection ............................................................................... 91 Form 1: Parent palm selection – data on coconut plantations ............................................ 92 Form 2: Parent palm selection – location of coconut palms ................................................ 93 Form 3: Parent palm selection – colour, variety and stem (trunk) ...................................... 94 Form 4: Parent palm selection – mature fruit analysis, method 1 (complete) .................... 95 Form 5: Parent palm selection – mature fruit analysis, method 2 (pig-feeding) ................. 96 Form 6: Coconut seed system – tender nut fruit analysis, Polynesian method .................. 97 Form 7: Coconut seed system – nursery test for hybrid removal ........................................ 98 Annex 3. Standard coconut fruit measurements ..................................................................... 99 v List of figures Figure 1: Variation in shape and colour of coconut inflorescences / Photos: R. Bourdeix ....... 3 Figure 2: Diversity of coconut varieties and forms in French Polynesia / Photo: R. Bourdeix .. 4 Figure 3: Various types of hybrid coconut palms / Photos R. Bourdeix .................................... 6 Figure 4: Identification of main coconut varietal types by the shape of the stem (trunk) / Photos R. Bourdeix ..................................................................................................................... 8 Figure 5: Standardised description of a coconut variety: Example of the Rennell Island Tall (RIT) from Solomon Islands/ Photos: R. Bourdeix ............................................................................... 9 Figure 6: Two ways of naming the same coconut variety in the same village ........................ 10 Figure 7: Female and male coconut flowers on spikelets / Photos R. Bourdeix ..................... 13 Figure 8: Coconut pollen under the microscope (with germinating pollen tubes) / Photo: V. Johnson ..................................................................................................................................... 14 Figure 9: Compact Green Dwarf coconut palm with two inflorescences, with a delay of about one week between them / Photo R Bourdeix .......................................................................... 15 Figure 10: The floral biology of a Tall-type coconut variety / Photos: R. Bourdeix ................. 16 Figure 11: Male and female phases of the inflorescences of four varieties / Photo: R. Bourdeix .................................................................................................................................................. 17 Figure 12: The four basic colours of young seednut shoots / Photos: R. Bourdeix .................. 18 Figure 13: Coconut colours / Photos: R. Bourdeix ................................................................... 20 Figure 14: Illustration on how crossing green and red coconut palms simplifies and optimises the process / Photos: R. Bourdeix ............................................................................................ 21 Figure 15: Illustration of the pink colour found in some coconut varieties / Photos: R. Bourdeix .................................................................................................................................................. 22 Figure 16: Mendelian genetics of coconut embryo sprout colour / Photos: R. Bourdeix ....... 24 Figure 17: Coconut embryos growing in glass tubes in a laboratory / Photo: R. Bourdeix ..... 27 Figure 18: An example of farmer selection / Photos: R. Bourdeix ........................................... 29 Figure 19: Emasculation of a Dwarf coconut palm inflorescence / Photos: R. Bourdeix ........ 30 Figure 20: Seed-garden design planted with two varieties / Image: R. Bourdeix ................... 31 Figure 21: Seed-garden design planted with three varieties / Image: R. Bourdeix ................. 32 Figure 22: Seed-garden design planted with four varieties / Image: R. Bourdeix ................... 32 Figure 23: Destemming spikelets harvested from male parents to collect male flowers for pollen production / Photo: R. Bourdeix .................................................................................... 35 Figure 24: Apparatus for making coconut pollinations from the ground / Photo: R. Bourdeix .................................................................................................................................................. 37 Figure 25: Bags used in controlled pollination: female bag (left); male bag (right) ................ 38 vi Figure 26: Turning a coconut plantation of Tall-type coconut palms into a seed garden producing seednuts of Talls, Dwarfs and Dwarf x Tall hybrids / Photos: R. Bourdeix ............. 41 Figure 27: A receptive female coconut flower with nectar on its tip, and a bee / Photo: R. Bourdeix ................................................................................................................................... 43 Figure 28: Setting a homemade pollination bag on a coconut inflorescence, made from breathable supermarket bags that can be prepared by farmers at low cost for controlled pollination / Photo: C. Pilotti .................................................................................................... 44 Figure 29: Bagging a single female flower / Photo: C. Pilotti and R. Bourdeix ........................ 45 Figure 30: Sensitivity of the coconut palm to environmental variations: the case of the Madang Brown Dwarf (MBD) / Photo: R. Bourdeix ............................................................................... 49 Figure 31: Sensitivity of the coconut palm to environmental variations: the case of the Malayan Green Dwarf (MGD) in Côte d’Ivoire/ Photos: R. Bourdeix ...................................................... 50 Figure 32: Fruit-size variation within a Laccadives Micro Tall (LMT) palm variety from India / Photo: R. Bourdeix .................................................................................................................... 51 Figure 33: Selecting seednuts based on two categories of “female” and “male”, where the “female” is generally kept for replanting / Photo: R. Bourdeix ................................................ 52 Figure 34: Illustration of the Polymotu concept (1) Tahaa, French Polynesia / Photos: R. Bourdeix ................................................................................................................................... 59 Figure 35: Illustration of the Polymotu concept (2) Samoa / Photos: R. Bourdeix ................. 61 Figure 36: A first design for a community genebank ............................................................... 63 Figure 37: A second design for a community genebank. ......................................................... 64 Figure 38: Example of a four-page description of a variety in the French Polynesian catalogue / Photos: R. Bourdeix ................................................................................................................ 67 Figure 39: Views of the Fakahina Green Dwarf / Photos: R. Bourdeix .................................... 68 Figure 40: Schematic representation of a model of a large coconut ecomuseum / Photo: R. Bourdeix ................................................................................................................................... 70 Figure 41: Taking quality photos of the whole palm / Photos: R. Bourdeix ............................ 84 Figure 42: Taking quality photos of the inflorescences and bunches / Photos: R. Bourdeix .. 86 Figure 43: Taking quality photos of the 12 fruits / Photos: R. Bourdeix .................................. 88 Figure 44: Taking quality photos of de-husked coconuts / Photos: R. Bourdeix ..................... 90 vii Acronyms ACIAR Australian Centre for International Agricultural Research CBO community-based organisation CDM Coconut Data Management software CGRD Coconut Genetic Resources Database CPCRI Central Plantation Crop Research Institute (India) CGIAR Consultative Group on International Agricultural Research CIDP Coconut Industry Development for the Pacific Project CIRAD Centre international de recherche en agriculture pour le Développement COGENT International Coconut Genetic Resources Network DNA deoxyribonucleic acid FAO Food and Agriculture Organization of the United Nations ITPGRFA International Treaty on Plant Genetic Resources for Food and Agriculture KIK Kokonas Indastri Koporesen (Papua New Guinea) NGO non-governmental organisation SPC The Pacific Community 2 1. INTRODUCTION Over centuries, coconut farmers have randomly or intentionally created hundreds of magnificent coconut varieties, as illustrated in Figure 1 and Figure 2. However, in most coconut-producing countries, farmers are mostly unaware of improved coconut varieties (see “Hybrid” in the Glossary of key technical terms) as these seednuts are scarce, unavailable, and/or too costly [1]. In 2012, the International Coconut Genetic Resources Network (COGENT) recommended that farmers play a primary role in developing and choosing coconut varieties for their own and others’ use [2]. Specifically, COGENT recommended the following: i) Encourage local stakeholders (men and women farmers, private enterprises, nongovernmental organisations [NGOs], and community-based organisations [CBOs]) to become more involved in supplying quality seednuts. ii) Teach farmers and other stakeholders how to produce quality seedlings of varieties and hybrids using the Polymotu5 concept (isolation) or any other suitable method. iii) Assess farmers’ knowledge of coconut seednut production, including the use of simple markers (e.g. fruit colour) for breeding purposes. iv) Increase farmers’ knowledge about seednut production. As of June 2025, none of these recommendations had been fully implemented. In 2018, these coconut conservation recommendations were refined within the framework of an ACIARfunded, Pacific Community (SPC)-managed project, “Coconut Industry Development for the Pacific Project” (CIDP).6[3] These refined recommendations include giving farmers a primary role in making their own decisions with respect to the varieties to be planted in their respective fields [4]. It is important to recognise that farmers already share coconut-planting material [5], mostly among their social networks (family, clan, village). Some also sell small quantities as seednuts, although these have often been produced with limited technical skills in coconut breeding. The wealth of local traditional knowledge, especially among elders, must be preserved and shared by elders with other community members, including youth [5]. Technical knowledge should be more accessible and user-friendly for farmers. Farmer field schools that educate farmers in selection and simple seednut production will strengthen on-farm conservation, as highlighted by Konelio et al. (2018): “There is… a great interest [in] developing a simpler controlled hand-pollination method [5] …for farmers, and small research centres lacking pollen-processing labs.” In 2023, under a project of the Food and Agriculture Organization of the United Nations (FAO), called the Safeguarding threatened coconut diversity within the upgraded international genebank for the South Pacific, workshops were held to educate Pacific farmers and extension workers on coconut biology and informal conservation. This guide draws on the subject areas covered in these workshops, as well as information provided in references on coconut conservation, and aims to provide guidance to farmers on reproducing coconut palms for 5 Based on an isolation of sufficient distance to prevent unintended pollen contamination. (See “Polymotu” in the Glossary of key technical terms.) 6 A joint initiative of the SPC, the European Union and the African, Caribbean and Pacific Group of States. 3 breeding, seednut production and in situ conservation. This will allow them to better conserve, breed, use and market their own coconut seednuts. This guide provides a review of the types of coconut varieties available, their ways of producing seeds, and the means of distinguishing these varieties and forms from each other. Following an overview provided on the classical methods of coconut seednut production, methods that can be simplified for farmers’ use are explored. The importance of farmers in selecting varieties and cultivars is discussed, as well as the risks of farming intensification. Other ways to boost farmers’ capacity to produce their own seednuts with desired traits and conserve preferred types are also discussed. The potential for farmers to produce and test their own seednuts through simple crossing methods is introduced, acknowledging that any seednut production or trials for commercial gain should be done with the consent and knowledge of local government and/or other responsible authorities that could help develop a quality seed-supply system for farmers. Figure 1: Variation in shape and colour of coconut inflorescences / Photos: R. Bourdeix From left to right: Malayan Red Dwarf (MRD), Niu Leka Dwarf (NLAD) from Fiji, Tacunan Green Dwarf (TACD) from the Philippines, and Rennell Island Tall from Solomon Islands 4 Coconut palms show extraordinary varietal diversity, expressed by the colour, size, shape and composition of their fruits. Varieties also differ by the size and growth of the trunk, their fruit and sap-producing capacities, and their tolerance to diseases, pests and climatic hazards. There remains limited knowledge of this diversity among the public. It is underexploited at cultural, commercial and industrial levels. Over time, farmers have created this array of diversity and of varieties. However, their selection has been long and difficult due to the unique way in which coconut palms reproduce. Figure 2: Diversity of coconut varieties and forms in French Polynesia / Photo: R. Bourdeix 5 2. BIOLOGY AND DIVERSITY OF COCONUT VARIETIES 2.1. Four kinds of coconut palms and their possible hybrids Most coconut palms (90‒95%) are characterised as Tall types (Talls). Because they readily cross-pollinate, they generally form mixed and heterogeneous populations that are fastgrowing, to a maximum height of about 30 metres, with an economic life span of 60–70 years. Around 5‒10% of coconut palms are characterised as Dwarf types, often called Dwarfs, Fragile Dwarfs or Malayan-Type Dwarfs. They grow more slowly, generally reaching a maximum height of about 15 metres. They have a 30–40-year productive lifespan and usually start flowering 12–30 months after field planting. Male and female flowers open at the same time; self-pollination occurs within the same palm. Because of their prolific fruiting and rapid flowering, they play an important role in hybrid seednut production. The remaining 5‒10% of palms are described as either Compact Dwarf types or Semi-Tall types. In Compact Dwarf types, the male and female flowers usually open at different times. They have a short thick stem and wide leaflets and are mainly found in the Pacific region. The Semi-Tall types possess features that lie between those of Dwarfs and Talls. The most famous is the King Coconut from Sri Lanka. Table 1 summarises the main features of these four types, as described in the literature [6] Farmers, gardeners and scientists can make crosses using parents from either within or between these four different types of coconut palms. Crosses are generally made by using two distinct coconut varieties, with the pollen recipient the mother palm and the pollen donor the father palm. Such crossings are generally called “hybrids”. Hybrids can be made using two different types of coconut palms, or within the same type. The most common hybrids are crosses between Dwarf and Tall types, followed by Tall x Tall. Dwarf x Dwarf hybrids, and hybrids using the Compact Dwarfs7 and Semi-Tall types as parents, have not yet been released to farmers, although some have been tested experimentally by breeders in research centres. Hybrids are generally not foreign varieties (which are not well adapted to local conditions). They are generally made using local or long-established varieties, well-suited to local conditions, as shown in Figure 3. It is relatively easy to teach farmers how to create their own, better-performing hybrids from varieties already present in their fields and gardens. Producing and selling hybrid seednuts offers an attractive business opportunity, in particular (but not only) in countries where official agricultural services produce few or no hybrids. 