World Conference on Horticultural Research - 17-20 June 1998 in Rome, Italy
WCHR Home   WCHR 1   WCHR 2   WCHR 3   WCHR 4   WCHR 5   Search  

BREEDING FOR FRUITS

Yves Lespinasse1, Frédéric Bakry2
1 INRA BP.57, 49071 Beaucouzé Cedex (France)
, 49071 Beaucouzé Cedex (France)
2 CIRAD-FLHOR BP.5035, 34032 Montpellier Cedex 01 (France)

This review is resulting of the contributions of
F. Laurens (INRA-Angers)
J.M. Audergon, J. Kervella, H. Duval (INRA-Avignon)
G. Salesses, J. Chat, E. Germain, J. Claverie et R. Saunier (INRA-Bordeaux)


Introduction

This review of the research undertaken on the most important fruit species is the result of the contributions of some French fruit breeders working at INRA (Institut National de la Recherche Agronomique) and CIRAD (Centre de coopération Internationale en Recherche Agronomique pour le Développement) This review aims at proposing some trends for the future as well as emphasizing questions to be answered by the scientific community in the short term. The review deals mainly with banana, citrus, pineapple, mango for tropical species, apple, peach and apricot for temperate species. Other species are also considered and the information obtained is synthetically summarized : this is the case of strawberry, sweet cherry, almond, kiwi, chestnut and walnut. For each species, objectives and breeding methods are considered first, then leading teams in the world, and lastly, the main results of the last years and perspectives for international cooperation.
and perspectives for international cooperation.


Part 1 : TROPICAL SPECIES

BANANA

Main research objectives: Since the beginning (1920's), the objectives of banana breeding were to create new varieties resistant to diseases: research on new clones of export bananas in the 1940's for replacing the "Gros Michel" variety, highly susceptible to Panama disease. Today, after a wide replanting programme with the Panama resistant Cavendish clones, the main objective is the research of hybrids resistant to Black Leaf Streak disease (BLS). BLS is a worldwide fungus producing large necrotic areas of the banana leaves and, consequently, important yield losses.

The recent development of biotechnologies with bananas (genetic transformation, molecular biology) opens the way to new objectives such as resistances to viruses, nematodes, banana borer and modification of the fruit ripening patterns. Finally, some authors are planning to use transgenic bananas for human vaccinations in developping countries.

Breeding methods: Two different strategies were developed in breeding. The first aims at producing tetraploid hybrids by pollinating triploid varieties with wild or improved fertile male parents. Originating from female gametic restitution, the few selected tetraploid hybrids aim at combining the agronomic and commercial featd hybrids aim at combining the agronomic and commercial features of the variety with disease resistances contributed by the male parent. The second strategy developed by CIRAD, aims at producing triploid hybrids but starting from natural or improved diploid germplasm. After selection, diploid clones are doubled with a colchicine treatment. Thereafter, the auto- or allotetraploids are backcrossed with another diploid clone in order to create large segregating progenies of triploids. This last strategy offers the opportunity to benefit a larger natural variability than the first option.

The development of biotechnologies with bananas by the beginning of the 80's led to significant results in the in vitro multiplication of banana plantlets (about 20.106 vitroplants /year are used today worldwide as planting material), in tissue, cell and protoplast culture and regeneration through somatic embryogenesis (France, Belgium, Australia, USA), genetic transformation (Belgium, France, Australia, USA) and molecular biology (marker assisted selection, research of genes of agronomical interest).

Leader teams: Historically, banana breeding started in the new world, first in the Bristish West Indies (Trinidad and Tobago; Jamaïca), and later in the 1960's in Central America (Honduras), which is high producing region for export bananas. Other countries later (in the 1980's) developed their own sweet banana breeding rs) developed their own sweet banana breeding research: national programmes in Brasil and in the French West Indies (Guadeloupe), regional and international programmes on cooking bananas in Africa (Cameroon; Nigeria). It should be noticed that, today, there is no identified breeding research work located in the primary centre of diversity of the species.

