LUPIN BREEDING

LUPIN BREEDING

B.S. Kurlovich and L.T. Kartuzova

Tendencies of lupin breeding

The most promising trend in breeding work is the complex approach employing the knowledge in the sphere of private genetics of qualitative and quantitative characters, practical utilization of the developed or identified genetic sources (donors) of economically valuable characters, and application of the models of statistical genetics and biometrics. It will help to solve the following problems:

· To study the breeding value of separate genes, their groups and donors developed on their basis with subsequent utilization of the best of them in the breeding programs (Мережко, 1994);

· To obtain transgressive forms using accessions with allelic distinctions in the control of characters in the breeding process. Transgressive forms of lupin can be effectively identified with the help of the eco-geographical approach discussed in the previous section and in other publications (Kurlovich et al, 1995, 2000b; Dragavtsev, 1997; Репьев and Барулин, 1998).

· To develop and identify initial breeding materials possessing group or horizontal resistance to different strains, races and species of diseases at the expense of accumulation of different resistance genes in the developed genotypes by testing the initial materials in contrast environments and selection of pairs for crossing on the basis of the eco-geographical approach (Kurlovich et al., 1995). Breeding of lupin cultivars resistant to different diseases (especially anthracnose and Fusarium) and pests is an extremely burning problem of breeding practice for all species of lupin.

· To study qualitative structure of populations of definite disease agents under definite conditions. This will make possible targeted breeding of resistant cultivars for each region (Ларионова and Курлович, 1990);

· To obtain different mutants with new economically valuable breeding characters and effectively uses them in breeding practice.

Biological nitrogen fixation is the problem of the highest priority in biological and agricultural sciences as well as in the attempts to develop sustainable agricultural production. Development of this trend is a response to ecological and energy problems recently encountered by agriculture. For effective utilization of biological nitrogen fixation, it must be optimized on the basis of wide involvement of the diversity of lupin forms and strains of Bradyrhizobium sp. (Lupinus). We have formulated the following major problems of this breeding trend:

· Selection of the strains of Bradyrhizobium sp. (Lupinus) most effectively compatible with already existing lupin cultivars;

· Breeding of high-yielding lupin cultivars with high nodulation ability (Nod), activity (Fix) and efficiency (Eff) of symbiotic nitrogen fixation, capable of interacting with appropriate strains of nodule bacteria in the natural conditions of low content or complete absence of mineral nitrogen in soil;

· Search for lupin plants capable of selecting a definite and effective strain of bacteria from the soil with maximum efficiency;

· Breeding of lupin cultivars with the ability to avoid competitiveness with spontaneous microflora and positively react to purposefully selected strains of bacteria;

· Identification of lupin forms ensuring high nitrogen fixing ability under interaction with spontaneous microflora (without application of commercial preparations);

· Breeding of plants with the ability to form large number of nodules with high activity and efficiency of biological fixation, and friendly to sufficient mineral nitrogen content in soil, without using it;

· Search for lupin samples demonstrating positive results on mineral nitrogen with application of the preparations of nodule bacteria;

· Creation of an effective complementary symbiosis of micro- and macro-symbionts for different breeding problems.

At the first stages of realization of the above-indicated tasks, it is recommended to use the source materials that we have presented in this work and in our previous publications (Kurlovich et al, 2000a). An important task of annual lupin breeding in Russia is selection for earliness and stable seed yield. Early-ripening cultivars should have a number of other positive qualities. First of all, they need to have the habit of rapid growth, which can facilitate weed control. All new cultivars of narrow-leafed lupin should have non-dehiscent pods and determinate or sympodial type of branching. It is desirable to remove the trait of premature falling of flowers and pods in new yellow lupin cultivars.

Another important direction in the breeding of all species is further increase of the contents of protein and oil and improvement of their quality in seed and green matter. It will open the way to use lupin seed not only for animal feed, but also for human food. Besides, it is necessary to breed new cultivars with thin pod valves (to increase the ratio of seed weight to pod weight) and to enhance the factor of seed duplication at the expense of their weight reduction. A new promising trend in the breeding of all species is the development of forms with the determinate type of branching.

While solving all these urgent problems, it is necessary to search constantly for new and stable low-alkaloid forms of various species. It may open new opportunities and prospects. However, it seems impractical to reject completely the opposite direction, i.e. breeding of alkaloid cultivars. They are especially promising for green manure production. Besides, modern industrial technologies make it possible to remove alkaloids and use them in medicine and cosmetics (Mohr, 1988).

For the conditions of Russia, the basic directions of lupin breeding are as follows:

· Yellow lupin. Breeding of semi-early and mid-ripening cultivars of an intensive type with high and stable seed yield up to 2.5-3.0 and green matter yield up to 40.0-60.0 t/ha, with protein content in seed up to 50% and in green matter up to 17%, and with resistance to virus and fungus diseases.

· Narrow-leafed lupin. Breeding of early cultivars of an intensive type, with high and stable seed yield up to 2.5-3.0 and green matter yield of 35.0-40.0 t/ha, with high protein content in seed (up to 35%) and green matter (17-20%), with non-dehiscent pods, and resistant to fungus and virus diseases, frost (down to -10оC) and short-term drought.

· White lupin. Breeding of mid-ripening and semi-early cultivars of an intensive type with high and stable seed yield (up to 3.5-4.0 t/ha), with protein content in seed up to 40%, with high and stable green matter yield (up to 50.0-60.0 t/ha), with protein content of 16-18 %, with resistance to fungus and virus diseases and extreme environmental factors.