7 There is an exception to this. In 1926, in Fiji, the agronomist Marechal, forced by the economic crisis to leave Fiji, distributed to farmers seeds of hybrids between the Malay Red Dwarf and the Niu Leka (Compact) Dwarf. This is one of the most strategic actions that a coconut breeder has undertaken, as the farmers and gardeners have continued this breeding work for 100 years. From this hybrid, and perhaps other sources, these farmers created the magnificent range of Compact Dwarf varieties existing today in Fiji, some of which have been exported throughout the Pacific islands. This example of collaboration with farmers deserves to be reproduced worldwide. 6 Figure 3: Various types of hybrid coconut palms / Photos R. Bourdeix 7 Table 1: General forms of the four main types (varieties) of coconut palms in normal growing conditions Coconut types Breeding method Growth rate Bole size Time to flowering (years) Bunches per year Fruit size Economic life (years) Type Other names Speed cm/yr Tall Normal or ordinary coconut palm Out-crossing; mainly between different palms Fast growing 50– 100 Thick 5–10 12–14 Small to large 60‒70 Semi-Tall King coconut in Sri Lanka Outcrossing or selfcrossing Medium growing 30–40 Med. 5–10 12–14 Small to medium 30‒40 Dwarf Fragile Dwarfs or MalayanType Dwarfs Self-crossing; inbreeding Slow growing 15–30 Thin 3–4 12–16 Small to medium 30‒40 Compact Dwarf Niu Leka-type Dwarfs Out-crossing; mainly between different palms but can be inbreeding Very slow growing for some varieties 6–20 Thick 4–5 12–16 Small to large 30‒70 8 2.2. Differentiating between coconut palms The coconut palm consists of a trunk (or stem) topped with a broad crown of fronds. In the axil of each frond, there is usually an inflorescence that develops into a bunch loaded with coconuts. Depending on the coconut types, as shown in Figure 4, the stem may widen at the bottom to form a bole that increases its resistance, particularly to cyclones. The stem is relatively smooth and pale in colour, with regular markings: each frond produced by the palm leaves a crescentshaped leaf scar. The red dots and yellow arrows in Figure 4 show the distance between 11 leaf scars, corresponding to 10 internodes. It is possible to distinguish between the two main types of coconut palm, Talls and Dwarfs, by the gaps between the scars. In Tall palms, the gap between two leaf scars is over five centimetres, whereas it does not exceed twoand one-half centimetres in Dwarfs. Figure 4: Identification of main coconut varietal types by the shape of the stem (trunk) / Photos R. Bourdeix The yellow arrows show the length of 10 internodes, or the distance across 11 leaf scars. Four varietal types are presented from left to right: Compact Dwarf; Malayan Type Dwarf or Fragile Dwarf; Tall from Africa and Indian Ocean; Tall from Asia and the Pacific region. Where the stem enters the ground, it takes the form of an upturned cone, called the “root bulb”. Several thousand fine roots grow from all over the root bulb, forming a dense felt mat that is mostly distributed in the first metre of soil, although some roots reach depths of four to five metres. It is generally easy to distinguish between these four types of coconut palm. However, distinguishing between all coconut varieties and forms tends to be more challenging. It is typically easier to identify varieties for Dwarf, Compact Dwarf and Semi-Tall types, as these tend to be homogeneous. For some Tall types, which are generally more heterogeneous because of their reproduction pattern, however, it can be more challenging to distinguish between the varieties. For farmers and gardeners, well-illustrated catalogues of coconut varieties and forms are the most useful tools for recognising distinct coconut varieties. Figure 5 provides an example of a variety in the COGENT Catalogue [7]. The description includes a page of text, with standard sections, and a one-page photo plate with six to seven standardised images of the variety. The 9 example provided in the figure is the Rennell Island Tall, a famous variety from the Solomon Islands. The importance of cataloguing coconut varieties is discussed further in Section 5 of this guide. Figure 5: Standardised description of a coconut variety: Example of the Rennell Island Tall (RIT) from Solomon Islands8/ Photos: R. Bourdeix 2.3. Naming of coconut varieties and populations Farmers’ naming conventions Many farmers are only familiar with common names of coconut varieties and forms and may not know the international systems of naming these varieties. It is appropriate, after all, for farmers to refer to plants using the names that are most familiar and useful to them. In addition, farmers use their own language and dialects to refer to plants and their varieties. Often, they use pleasant-sounding, poetic or remarkable names that are easy to remember, like the examples provided below. • “Niu potopoto” designates a Red Dwarf, producing large fruits in Cook Islands. • “Nei moto” designates a Tall type producing an array of fruits in Beru island, Kiribati. • “Ni biroro kadru” designates a Red Dwarf in Arno Atoll, the Marshall Islands, a name that was likely introduced from Fiji but originating from Malaysia. • “Niu eisefou” designates the Dwarf x Tall hybrid produced at the Elisefou Seed Garden in Tuvalu. The different forms of coconut palms are also often referred to by their colours. Farmers sometimes use the term "red" to refer to the Tall-type coconut palms that display various shades 8 See: pp. 243–244 of the COGENT catalogue of conserved germplasm [7]. 10 of brown. The main use of a variety is also often incorporated into its name. For example, "Niu afa" in Samoa and "Niu kafa" in Tonga designate the same type of tall coconut palm with huge, elongated green nuts, whose fibre (called “afa” or “kafa”) is used to make ropes. However, in Tuvalu, "Niu aafa" designates a dwarf coconut palm with light red fruits. Additional discrepancies were observed in 2000 in the Cook Islands (Figure 6). In Aitutaki Island, Vaipeka Village, male and female farmers were noted using different names for the same variety, the immature fruits of which had a red ring around the peduncle. The men called this variety "Red eyes" and the women, more poetically, called it "Red lips". Another example comes from India. At the beginning of the process of creating the nomenclature, Indian researchers and farmers designated one of their main varieties "West Coast Tall". However, given there are “west coasts” throughout the world, including on each island, the potential for confusion led to the renaming of this variety at the international level – to "Indian West Coast Tall". Nonetheless, in everyday conversations, Indian farmers and scientists continue to call the variety "West Coast Tall", which again is appropriate. Figure 6: Two ways of naming the same coconut variety in the same village The variety displays a pink ring around the peduncle of its immature fruits. Named as “Red lips” (Women) and “Red eye” (Men) / Photos: R. Bourdeix In conversation with other farmers or scientists, especially at the international level, it is important to keep in mind that the names one finds familiar and useful may not be familiar or useful to others. In two distinct regions, the same name may be used to designate two different varieties or, conversely, the same variety may be designated by different names. Scientists and breeders’ naming conventions Since the 1980s, scientists and breeders have been working to standardise the names of coconut varieties, so they are more broadly understood. In 1986, Dr Narong Chomchallow, a famous Thai researcher, started the process9 [8] and laid the foundations for the current classification system. The International Workshop on Coconut Genetic Resources, held in Indonesia in October 1991, recommended setting up the International Coconut Genetic Resources Network (currently known as COGENT) and creating a Coconut Genetic Resources Database (CGRD). In 1992, researchers and curators from many national coconut genebanks defined the international nomenclature system, designating coconut varieties and populations conserved as “accessions” (see Glossary of key technical terms) in their field genebanks. An accession is generally named immediately after the material has been planted in the ex-situ genebank. These names typically become “international” only when the data on the collected germplasm has been transmitted to the CGRD, serving as the international reference. The 9 The unpublished list provided in 1986 remains available [8] 11 COGENT secretariat, which has previously acted as CGRD manager, must ensure that any new name meets the nomenclature rules. If necessary, the secretariat should interact with national and international researchers to adjust the proposed name accordingly. For instance, a new international name must not duplicate any previously recorded name (including synonyms and abbreviations). Naming convention for international varieties A variety name should consist of two parts, not exceed 30 characters, and be in English (except perhaps the first part). A new international name must not duplicate any previously recorded name (including synonym names). The first part of the name includes at least one of the following: • a vernacular variety name; • a place or region or country of origin; • a prominent biological trait of the variety; • an ethnological trait linked to the history of the variety; • a colour, if the cultivar is homogeneous for fruit colour (Self-pollinated Dwarfs); or • a combination of these. The name of the country of origin (or the corresponding adjective) is optional but may be the only element of part one. If it is not the only element of part one, the other elements must be placed after the country name. Unnecessary words, such as “local”, “ordinary” or “coconut”, should be avoided. Examples of acceptable names are provided below. • Agta Tall is derived from “Agta”, a vernacular name describing a prominent trait (blackish necrotic pericarp). • Markham Valley Tall uses the name of a valley in Papua New Guinea. • Indian West Coast Tall was renamed from the local name, “West Coast Tall”. “Indian” is desirable here because “West Coast” alone can refer to many places in the world. • Andaman Giant Tall includes a combination of a place (Andaman) and a prominent trait. • Malayan Yellow Dwarf includes a combination of a place (country) and a colour. • Raja Brown Dwarf includes a combination of an ethnological trait and a colour. The international abbreviation of a variety consists of three to four letters, followed by two numerical digits. An abbreviation must not duplicate any previously recorded abbreviation (preferably including synonym abbreviation). The abbreviation also indicates the type, by using the letters T or D when applicable. For instance: PRD for Pemba Red Dwarf (autogamous Dwarf with homogeneous colour); WAT for West African Tall; and RTB for King Coconut (synonymous with Rath Thembili, semi-tall). The first part of the abbreviation is a mnemonic10 of the full name, as shown in the examples below. • WAT stands for West African Tall (cross-pollinated Tall with various colours within the cultivar). • BAYD stands for Bali Yellow Dwarf. This acronym is preferred over the previous AYD, which was a poor mnemonic, and over BYD, which is already used to designate Brazilian 10 A device, such as a pattern of letters, ideas or associations, that assists in remembering something (e.g. Richard Of York Gave Battle In Vain to help recall the colours of the spectrum [red, orange, yellow, green, blue, indigo, violet]). 18 2.6. Colours of coconut fruits and sprouts Depending on the coconut variety, the colour of the young fruits, inflorescences and small seed sprouts varies from pale yellow to red-orange, and many shades of green or brown, as shown in Figure 12. Figure 12: The four basic colours of young seednut shoots / Photos: R. Bourdeix The four colours are represented in the photos as follows: Top left to top right: green, brown, red, yellow Bottom: the corresponding fruit and inflorescence colours There are more subtle variations in shade, but the basic colours can still be recognised. However, it is important to record observations before the sprouts are five centimetres long, as the effect of the sun can cause discolouration or colour changes. 19 The colours of coconut fruits13 and shoots are important for two main reasons, outlined below: • Visual and cultural appeal: The colours are beautiful to see, and the coconut palm is an ornamental plant, widely appreciated for its decorative appearance in gardens and public spaces. People enjoy seeing yellow, orange, red or deep green coconut fruits. In several Pacific region traditions, some of its colours are even associated with medicinal or magical properties. • Differentiation between varieties and hybrids: The colour of the young sprout is an early marker that can allow for differentiation between the different varieties and hybrids just after their germination. Inserted in the coconut meat, an embryo measuring around five millimetres in length lies beneath one of the three germination "eyes" of the coconut. At germination, part of this embryo will grow inside the coconut and feed on coconut meat and water. The other part will pierce the germinative pore (one of the “eyes”), pass through the husk and start to point outside as a sprout. Not all varieties of coconut germinate at the same speed or with the same success rate.14 A quality control step can be introduced in the nursery to improve farmers’ selection of seednuts for sale or use, based on the colours of the young shoots of the seednuts (see Figure 12). By using these sprout colours, some illegitimate seedlings can be removed or separated. However, this is only effective if observed on small sprouts under four centimetres in length. If the seedlings exceed five centimetres, this method should not be used for identification. We recommend marking the seedlings in the nursery with a trace of coloured paint as soon as the sprout appears. This will save the nursery staff time and ensure the best identification. Before planting seednut gardens and teaching controlled pollination to farmers, it is important to agree on how to name the different colours of the coconut palm and, more specifically, the colours of the young fruits and young sprouts, before they change. As shown in Figure 13, the red colour can be problematic. Very often, farmers call coconut palms with brown or reddish-brown nuts "red", but these are not a true red colour. For their part, scientists call coconut palms “Red Dwarf” which are also not truly red but, rather, different shades of orange. Nevertheless, for simplification, these shades of orange are nonetheless referred to as “red” for the purposes of this guide. 13 The fruits of all coconut varieties are organised in a similar way. An epidermis (skin) is first coloured in yellow, orange, red, green or brown, or all intermediate shades. There is generally only one fruit colour per palm, or much more rarely two colours. After 11–14 months, when the fruits mature, their skins turn to different shades of a greyish-brown colour. 