Noticeable results: Despite so much work during this century, we must state that no new banana hybrid is available to replace the current Cavendish varieties for export markets. For the more diversified local markets which require less constraints of commercialisation, some oustanding dessert and cooking tetraploids hybrids (FHIA01, FHIA21) were obtained but none of them reaching the fruit quality of the current natural varieties. Some good hybrids were also obtained by the triploid strategy (IRFA909, IRFA910, IRFA914) but they still need to be extensively evaluated before release.

Conclusion and international cooperation: Since 1994, all these initiatives in banana improvement are coordinated by the "International Network of Banana Breeders", organized and supported by INIBAP (International Network for the Improvement of Banana and Plantains.

This organisation, which is a programme of IPGRI (International Plant Genetic Resources Institute), has a mission (among others) of organizing and coordinating a global research) of organizing and coordinating a global research effort on banana and plantain, aimed at the development, evaluation and dissemination of improved cultivars and at the conservation and use of Musa diversity, as to coordinate, facilitate and support the production, collection and exchange of information and documentation related to banana and plantain

Owing the high sterility of the cultivated bananas, it is commonly accepted today that it will be fairly unpossible to obtain a "super disease resistant Cavendish" through crosses. In the other hand, facing the high pathogen variability and given our current knowledge about resistant genes, there is little chance of conferring durable resistance to BLS by genetic transformation only. In the near future, success in banana improvement will come to pass through a close interaction and cooperation between all the partners of different disciplines. The current undergoing collaboration between the biotech-laboratories and field experimental stations e.g. for linking crosses and new biotechnologies, illustrates the necessity for an integrated approach as a key for success in banana breeding.


CITRUS

Main research objectives: the Citrus industry is currently facing many increasing biotic (diseases) and abiotic (drying, high salinity soils) constraints justifying the research of new rootstocks as support for a sustainable citriculture in rootstocks as support for a sustainable citriculture in developing countries. The increasing demand of citrus markets throughout the world encourages diversification strategies with new varieties, especially for small fruits. Owing the diversity of the climatic and ecological conditions in citriculture, a variety selected for its proper qualities should have an international appeal whereas a rootstock would be more adapted to specific soil conditions and soil pathogens present in a given region. Moreover, it must be considered that rootstocks should modify the variety features for many characters.

Selection priorities for the varieties are disease resistance (tristeza, greening, Phaemularia angolensis, variagated chlorosis) which generate high yield losses in orchards. Others criteria of selection should be retained according to the final products. Juice processing relies on varieties having good productivity and high sugar content. For fresh fruit markets, variety selection is oriented towards a longer harvest period, of small fruits displaying higher pomologic and organoleptic qualities (easy-peel seedless mandarins for example). Cold tolerance is also subject of many breeding objectives in citrus.

Breeding methods: Most present-day cultivars of oranges, mandarines, lemons grown in the main commercial producing areas arise from sports selected by the productors themselves and propagated by graftingroductors themselves and propagated by grafting (Spain, Morocco, ....). In the same way, pomelo varieties were derived from a common natural ancestor pummelo/orange hybrid originated in the Carribean and the common ancestral clementine (natural mandarin/orange hybrid) was found in a backyard by the Franciscan Clément Rodiez in 1902 in Algeria.

Germplasm characterization showed some sources of rootstocks to be tolerant and/or resistant to various strains of tristeza (virus), to Phytophtora and nematodes, resistant to drought and high salinity soils.

Disease and pest resistance of the scion cultivars (particulary to greening in Asia and blight in USA) are desirable but is difficult to accomplish because gene resources either are not available or are so distantly related to the scion cultivar that recovery of acceptable cultivars is unlikely.

Classical breeding programmes rely on intercrossing varieties which usually do not display hygh levels of sterility. Rarely, backcrossing was carried out. Identification of zygotic seedlings in progenies containing high percentages of nucellar seedlings is a problem. Analysis of leaf or bark isozyme showed to be the most usefull technique for distinction because of the advantage of specific codominant alleles of known inheritance. In another way, somatic hybridization and regeneration of protoplasts involving a wide range of citrus varieties is now applied for the se range of citrus varieties is now applied for the synthesis of tetraploid rootstocks and varieties and also triploid scions by fusion of haploid and diploid material.