· Multifoliate or Washington lupin (Lupinus polyphyllus Lindl.). Breeding of cultivars having stable low alkaloid content, different types of pollination, non-dehiscent pods, winter hardiness and frost tolerance, and capable to produce high yields of green matter (40.0-50.0 t/ha) and maintain shading of cover crops.

Breeding Methods

Methods of selection. The mankind has been using for ages the natural populations of lupin, which always contained large diversity of morphological, biological and economic characters. On the basis of such populations, folk breeding produced high-yielding local varieties of different lupin species under the influence of natural and artificial selection in the Mediterranean and American countries. Nowadays, many cultivars have been bred analytically with the help of individual and group selection (yellow lupin cvs. Academichesky 1, Быстрорастущий 4 and others). Very important is recurrent selection on the level of populations. It helps to distribute a population into separate and more homogeneous groups. Further crossing seems the most effective when carried out between their best forms (Таранухо, 1980). Different types of selection are an integral element at all stages of the breeding process.

Intraspecific hybridization. Now, the basic breeding method is experimental and spontaneous intraspecific hybridization. Step-by-step crossing is used more often, enabling the breeder to unite in one genotype valuable characters and properties inherent in many initial forms. Multiple, back, reciprocal, diallelic, polyallelic and others crossings are used in recurrent schemes of hybridization.

A classical example of most effective application of step-by-step hybridization is the breeding of the series of German cultivars Weiko (Майсурян and Атабекова, 1974; Таранухо, 1979). However, selection is also continuously used in the process of hybridization. It is possible to produce new stable forms, as a result of combinative variability, already in the third generation by using the pedigree method. In recent years, the method of selecting one seed in posterity (SSD method) has become popular.

To avoid restoration of alkaloid synthesis in cross-pollinated species of lupin, a new approach has been developed on the basis of specific crossing. It applies, in particular, to multifoliate or Washington lupin (Lupinus polyphyllus Lindl.). The principal distinctive feature of the developed approach is that all initial sources of low alkaloid content, whether identified or bred, are not united to ensure cross pollination, as it was done earlier (Voronov, 1974). Only compatible forms are involved in hybridization, with their low alkaloid content controlled by one and the same genetic system. For practical application of this approach, two methods were worked out. The first method of obtaining fodder (low-alkaloid) forms of L. polyphyllus Lindl. was as follows: the selected low-alkaloid plants were crossed with each other and progeny from every crossing was grown separately. Populations entirely consisting of low-alkaloid plants were selected as initial materials for breeding (Чекалин and Курлович, 1989). Regarding the second method, each low-alkaloid plant was crossed with a productive alkaloid-containing form possessing other valuable characters, and populations with stable low alkaloid content were selected in F2 (Курлович and Чекалин,1992). Comparative characteristics of the produced fodder (low-alkaloid) forms witness that both methods can be used for breeding low-alkaloid varieties. The negative feature of the first method (when the selected low-alkaloid plants were crossed with each other) is that it is too labor consuming. The experiment required a large number of low-alkaloid plants, which very often had some negative properties (low productivity, susceptibility to diseases, etc.). The aim of the second method was to facilitate the process and obtain high-yielding forms. Populations bred by the second method (when each low-alkaloid plant was crossed with productive alkaloid forms) were more high-yielding and resistant to Fusarium wilt. The productivity of these populations was practically the same as in the initial alkaloid forms (Tab. 19 in the book!).

The suggested ways exclude the possibility of alkaloid plants’ occurrence at the expense of genetic homogeneity in the obtained low-alkaloid populations and the absence of complementary interaction of non-allelic genes leading to reconstruction of high alkaloid content. However, in such fodder (sweet) crops as yellow, white and blue lupin, as well as in fodder low-alkaloid forms of multifoliate Washington lupin, bitter plants may appear from time to time as a result of inverse mutations or re-pollination of low-alkaloid (sweet) forms with the pollen of wild high-alkaloid (bitter) plants. In all phases of seed production, involving fodder low-alkaloid forms of multifoliate Washington lupin, it is necessary to ensure strict observance of spatial isolation between different samples, as well as constantly to perform checkup procedures to control the level of alkaloids. Bitter plants, if identified, should be removed before blossoming. Fodder low-alkaloid multifoliate Washington lupin is a perennial entomophilous plant. In view of this, for efficient seed production it seems reasonable to establish a special nursery for a term of many years, and to remove wild bitter lupin plants from the adjacent area. In the first year, all plants in the nursery need to be tested for the presence of alkaloids by pressing leaves and petioles against alkaloid-sensitive paper. All identified bitter plants should be removed before blossoming. Subsequently, such nursery will satisfy the demand for seed for many years on greater areas.

Using the developed methods and the studied source materials made it possible to produce fodder (sweet) forms of multifoliate Washington lupin with sufficiently stable low alkaloid content (VIR-1, VIR-3, VIR-5, VIR-6 ). For the conditions of the Ukraine, Truvor, a fodder cultivar of multifoliate lupin, was bred there. Now it is passing the Ukrainian state variety trials. In the conditions of the northwestern Russia, positive results were demonstrated by the commercial cultivar Pervenec (the first sweet variety that we have bred) which has been listed in the State Catalogue of Breeding Achievements in Russia.

The developed methods may also be applied to such cross-pollinated lupin species as L. nootkatensis Doon., L. mutabilis Sweet., L. arboreus Sims., L. perennis L., L. elegans H.B.K., L. hartwegii Lindl. and other forms promising for agricultural production.

The approaches designed by Dragavtsev (Драгавцев, 1993) are effective in computer-aided selection of the best two (or several) parents for hybridization, when the breeding process is targeted at increasing productivity, stability, quality and complex quantitative characters. For these purposes, the breeder is expected to select unique plants carrying both genetic systems of their parents, being transgressions and ancestors of valuable cultivars, by means of identifying genetic systems in F2 either in glasshouses or in the field (with the same typical dynamics of the limiting factors).