14 A few rare forms germinate on the palm, even before the coconut falls to the ground. Most varieties take between two and four months to reach 90% germination, but some Mexican varieties have had nuts germinate 238 days after sowing, or nearly eight months. 20 Figure 13: Coconut colours / Photos: R. Bourdeix The varying shades of coconut colours are depicted in the figure as follows: Top left: colours of the immature fruits of the coconut palms Top right: perception of red colour by farmers and scientists Bottom left: shades of orange Bottom right: shades of red The denomination of the colour “red” is highly debated by scientists and farmers. Scientists tend to refer to shades of orange as “red”, while farmers tend to refer to shades of brown as “red”, except for the "brick" red shade. As a result, farmers generally make fewer mistakes than scientists with respect to the colours of coconut fruits. Nonetheless, out of respect for past work and to avoid duplication, this guide still refers to the orange fruit colour of the famous variety, incorrectly called "Malayan Red Dwarf", as "red". The simpler method, for farmers as well as for breeders, is to cross (or plant together in the same field) a “red” variety with a green variety, excluding all shades of brown (Figure 14). In this case, when the coconuts first start to germinate, the sprout colour enables the varieties to be distinguished from one another: 21 • Red sprouts: the red variety. • Green sprouts: the green variety; and • Brown sprouts: hybrids or crosses between the red and the green varieties. Figure 14: Illustration on how crossing green and red coconut palms simplifies and optimises the process / Photos: R. Bourdeix The system is simplified and optimised by producing Brown hybrids that can be selected true-to-type in the nursery as a result of the sprout colour. When crossing or planting together a yellow and a green or brown variety: • Yellow variety → yellow or hybrid seednuts: Seednuts harvested from the yellow variety will be mainly: 1) the yellow variety if the sprout is yellow; or 2) the hybrid if the sprout is green or brown. • Green and brown variety → unknown seednuts: It is not possible to determine if seednuts harvested from the green and brown variety will be hybrids or if they will be of the green or brown parent variety. When crossing a green variety with a fixed15 brown variety: • Green variety → green or hybrid seednuts: Seednuts harvested from the green variety will be: 1) the green variety if the sprout is green; and 2) the hybrid if the sprout is brown. • Brown variety → unknown seednuts: It is not possible to determine if seednuts harvested from the brown variety will be hybrids or the parent variety. In several Dwarf and Tall coconut varieties, a pink colour can be found in the young-fruit husk, the centre of young roots and the very young sprout. These characteristics, for the roots and the 15 To know if a brown-fruited coconut palm is fixed for brown colour, take 20 open-pollinated seednuts and observe the germination in the nursery. If all the sprouts are not of various shades of brown (meaning some are of yellow, red, or green colours), it is not a fixed colour. 22 young shoot, can also be used as a recognisable genetic marker for its desirable characteristics (Figure 15). Many of these pink-coloured coconut varieties are considered medicinal, and their tendernuts are generally sold at about twice the price of “normal” coconuts. It, thus, could be a profitable business for farmers to produce these coconut varieties. Figure 15: Illustration of the pink colour found in some coconut varieties / Photos: R. Bourdeix The pink colour is depicted in the photos, as follows: Top left: the pink young sprout and root tips Top middle: another younger pink young sprout Top right: the pink base of the young female flowers Bottom: the pink colour of young coconut husk. Note: this colour is genetically recessive and serves as a genetic marker in the nursery. If a variety has this pink colour, its legitimate progeny must also be pink inside. If not, this is an illegitimate seednut. Although the root tip is not always as pink, there is always a central pink line inside the root tip. 23 3. WAYS FARMERS CAN PRODUCE BETTER, MORE DIVERSE SEEDNUTS Drawing on workshop discussions between farmers, practitioners and scientists, and the technical considerations outlined in this guide, a two-fold strategy is recommended: 1. Local production: Farmers can produce quality seednuts independently, using a series of simple and cheap techniques. 2. Commercial production: A few larger seednut gardens can be established on a commercial scale, which will have higher production costs but will ensure a more regular and larger supply of quality seednuts of Dwarf, Tall-type and hybrid varieties. We propose new designs, in recognition of the varieties using sprout colours, allowing production beyond just hybrids in these seed gardens. Teaching farmers to produce their own seednuts will also increase their acceptance of hybrids produced elsewhere. As part of a more sustainable agroecologically intensification approach, which allows farmers to more efficiently select and improve their own coconut varieties, coconut-based livelihoods and the whole coconut value chain will be enhanced. Section 2 explains coconut biology and how coconut palms reproduce in nature. In the context of coconut improvement and conservation, this third section explains how breeders and scientists produce coconut seednuts. It also shows how their methods can be adapted and simplified for farmers to autonomously produce better quality seednuts for themselves. 3.1. Brief overview of genetics and plant breeding Like other plants, coconuts sexually reproduce when a pollen grain from the father palm fertilises the female ovule, located in the female flower (see Section 2). Each pollen grain carries a single “segregated” copy of about 30,000 father palm genes (called “alleles”). Each allele contains genetic information (together called “genotype”) that controls specific characteristics expressed in the field (phenotype) by the father palm, such as palm height, disease resistance, nut size, and seedling sprout colour. The female ovule also contains a single segregated copy of each of the mother palm’s alleles. The fusion of the pollen and ovule give rise to a new embryo palm carrying paired/de-segregated alleles (called genes) from both parents. These are the distinct hereditary units that are passed from parents to offspring (embryo). For example, as illustrated in Figure 16, a “purebred” father palm with “green” embryo sprouts will have a pair of alleles that can be labelled “G/G”. A purebred mother palm with “yellow” sprouts has alleles that can be labelled “g/g”. When the parental generation “G/G” and “g/g” cross, combining their genes, the seednuts – labelled the F1 (first progeny) generation – will have the genotype G/g and the phenotype green sprout. The allele G coding for the green colour is called the dominant allele, and g the recessive one, because G wins over g and the genotype G/g gives a green sprout phenotype and not a yellow one. 24 Figure 16: Mendelian genetics of coconut embryo sprout colour / Photos: R. Bourdeix 25 All the F1 progeny will have the same paired alleles G/g, one coming from the mother and the other from the father. If we plant these seednuts and, after three to five years (when they produce pollen and ovules), we cross these F1 plants with each other, they will produce a F2 (second progeny) generation. Approximately three quarters of the F2 progeny will produce green sprouts (genotypes G/G and G/g); the remaining quarter will produce yellow sprouts (see Figure 16). Although the genetic process for a single gene of the coconut palm (i.e. the process that controls the green and yellow colours of young fruits and sprouts) has been outlined above, it is important to note that the entire coconut palm genome has been estimated to contain more than 28,000 different genes. [11] Thus, each time two coconut palms are crossed, the process described for the yellow and green colours will occur for all of the other genes as well – about 28,000 times. This offers some insight into the complexity of life’s reproduction mechanisms. When the ways in which traits are inherited is understood, selections and hybrids/crosses can be made to produce offspring that will be “true to type” and express the traits desired. The goal of plant (coconut) breeding is to produce (coconut) varieties that possess unique and superior traits of interest [12]. Different breeding approaches are required for self-pollinating, cross-pollinating and clonally propagated plants, and coconut falls into all three categories. 3.2. Brief overview of coconut seednut production methods The topic of open pollination, the natural way that coconut palms reproduce, is covered in Section 2.4. A more detailed description is provided in the following two sections. There are other, more controlled methods of seednut production that require greater input in terms of labour, skill, time and money. The method used will depend on the resources available to the farmer. Five possible methods for producing coconut seedlings are described below. 1. Open pollination: • Coconut palms reproduce naturally through open pollination. • However, natural reproduction brings together various coconut palm reproduction strategies, ranging from systematic self-fertilisation to systematic crossing between two palms, through all of the intermediate balances. 2. Controlled natural pollination: • Two or more varieties are planted in the same isolated seed-garden field. • As soon as the inflorescences open (and even one or two days before, for some Dwarf varieties like Cameroon Red Dwarf,16 as outlined in Section 3.4), the male flowers are removed from all of the varieties except one. • The pollen of the variety chosen as the male parent is allowed to naturally pollinate the other varieties in the field. 3. Assisted hand pollination: • One variety or more is planted in an isolated seed-garden field (as generally, but not always, is done with Dwarf types). 16 In some Dwarf coconut varieties, male flowers may mature and open even before the inflorescence has opened (still enclosed in the spathe). In this case, the spathe must be opened before its natural opening to avoid pollen contamination. 26 • As soon as the inflorescences open (and even one or two days before, for some Dwarf varieties), the male flowers are removed from all of the varieties in the seed garden. • Pollen from another field planted with the chosen male parent is then harvested and prepared. • A mix of pollen and talcum powder (or another diluting powder, such as flour) is applied to the receptive female flowers in the isolated seed garden. 4. Controlled pollination with bagging: Until now, this technique has mainly been used by breeders and researchers. They favour a heavy and expensive, but extremely reliable, variant of the technique. This is justified because, if the objective is to conduct planting experiments in the field and monitor them for at least 10 years, it is necessary to ensure that the crosses planted are legitimate (i.e. the intended crosses). The methods used by scientists and breeders generally consist of bagging entire inflorescences using the following steps: • For the female parent, the inflorescence is bagged just after completely removing and destroying all of the male flowers (emasculation). • For the male parent, the whole inflorescence is also bagged, and later the male flowers are harvested inside this special bag and transferred without any risk of contamination. • The pollen is extracted and dried, palm by palm, following a strict protocol avoiding contamination. • The pollen is then spayed for pollination inside the bag containing the emasculated inflorescence used as female. One of the main objectives of this guide is to propose simpler and cheaper methods of controlled pollination with bagging that can be used by farmers in their fields, even if these methods are somewhat less reliable than those used by researchers. 5. Cloning: Plant cloning means producing plants that are genetically identical to their original parent plant. Many crops can be cloned naturally through different methods. For instance, the domesticated banana and cassava plants are mainly reproduced by cloning, while the coconut palm naturally reproduces by seednuts. Some common methods of cloning plants are cutting, grafting, layering, and tissue culture. In the case of coconut palm, researchers have been trying for decades to produce clones by growing small fragments of a living coconut palm in glass tubes containing a nutrient medium (Figure 17). Although recent progress has been made, the technique is not yet ready for largescale production. For now, researchers can only reproduce coconut palms from fragments of their embryos. As a result, they are not able to reproduce identical coconut palm seedlings/plants – only the progeny of that coconut palm, which may not be genetically identical. This is the primary limitation of the present technique. 27 Figure 17: Coconut embryos growing in glass tubes in a laboratory / Photo: R. Bourdeix 3.3. Dealing with open-pollinated seednuts Natural (open) pollination is the natural way by which coconut palms produce seednuts. Currently, over 90% of the seednuts used by farmers still come from open pollination. Even if there is an interest in changing how seednuts are produced, it is necessary to adapt to natural (open) pollination and, where possible, propose methods to improve the value of seednuts obtained by open pollination. As already mentioned, 90–95% of Dwarf-type coconut palms reproduce through self-pollination. This means that, if red-, yellowor green-fruited Dwarfs are planted, mostly red, yellow and green sprouting seednuts will be produced. That is, the seednut sprouting the same colour as its mother palm will be “true to type”, or “the same variety as the parent palm”. For this reason, open pollination is perfectly appropriate for reproducing most Dwarf-type coconut palms. Tall-type coconut palms generally cross-pollinate17 but can also self-pollinate. When Tall-type coconut palms self-pollinate, the outcomes are comparable to that of reproduction between two members of the same human family, where their offspring can suffer from “inbreeding depression”. This includes physical problems of weakened vigour or health (and mental problems in the case of humans and other mammals). Because of inbreeding depression, self-pollinating Tall-type coconuts typically produce between one quarter and one third less than inter-crossing coconuts. When using open-pollinated seednuts harvested from the same coconuts, the proportion of self-pollination, and therefore the value of the seednuts, varies with the seasons. In fact, when the coconut quickly produces 17 As explained above about the flowering times, if male and female flowering overlaps, the palms may self-pollinate. If the flowering times do not overlap, the female flowers can only receive pollen either from another palm or from a later-flowering inflorescence on the same palm. 34 The father palms should be planted in a separate field, but one does not necessarily have to personally own that field. For example, one could negotiate with nearby farmers to collect pollen from their best palms, or one could request pollen of good varieties from the national agricultural research institution (as is done in the Philippines). Sometimes, good pollen can be purchased from specialised private companies or national institutions. Harvesting the male flowers Harvesting is undertaken with the variety chosen as the male parent. If all coconut palms for pollen collection are grouped together, each pollen collector can visit about 150 palms per day and collect 20–30 kilograms of fresh male flowers. The harvester cuts the spikelets as follows: • from the spikelet top and the middle part six to eight days after the natural opening of the spathe; • from the spikelet bottom between 10 and 14 days after the opening; or • in a single sampling between eight and 10 days after the opening, but with a slightly lower flower/pollen yield. Once the technician has finished visiting and harvesting pollen from all the palms, s/he removes the spikelet stems while still in the field (Figure 23). To do this, s/he places the spikelets on a tarpaulin and de-stems them by hand (ideally protected by a glove) in order to obtain male flowers detached from the spikelets. The male flowers are then placed in a plastic bag, which is carefully closed, and subsequently transported to the place where the male flowers will be dried and the pollen will be extracted. 