Leader teams: Historically, first crosses on citrus started at the USDA in Florida by the end of the last century (1893). This programme extended later in California in 1948. The University of Florida (1923) and California (since 1914) developped also their own breeding programmes. Today, USA, Israël and Japan are the most active countries in citrus improvement. Spain, Itali and Morocco which have a long tradition of sport selection, are now initiating breeding projects as in Australia, Brazil (EMBRAPA), China, Georgia, Italia and France (CIRAD and INRA).

Noticeable results: Programmes developed in the USA resulted in the creation of some good new rootstocks: citrumelo hybrids (Poncirus x pomelo, USDA) and Troyer and Carrizo citranges (Poncirus x orange, University of Florida). Nevertheless, more rootstock varieties are still needed with tolerance to tristeza.

For market purposes, out of sport selection, some new varieties were released from crosses between pomelos and mandarins (tangelo Nova, Orlando, and Minneola). Some mandarin hybrids were also selected for small fruit diversification (Fairchild, Frémont, Fortune, Wilking and Honey). All these new varieties, obtained than mond Honey). All these new varieties, obtained than more 25 years ago, are not yet well developed on the international markets with one exception: the Sunburst cultivar (Florida, 1961) which is the main small fruit variety cultivated in USA today.

Conclusion and international cooperation: An oustanding new rootstock is still needed with the tolerance to tristeza, Phythophthora, and nematodes displaying also a good adaptibility to high salinity and calcaerous soils, and drought.The research of new small fruits clones needs to be maintained looking for easy peel seedless mandarins harvested on a wider period of time.

Finally, it has to be recognised that no work was reported on breeding of orange which still represents about 70 % of the world citrus production. Moreover, most of the researchs and results reported here were developed in a competitive context with an apparent lack of global organisation at the international scale. Within a worldwide network covering all the types of production, it must be considered that citrus improvement has to evolve towards new breeding strategies including the current available technologies (sport selection, crosses, somatic hybridization) with new tools such as molecular markers (and linked methodologies), haplomethods and genetic transformation.


PINEAPPLE

Main research objectives: 'Smooth Cayenne' fruits are fragileobjectives: 'Smooth Cayenne' fruits are fragile and poor in ascorbic acid content. The clone is sensitive to many known pests and diseases (fusariosis, nematodes, Thecla fruit borer) and is a poor producer of planting stock. A first objective of pineapple improvement is the creation of hybrids close to 'Smooth Cayenne' with resistance to the main diseases and high quality fruit, preserving the good agronomic features of the variety (high productivity; leaf margins with very few spines). A second objective is the creation/selection of completely new varieties for market diversification in developed countries in order to increase the choice for consumers. A third objective, as result of wide crossing, should be the selection of some exotic forms as ornamental plants. Favourable traits have been identified in pineapple germplasm: pulp firmness, completely smooth ("piping") leaves, high ascorbic acid content, resistance to fusariosis in the Colombian cultivars Perolera and Manzana, as well as in in the Brazilian 'Primavera', and resistance to fruit borer in the Peruvian 'Samba'.

Breeding methods: As a vegetatively propagated crop, pineapple has been mainly maintained

through clonal selection. However some new characters, as resistance to wilt were identified in 'Smooth Cayenne' in Hawaï and in Mexico. Breeding methods rely on a simple strategy. Pineapple is an allogamous species. Owing the absence o is an allogamous species. Owing the absence of self-pollination (autoincompatibility), it is not necessary to emasculate the seed parent of a cross. The female parent inflorescence is protected before flowering by placing a mosquito net cover of a weather-resistant paper bag on the inflorescence.

Flower induction using etephon must be programmed for synchronying flowering of the parent plants. Germination rates of the seeds vary between 80 and 100 % among species and cultivars according to different methods reported by several authors. In the world, the breeding strategies undertaken by almost all the programmes are based on intercrossing two clones followed by three to four cycles of field observation on the progenies to evaluate the potential of the best clones. These clones are then multiplied and submitted to extensive field trials including acceptance or canning tests. With few exceptions, no strategy of recurrent selection nor production of inbred lines has been reported.