Interspecific hybridization. The polymorphic genus Lupinus L. unites two independent and detached from each other groups of species occurring in the eastern and western hemispheres. Divergent ways of their evolution have set between them an isolating barrier against interbreeding. Lupins of these two groups cannot produce interspecific hybrids. Also fruitless were the attempts to make intercrosses with a majority of species with different chromosome numbers from the eastern hemisphere. However, many lupin species from the western hemisphere with 2n=48 are easily crossed with each other and produce entirely or sufficiently fertile progeny. Besides, many of the now recognized species of lupin from the western hemisphere are latent hybrids (Майсурян and Атабекова, 1974).

Most amply described in literature are the following hybrids: L.mutabilis x L.polyphyllus, L. ornatus x L. mutabilis, L. pubescens x L. hartwegii, L. mutabilis x L. elegans, L. mutabilis x L. albococcineus, L. nootkatensis x L. arboreus, L. arboreus x L. hartwegii, L. mutabilis x L. douglasii(Kazimierski, 1960, 1963; Kazimierski and Nowacki, 1961b; Kazimierski and Kazimierska, 1970; Kazimierska, 1970; Maissurjan and Atabekova, 1974; Schäfer-Menuar et al., 1988; Almut Busmann-Loock et al., 1992). As to the species of lupin from the western hemisphere (subgen. Lupinus), by now it was possible to make successful intercrosses only with yellow lupin (L.luteus) including both subspecies of Spanish lupin (ssp. hispanicus andssp. bicolor) (Lamberts, 1955, 1958; Kazimierski and Kazimierska, 1970; Maissurjan and Atabekova, 1974; Swęcicki, 1988). More viable progeny of breeding value is obtained by crossing L. luteus with L. hispanicus ssp. bicolor (synonym - L. rothmaleri), and also L. luteus with hybrids ssp. hispanicus x ssp. bicolor (Cordero et al., 1988). Also described are the results of interspecific crosses among the species of section Scabrispermae: L. cosentinii, L. digitatus, L. atlanticus, L. pilosus and L. princei (Roy and Gladstones, 1985, 1988; Carstairs et all, 1992). In a number of crossings it was possible to produce sterile F1 hybrids. More successful results in this area were described by Buirchell (1994) and Atkins et al. (1998). Besides, the following hybrids were also described in publications as interspecific: L. angustifolius x L. linifolius, L. angustifolius x L. opsianthus and also L. albus x L. vavilovi and L. albus x L. jugoslavicus(Kazimierski and Novacki, 1961; Atabekova and Maissurjan, 1962, 1968; Maissurjan and Atabekova, 1974). Later, however, it was found out (Gladstones, 1970, 1974; Курлович and Станкевич, 1990; Kurlovich et al., 1995) that such hybrids were not interspecific, as L. linifolius and L. opsianthus appeared to be nothing more than separate varieties and ecotypes of L. angustifolius, while L. vavlovi and L. jugoslavicus were synonyms of a subspecies of white lupin (subsp. graecus). Low fertility of some interspecific hybrids is a consequence of deviations in meiosis and various anomalies during development of the embryo sac (Kazimierski, 1961b; Kazimierska, 1970). Meanwhile, positive results undoubtedly represent theoretical interest and specify about relationships and degree of affinity between various species of lupin.

There are usually the following failures after pollination in hybrids (Atkins et al., 1998):
· imply that the pollination failed for „environmental” reasons;
failure of роllеn to germinate; · failure of pollen to develop pollen tubes that penetrate the stigma; · failure of pollen to grow, or grow far enough, down style;
· failure of роllеn to find ovary and penetrate ovule;
· failure of ovary stimulation and early embryo;
· failure of later embryo development;
· inviability оf the developed embryo.
Achievements of modern and future biotechnology will undoubtedly help to overcome barriers between many species of lupin. That will open new opportunities and prospects for this crop (Sator, 1984; Atkins et al., 1998).

Mutagenesis. A majority of lupin cultivars were produced with the use of spontaneous or induced mutants. The first fodder (sweet) cultivars of yellow, narrow-leafed and white lupin were bred in Germany in 1927-1928 as a result of reproducing the identified mutant plants with low alkaloid content (Sengbusch, 1931, 1942; Майсурян and Атабекова, 1974; Коновалов et al., 1990). Natural mutants were later used in hybridization for improvement of many parameters (white coloring and fast swelling of seed, non-dehiscence and non-pubescence of pods, fast rates of initial growth). A series of famous German cultivars of yellow lupin, under the name Weiko, was produced by step-by-step hybridization of the identified mutant plants. In recent years, diverse chemical substances and various kinds of ionized radiation were widely used to induce mutations. Promising for lupin is the method of artificial mutagenesis, because it helps considerably to expand the range and increase the occurrence of the forms with changed heredity bases in this young crop. More useful mutations were observed the effect of chemical mutagenes, as compared with physical factors (Солодюк, 1971; Дебелый and Зекунов, 1977). Their concentration and performance depend on the object (varietal features, dry or swelled seed, fodder or bitter forms, different species of lupin). The best results were reported during our experiments after soaking the seed in distilled water for 8 hrs and 3 hrs in MNU of 1.5 mM concentration. One should also take into account different intensity of radiation when physical factors are used (x-ray or gamma-ray). Its dosage depends on the biological features of accessions and the selected characters (Коновалов et al., 1990). The influence of radiation on the growth, development and fruit bearing in M1, on the occurrence of morphological mutations in М2 and М3 and biochemical mutations (including low content of alkaloids) in M3 have been studied in white lupin (Солодюк and Головченко, 1984). The probability of mutations with economically valuable characters has been determined.