35 Figure 23: Destemming spikelets harvested from male parents to collect male flowers for pollen production / Photo: R. Bourdeix Drying and extracting pollen from the male flowers The number of people needed to prepare the pollen depends on the daily amounts of fresh flowers to be processed. In most cases, one person is enough to ensure pollen conditioning and quality control. The flowers are crushed and then placed in a drying room with electric dehydrators. The next day, the dried flowers are sieved. Crushing and sieving can be mechanised, reducing the personnel required. The steps are outlined below. 1. The male flowers are crushed with a simple roller above a metal grid with two to three millimetres mesh, which prevents male flowers from passing through it. Where several male flowers must be treated, it is possible to pre-crush the male flowers mechanically. 2. The ideal conditions for pollen drying: • For large quantities of male flowers: dry at ~30°C and 45–50% relative humidity. • For small quantities of male flowers: a ventilated oven set at 39°C will suffice. In this case, the drying time should not exceed 48 hours, and the process should produce pollen with a moisture content of ≤13%. 3. The crushed and dried male flowers are removed from the room or oven to be sieved. The handler then places them in a sieve and, by repeatedly shaking, obtains the pollen. From this, between two and two and a half per cent of the pollen is extracted, relative to the weight of fresh male flowers treated. At this stage, the pollen has a moisture content of 10–12%. 36 Pollen conditioning The sieved pollen should be used within 15 days. The pollen is immediately placed into sealable plastic bags, in five-gramme doses. It is then stored in the freezer at -20°C and removed from the freezer on the morning it is to be used. If the pollen needs to be conserved more than 15 days, an immediate second phase of drying will be needed, followed by vacuum packaging. It is also possible to freeze-dry the pollen, according to the protocol presented in the STANTECH Manual [9] Emasculating the female parents Emasculation consists of removing all the male flowers from an inflorescence. (See Section 3.4 where this technique is described.) Carrying out pollination Five grammes of pollen are mixed with 100 grammes of talcum powder. This mixture is sufficient to pollinate about 150 coconut inflorescences in the same day. The pollination technician can visit 1500 palms per day. S/he sprays a mixture of talcum powder and pollen on the inflorescences that are showing receptive flowers.19 If the inflorescences are located more than two metres above ground, the technician may carry out these pollinations from the ground, with an apparatus consisting of a squeeze bottle, plastic pipe and rubber bulb, all attached to a long pole (Figure 24) generally made from long, thin bamboo.20 In large seed gardens, monitoring is carried out by supervisors who oversee all of the abovedescribed operations: male flower harvesting; pollen preparation, conditioning and control; and emasculation and pollination. Each supervisor oversees seven emasculation technicians or four to five pollination technicians. A general supervisor oversees the supervisors and, thus, is responsible for the quality of all technicians’ work in the seednut garden. 19 The technician is trained to recognise precisely when a female coconut flower becomes receptive to pollen – those that already have nectar drops oozing from their lower ends (as shown in Figure 27, Section 3.7). 20 For both emasculation and pollination in the seed garden, see our video: https://www.youtube.com/watch?v=mVB1TOr97OE. 37 Figure 24: Apparatus for making coconut pollinations from the ground / Photo: R. Bourdeix The apparatus consists of a squeeze bottle, plastic pipe and rubber bulb, all attached to a long pole generally made from bamboo. 3.6. Controlled pollination with bagging The version of this method currently proposed in many publications is used only by researchers and breeders who apply it for scientific experiments or when replanting varieties conserved in a field genebank. However, this extremely reliable method is cumbersome, unproductive and expensive. As noted above, when setting up a field experiment that will last 10 years, it is essential to ensure legitimate crosses. In this section, the method used by scientists and breeders is briefly explained and, in the next section, ways to adapt and simplify this method for farmers’ use are provided. Polythene squeeze bottle Rubber bulb Fixings Flexible tube Pole Pollen-Talcum powder mix Fixings 38 Controlled pollination with bagging requires the following steps: • Prepare two types of bags with a special fabric permeable to air and impermeable to pollen (e.g. see Figure 25). One, the “female bag”, will be used to protect the emasculated inflorescences. The other, the “male bag”, will be used to bag the inflorescences of the male parent for collecting the pollen. Figure 25: Bags used in controlled pollination: female bag (left); male bag (right) (This example is adapted from: https://www.pbsinternational.com/our-products/.) • Emasculate the female parent as soon as the inflorescences open (and even one or two days before for some Dwarf varieties) to remove all the male flowers from the inflorescence. • Immediately after the emasculation, isolate this inflorescence in a female bag. This bag should have a plastic transparent window to assess the stage of maturity of female flowers, as well as a small hole in the centre of the transparent window covered with adhesive tape. • Separately, use the male bag to isolate the inflorescence of the male parent. When the male flowers become mature, harvest them, dry them and extract their pollen by following an isolation process to prevent any external pollen contamination. • Prepare a mix of dehydrated pollen and talc (talcum powder) from the palm chosen as the male parent in a squeeze bottle (see Section 3.5). • When the female flowers are receptive, briefly detach the adhesive tape, insert the tip of the squeeze bottle into the hole, spray the female flowers with the mixture of talc and pollen, and again plug the hole with the adhesive tape. • When all the female flowers are necrotic and have a black tip, remove the bag. 3.7. Toward seednut production methods tailored to farmers For their own needs, most farmers do not require large quantities of coconut seednuts. A typical farm size ranges from one-half to two hectares. Thus, to replant this entire area, farmers would need about 100–400 seednuts. Replanting can be undertaken all at once, or gradually over a period of 30–40 years. In the latter case, only 10–50 seednuts per year may be required. Farmers can produce an estimated 50 seednuts by spending a maximum of one full day of labour per year, 39 and by using inexpensive equipment. Because farmers need only a few seednuts, they can apply high selection rates on parent palms. For instance, they can cross their best palm with their second-best palm, or with the neighbour’s best palm. This will increase breeding efficiency. Where the farmer seeks to develop a business to sell coconut seednuts, the seednut quantities needed would be greater. The ways in which the four different seed production methods described in the previous four sections (Section 3.3–Section 3.6) can be adapted to small planters are outlined below. 3.7.1. Natural (open) pollination The advantages and (many) disadvantages of natural (open) pollination are discussed in Section 3.3. Tips are also provided, often identifying germ colour in the nursery seedbeds to try to improve the efficiency of this method. Natural (open) pollination can be used in certain cases by farmers as a technically simple, albeit not particularly efficient, method for producing hybrids. For this method to be feasible, the farmer must have Yellow or Red Dwarf coconut palms, surrounded by Tall-type coconut palms , without any hybrids nearby. In this case, the farmer will collect Dwarf seednuts from open pollination, germinate them, and use the colour of the sprout to recognise the hybrids: o n Yellow Dwarfs, any seednuts harvested that do not germinate yellow are hybrids; on Red Dwarfs, any seednuts harvested that do not germinate red are hybrids. However, with this technique, the farmer will only obtain 5–15% hybrids from unknown father palms among the seednuts s/he has germinated. This method has been applied by Indian farmers in Kerala. They attribute great value to the rare, naturally brown sprouting seednuts harvested from their Chowgat Orange Dwarfs (COD) in the villages [16] but rejected the idea of COD x Tall hybrids proposed by researchers. However, these two resulting crosses – natural and research-made hybrids – are genetically similar. Thus, the preferences of farmers are often less correlated with the palm’s biological characteristics than with its qualities as a cultural entity within a human community. 3.7.2. Controlled natural (open) pollination Controlled natural (open) pollination is described in Section 3.4. It is important to consider the limiting factors that could prevent farmers from using this method and how these limitations can be reduced. 1. Isolating the seed garden can be difficult for farmers to achieve. However, even if the isolation is less than perfect, a proportion of at least 50–80% of hybrids will still be obtained in the progenies, far more than from completely open pollination. The main precaution is to avoid using this method when hybrid palms are already growing close to the coconut palms intended as female parents, as this may compromise the genetic purity of the seednuts. 2. Emasculating the inflorescences is rather easily achieved if the coconut palms are still young. If the coconut palms are already tall, however, it can be difficult to reach the inflorescences for performing emasculation. 3. Identifying the coconut sprout colours in the nursery seedbeds is relatively easy, if done early when the sprouts do not exceed five centimetres in length. If too much time has passed, some brown or green sprouts will take on a bright orange colour under the sun's 40 rays. It is advisable to mark the sprouted seednuts early using different colours of paint to differentiate the sprouts and more easily identify the varieties when the plants have grown. 4. When planting a new seed garden, where Dwarfs and Tall types are customarily planted together, the Dwarfs start producing two years before the Tall types. Thus, during these two years, a seed garden producing female parents will result but no pollen will be available from the male parents. A solution would be to bring external pollen to the seed garden for assisted hand pollination, as described in Section 3.5. Another option is to modify and organise an already existing plantation of Tall types to turn it into a seed garden, as shown below and in Figure 26. Even as a small farmer, unable to create a large seed garden, the method of assisted (open) pollination can be applied. Plant some Yellow or Red Dwarfs near Tall types, but without coconut hybrids around them. Make the emasculations on the Dwarf coconut palms. On Yellow Dwarfs, the seednuts harvested that do not germinate yellow are hybrids. On Red Dwarfs, the seednuts harvested that do not germinate red are hybrids. With this technique, if the emasculations are well done, the farmer will obtain 90–95% of hybrids among the seednuts s/he has germinated. As shown in Figure 26, it is possible to convert a plantation of a single good Tall-type variety into a seed garden producing Tall types, Dwarf x Tall hybrids, and Dwarf seednuts. Generally, the Talltype variety has fruits that are brown, brown-green, and green. To turn the plantation into a seed garden, the following steps should be taken. 1. Choose an area for the garden: In the middle of the field, choose an area containing 50– 500 coconut palms (depending on the seednut needs), surrounded by other palms of the same variety. 2. Cut down all coconut palms not producing green nuts: In this area, and around this area within a 20–30 metre radius, cut down all coconut palms not producing green nuts. Where green-fruited palms produce little or poor-quality fruit, cut these palms too.21 3. Plant red or orange Dwarfs in the area chosen for the seed garden: As indicated before, if Dwarf x Tall hybrids are not desired, instead of Red Dwarfs, use red Tall types. However, the latter are more difficult to find and will flower about two years after the Dwarfs.22 4. Consider options: It is possible to plant these dwarfs before cutting the Tall brown or green-brown coconut palms, and to wait for the red Dwarfs to grow or even to produce before cutting. However, in this case, the red Dwarfs will require much more time to grow. 5. Complete the planting: Add good, green-fruited coconut palms of the same variety as in the unplanted areas. To avoid inbreeding, it is best, if possible, to source these green fruits from the same variety but from another plantation. Waiting only for all new red palms to start producing, the seed garden will be ready to be managed using the controlled natural pollination technique. 6. Note the results: These palms will produce Red Dwarf, Green Tall, and Brown Hybrid seednuts. 21 See also the video: https://www.youtube.com/watch?v=j0hqT3eEnJ8 at 12.41 minutes. 22 It is also possible to set up a plantation of Green Dwarfs or Compact Green Dwarfs with this system, resulting in the production of Dwarf x Dwarf hybrids. 41 7. Be careful: Avoid replanting brown coconut palms in or around the seednut garden afterwards. This would make the seednut garden produce poor seednuts. (See Section 4 for more information on seednut quality.) Figure 26: Turning a coconut plantation of Tall-type coconut palms into a seed garden producing seednuts of Talls, Dwarfs and Dwarf x Tall hybrids / Photos: R. Bourdeix 42 3.7.3. Assisted hand pollination. Section 3.5 describes the scientific version of this method, which consists of emasculating the inflorescences in an entire isolated field, the seed garden, and bringing pollen from father palms located outside the seed garden. Issues regarding isolating the seed field have already been addressed in Section 3.4 and Section 3.7.2. To simplify this method, the most difficult step for a farmer will be the preparation and use of pollen. To prepare the pollen, the following steps must be undertaken: 1. Harvest male flowers, as indicated in Section 3.5, by placing them in a small cloth or nonwoven bag23 (like the light-coloured, air-permeable bags often given in supermarkets). 