Leader teams: Pioneer hybridization work started in Florida in the beginning of the 20th century succeeded by other initiatives (Philippines, in 1921; Formosa, in 1926). But in fact, it might be considered that systematic improvement really started in Hawaii where the main breeding programme was implemented from 1942 to 1975 by the Pineapple Research Institute (PRI).In Asia, the Malaysian Pineapple Industry Board (MPIB) developalaysian Pineapple Industry Board (MPIB) developed in the 1970s its own improvement programm relying initially on clonal selection but rapidly shifted towards hybridization of Smooth Cayenne and Singapore Spanish. The programme was continued by the Malaysian Agricultural Research and Development Institute (MARDI) which completed the selection of the newly created hybrids.In West Africa, the Institut de Recherche sur les Fruits et Agrumes (IRFA) established in 1978, in Ivory Coast, a breeding programme aimed at creating new hybrids well adapted to fresh export markets as well as processing. In the late 1980s the programme involving 'Smooth Cayenne' and 'Perolera' clones was progressively split between the fruit departement of the Ivorian "Institut des Forêts" (IDEFOR) and the Centre de Coopération Internationale en Recherche Agronomique pour le Développement (CIRAD) in its experimental station of Martinique (F.W.I.) where selection was completed. Since 1978, the Empresa Brasileira de Pesquisa Agropecuaria (EMBRAPA) in the Bahia State has been involved in a breeding programme for the creation of resistant hybrids. Other research institutes active in pineapple improvement are in Philippines, Cuba, South Africa and Australia.

Noticeable results: In Hawaii several hybrids were released to PRI member companies for potential commercial use. Finally, a cross between two of these hybrids linally, a cross between two of these hybrids led to the Golden Ripe variety commercialized by Del Monte in 1995. In Malaysia a first hybrid, 'Nanas Johore' was released by MARDI but failed to compete because of its susceptibility to marble disease. A second hybrid, 'Josapine', was officially released last year. In France a clone with good potential has been released but it needs to be further evaluated at the field scale. In America, the Brazilian production relies on the cultivars Pérola and Smooth Cayenne, both highly susceptible to the fusariosis present throughout the country. Some existing clones were distributed (Perolera, Primavera) but not yet hybrids.

Conclusion and international cooperation: This review of the main breeding programmes shows that most hybridization work is based on the same crossing strategies using clones from the few world leading cultivars, neglecting the large genetic pool available. Noteworthy is the extreme paucity of the results. As already stated, despite millions of seedlings produced since 1905, breeding work has been done over all the continents resulting in a significant contribution of knowledge in cytology and genetics but not to the list of new cultivars. This statement underlines the necessity for redefining the objectives of pineapple improvement (e.g. more focus on a "Super Cayenne" ?) and the needs to rethink the improvement methodologies: is hybridizatiothe improvement methodologies: is hybridization the most efficient strategy ? If the answer is positive, what crossing scheme and parent choice have to be done ? Moreover, more communication is needed between the different actors worldwide in order to avoid duplication of works and to favour cooperation. Finally, a relevant feature of pineapple breeding is the lack of significant progress in advanced biotechnologies (tissue culture, cell regeneration systems, haplomethods, genetic transformation). More focus has to be placed on such methodologies to increase the chance of obtaining satisfactory new varieties in the future.


MANGO

Main research objectives: Breeding objectives vary from region to region, depending on specific traits. However, they can be broadly listed as following: For varieties: 1. Regular bearing - 2. Dwarf tree habit with precocity - 3. Attractive colour, good sized and shaped fruits and flesh quality - 3. Resistance to major diseases and pests - 4. Freedom from physiological disorder - 5. Good shipping qualities (for export) and shelf life - 6. Enhancement of the production period. For rootstocks, the main desirable traits are : 1. Polyembryony- 2. Dwarfing - 3. Tolerance to biotic stresses and to physical and chemical adverse soil conditions - 4. Good scion compatibility.