Mutants of narrow-leafed lupin are subdivided by Hackbarth (1957b) into eight groups: 1) mutations of the leaf, with variations in the intensity of color, pigmentation and size of this organ; 2) mutations of the flower, with variations of its color; 3) seed, with variations of shape, size and color; 4) pod, with variations of the pod size and various degrees of resistance to dehiscence; 5) physiological mutations with late and early maturity and with variable rhythm of growth; 6) biochemical mutations with variations in the content of protein and alkaloids; 7) mutations of resistance to diseases and other stresses; 8) mutations of the whole plant including very tall mutants, dwarf mutants, and mutants with variable habits (compact, without branches, standard). Low alkaloid mutants of Lupinus digitatus Forsk. were received in Australia (Gladsones and Francis, 1965a). Effect of dose and seed moisture content on mutation production in L. angustifolius by X-rays was also discussed (Gladsones and Francis, 1965b). Utilization of natural and artificial mutants in combination with planned hybridization helped to produce the majority of new lupin cultivars.

Polyploidy. This method, capable to yield positive breeding results with many agricultural crops, has yet been underused in breeding practice with lupin. It has happened owing to the difficulties induced by large chromosome numbers in the cells of this crop. Meanwhile, in Poland there was a publication (Kazimierska and Kazimierski, 1969) dedicated to the study of biological features in the haploids of yellow lupin, which were identified among chimeric diploid plants. Those haploids, as well as the dihaploids derived from them, appeared less viable as compared with the common dihaploid forms.

Spontaneous polyploids of L. rothmalery (a synonym of L hispanicus, ssp. bicolor) were also reported by the authors (Kazimierski and Kazimierska, 1981). Tetraploid plants, with 104 chromosomes in somatic cells, were isolated from diploid forms. Assessment of morphological characters and analyses of fertility showed that tetraploid plants, as compared to diploid ones, had larger vegetative and generative organs, larger stomata cells, pollen diameter and seed. Regarding pollen viability, tetraploid plants were slightly inferior to diploid plants, but appeared to be less fertile.

Subsequent researches (Kazimierski and Kazimierska, 1992, 1994) revealed haploids and tetraploids in white lupin. The most interesting from theoretical and practical viewpoints were tetraploids (2n=4х=100). The studies focused on the mechanism of autotetraploid plant occurrence in a hybrid progeny obtained from crossing geographically distant white lupin subspecies. It became clear that autotetraploid plants occurred (Kazimierski and Kazimierska, 1994): a) parthogenetically – through duplication of chromosome number in an unreduced megaspore; b) amphimictically – due to duplication of chromosome number in part of somatic cells; in sector with a doubled chromosome number there occur 2n gametes. Tetraploid plants which had occurred parthenogenetically died in the next generation, whereas those that had occurred amphimictically on chimeric plants appeared to be viable. Autotetraploid plants, if compared with diploid forms, flowered and matured later, were less fertile, and characterized by larger seed. Best of them were no less productive than the usual diploid forms, thus representing interest for future breeding.

Natural polyploids may sometimes also occur among wild lupins of the western hemisphere. Tetraploid plants were found in Lupinus perennis L. (Майсурян and Атабекова, 1974).

Evaluation of breeding materials on infested backgrounds and by artificial infection. Lupin in Russia and other adjacent countries is exposed to numerous diseases, among which the most widespread and devastating are Fusarium wilt and anthracnose. These diseases are caused by the agents belonging to different species and strains of Fusarium Lk. and Colletotrichum gloesporioides Penz. Their incidence and forms are not identical in different soils and climatic zones. Different lupin cultivars resistant to Fusarium wilt and anthracnose in one region frequently appear susceptible in others. Evaluation of source materials and new breeding cultivars of lupin for their resistance to diseases should be carried out on the plots with infested soil. The technique of making backgrounds infected by soil pathogens (Fusarium) is based on accumulation of infection in the soil (Гешеле, 1964; Корнейчук, 1985). The most widespread methods are setting two- or three-year monocultures of highly susceptible variety, entering crushed parts of the affected plants into the soil, and also the method of artificial infection of the soil with pathogens reproduced on nutrient substances (Корнейчук, 1985) when a nutritious substratum may be made from cereal grain. It is also possible to apply the pure culture of a pathogen prepared on a liquid nutritious substratum (Овчинникова and Андрюхина, 1984). Different approaches may be used to study resistance depending on the task posed. For example, our research programs aimed at breeding Fusarium resistant lupins were in­itiated in 1971. The first step of the research was to identify resistance sources in the lupin collection of VIR, including over 50 species of both American and Mediterranean origin.

One of the ways to control Fusarium wilt is to breed cultivars with a group resistance to different strains and races of the pathogen. Source materials were examined for this purpose on the plots with infested soil in contrasting environments at three different sites. By this approach, variation in Fusarium resistance among the acces­sions was recorded, resistance sources were identified, and new forms with a high resistance were selected (Курлович et al. 1990a). The obtained data confirmed the known polygenic nature of Fusarium wilt. Significant variation in the harmful effect produced on a majority of lupin varieties in different regions and against different backgrounds proved the need to study also the qualitative structure of a population of the disease agents for definite environments, and the degree of their virulence. Such information and the data on the resistance of lupin material to this or that population of agents will make it possible to carry out targeted breeding of resistant lupin cultivars for each given area.

Other published works describe in more detail how to set up different infected backgrounds and how to evaluate breeding materials (Хотянович and Муромцев, 1975; Корнейчук, 1985; Ларионовa and Курлович, 1990).