2. Open the male flowers by crushing them when still inside the bag with a bottle or rolling pin on a flat surface (e.g. a table or a wooden board) or on a metal grid. 3. Place the bag containing the male flowers flat and in full sun on an aerated surface (e.g. a pile of branches or a dry coconut leaf). Then, transfer the bag to the driest and most ventilated place possible to continue drying the male flowers. A fan can then be used, if available, or an oven or a drying box; however, the drying temperature should not exceed 39°C. Drying the pollen will take one to two days. 4. Transfer the male flowers into a fine mesh sieve,24 preferably with a lid so the flowers can be shook vigorously; collect the pollen and transfer it to an airtight container. It should be possible to dehydrate the pollen further by adding a sachet of silica gel, a product that absorbs moisture, but this is not essential and has not yet been tested. 5. Store the pollen in a freezer or, if unavailable, a refrigerator. Use this pollen within 10 days. 6. Mix one part pollen with 20 parts talc, on the day of use.25 The authors of this guide believe that it is possible to use all kinds of grain flour (wheat, corn, rice) for the mixture but have not yet found the funding to test these mixtures. 7. Pollinate the female flowers according to the descriptions below. The farmer must also be able to recognise precisely when a female flower becomes receptive to pollen. This is relatively simple and is explained further in Section 2.4 and depicted in Figure 7. In the morning only, carefully examine the female flowers and find those that already have nectar drops oozing from their lower ends, as shown in Figure 27. 23 The size of the bag used to crush the male flowers should be about 20 centimetres x 30 centimetres or slightly less but not larger. 24 We have imagined, but not yet tested, a method that would prevent having to sift the pollen. This method would consist of vigorously mixing the dried male flowers with talc, then transferring this mixture to the wash bottle, and using it to carry out the fertilizations by spraying. It is important to note that we do not know if this approach would work, because small fragments of male flowers could block the operation in the wash bottle. 25 It is important to exercise caution, as some talc brands contain chemicals that reduce the germination of the pollen. It is necessary to test with several brands of talc, favouring the most natural ones. 43 Figure 27: A receptive female coconut flower with nectar on its tip, and a bee / Photo: R. Bourdeix Another technique, which is a variation of the assisted hand pollination method, was used in Indonesia by a farmer who produced toddy, where the sap is obtained by regularly incising the top of the inflorescence. The method used by this Indonesian farmer involved the following steps: • Prepare the pollen from the parents chosen as males. • Then, in the plantation, emasculate some inflorescences on the parents chosen as females, without worrying about the isolation of these female parents. • Finally, apply pollen with a brush to the female flowers, as soon as they became receptive. With this technique, the authors of this guide believe there is a high probability of obtaining a fairly large number of illegitimate seednuts: the inflorescences are not protected, and the bees move quickly with pollen in the plantation. By selecting seedlings based on sprout colour, the farmer, who used Red Dwarfs as mother palms, was able to eliminate some illegitimate plants, but not all. To use this kind of technique, it is better to bag the inflorescence, even if the bag is removed regularly to apply the pollen with a brush. It is even better to bag the inflorescence in a bag with a transparent window to see when the female flowers become receptive, and to spray a mixture of talc and pollen with a squeeze bottle through a small hole in the bag. In this case, however, the assisted hand pollination method is no longer being used but, rather, that of controlled pollination with bagging, a method described in the following section. 50 Dwarf varieties often show alternative biennial bearing, as illustrated by the second example, the Malayan Green Dwarf (MGD) in Côte d’Ivoire. (See Figure 31.) Figure 31: Sensitivity of the coconut palm to environmental variations: the case of the Malayan Green Dwarf (MGD) in Côte d’Ivoire/ Photos: R. Bourdeix A seemingly healthy coconut was photographed but, one year later, the number of coconuts in the crown had dramatically reduced. The leaf canopy turned from X-shaped to spherical, because the palm could no longer support the weight of its heavy bunches. The third example is an Indian Tall-type variety (shown in Figure 32) that produces large or small fruits depending on the season, the number of coconuts in a bunch, and the origin of the pollen of the parent palms. The authors of this guide hypothesise that, within a coconut bunch, larger coconuts are frequently the result of cross-pollination – where the fertilising pollen originates from a male palm of a different variety – whereas smaller coconuts are more likely the outcome of natural self-pollination by the same palm. However, this is not yet scientifically proven. 51 Figure 32: Fruit-size variation within a Laccadives Micro Tall (LMT) palm variety from India / Photo: R. Bourdeix Fruit size is often inversely proportionate to its number of fruits. The nature and origin of the pollen can also influence fruit size. When the coconut palm self-pollinates, the fruits produced are sometimes smaller. 52 4.2. Selection of parent palms Traditional ways of palm/seednut selection In many Pacific Islands, farmers know and use traditional methods of selecting parent palms or seednuts. Their methods are often based on classifying coconut palm and coconut into two categories of “female” and “male”, and the “female” is generally preferred and kept for replanting (Figure 33). Each of these traditional selection processes is not described in this guide, as the authors of this guide have already published on them in other books31 and articles, available for free online. Nonetheless, it is the belief of the authors of this guide that these selection processes have some effectiveness, and this knowledge should be preserved. The most effective selection may consist of combining traditional methods with the more modern methods presented in this guide. Figure 33: Selecting seednuts based on two categories of “female” and “male”, where the “female” is generally kept for replanting / Photo: R. Bourdeix The right one is called “male”; the left one is called “female” and preferred for replanting, except for the varieties called Niu kafa, Ni afa, Nape or Magi-Magi, as these varieties selected for husk always germinate as “male”. With respect to coconut, scientists have long debated what they call "mass selection in open pollination". This refers to the method of assessing the productivity of coconuts, selecting the best palms, and then harvesting the seednuts naturally produced on these palms. This method is described in Section 3.3 and Section 3.7.1 (although it is not identified as mass selection in open pollination in these sections). The experiments that have given the best results have consisted of closely monitoring a Tall-type plantation to record the number of fruits and their composition over four years, carefully analysing the data, and selecting the single best-performing palm from every 20 palms (i.e. 5%). Despite the intensive effort of monitoring every single palm in the plantation for four consecutive years, the progeny of the selected individuals achieved only a modest yield gain, averaging 14%. Sometimes, agricultural technicians have been observed walking around the plantations and scanning the palms to identify the ones most loaded with fruits and immediately selecting one palm out of five (or 20%) or one palm out of three (or 33%) as the mother palm, without considering the opinions of the farmers. Such selection is inefficient, as it is unlikely to increase the value of the seednuts harvested from these selected palms. 31 See: https://agritrop.cirad.fr/576084/1/Bourdeixetal_Biodiversity.pdf. 53 Often, coconut grove owners know relatively little about their palms, whereas even the humblest farmer knows his/her coconut palms best from spending significant time on the plantation. Those who prepare coconuts for food also have extensive knowledge. A proposed alternative selection method In this section, an alternative selection method is proposed to identify the best palms in a plantation more efficiently. Once the best palms are identified, farmers can then breed them and use them to create hybrids, choosing one of the pollination techniques described earlier in this guide. The following is proposed as a better way of selecting parent palms: 1. Pre-select 20–30% of the best coconut palms in the plantation, based on information provided by the farmer and other farmworkers, and general aspects of the palm including: 1) absence of disease; 2) crowns bearing the most fruits compared with other palms; and 3) palms with fruits that are not too small and (where visible under the palm) do not have a high husk content. Avoid selecting only the largest fruits32 without considering these other criteria. Palms that produce the largest fruits often produce fewer fruits, lending to a meagre yield. 2. Mark the pre-selected palms with a strip of paint, at least four centimetres wide and light in colour (yellow or white), at 180 centimetres from the ground so the strip is visible from afar. Number the palms. This can be done by painting numbers on the stem, with all numbers noted on the same side of the stem. This numbering of the palms is crucial. If affordable, durable tags (e.g. aluminium ones) should be used in addition to painted labels. By using these tags, if an agricultural officer or the farmer’s children return 15 years later, they will still be able to identify the palms. 3. Harvest (if possible) four mature fruits and two tendernuts for drinking, on these pre-selected coconut palms. The mature fruits must be completely dry, with brown-grey skin and water inside, and without visible disease or wounds. Harvest tendernuts at an optimal stage, depending on the country (with an average kernel thickness of between one and six millimetres). 4. Carefully record and duplicate any information, measurements or other data collected. Data can best be shared with the local agricultural institution, for both an efficient analysis of the results and safe-keeping. Forms for recording all of the required information, including site data, palm location, and palm characterisation, are available in Annex 2. 5. Calculate two fruit quality indexes, potentially with the help of an agricultural technician. These fruit quality indexes include: the mature fruit quality index – kernel weight divided by the weight of the whole fruit without the free water; and the young fruit drinking quality index – weight of the water divided by the weight of the whole young fruit. 6. Use all of the collected data to select between five and 10% of the plantation’s best palms. The higher selection rate must be applied, considering the number of seednuts or the quantity of pollen needed. The higher the selection rate, the better the value of the progeny and the efficiency of the selection. 32 The selection must also be adapted to the intended use of the fruits. If, for example, the goal is to produce tendernuts to drink and if the fruits are sold individually, it is better to have more medium-sized fruits than a few very large fruits. 54 7. Mark the selected palms with a second strip of paint, at least four centimetres wide and red or orange in colour, at 170 centimetres from the ground, in order to see this strip well from afar. Number the palms. 8. Carefully record, duplicate and transmit the entire set of breeding data. (Refer to the forms provided in the Annex 2). 9. The selected palms are now ready for use, both for seednuts or pollen production, by applying the pollination technique chosen from those presented in this guide. Sometimes, farmers do not know the origin of the varieties that are grown in their plantations. Some of these plantations are made with Dwarf x Tall hybrids, with mixes of these hybrids and Tall types, or even from progenies of Dwarf x Tall hybrids. It is better to avoid selecting parent palms among hybrids or in progenies of hybrids, because the selection will not be effective in these cases. These hybrids are often made from Yellow and Red Dwarf mother palms, and these colors are normally rare among Tall types. For this reason, a clever technique33 is recommended to verify whether the parent palms chosen are hybrids or not. Although it can be challenging, the procedure to check whether parent palms are hybrids or not is as follows: 1. Pack, label and send the seednuts to the nursery. For each selected palm, pack a bag with at least 15 seednuts, label the bag with the palm number, and send it to the nursery. 2. Germinate the seednut lots in separate, small seednut beds labelled with the palm number and date of sowing. 3. Observe the colour of the small sprouts, as soon as the seednut sprouts are visible. If the mother palm has brown or green coloured coconuts, and if there is more than one yellow or red sprout in its progeny, the palm is probably a Dwarf x Tall hybrid or a hybrid progeny. In this case, remove this parent palm from the selection and discard its seednuts. Special varieties, such as sweet husk or medicinal ones, which are rare, should be planted separately. To avoid inbreeding, it is preferable to procure the seednuts of the same variety from different locations. Coconut exchanges between farmers are strongly encouraged and should increase. 4.3. Efficiency of pollination techniques In this guide, simplified adaptations of different seednut production methods are proposed for farmers. The techniques described are still experimental and need to be evaluated more systematically. Therefore, farmers must proceed by successive tests in order to determine which pollination techniques and which seednut production methods are best adapted to their particular situation. Farmers are invited and encouraged to communicate their results to the authors of this guide in order to help improve these techniques. As mobile telephone coverage expands, simpler, more cost-efficient and information-rich information and communications technology (ICT)-based systems become possible. A relatively new “farmer-citizen-science” approach allows farmers to participate in testing technologies, such as crop varieties and management practices, including the pollination techniques proposed in 33 This technique will not work in French Polynesia, where farmers use a hybrid created with a Green Dwarf as the mother palm. 55 this guide, and to share information by mobile phone. Large numbers of farmers can be involved in testing to help more cost-effectively and statistically validate such technologies.34 The authors of this guide have seen that the most delicate and difficult phase of several methods is the pollen preparation. In large seed fields, which often have small laboratories, two tests are carried out to check pollen quality: 1. Test the pollen to ensure it has properly dried: A first test is undertaken to check the pollen’s water content and ensure the pollen has been properly dried. A sample of a few grammes of pollen is weighed, then completely dehydrated in a special oven, and finally weighed again to determine the water content. If this water content is less than or equal to 13% (for short-term pollen conservation up to 15 days), or 4–8% (for longer vacuum conservation), the pollen is considered properly dried. 