Breeding methods: Mango breeding is undertaken by a few countries around theo breeding is undertaken by a few countries around the world. There are two main types of mango. The Indian types have monoembryonic seeds and tend to be susceptible to anthracnose. The Indo-chinese types have polyembryonic seeds and might be more tolerant to anthracnose. As other tree crops, many constraints have to be faced in breeding: high clonal heterozygocity ; early post zygotic autoincompatibility; recalcitrant seeds; shortening of the juvenility phase for earlier field evaluation by grafting the new seedlings. Moreover, hand pollinations are problematic and do not lead to consistent fruit setting. Therefore, parent neighbour trees have to be cagged with bees for undertaking controlled cross pollinations. At present, most of the hybrids arise from selection among two varieties or primary hybrids and no strategy of recurrent selection is reported. In another way, with regard to natural mutations, important agronomical traits (such as precocity, yield, regular bearing and resistance to diseases) might be improved by sport selection.

In a parallel direction with breeding, tissue culture was developed to assist mango improvement through somatic cell biotechnologies. During the past 15 years of research in this areas, significative results have been obtained in the regeneration of somatic embryos recovered from nucellar tissue derived callus and cell suspensions of the same origin. Nevertheless, entire plant recovery has been problematis, entire plant recovery has been problematic and no report has been published on the agronomic behaviour of the vitroplants, especially in relation to the occurrence of somaclonal variations in the field. In another area, co-infections of A. tumefaciens with embryogenic tissus led to the development of stable transformed plantlets. These new technologies might be used in the future for the improvement of specific traits such as fruit ripening, tree size, insect and insect resistance and so on.

Leader teams: In India, following last 4-5 decades of intensive breeding, many new hybrids have been released from several research centres. Mango breeding in Israël has resulted in the identification of 15 hybrids. In Australia, very promising progenies were obtained from crosses between the referenced clone 'Kensington' (with typical good flavour) and 'Sensation' (bringing favourable agronomic traits). In South Africa, past genetic improvement was achieved by selection and released in 1990 of four new cultivars.

Noticeable results: Baring a fex hybrid varieties resulting from planned hybridization programmes, almost all known cultivars have resulted from the selection chance seedlings from natural cross pollination mainly in India. However, in Florida, following intensive introduction by the end of the 19th century, some important export varieties have resulted from seemportant export varieties have resulted from seedling derived from open pollinated (or not) identified mother plants. Today, most of the new Indian hybrids are regular bearing, with good quality fruits (free from spongy tissue) and attractive skin colour. Compared to the cultivars of reference, all the hybrids have a marginally higher pulp yield and possessed lower peel, stone and fibre content. Nevertheless, owing problems of commercialisation and promotion in markets, the adoption of these new varieties is still fairly low. In Israel, interesting seedlings were selected for peel colour, fruit quality and favourable harvesting season. With regard to rootstocks resistant, or tolerant to calcaerous and high salinity soil, several monoembryonic and polyembryonoic hybrids have been identified but none performed better than '13-1', the currently preferred rootstock in Israël. In South Africa, the oustanding new variety "Heidi" was released in 1990 and is already commercialized at the international level. Thereafter, a comprehensive breeding programme was initiated on a sound genetic strategy.

With regard to sport selection, there are two reports of somatic mutants giving rise to new cultivars. In USA, "Davis Haden" has been identified as being larger and flowering one month earlier than the donor parent "Haden". Another cultivar is "Rosica" from Peru which is a bud mutant of the Peruvian cultiru which is a bud mutant of the Peruvian cultivar "Rosado de Ica". Unlike the parent, this mutant is high-yielding and regular bearing and does not produce seedless fruits.