As to anthracnose, the infection of this pathogen is dispersed basically by seed and survives the winter on infested plant residues. This disease provokes appearance of peculiar spots on leaves, which would turn into ulcers on stems and pods. After the rain, mucous conidia may be observed in the place of lesion. The inoculum is diffused by raindrops and infects adjacent healthy plants. The fungus can also pass over onto seeds (Кривченко et al., 1975).

The agent spends the winter together with plant residues left after harvesting, or in the form of microsclerotium, or in the form of fruit bodies. With this in view, the technique of making an infected background for this disease is slightly different. Tested plants are usually sprayed by a solution containing in plenty of the pathogenic strains most virulent in the given region. Races and strains of the agents of anthracnose are not similar in different areas, like those of Fusarium wilt. A search for resistant materials should be conducted in the same place where the breeding process is expected to take place, and the most virulent under local conditions races and strains should be used for infection. This was confirmed by the results of testing white and yellow lupin accessions collected by us during explorations in Brazil (those accessions were a kind gift made by Dr. Augusto C. Baer).

There had been an outburst of this disease in this country, and a number of resistant forms were identified (Baer and Linhares, 1998). However, testing these accessions in different eco-geographic conditions of the ex-USSR republics in most cases failed to reveal high levels of resistance to the disease, when infection was induced by local strains and races of pathogen. BYMV – Bean yellow mosaic virus (Phaseolus virus 2, Smith.) is the most harmful virus disease of lupins worldwide. Two strains of this virus differing in the degree of virulence (the first strain is weakly and the second one is strongly virulent) were identified in the Ukraine when the biological, morphological, physiological and other properties of this disease had been studied (Курлович et al., 1989). Spreading of the mixed infection of BYMV and CMV (Cucumis virus 1, Smith.) also took place. Analysis of accessions for their resistance and search for resistance breeding sources are usually performed in two stages: first, against the background of natural infection (primary assessment), and then, against the background of artificial infection with BYMV (basic assessment).

Source Material for Breeding

Plant Yield

One of the major parameters characterizing the value of this or that accession is its yield. It is assessed when matched with the yield a reference variety. At present, lupin is cultivated and used in Russia both for grain and for green matter production for forage purposes. As witnessed by the researches, the range of variability in seed weight per plant is from 2.0 to 24.6 g for yellow lupin accession, from 2.2 to 26.1 g for white lupin accessions, and from 1.5 to 30.6 g for narrow-leafed lupin. Green matter yield in the phase of flowering varies from 8.0 to 245.0 g for yellow lupin accessions, from 9.0 to 250.0 g for white lupin, and from 5.0 to 120.0 g for narrow-leafed lupin. Sources of high productivity have been identified with different species of lupin.

Yellow lupin (L. luteus L.)
Sources of high seed yield (seed weight per plant is 15.0 - 24.6 g): k - 2004 from Morocco, k.k. - 2050, 2271 from Portugal, k.k. - 2315-2317, 2521, 2529, 2647 from Belarus, k.k. - 2305, 2600, 2616 from Ukraine, and k.k. - 2614, 2615 from Moldavia.
Sources of high green matter yield (weight of green parts of a plant is 180.0 - 245.0 g): Cv. Припять (k - 2652) from Belarus, k.k - 2290, 2292, 2869, 2870 from Spain, and k - 3099 from Portugal.

White lupin (L. albus L.)
Sources of high seed yield (seed mass per plant 20.0 – 26.1 g): k - 495 from Ethiopia, k - 2094 from Libya, k - 2357 from Spain, cvs. Лотос (k - 2222) and Дружба (to - 2603) from Ukraine, Line 09 (k - 2575) from Belarus.
Sources of high green matter yield (mass of green part of a plant 180.0 - 250.0 g): k - 516 from Spain, cv. Снежинка (k - 1596) from Germany, El.Harach-1 (k - 3110) from Algeria.

Narrow- leafed lupin (L. angustifolius L.)
Sources of high seed yield (seed mass per plant 20.0 – 30.6 g): La V (k-1461), La VIII (k-2167), Mirela (k-2575) from Poland, k.k. - 2500, 2501, 2573, 2574, Line 5Ì2-253 (k - 2979) from Russia, БСХА-240 (k - 2748), Лаф-рбс/з from Belarus.
Sources of high green matter yield (mass of green part of a plant 100.0 - 120.0 g): LAKS-1 (k-2573), LAKS-2 (k-2574), LAKS-4 (k-2576), Галина (л - 2705) from Belarus and k.k.-3083-3087 from Portugal.