2. Test the pollen to ensure it is alive and capable of pollination: A second test consists of germinating the pollen in transparent boxes containing a nutrient medium, to check with the aid of a microscope that the pollen is alive and capable of pollination. Good pollen shows a germination percentage of at least 30%. Unfortunately, the authors of this guide have not found a creative strategy to adapt these quality controls to the case of farmers without access to a laboratory. In some cases, national agricultural institutions or other structures may help farmers by performing these tests for them. Sometimes, these institutions can even provide good pollen to farmers, as is done in the Philippines. 4.4. Selection at seedbed and nursery stages In 2018, a video was made on coconut seedbeds and nurseries around the world. It showed that the quality of these seedbeds and nurseries varied depending on the location and situation. Creating good seedbeds and nurseries is crucial. This allows for effective selection and elimination of plants that will not produce or will produce little, throughout the entire plantation period. In the case of Dwarf x Tall hybrids, it is estimated that good selection at this stage can increase plantation productivity by at least 10%. In the case of traditional Tall-type varieties, harvested in open pollination, we know that the effectiveness of nursery selection is even more important. Indeed, Tall types sometimes self-fertilise; in this case, some of the offspring suffer from inbreeding and are genetically weak. Selection in germinators and nurseries, if done well, can eliminate most of the offspring resulting from self-fertilisation. Take care of seedbeds and nurseries, and practice effective selection! Early selection is believed to help increase plantation yields by 10–20%, enabling the plantation to yield for at least 40 years, making it worth the investment. The worst technique is to take plants from a plantation where they have grown naturally from abandoned nuts, and to replant them. The forgotten nuts are often the smallest nuts, or nuts attacked by diseases or from sick palms. This is exactly how a counter-selection of the worstperforming palms is carried out. It is believed that some farmers have practiced this for over 100 years, gradually reducing the overall size of nuts from their traditional Tall-type variety. 34 See: https://ccafs.cgiar.org/research/projects/citizen-science-approach-climate-adaptation. 56 Germination beds and nurseries must be organised in batches according to the seed harvest date and the variety. This makes it possible to obtain homogeneous germinators and nurseries, facilitating the identification of plants that grow poorly or too slowly and eliminating them. The ideal is to identify the germination beds with a small durable sign bearing the variety name, the seed harvest date, and the number of seeds germinated. In the case of oil palm, experts have noted that some farmers selling plants were reluctant to eliminate the weakest plants. These farmers bought the seeds and paid for the elimination. This bad practice must be avoided in the case of coconut palms as it introduces low-yielding palms into the plantation that will then persist for about 50 years. 4.5. Traceability of seednut production, use and sales The origin of the seednuts and plants that farmers produce, use and, above all, sell should be recorded in an accessible and sustainable way. There is a framework of laws and rights on all cultivated plants, which regulates seed markets, seed marketing, intellectual property rights on plants, and farmers' rights, among other considerations. These operate at local, national and international levels (e.g. European Union). It is important not to be deterred by the complexity of this framework, as avenues exist to support producers of farmers’ seeds (peasants, gardeners, seed artisans). In some countries, such as France, associations offer regulatory kits that provide the basics for understanding the existing framework to collectively build farmers’ seed stores, share seeds and associated knowledge, and even sell "non-standardised" seeds and plants that are free of intellectual property rights. Nonetheless, the legal situation differs depending on whether the plant is one of the 50 species considered crucial for humanity and listed in the International Treaty on Plant Genetic Resources for Food and Agriculture (ITPGRFA) (“Annex 1”).35 The coconut palm is one of these 50 species. Under the ITPGRFA, signatory governments agree that “ITPGRFA Annex 1 crops” can be exchanged under a standard agreement that allows the exchange and use of the plant genetic resources (PGR) for research and breeding. Where the PGR are further developed through breeding and/or used for commercial gain, royalties are paid to the Benefit Sharing Fund under the Food and Agriculture Organization of the United Nations (FAO). This fund supports projects on in situ conservation and other farmer-based incentives for the improved utilisation of ITPGRFA Annex 1 crops. It should be noted that farmers have rights to breed and benefit from producing their own coconut hybrids or ecotypes of known varieties selected by farmers and propagated for their own use (see Article 15 CBD and Article 9 ITPGRFA36). The following points should be considered: 1. Understand local laws on the sale of seednuts: In some countries, you cannot sell seednuts without legal authorisation from the Ministry of Agriculture. It is important to obtain more information within the respective local administration/Government, preferably through a farmers’ organisation. 2. Understand local laws on the movement of plants: Local laws may prohibit the movement of plants from one region to another, or from one island to another. For example, in French 35 See: https://www.fao.org/plant-treaty/areas-of-work/the-multilateral-system/annex1/en/. 36 These articles affirm “...that the past, present and future contributions of farmers in all regions of the world, particularly those in centres of origin and diversity, in conserving, improving and making available these resources, is the basis of Farmers' Rights.” See: https://www.fao.org/plant-treaty/areas-of-work/farmers-rights/en/. 57 Polynesia, it is forbidden to transfer coconut plants from the island of Tahiti to the Tuamotu archipelagos, in order to prevent the spread of harmful insects. Consider the example in Madang, PNG with respect to Bogia Coconut disease [17] 3. Include relevant information on seedlings: Seedlings that local authorities permit to be sold should be accompanied by minimum data from the nursery process (i.e. the age or sowing date, the origin and type of the planting material [Dwarf, Compact Dwarf, Tall type, hybrids of different kinds]). 4. Include relevant information on germinating seednuts: Germinating seednuts that are permitted to be sold should be accompanied by the minimum information: i) sowing and germination dates; ii) origin (farmer group or name); and iii) variety or hybrid cross. 5. Keep records on any seednuts sold to the local agricultural authority for distribution: If seednuts are sold to the local agricultural authority for distribution to other farmers, these records should be kept, and information channelled back to farmers so they can refine their field selections of mother palms and aim for continuous improvement. (Examples/templates of relevant forms are provided in Annex 2.) 6. Monitor and record seedling germination and growth in the nursery: Strict culling percentages should be applied to: i) maintain high-quality; ii) ensure the seedlings sold will be successfully transferred to the field; and iii) enable high future yields. 7. Ungerminated seednuts sold are at the buyers’ risk: If seednuts are sold ungerminated, this should be at the buyers’ risk, where no seednut certification system is in place. Farmers should encourage buyers to provide feedback on the number that germinate and any other agronomic data that is generated. Normally, at least 80% of the seednuts should germinate. 8. Collect seed only from healthy coconut palms throughout the selection process: If parent palms (male or female) show any signs of disease, they should be removed from further selections. Good hygiene should also be observed in the nursery to avoid losses from diseases and pests. 9. Collaborate with the local agricultural authority responsible for the coconut industry, in all situations, to develop farmer-led seed gardens and seednut production and sale. This way, the seednuts can be certified under a local system of quality control. This could mimic other systems used for other crops in different countries.37 [18] An example of seednut label is provided below. Several coconut diseases are prevalent in both nurseries and in mature palms. A common and sometimes severe nursery disease in the wet tropics is embryo rot caused by Marasmiellus spp.; germinating seedlings showing signs of this disease and any other disease-like symptoms should be removed and destroyed. Readers should refer to the SPC Pest List database and Pestnet38, which provide more information on both major and minor coconut pests and diseases. 37 See, e.g., farm-saved seed designations for UK farmers: https://www.gov.uk/guidance/farm-saved-seed. 38 See: https://www.pestnet.org/. 58 5. OTHER INITIATIVES TO PROMOTE THE CONSERVATION OF FARMERS’ VARIETIES As emphasised above, conservation is critical and the best way to conserve diversity is to use it and breed with this diversity. This section presents six additional initiatives that can help increase farmers' capacity to produce quality seednuts, better value their breeding work, and make their businesses more profitable if they decide to market seednuts and plantlets. 1. Polymotu (Poly=many, motu=island) uses geographical isolation to conserve coconut diversity and produce seednuts.39 2. Community or private coconut genebanks offer ways to both conserve diversity and increase seednut availability. This guide suggests a variety of designs for these new kinds of genebanks. 3. The national agricultural institution and national coconut genebank, if available in the country, offer important resources to farmers. 4. Using social media to communicate about coconut varieties can help increase the sales of seednuts and, in some cases, help sell special seednuts at much higher prices. A success story of the Fakahina Green Dwarf illustrating this is provided in Section 5.5 below. 5. Comprehensive and well-illustrated catalogues of coconut varieties raise awareness among farmers, gardeners and the general public of the magnificent diversity of coconut and, in turn, help boost the use, exchange and sale of seednuts. 6. A coconut ecomuseum – including small ones created at the farm level and larger ones at the country level – offers an excellent opportunity to link agriculture, science, education, craftsmanship, tourism and business, and adds value to the geographic origins of coconuts (e.g. as champagne has done for the Champagne region of France where champagne originates). 39 This can be combined with various profitable tourism activities, such as guided visits to small islands showcasing different coconut varieties and demonstrating nursery operations. 59 5.1. The Polymotu concept Polymotu is a concept inspired by ancient traditional practices from the Tongan Islands.40 The elders created, consciously or accidentally, traditional conservatoires by planting unique coconut varieties on small, isolated islands, such as those shown in Figure 34. Figure 34: Illustration of the Polymotu concept (1) Tahaa, French Polynesia / Photos: R. Bourdeix Around the island of Tahaa, in French Polynesia, a string of motus (small islands) seems suitable for conserving coconut palm varieties according to the Polymotu concept. Some isolated inland locations are also suitable because no coconut palm or pollen are present there. Planting coconut palms in areas completely isolated from other coconut palms offers an inexpensive, practical way of conserving and producing seednuts from a single coconut variety. For example, if special coconuts are planted on a small remote island, then the coconut palms planted that will grow on this island will interbreed with each other. Under these conditions, there is a much better chance that farmers will succeed in creating and maintaining a new variety. This has already happened in the past, including on Rennell Island in the Solomon Islands and on the islands of Niuafo'ou and Onoiki in Tonga.41 The original Polynesian practice used for traditional conservatoire was to plant only one coconut variety on each small island. When compared to this initial model, the Polymotu concept has evolved in three ways [19]. 40 In the 2000s, we visited numerous Pacific islands in the framework of surveys organised by Bioversity International (now part of the Bioversity-CIAT Alliance, formerly IPGRI). L.M. Fili and T.H. Hoponoa, from the Ministry of Agriculture and Forestry of Tonga, provided information on the traditional coconut variety called “Niu 'utongau”. This variety belongs to a group of rare, highly threatened coconut forms, known as “Sweet husk”. In most coconuts, this husk is fibrous and inedible. However, sometimes, the whole husk of the young fruit is sweet and can be chewed like sugar cane. Its taste resembles that of coconut heart. Once the fruits are ripe, the husk fibres are white and notably slender. Although there are various names and types for this traditional coconut variety, the “Niu 'utongau” coconut variety can only be found abundantly on the small coral islet of Onoiki in the Ha’apai group, an islet that is so small it does not appear on most maps. Tongans still sometimes take seedlings from that islet. 41 See: https://polymotu.blogspot.com/2009/02/lesson-from-ancient-and-some.html. 66 5.4. Importance of compiling and publishing varietal catalogues The idea of publishing an illustrated coconut germplasm catalogue was first presented to the COGENT Steering Committee by R. Bourdeix in 1993. A first catalogue was published by R. Bourdeix et al. in 2005, with varieties conserved in the African coconut genebank. In 2010, the COGENT Catalogue of Conserved Coconut Germplasm was published with more than 50 participating authors from 20 countries [7] To raise awareness of this diversity among farmers, gardeners and the public, and to create a demand for seednuts that can foster conservation, the compilation and publication of local or national catalogues of coconut palm varieties is critical at the country level. One example illustrates the importance of cataloguing varieties for more efficient conservation and use of varieties. In Indonesia, there are many Red and Orange Dwarf varieties, with at least seven names to describe them: Bali; Jombang; Malayan; Nias; Sagerat; Soasio; and Waingapu. Tests conducted at Balit Palma, Indonesian palm research institute in Manado, clearly showed that some of these varieties produce more coconuts or sap than others. In addition, the Red Dwarfs are important for the establishment of seednut gardens to produce hybrids. At present, only a few individuals can recognise these varieties in the field. The procedure for making clear, representative photos of coconut varieties to be included in the catalogue is difficult. Illustrated guidelines were developed and sent to participating members, and they are now available online and reproduced in Annex 1 of this guide. A catalogue of farmers’ varieties was recently produced for French Polynesia.46 The authors of this guide believe that this is the first catalogue of farmers’ varieties ever published for coconut. As of 2024, this book was the most accomplished and modern version of the coconut varietal catalogues. The layout is aesthetically appealing, with many illustrations, making it accessible to a wide audience. The format of the varietal descriptions has been made more flexible and is no longer limited to only two pages. In Figure 38, an example is provided of an amazing Polynesian variety of Compact Dwarf, with a sinuous, snake-like trunk. It was described over four pages, to encapsulate the valuable extra detail. The publication of this catalogue for French Polynesia had a very positive impact, as shown in the success story presented in the next section. 