Conclusion and international cooperation: As a result of the implementation of controlled pollinations, good procedures of evaluation and selection in the field, new outstanding hybrids of mango were released. Nevertheless, as stated by Israeli breeders, mango breeding is still in their infancy phase; thus , it must be emphasized that considerable genetic and varietal progress should be expected from long term integrated programmes including all available strategies. In the next few years, mango improvement will have to take into account also the improved knowledge of inheritance of specific characters (recessivity for polyembryony, dwarfism, regular bearing and precocity for example) and the recent findings on heritability pointing out that the additive genetic variance was small and non-significant whereas the non-additive variance was found to be large and significant in most traits. Moreover, the integration of protein (isozymes) and molecular markers (RAPD's, AFLP's, VNTR) will bring more information regarding the global genetic diversity in mango and the identification of genetic relatedness between clones. The use of this markers might allow the breeder to predict the parent combination which will led to superior progeniebination which will led to superior progenies and in the future, and might assist selection in early screening in nurseries.

In the future, it might be expected that better efficiency in mango breeding will rely on planned hybridization assisted by the new tools offered by biotechnologies. The recent emergence of molecular markers and the application of somatic embryogenesis to genetic transformation will enable the integration of specific genes from cultivated varieties or wild species into popular current cultivars.

Part 2 : TEMPERATE SPECIES

APPLE

Main research objectives - breeding methods.

From the objectives that breeders are mainly looking for, two are most often mentionned : fruit quality and disease resistance. Adaptation to climatic conditions is also of prime interest for the countries located in extreme areas. Many apple breeders are also involved in tree habit studies; their objective is to obtain productive and regular cropping trees. Storage ability, and harvesting period are also objectives developed in some programmes. Several institutes are involved in breeding for fruit processing cultivars; two have a specific activity on breeding for cider apples.

Fruit quality

Selection criteria

Releasing apple cultivars with high fruit quality is obviously the major aim of each breeder. All arounduit quality is obviously the major aim of each breeder. All around the world, the same criteria are taken into account to assess fruit quality:
fruit appearance : ground colour, overcolour and type of colour are the first criteria. Fruit shape and fruit size are also frequently assessed.
fruit taste is appreciated by various characters. The most important are quality of the flesh texture, firmness, juiciness, sugar and acid content. Flavor also is a very important criteria. No objective measurements are carried out to assess it.

In Finland and Russia, emphasis is also put on vitamine C content.

Breeding for cultivars intended to processing market requires very specific criteria. Fruit appearance is not taken into account but sensory and technological characters are carefully assessed and measured. Criteria are closely dependent of the type of final product : juice, cider, dried fruits, compote.

Breeding objectives

Fruit quality perception is very subjective and its assessment varies between people and countries. Breeding programmes in China, Japan, Brazil or India focus on sweet fruits : Fuji, Gala, Red Delicious are the references. People from Northern countries prefer more acidic fruits as Braeburn, Jonagold or Elstar.

But most of the apple breeders don't focus on only one type of cultivar . They aim to release a wide range of selections, including various coloured types, taste, harvesting periods. However some coloured types, taste, harvesting periods. However some breeders select very specific types of fruit adapted to the preference of local consumers or growers. In India, for example, the breeding aim is to release resistant sweet-flavored cultivars similar to Red Delicious. Many apple breeders are concerned with storage ability, but very few advanced studies which are performed on this character.

Parents

All over the world, the great commercial cultivars are included on a large scale in breeding programmes to improve fruit quality. Golden Delicious and its derivatives, as Gala, are the parents the most often used in the crosses. Jonathan, Red Delicious, Granny Smith, Idared and more recently Fuji and Braeburn are also frequently used.

In addition cultivars more adapted to local preferences, and their derivatives, are used : Mac Intosh in Canada, Cox Orange Pippin in Great Britain, Lobo and Aroma in Nordic countries, Elstar, Gloster, Melrose in the North of Europe.

Some institutes include also landraces or old cultivars to improve fruit quality, especially in former USSR, Nordic countries, Belgium, and in Spanish and French cider and juice breeding programmes. Generally, these local cultivars besides good fruit quality, transmit also a good environmental adaptation.