Biochemical Structure

Lupin, like other grain legumes, is cultivated as a source of high-quality protein, amino acids and oil; so further improvement of seed quality is a priority for breeders. As a result of studying the biochemical structure of lupin genetic resources, breeding sources able to improve seed quality in future cultivars have been identified.
Yellow lupin (L. luteus L.)
Protein content variability in seed is 34.0–55.0%. Correlated to protein, methionine content variability is 0.34–0.94%; lysine variability is 3.0–6.2%; and variability of the sum of essential amino acids is 32.9–50.5%.
Accessions with high protein content in mature seed (42 % and higher): cvs. Кастрычник (k - 2651) from Belarus, Кoрмовой 875 (k - 2151), № 775 (k - 2152), Мартин 2 (k - 2370) from Ukraine, Weiko (k-1310), Weiko II (k-1495), Aschersleben (k-2079), LLU 5855/70 (k-2255) from Germany, k - 2004 from Morocco, k.k. - 1551, 1556, Martini (k-1796), Batavo (k-1798) from the Netherlands, Popularny (k - 1511) from Poland.
Accessions with increased content of methionine ( higher than 0.6 % to protein): cvs. Академический 1 (k - 1947), Кастрычник (k - 2651) from Belarus, Копыловский (k - 2601) from Ukraine, Refusa nova (k - 2071), Borluta (k - 2200), LLU 5855/70 (k - 2255) from Germany, Аfus (k - 2084), Тоmik (k - 2146), Cyt (k - 2398) from Poland and k – 2289 from Portugal.
Accessions with increased content of lysine (higher than 5.0 % to protein): cvs. Академический 1 (k - 1947) from Belarus, Боец (k - 1483) from Latvia, Fodder 190 (k - 1636), Раннеспелый 301 (k - 1726), Юбилейный (k - 1995), k.k. - 2064, 2065, Fodder (k - 2151), k - 2262, Мартин 2 (k - 2370), Искорость (k - 2437) from Ukraine, Искра (k - 2302 from Bryansk area, Schwako (k - 1835) from Hungary, Гюльцо 2 (k - 1400), Weiko II (k-1841), Aschersleben (k - 2079) from Germany, k - 2004 from Morocco, k - 1562, Batavo (k - 1798) from the Netherlands.

White lupin (L. albus L.)
Range of variability of the contents of protein in seed is 35.0-53.7 %; range of variability of methionine is 0.35-0.74 % to protein, and lysine 3.0-6.2 % to protein; contents of irreplaceable amino acids changes from 32.9 to 50.5 % to protein and contents of oil changes from 6.2 to 12.0 %.
Accessions with high protein content in mature seed (45 % and higher): k - 303, k - 2209 from Russia, Sweet White (k - 2288), k - 2334, Turmus (k - 2335), k - 2338, 2375 from Great Britain, k - 1491, Kisvardai-529 (k - 2020) from Hungary, k - 2173, Нерквелл (k - 1496), Hadmerslebener Kraftquell (k - 1661) from Germany, k.k. - 240, 1600, 1601 from Italy, k - 2094, k - 2095 from Libya, k.k. - 294, 295, 298 from Palestine, k - 1513, k - 1538 from Poland, k - 1547 from France, k.k. - 486, 494, 495 from Ethiopia, k - 682, k - 2374 from Yugoslavia.Accessions with increased content of methionine (higher than 0.58 % to protein): k - 2209 from Belarus, k - 2373 from England, k - 828 from Germany, k - 2229 from Syria, k - 495 from Ethiopia.
Accessions with increased content of lysine ( higher than 4.0 % to protein): k - 2173, k - 1661 from Germany, k - 2093 from Libya, k - 1538 from Poland, k 2374 of Yugoslavia. Accessions with the high contents of oil (10,5 % and higher): k - 303, cv. Kiev mutant (k - 1904) from Ukraine, k - 2605, Mutant 5 (k - 2639), Mutant 28 (k - 2640), Mutant 47 (k - 2641), Mutant 52 (k - 2643), Start x Kiev generous (k - 2644) from Russia, k - 2191, Hamburg 30 (k - 2587) from Australia, k - 2339 from England, Kisvardai Feherviragu (k - 1810) from Hungary, k - 243, k - 474, Hansa (k - 2172), Kandanos (k – 1784) from Germany, k - 2297 from Spain, k.k. - 290, 298, 302 from Palestine, Pallidus gorzki (k- 1984) from Poland, k - 2620, k-2624 from Portugal, Lublanc (k - 2589) from France, k - 1649 from Yugoslavia.

Narrow- leafed lupin (L. angustifolius L.)
Range of variability of the contents of protein in seed is 18.0-39.2 %, oil – 3.2-8.5 %.
Accessions with high protein content in mature seed (36 % and higher): № 7816 (k - 2563), Стодолищенский Е-541 (k - 1908) from Russia, Local (k - 1526) from Ukraine, Вада-15 (к-2685), Вада-65 (к-2684), Вада-14 (к-2686), Anba-21 (к-2688), Вада-10 (к-2681), Лаф-рбс/1 (к-2839) from Belarus, k - 1535 from Lithuania.
Accessions with high oil content in mature seed (7,6 % and higher): ГЛ 177 (k - 3152), ДМ-52 (к-2700), дм-117 (к-2846), ДМ-396/1 (к-2844), ДМ-50 (к-2704) from Belarus, N 483 (к-2252), N 7819 (к-7819), Ранний 79 (к-1873) from Russia, LA-V (к-2166) from Poland.

Perl lupin (L. mutabilis Sweet.)
Range of variability of the contents of protein in seed is 42.7-51.2 %, contents of methionine changes from 0.64 to 1.11 % to protein and contents of oil changes from 10.5 to 16.3 %. Accessions with high protein content in mature seed (47 % and higher): k - 2160 from Russia, CP-54753 (k - 2280) from Australia, Rink Perl (k - 2274), k - 2275 from England, k.k. - 2128, 2129, 2132, 2138 from Peru, k - 1737 from Poland, k - 1566 from Slovakia.
Accessions with increased content of methionine (higher than 1,0 % to protein): k - 2160 from Russia, k - 2275 from England, k - 2017 from Hungary, k - 1918 from Germany, k - 2129, k - 2133 from Peru, Otavalo (k - 2051) from Ecuador.
Accessions with the high contents of oil (15,0 % and higher): k - 2349 from England, Argentina II (k - 2344) from Argentina, Huanoayo 5 (k - 2126), k - 2444, Sog-25 (k - 2459), k - 2460, Bg 309 (k - 2595) from Peru.