46 See the English version at : https://www.calameo.com/read/00566921277dcc9a57440 and French version at: https://www.calameo.com/read/0056692121c8a46c70005?authid=SsoSGy7Dq2z2. 67 Figure 38: Example of a four-page description of a variety in the French Polynesian catalogue / Photos: R. Bourdeix This is an extraordinary “snake” coconut palm with the genetic specificity of a trunk that naturally curves toward the rising sun. There is no longer a need to climb the trunk to harvest drinking coconuts! This new variety has been named “Compact Green Dwarf with flexuous trunk”. 68 5.5. Communicating about varieties and seednuts on social media A survey of the Fakahina Atoll, Tuamotu archipelago, French Polynesia identified a female farmer with a unique small green coconut palm variety. She had planned to cut down this palm. The survey observed that the palm bore extraordinary fruit,47 containing water and such buoyant and optimised natural “packaging”. Figure 39 makes a comparison between this Fakahina Green Dwarf and the Brazilian Green Dwarf, which is planted in tens of thousands of hectares to produce coconut water, particularly in Brazil. This figure shows that the Fakahina Dwarf has a higher fruit composition than that of the Brazilian Green Dwarf. The round and thin-husked Fakahina Dwarf fruits are almost perfect natural bottles. Figure 39: Views of the Fakahina Green Dwarf / Photos: R. Bourdeix The Fakahina Green Dwarf is depicted next to the Brazil Green Dwarf for comparison: Top right: Fakahina Green Dwarf Bottom right: Brazil Green Dwarf Left: The successful female farmer shown next to a Fakahina Green Dwarf in French Polynesia The survey results were published in the varietal catalogue. In addition, the farmer’s son posted about this palm on social media, attracting the interest of many people who subsequently ordered seednuts from this palm. As a result, the farmer is now selling her seednuts for about USD 90 each and no longer wishes to remove this palm, even applying fertiliser to the palm base to support its longevity. Although her selling price is too high to be affordable for many farmers, 47 The green nuts weighed about 1800 grammes and gave 500–600 millilitres of sweet, pleasant coconut water. The kernel weight of mature nuts varies between 400 and 600 grammes. For a 565 gramme kernel, the weight of the shell and the husk only represented 460 grammes, or 81% of the kernel weight! The kernel came off the shell very easily. 69 this story is a success for breeding. When people are willing to pay USD 90 for a seednut, they are more likely to conserve the genotype, negating its risk of extinction. 5.6. Creating coconut ecomuseums and learning centres Ecomuseums originated in France, and about 300 now operate worldwide, with 200 in Europe. An ecomuseum focuses on the identity of a place, or a cultural or agricultural theme linked to the place. Largely based on local participation, ecomuseums aim to enhance the welfare and development of local communities. The concept of a coconut ecomuseum is to encourage local artisans and processers to extend the range of their coconut-based products. These ecomuseums are designed to be profitable, and combine agriculture, culture, tourism, science and business. They are comprised of the following: 1. exhibition galleries of objects (whether old, rare or precious) related to the coconut palm; 2. large educational and recreational panels and posters featuring the coconut palm, its agriculture, and its local, national, regional and global uses; 3. a living collection, presenting various types of Dwarfs, Talls and rare coconut varieties, and including seedling sales, and opportunities to taste or drink nuts, sugar and derived products; 4. training activities (e.g. traditional oil extraction, toddy extraction, sugar-making, cosmetic oil production, hat-weaving, making of other objects, use of coconut wood, etc.); 5. a restaurant specialising in preparations made from coconut palms (salty and sweet) and/or dishes with coconut images on them; 6. a shop selling coconut products (e.g. food, body care, artifacts); and/or 7. a health centre using coconut oil and other coconut-derived products. Such ecomuseums could be sponsored by the private sector, or by national institutions, for communication and training purposes. They could be located near a coconut research centre and/or seednut garden and could play a key role in catalysing social development. They are suitable for all audiences, including both locals and tourists, with a particular focus on young people and schoolchildren in the country, and acting as future repositories of knowledge. Figure 40 illustrates a possible design of large coconut ecomuseum and learning centre. Its architectural design takes the emblematic form of a coconut, and it contains the seven components described above. In the centre of the building, a basin with a fountain symbolises the water inside the coconut. Around this fountain, tables are arranged so customers can eat or drink. Around the outdoor gallery, there is a small living collection of coconut palms and a nursery where coconut seedlings are sold and where visitors can be trained in the operations of germination and varietal recognition by the colour of the germ. 70 Figure 40: Schematic representation of a model of a large coconut ecomuseum / Photo: R. Bourdeix For more information, see this video: https://www.youtube.com/watch?v=7HXPXfWLZ_0 71 6. CONCLUSION Even as researchers and breeders continue to propose and develop improved coconut varieties, it is important for farmers to retain their capacity to choose the varieties they cultivate and, where possible, to engage directly in the process of breeding. Such empowerment encourages greater on-farm coconut diversity and allows farmers both to evaluate different varieties under their own agro-ecological conditions, and to carry out cross-breeding themselves, often using Red and Yellow Dwarf varieties as female parents. Some farmers may prefer to focus on Tall-type varieties, and the techniques described in this guide may also serve this purpose. Most farmers will also be interested in planting Dwarfs, Dwarf x Dwarf hybrids, and Dwarf x Tall hybrids.48 To be most effective, farmers could collaborate with local extension workers, especially if the seednuts are to be sold commercially. Although training farmers to carry out coconut pollination may appear straightforward, applying it in the context of coconut breeding represents a fundamentally original and disruptive innovation. This idea has taken over two decades to mature, reflecting a substantial shift in the conceptual framework of coconut improvement. The first conceptual breakthrough was the recognition that geographical reproductive isolation and genetic drift have often been key drivers in the emergence of distinct coconut varieties. The fact that several major cultivars originated on small, remote islands – such as Niuafo’ou or Rennell – was not incidental. It underpinned the development of the Polymotu concept, which leverages such isolation to conserve genetic diversity and enable decentralised hybridisation strategies. The second breakthrough has been more epistemological: scientists had to move beyond rigid methodological idealism, recognising that perfection is not always a prerequisite for impact. Traditional breeding approaches, grounded in scientific precision, often result in methods that are too complex or costly for widespread adoption. In contrast, if farmers can apply simpler pollination techniques with 80–95% reliability, this already constitutes significant progress over uncontrolled open pollination. In many contexts, "good enough" methods tailored to farmers’ realities can outperform “perfect” but inaccessible techniques. The approach described in this guide represents more than a technical advance; it proposes a paradigm shift in how coconut diversity could be managed and improved. For centuries, most coconut farmers have chosen the mother palms of the seednuts but have not chosen the father. By giving them access to practical knowledge and feasible pollination methods, we offer coconut growers the unprecedented possibility of directly shaping the genetic composition of their future plantations. Research is still needed to adapt these crossing methods for farmers. As discussed in Section 3.1.3 of the COGENT Global Strategy [20] once the process has been validated, videos can depict these techniques and help farmers more easily apply them to their own contexts. By making breeding more accessible, this approach encourages farmers to take greater ownership of the hybrid varieties introduced by public and private breeding programmes, leading to broader acceptance and more informed use of improved planting materials. In many coconutproducing countries, coconut seed gardens are insufficient to meet rising demand, and coconut 48 See the video: https://www.youtube.com/watch?v=5BoWreQEcto. 72 hybrid seednuts are often too costly for farmers. The low-cost, farmer-led pollination techniques described in this guide offer a scalable and participatory alternative. By simplifying access to breeding practices, this approach empowers farmers to better understand and adopt hybrid varieties developed by public and private institutions, thereby increasing their acceptance and relevance at the local level. Given that, in many coconutproducing countries, existing seed gardens are unable to meet demand, and hybrid seednuts remain prohibitively expensive for many growers, the farmer-driven, low-cost pollination methods presented in this guide provide a viable, scalable solution that addresses both supply constraints and affordability. These methods offer a participatory pathway to enhance local seed sovereignty and meet growing planting needs. 73 References 1. Bourdeix, R., Sourisseau, J. M., & Lin, J. (2021). Coconut risk management and mitigation manual for the Pacific Region. Pacific Community, Suva, Fiji. 2. Recommendation 3 of the 2012 COGENT Steering Committee meeting 3. Bourdeix, R., Nat Tuivavalagi, N., Mataora, V. Jerard, A.B., Hussein N. (2018). Germplasm and incentives for boosting coconut production: case studies from the Pacific region and some other countries. CORD, 34(1), 14-14. 4. Kete, T., Delamata Delaisla, D. & Bourdeix, R. (2018). 3.1.2 Concepts for communication - Chapter 3. Where we need to be to secure diversity and promote use. In R. Bourdeix & A. Prades (Eds.), A Global Strategy for the Conservation and Use of Coconut Genetic Resources 2018-2028. (pp. 116). Montpellier, France. Bioversity International. https://cgspace.cgiar.org/handle/10568/96540 5. Konelio, M., Kanniah J., & Sileye, T. (2018). 3.7.2 Promoting farmer-made planting material - Chapter 3. Where we need to be to secure diversity and promote use. In R. Bourdeix & A. Prades (Eds.), A Global Strategy for the Conservation and Use of Coconut Genetic Resources 2018-2028. (pp. 157-158). Montpellier, France. Bioversity International. https://cgspace.cgiar.org/handle/10568/96540 6. G K Ekanayake; Chandrika Perera; Nilanthie Dasanayaka; J M D T Everard (2012). Varietal Classification of New Coconut (Cocos nucifera L.) Forms Identified from Southern Sri Lanka, October 2012, COCOS 19(1). DOI:10.4038/cocos. v19i1.4749 7. Bourdeix Roland, Batugal Pons, Oliver Jeffrey T., George Maria Luz. 2010. Catalogue of conserved coconut germplasm. Rome: IPGRI ISBN 978-92-9043-831-1 https://www.cogentnetwork.org/conserved-germplasm-catalogue & https://agritrop.cirad.fr/561440/ 8. Chomchallow, N. (1986). – Code for coconut varieties. IBPGR. (unpublished). Available at the URL: https://www.coconutmuseum.org/2021/12/caga.html. 9. G.A. Santos, P.A. Batugal, A. Othman L. Baudouin and J.P. Labouisse (Eds) (1992) Manual on Standardized Research Techniques In Coconut Breeding, COGENT-IPGRI, Rome.https://cogentnetwork.org/oldsite/images/publications/StantechManual.pdf 10. Bourdeix Roland; Rivera R L.; Sileye Tiata; Doubi Bi Tra Serge; Kembu, A; Maot, J.; Thilakarathne, O (2024, in press). Structure and dynamics of the nomenclature of coconut cultivars. Towards an evolution to integrate the Compact Dwarfs varieties as a special type. 11. Xiao, Y., Xu, P., Fan, H., Baudouin, L., Xia, W., Bocs, S., ... & Yang, Y. (2017). The genome draft of coconut (Cocos nucifera). Gigascience, 6(11), gix095. 12. M. A Arterburn, S. S. Jones and K. K. Kidwell Soils, Plant Growth and Crop Production - Vol. I - Plant Breeding and Genetics, UNESCO (accessed 01/10/2024) https://www.eolss.net/ebooklib/bookinfo/soils-plant-growth-crop-production.aspx 13. Nuce de Lamothe, M. de, and Rognon, F. (1972b). La production de semences hybrides chez le cocotier par pollinisation assistée (The production of seednuts in coconut by assisted pollination). Oléagineux (France) 27, 11, 539-544. 14. Nuce de Lamothe, M. de, and Wuidart, W. (1994). La production de semences hybrides de cocotier : cas des semences hybrides Nain x Grand (Hybrid seednut production of coconut: the case of Dwarf x Tall hybrid seednuts). Oléagineux (France) 47, 2, 93-102. 15. Rognon, F., and Bourgoing, R. (1992). (The hybrid seednut production of coconut: the case of Dwarf x Tall hybrid seednuts. II. Exploitation of the seed-garden). Oléagineux (France) : 47, 7, 481-489. 74 16. Bourdeix, R., Leclerc, C., Thampan, P. K., Baudouin, L., & Joly, H. I. (2008). Modern and natural coconut hybrids in southern India: Natural, technical and social facts. Journal of ethnobiology, 28(1), 39-54 Bourdeix, R., Leclerc, C., Thampan, P. K., Baudouin, L., & Joly, H. I. (2008). 4 17. Bourdeix et al., Coconut palms and bananas threatened by a new lethal disease (2014) https://www.youtube.com/watch?v=aNbHW4N124U Bioversity/ COGENT (video) 18. Jones, SK (ed) (2006) Quality Declared Seed System: FAO Plant Production and Protection Paper 185. Rome: Food and Agriculture Organization of the United Nations (2006), pp. 243, ISBN 92-5-105510-6 - - Volume 43 Issue 2 https://www.fao.org/4/a0503e/a0503e00.htm 19. Bourdeix, R., Johnson, V., Baudouin, L., Tuia, V. S., Kete, T., Planes, S., Weise, S. F. (2011). Polymotu: A new concept of island-based germplasm bank based on an old Polynesian practice. Ogasawara research. Bourdeix, R., Johnson, V., Kapé, J., & Planes, S. (2013). Traditional conservation areas of coconut varieties and associated knowledge in Polynesian Islands (South Pacific Ocean). 20. Johnson, V., Delapresa, A. & Picq, C. (2018). 3.1.3 Implementing the communication strategy - Chapter 3. Where we need to be to secure diversity and promote use. In R. Bourdeix & A. Prades (Eds.), A Global Strategy for the Conservation and Use of Coconut Genetic Resources 2018-2028. (pp. 116-120). Montpellier, France. Bioversity International. 