Disease and pest resistances

In most of the countries, scab is the major apple disease. It is the first disease targetb is the major apple disease. It is the first disease targeted in the breeding programmes. Mildew resistance is also commonly introduced. The final aim of the breeder is to release cultivars carrying multiple resistance.

Example of scab resistance

Although various sources of scab resistance have been identified in different wild Malus species, Vf gene is widely used in all the scab resistant breeding programmes : more than 80% of the scab resistant cultivars released today and the parents currently included in the breeding programmes carry the Vf gene. One breeding strategy has been performed all over the world: a succession of modified backcrosses between scab resistant hybrids and commercial scab susceptible cultivars. For 50 years, several crosses have been made from the initial cross : 7 or 8 successives crosses for the last Vf selections.

Breeding for durable resistance to scab is for some institutes a main objective. This tendency has increased since the recent discoveries of new strains of scab which overcome the Vf gene. This leads apple breeders to change crossing strategies : the aim is now to combine several types of resistance in one genotype. Molecular markers will be essential to detect in the progenies, seedlings which carry more than one gene of resistance.

The other alternative is to use polygenic resistances. In almost every country, some landraces or old varieties which show good field resistancces or old varieties which show good field resistance have been included in breeding programmes. But, usual scab screening technics are difficult to adapt to this material : progenies are very difficult to screen in glasshouse (very high level of susceptible plants; difficult assessment of resistance symptoms); the correlation between leaf and fruit resistance is low. Thus, the use of polygenic resistance in apple breeding programmes needs new scab screening processes and new selection strategies.

Climatic adaptation

Main programmes are developped in the Nordic countries and in Canada. Objectives are to release winterhardy cultivars adapted to short and cool growing season. Most progenitors are local cultivars.

An important programme is also underway in Brazil : releasing cultivars with low chilling requirements adapted to the various major apple areas in South America as its main aim.

Usually, the aim of these programmes is not to develop worldwide cultivars but to release local cultivars not only adapted to local climatic conditions but also to the economic and industrial environment of the production area.

Tree habit

Almost all the apple breeders highlight their tree habit programmes on compact trees . Spur sports important commercial cultivars, i.e. spur mutants of Golden Delicious, Red Delicious, are widely used to transmit this character. Mc Wijcick, are widely used to transmit this character. Mc Wijcick, a sport of Mc Intosh with columnar habit is also often used. But this type of tree is very susceptible to biennial bearing.

The French breeding aim for tree habit is to create weeping and not vigorous trees, which give good and regular cropping and are easy to prune and harvest. Parents which transmit the character "one fruit per cluster" are also included.

Noticeable results

Many apple breeding programmes have been reviewed . In spite of very different locations and very different means, programmes show a great "uniformity" : breeding objectives are quite identical; except in countries situated in extreme climatic conditions, all the institutes aim to release scab resistant cultivars with high fruit quality. Progenitors are also very similar between the programmes : the pool of progenitors on which apple breeders have been working with is very limited : to improve fruit quality, all the great commercial varieties have been used; they are already derived from a very narrow genetic base; to improve scab and mildew resistances, very few progenitors have been used, despite the large diversity in the genus Malus. Furthermore, the main sources of resistance are mono or oligogenic, Vf for scab resistance and Pl2 for mildew resistance. But with the discovery of new strains of scab capable of overcoming the major genes of ref scab capable of overcoming the major genes of resistance, apple breeders begins to develop new strategies which aim to pyramid various types and sources of resistance in one individual. At the same time, great emphasis is put on the development of molecular markers. The mapping of different major genes for resistance is progressing. To enlarge the variability of the sources of resistance, Malus species and landraces are screened for various resistances.

Apple breeding programmes have already been very successful : among the varieties of interest today, several issued from controlled hybridization, i.e. Jonagold, Fuji, Elstar, Gala and also Quinguan , in China. They give significant improvement compared with older cultivars. Furthermore, many new selections are currently being tested in variety trials throughout the world. The number of promising scab resistant selections is increasing.