High nitrogen fixing ability

White lupin (L. albus L.)
Accessions ensuring an increasing of a crop yield under processing of Bradyrhizobium sp. (Lupinus) bacteria (strain 363a) without application of mineral nitrogen (in comparison with control): cvs. Start (k-2644) from Russia, Olezka (k-2980) from Ukraine accessions k-2989 and k-3250 from Portugal, k-2864 from Greece, El Harrach-1 (k-3110) from Algeria.
Accessions with high activity of nitrogenase : Lines 802-15 (k-2623) and 48B (k-7986) from Portugal, accessions k-507 from Egypt, El Harrach-1 (k-3110) from Egypt, k-1602 from Poland.
Accession described by increasing of nitrogenase activity under artificial processing of Bradyrhizobium sp. (Lupinus) bacteria : cvs. Snezinka (k-1596) and Tambovsky 86 (k-2806) from Russia, k-1601 from Italy and Lublanc (k-2589) from France.

Narrow-leafed lupin (L. angustifolius L.)
Accessions ensuring an increasing of a crop yield (in comparison with control) under processing of Bradyrhizobium sp. (Lupinus) bacteria (strain 367A) without application of mineral nitrogen: k-3065 from Australia, Determinant-2 (k-3365) and Determinant-3 (k-3366) from Russia, Apva (k-2950), Vika 65 (k-2954), DG-94 (k-3351) and DG-95 (k-3352) from Belarus, wild forms k- 3076, k-3079 from Spain, k-3083 from Portugal and k-3093 from Marokko.
Accessions ensuring an increasing of green and dry matter yieldunder processing of Bradyrhizobium sp. (Lupinus) bacteria (strain 367A) also in a combination with application of mineral nitrogen: cv. Unicrop (k-2096), Lines 75A/326 (k-3061), 75A/330 (k-3064) from Australia, cv. Melkosemianny (k-1354) from Latvia, Mut-1(k-2803 from Poland, accessions Vada 10 (k-2681), Jniven (k-2953) from Belarus, Nemchinovsky 846 (k-1981), Determinant 4 (k-3367) from Russia, wild forms k-3079, k-3081, k-3082 from Spain, k-3083, k-3084, k-3087, k-3090 from Portugal, k-3093, k-3094, k- 3097 from Morocco.

Yellow lupin (L. luteus L.)
Accessions ensuring an increasing of a crop yield under processing of Bradyrhizobium sp. (Lupinus) bacteria (strain 375A) without application of mineral nitrogen:
cvs. Sojuz (k-2610), Foton (k-2649) from Ukraine, T-12 (k-2000) from Sweden, k-2869 and k-3070 from Spain, wild form k-2292 from Portugal, cv. Cyt (k-2398) from Poland, and cv. Augy (k-2956) from Lithuania.
Accessions ensuring an increasing of green and dry matter yieldunder processing of Bradyrhizobium sp. (Lupinus) bacteria (strain 375A) in a combination with application of mineral nitrogen:
cvs. Sojuz (k-2610), Foton (k-2649) from Ukraine, k-2869 from Spain, wild form k-2292 from Portugal.

Duration of the Period of Vegetation

In Russia, Byelorussia and the Ukraine, duration of the period of vegetation varies in spring forms, depending on the cultivars, year and place of cultivation, from 72 to 170 days for narrow-leafed lupin, from 90 to 175 days for yellow lupin, and from 106 to 180 days for white lupin.

Narrow- leafed lupin (L. angustifolius L.)
Accessions belonging to the very early group (71-100 days): Lanedeks-1 (k - 2687), Lanedeks -2(k - 2696) from Belarus, Stodolishchensky Fodder L-677 (k- 2225), Stodolishchensky L-633 (k - 2226), Oryol Mutant (k - 2500), Ladny (k - 2648), Dikaf-14 (k - 3840) from Russia, La -V (k - 2166) from Poland, and k - 2119 from Sweden.

Yellow lupin (L. luteus L.)
Accessions belonging to the very early group (71-100 days): Zhitomir Anniversary (k - 2154), Limonnyi Belosemyannyi (k - 2153), Limonnyi Serosemyannyi (k - 2154), Plamennyi (k - 2155) from the -Ukraine, Akademik -1 (k - 1947), Zhodinsky (k - 2284), Kastrychnik (k - 2651), BSKhA-382 (k - 2735) from Belarus, Tedin II (k - 2579) and Baltyk II (k - 2163) from Poland.

White lupin (L. albus L.)
Accessions belonging to the early droup (101-115 days): Tel Karam (k - 290) from Palestine, B-2 (k - 2234), B -3 (k - 2235), B-4 (k - 2236), B-5 (k - 2237), line 09 (k - 2572) from Belarus; commercial cultivars in the ex-USSR countries: Start (k - 2498) and Gorizont (k - 2026). The earliest among the wild species of lupin in Russian conditions are L.nanus Dougl. and L.lindleuanus Agardh.