75 Glossary of additional technical terms Term Definition Abscission The natural detachment of parts of a plant, typically dead leaves and ripe fruit Agar A gelatinous substance obtained from certain red seaweeds and used both in biological culture media and as a thickener in foods Agronomic practices Any crop management operation the farmer implements to ensure the best performance of his/her crop Aluminium label tags The aluminium tags placed on bags to label the coconut variety Anther(s) The floral organ that forms male spores (microspores) or pollen; normally consists of two lobes (thecae), each with two pollen sacs (microsporangia) in which pollen development takes place (Kearns and Inouye 1993) Assisted pollination Any process where natural pollination is assisted (see also: “Pollination”) Autogamy When a plant makes seeds by itself, without needing pollen from another plant Axil The space/angle between the upper surface of a branch or leafstalk and the stem from which it grows Axillary bud/shoot A bud/stem emerging in the axil (see also: “Axil”) Bagging The method of surrounding female flowers with a mesh or paper bag to prevent the pollen of other flowers from reaching the stigma of the bagged flowers; this method can be used to determine if a plant has self-fertile flowers to prevent floral visitation, or to prevent pollen flow in transgenic crops Biological characteristics Any observable feature or trait of an organism, whether acquired or inherited Bole The base of a coconut palm trunk Bract(s) A small leaf or scalelike structure near the base of a flower (Roubik 1995) Breeder(s) A person who breeds animals or plants Bud Small side or end protuberance on a plant stem that arise from meristem (growing point) tissue and may develop into a flower, leaf, or shoot Bunch production Production of coconut bunches Castration/emasculatio n The act of removing male flowers or the anthers (male, pollenproducing parts of a flower) usually for controlled pollination/breeding and experiments involving these aspects Catalogue(ing) The process of making a systematic list of items of the same type; with respect to coconut, the development a descriptive list of coconut varieties Certification (of planting material) Process of providing a supplier with an official document attesting to a status or level of achievement or, in this case, planting material quality (e.g. purity or true-to-type) Characterisation The process of measuring characters or traits 82 Annex 1. Updated guidelines for producing technical coconut photos The procedure for producing high-quality coconut photos is complex. The main elements of these guidelines are provided below. While some of the steps may appear simple, it is important to follow them carefully to produce high-quality photos. Resources needed to produce high-quality palm photos 1. Harvest hook 2. Bamboo metre scale 3. Digital camera with 300-millimetre zoom lens and medium intensity electronic flash capable of taking the following quality: 8.7 centimetres x 13.9 centimetres (1029 x 1643 pixels) in the CYMK mode (Cyan, Yellow, Magenta, and BlacK) with a resolution of 300 pixels per inch or 118.1 pixels per centimetre 4. Twelve-metre aluminium ladder and/or trained coconut climber 5. Manual saw 6. Nut drying rack (thick piece of wood with nails) 7. Two workers for sawing and hanging fruit 8. Mixture of water and lemon juice (copra-cleaning fluid) 9. Scissors 10. Copra photo platform: A plywood sheet of 1.5 metres x 1.8 metres, with a thickness of 20– 30 millimetres covered with grey-blue fabric (any available clean grey-blue fabric can be used, as the photos will be further digitally processed) 11. Hammer 12 large nails (about 10 centimetres long) into the back of the plywood sheet (at distances as shown) to hold the fruits; add a 20-centimetre scale and a tag with the international name of the variety to the lower part 12. De-husked nut photo platform: Another plywood sheet (0.5 metres x 1 metres x 10–20 millimetres), covered with grey-blue fabric, pieces of plastic tubes (10–15 centimetres in diameter and two to five centimetres long) may be used to hold the coconuts; add a 10centimetre scale to the lower part. 13. Coconut eye-cleaning brush (wire brush or toothbrush) 14. Light-reflector: Two in situ white painted walls, forming a reflective corner; or two whitepainted 1 metre x 1 metre plywood sheets; or two large white umbrellas 15. Photo-editing software 83 Guidelines for producing the whole palm photo Figure 41 illustrates the technique for taking a high-quality photo of a coconut palm. The steps are outlined below. 1. Select an appropriate, representative true-to-type palm, ideally 10–12 years of age and, if possible, bearing more than 60 fruits, such as the palm presented in photos 1 and 2. It is preferable to portray healthy and productive coconut palms, as photographing an “average palm” is inappropriate: the same palm may bear a variable number of fruits according to seasons and years. 2. Remove the old, dry leaves and weeds around the palm. By using a harvest hook, or by climbing to the palm crown, cut two to three leaves (as close to the stem as possible), in the direction of the photographer, so the bunches and fruits are clearly visible from the ground. 3. Avoid taking the photograph in the middle of the day, when the light reduces image quality, as shown in photo 3. Wait for a clear, sunny day, and go to the site at 7:00 a.m. or 5:00 p.m., depending on the side of the palm that was prepared (see step 2) to be photographed. At these times of day, a golden, flat light should come from behind the photographer allowing for sharp photos, with bunches well illuminated and a blue sky as background, as shown in photo 4. 4. The base of the stem and the ends of the leaves must be visible in the photo to have a complete view of the palm. There is no scale in this photo. Ideally a one-metre scale (e.g. a bamboo cane) should be visible or an informal scale (e.g. a person). 5. Taking such a photo will require one to two hours, including the time spent to select and clean the palm. 6. The final output is a photo of 8.7 centimetres x 13.9 centimetres (1029 x 1643 pixels) in the CYMK mode (Cyan, Magenta, Yellow and Black). Its resolution should be 300 pixels per inch or 118.1 pixels per centimetre. 84 Figure 41: Taking quality photos of the whole palm / Photos: R. Bourdeix Note: detailed explanations are provided in the text of this section. 85 Guidelines for producing palm inflorescence and coconut bunch photos Figure 42 illustrates the technique for making inflorescence and bunch photos. As inflorescence and bunch photos need to be made in situ, climbing the palm may be required. As depicted in the photo, another way is to use an old car and climb on the roof! For palms less than 12 years old, it is possible to take some of the photos from the ground or even standing on the roof of an (old) car, as shown in photo 1. For taller palms, researchers and technicians generally use aluminium ladders to reach a height up to 12 metres. If ladders are not available, a coconut climber can be trained to do the following: 1. Climb the selected palm. 2. Clean it by cutting two leaves and a few leaflets from the other leaves. 3. Climb down. 4. Climb the closest neighbouring palm, well positioned to take the photo. 5. Use a good camera with a minimum of 300-millimetre zoom lens, as generally a distance of 8–10 metres separate the palms. Photos of bunches and inflorescences are ideally made during light cloudy days, avoiding the period from 11:00 a.m. to 3:00 p.m. when the light is hard and vertically orientated. Otherwise, the photos will not be as clear, such as those presented in photos 2a, b and 3a, b. In these photos, the palms were not cleaned, and the light was too hard/intense and bright. The best light for these photographs is generally obtained during the changeover from direct sunlight to cloud shadow. Additional electronic flash of medium intensity may be used. 86 Figure 42: Taking quality photos of the inflorescences and bunches / Photos: R. Bourdeix Note: Detailed explanations are provided in the text of this section. 87 Guidelines for producing photos of 12 fruits Figure 43 provides an illustrated guideline for taking a photo of 12 fruits. The steps are outlined below. 1. Harvest nine immature fruits (three samples of each nut [small, medium, large]) on nine different palms. All fruits must be free of disease lesions and insect damage. 2. Select three immature fruits among the nine harvested to portray colour, size and shape variability that may exist within the variety. 3. From the same nine different coconut palms, also harvest 21 mature fruits (seven each of small, medium and large nuts), each with a brown-grey epidermis and free water inside. All fruits must be free of disease lesions and insect damage. 4. Select three mature fruits among the 21 harvested to portray the colour, size and shape variability that may exist within the variety. 5. Split the remaining fruits, half longitudinally, the other half equatorially, using a manual saw, with one worker sawing and another worker, if available, hanging the coconut fruit on a thick piece of wood with nails. 6. Wash the split nuts with a mix of water and lemon juice for the meat to remain white while drying. Remove any loose fibres that go over the husk with a pair of scissors. 7. Mount the fruits on the pre-prepared coconut photo platform (see “Resources needed” above) on the nails, as shown in the figure. 8. Add a 20-centimetre scale and a tag with the international name of the variety to the lower part of the platform. 9. Take at least three similar photos. 10. Use a photo-editing software to optimise the final presentation. The final output is a photo of 8.7 centimetres x 13.9 centimetres (1029 x 1643 pixels) in the CYMK mode. Its resolution is 300 pixels per inch or 118.1 pixels per centimetre. The 20centimetre scale presented close to the fruits counts 200 pixels and is 1.69 centimetres long in the photo. The background colour is a standard colour referenced: Pantone 644 CVC. This colour is equivalent to: Red 163, Green 180, and Blue 207 in the RGB mode, and Cyan 39%, Yellow 9%, Magenta 22%, and Black 0% in the CYMK mode. Making such a photo will require two to three hours of work. A team of three people can make 10 of these photos in one (hard) working day. If the initial image is of good quality, the subsequent editing work is easier. However, even if the photo is only of medium quality, photo processing software is now so powerful that it is possible, although time-consuming, to achieve a final product that is of an acceptable quality. 88 Figure 43: Taking quality photos of the 12 fruits / Photos: R. Bourdeix Note: Detailed explanations are provided in the text of this section. 89 Guidelines for producing the photos of three de-husked nuts As shown in Figure 44, taking a photo of three de-husked nuts requires the following steps: 1. De-husk 15–20 fully mature fruits. The fruits need to be fully mature to avoid any unpleasant whitish appearance from dry immature coconuts. 2. Allow an interval of one to two days between de-husking and taking the photographs, to ensure a homogeneous russet colour. 3. Select three coconuts from the lot (one large, one medium, one small). 4. Dry-clean the “coconut eyes” (fungus-blackened germination holes) with a brush (e.g. toothbrush). 5. Place a pre-prepared nut photo-platform (see “Resources needed”) on the ground. 6. Mount the three coconuts on the pieces of plastic tubing, horizontally aligning the top (germination holes) of the three coconuts. This ensures that the three coconuts are at the same distance from the camera to enable a perfectly sharp photo to be taken. A 10-centimetre scale is added to the bottom. 7. Take three similar photos from the top, avoiding direct sunlight and any unidirectional lightning. Ideally, find a corner between two white walls, which function as reflectors, or two white-painted 1 metre x 1 metre plywood sheets, or two white umbrellas. The final output is a photo of 10.16 centimetres x 6.77 centimetres (1200 x 800 pixels) in the CYMK mode. Its resolution is 300 pixels per inch or 118.1 pixels per centimetre. The 20centimetre scale presented close to the fruits counts 400 pixels and is 3.36 centimetres long in the photo. Thus, compared to the other fruits from the 12-fruits photo, the de-husked coconuts are represented with a two-fold larger scale. Previous observations record great variability in shape and disposition of the three “eyes” among and between the coconut varieties. Figure 44 provides examples of good and bad photos for the three de-husked coconuts: - 1 and 2: good quality photos of single coconuts - 3 and 4: the coconuts are too young and not dry - 5: the coconut remains too young (it remains whitish when dry) - 6: the coconut eyes are not brushed (whitish) - 7: the coconut eyes are not brushed (black fungus) - 8: the coconut eyes are not well centred - 9: removal of husk fibres by scraping is excessive - 10: poor or hard direct sunlight (excess shadow) - 11: light is too directional (excess shadow) - 12: an example of a poor quality photo: two coconuts are of the same size, eyes and mouth are not well oriented, scale is missing, and the smaller coconut is blurred - 13: a good quality photo of three coconuts 90 Figure 44: Taking quality photos of de-husked coconuts / Photos: R. Bourdeix Photos 1 and 2: Quality photos of husked coconut Photos 3–11: Poor quality photos showing the kinds of mistakes that may be made with husked coconuts, whether the nuts are not at the right stage of maturity, or poorly cleaned, or poorly lit Photo 12: Properly prepared but poorly arranged and oriented nuts Photo 13: A good quality photo 91 Annex 2. Forms for parent palm selection In the framework of the CIDP project49, we have created seven technical forms to record all of the information needed when selecting parent palms: The first concerns farmers and farms, and the following are distributed: 2) location of coconut palms; 3) characterisation of coconut palms; 4) mature fruit analysis and 5) two methods of analysis of mature fruits, of which only one must be chosen; 6) analysis of drinkable coconuts; and (7) nursery testing for hybrid elimination. The Excel file for data entry and the seven data forms can be downloaded online by clicking on the links in this text. Implementing this process requires a team of two or three workers, including an agricultural officer and a climber/harvester. The total working time typically ranges from one to two hours per coconut palm selected, including the test in the nursery. If necessary, Agricultural Services may request the temporary recruitment of workers to carry out these campaigns. The seven forms are provided below. 49 The Coconut Industry Development Project for the Pacific (CIDP) was a joint initiative of the Pacific Community, the European Union and the African, Caribbean and Pacific Group of States, in placefrom 2016 to 2018. 98 Form 7: Coconut seed system – nursery test for hybrid removal Legends Variety type Colours Person in charge: A. Tall-Type Yellow Red-Orange Date: B. Semi-Tall Yellow Brown C. Compact Dwarf Orange Brown Note: This trial is used to remove selected brown and green parental palm hybrids. If more than 10% of the shoots are yellow or orange/red in colour, there is a high probability that the palm is a hybrid or the offspring of a hybrid and will have to be discarded. Release the good seedlings to farmers. D. Thin Dwarf Orange-Red Green E. Unknown Orange-Red Green Farmer’s last name Farmer’s first name Palm Number Colour of young palm fruits Sowing date Number of ripe fruits sown Number of germinated seednuts Percentage of germination Number of orange or yellow shoots Percentage of orange or yellow shoots Is this palm hybrid or not? 99 Annex 3. Standard coconut fruit measurements Recommended in: International Coconut Genetic Resources Network; Santos G.A., Batugal P.A., Othman A., Baudouin L., Labouisse J.P. (eds.) 1996. Manual on standardized research techniques in coconut breeding, 46 p. Available at: https://cgspace.cgiar.org/items/1cc1e742598c-44bc-a720-40c714c47ef1 Characteristics Immature fruit Mature fruit FRUIT Colour (chart) Number of fruits analysed (#) Fruit equatorial circumference (cm) Fruit polar circumference (cm) Fruit weight (g) NUT Nut equatorial circumference (cm) Nut polar circumference (cm) Nut weight (g) SHELL, HUSK & ENDOSPERM Kernel texture Endosperm weight (g) Endosperm thickness (cm) Shell and meat weight (g) Shell thickness (cm) Shell weight (g) Husk weight (g) WATER Water weight (g) Sugar level in water (brix) 100 101 102