International Cooperation

These successes must not hide problems for the future. Apple breeding programmes could be more efficient if more real scientific collaborations could be developped; on the one hand with scientists from other disciplines (pathologists, zoologists, specialists in fruit quality , of storage...). Apple breeding institutes must not only to exchange budwood but also to manage more efficiently the breeding programmes and the diversity of sources of resistance. But the increasing privatizaties of resistance. But the increasing privatization of breeding programmes with the related strong commercial interest should not be a constraint for future collaboration.


PEACH

1. Main research objectives - breeding methods

Main selection criteria in breeding programmes deal with fruit characteristics such as large size, firmness, attractive shape and color, good quality and transpor-ability. Breeding programmes also often aim at extending the ripening season both toward earlier and later productions. Adaptation to specific environmental conditions (low chilling requirements for subtropical areas, winter hardiness for northern areas, resistance to spring frost through for example high flower bud density) appear to be important concerns for the setting of national or regional breeding programmes. Tree habit has arisen as a possible breeding goal in the eighties. Reduced-tree size and growth have been looked for, using dwarf or compact mutants as progenitors. Surprisingly, apart from a few exceptions, disease and pest resistance is not directly evaluated. In most cases, hybrid susceptibility is only assessed in experiments designed for fruit evaluation.

The genetics of peach fruit and tree characteristics has been studied for many years. These studies have mainly lead to the identirs. These studies have mainly lead to the identification of major genes governing fruit characteristics (yellow/white flesh, melting/non melting flesh, cling/free stone, peach/nectarine skin) that correspond to distinct commercial fruit types. Major genes governing tree habit and some flower and leaf characteristics have also been identified. Conversely, very few resistance genes have been identified and the inheritance of complex characters has been given comparatively little attention.

Released cultivars are mainly obtained from control crosses, sometimes after a further generation of selfing, but selection among seedlings from unknown origin or open pollination is not uncommon. A few cultivars are mutants ; they appeared in production orchards. As a very general rule, individual selection is performed. Progenitors are mainly choosen on empirical knowledge about their own characteristics and, when available, their past performance as progenitors. Biotechnology, except for the rescue of immature embryos, is hardly used in breeding programmes, due to difficulties in the regeneration of explants from mature plants.

2. Leader teams

Peach cultivar development has been and is still mainly performed by USA teams, and noticeably USDA laboratories and Zaiger and Bradfoticeably USDA laboratories and Zaiger and Bradford (private breeders). USA cultivars are cultivated worldwide and used as progenitors in most breeding programmes, sometimes for decades. However, significant releases (54 %) are from other countries. In several countries, a trend toward the separation of basic research and strict cultivar development is now being observed.

Accordingly, the relative importance of private breeders' introductions increases. Public programmes more and more aim at understanding the physiological and genetic basis of some characteristics (fruit ripening , fruit color, chilling requirement, tree growth and development, diseases and pests resistance). No clear leadership is observed in that respect, as several teams, often in interdisciplinary associations, contribute significantly to knowledge improvement.

3. Noticeable results of last years

In the field of cultivar development, steady improvement of fruit characteristics is observed. For the development of genetic studies, the development of saturated maps with RFLP markers is an important step, that is being achieved for peach.

4.Possibilities for international cooperation

International cooperation could develop between countries encountering the same specific problems. The major pests and diseases are not the same in all productions areas,nd diseases are not the same in all productions areas, but some (Sphaerotheca pannosa, plum pox virus in Europe, Myzus persicae in Europe and Asia, Monilia laxa or fructicola in most countries) are found in extensive areas. Exchange of germplasm, development of screening tests, assessment of germplasm in several countries could be helpful to design breeding programmes for durable resistance. Similarly, germplasm exchange exist and could intensify between countries with similar climatic environments.

Physiological and genetic research about fruit quality are developing in several fields (sugar and acid contents, polyphenols, evolution of firmness, anthocyanins). The application of the results obtained for the breeding of fruit quality would benefit from collaboration between breeding teams involved in the study of the various aspects of fruit quality.


APRICOT

1.Main research objectives ­ Breeding methods

The main research objectives in apricot breeding programmes deal with different types of products :

The main common selection criteria are :