Determinate Branching

The forms with determinate branching have been selected and described in yellow lupin (Валовненко, 1974); narrow-leafed lupin (Дебелый and Зекунов, 1977; Купцов, 1984; Коновалов, Клочко and Аникеева, 1985; Delane, Hamblin and Gladstones, 1986; Debely and Derbensky, 1988; Kurlovich, 1988; Курлович, 1991; Конарев, Клочко and Аникеева, 1991; Kurłowicz, 1992); white lupin (Micolajczyk J., 1961; Swęcicki W., 1988; Курлович, 1988; Huyghe, 1990; Huyghe et al., 1994), and also L. mutabilis (Römer, 1994).
Many of them are available in the collection of VIR:
Yellow lupin (L. luteus L.):
Cvs. Yubileynyi (k - 1935), Zhitomirsky Yubileynyi (k - 2149), Limonnyi (k - 2154), Iskorost (k - 2437) from the Ukraine; k.k - 2333, 2815-1817, 2827-2829 from Belarus.
Narrow-leafed lupin (L. angustifolius L.):
Ladny (k - 2648), Dikaf-14 (k - 3840) from Russia, Lanedeks-I, Lanedeks -2, Lanedeks-3 from Belarus, Mut-1 ( k- 2803) from Poland.
White lupin (L. albus L.):
EP-1 (k - 2890), EP-2 (k - 2891) and other forms from Poland.
These accessions are widely used in breeding in many countries (Belarus, Germany, Latvia, Lithuania, Poland, Russia, and the Ukraine) as a source of early maturity. Phenotypic manifestation of determinate branching includes the influence of the genotype, influence of the environment, as well as genotype-environment interaction. More detailed information on this problem is presented in the section «Genetics of quantitative Characters».

Non-Dehiscent Pods

In the conditions of Russia, narrow-leafed lupin has such valuable qualities as earliness and high seed yield. However, pods of the domestic cultivars of this species dehisce at maturity, which leads to big losses in seed yield. The character of non-dehiscence in pods was discovered in 1960 in Australia (cv. New Zealand Blue, as a natural mutant). According to Gladstones (1974, 1977), this property is caused by two genes: tardus (ta) and lentus (le). The gene tа limits dehiscence at the expense of fastening pod valves by formation of a solid beam of sclerenchymal cells along all perimeter of the pod. The gene lе reduces shattering at the expense of structural changes in the valves. The layer of dense pigmented grid-like tissue is developed inside them, thus impeding curling of the valves. The valves of such pods acquire an expressed reddish hue by the beginning of seed ripening.

Results obtained by the analysis of F2 hybrids produced by crossing domestic Russian samples with Australian cvs. Uniharvest, Unicrop, Marri, Illyarrie, Chittick and Yandee made it possible to identify 4 phenotypic classes of plants: the first with dehiscent and non-pigmented pods; the second with limited dehiscence in non-pigmented pods; the third with limited dehiscence in pigmented pods (reddish hue); and the fourth with non-dehiscent pigmented pods. Thus, the factual correlation satisfied the theoretical parity of 9:3:3:1. Hence, in this case dihybrid crossing took place (Курлович Б.С., 1988). On the basis of the results produced, practical recommendations for breeders are offered. Australian cvs. Uniharvest, Unicrop, Marri, Illyarrie, Chittick, Yandee, Danja, Yorrel, Warrah, Cungirri, and others, possessing two independently working genes of non-dehiscence (ta and le), are recommended as source material for hybridization with disease-resistant forms adapted to local environments. It is possible to produce plants with practically non-dehiscent pods solely by combining both dehiscence controlling genes in one genotype. It seems expedient to search for non-dehiscent material in F2 among the forms with pigmented pods, because one of the genes (le) provides for this marker attribute (reddish pigmentation of a pod and its thickening). Research efforts helped to disclose the properties of the Australian cultivars and identify Fusarium-resistant and early-maturing sources of non-dehiscence in pods, namely VIR-2 ang (k - 3290) and VIR-3 ang (k - 3291). Efficient secondary donors of non-dehiscence were also released on the basis of the Australian cultivars (k.k - 2677-2682, 2685) in other institutions (Byelorussian Institute of Arable Farming, Enterprise ’’Podmoskovie’’, Timiryazev Agricultural Academy in Moscow). Of special interest are the mutants Lanedeks-1 (k - 2687) and Lanedeks -2 (k - 2696) developed by Dr. Kuptsov (Byelorussian Institute of Arable Farming). They are also characterized by determinate branching and easy transferability of all these properties to their progeny. Mechanisms and features of pod dehiscence habit in other lupin species are discussed in more detail in the section «Anatomic Structure».

Resistance to Diseases

Fusarium Wilt
The accessions of lupin from the collection of VIR were evaluated for susceptibility to this disease on different infectious backgrounds. The results are discussed in the sections «Evaluation of breeding materials against infested backgrounds and by artificial infection» and «Diseases and Pests». The following accessions are most widely engaged in different breeding programs:
Yellow lupin (L. luteus L.):
Cultivars and accessions: BSKhA-382 and Line 95/3 (Belarus), Kopylovsky, Pripyatsky and Soyuz (Ukraine), Borluta and Trebatch (Germany), Cyt and Afus (Poland), Novozybkovsky 653 (Russia). 
Narrow-leafed lupin (L. angustifolius L.):
Accessions: Line h56-23 and line 56-10 (Russia), Vada-14, Laf-rbs/3, Laf-rbs/5 and BSKhA-892 from Belarus. White lupin (L. albus L.): Accessions: k - 507 (Egypt), k - 682 (Yugoslavia), k - 2857 (Brasil), No. 206 (Ukraine)

Anthracnose 

Many species of lupin have become severely affected by аnthracnose (Colletotrichum gloesporioides Penz) in the past years. Our research has shown that samples of L. mutabilis and others wild species from America are most devastatingly affected by this disease. It has also been observed that cultivated varieties of white and yellow lupins are also threatened by high incidence of аnthracnose. The main source of infection in Russia and other countries are susceptible forms of L. mutabilis and other small-seeded wild species of lupin. It is recommended to grow them in isolation from the plantings of white, yellow and narrow-leafed lupins. The accessions of white lupin that we brought from the Isle of Madeira may be put to a test as a source of anthracnose resistance for breeding. Concerning yellow lupin, accessions of the Iberian geotype might be tested in the long term as resistance sources. It should be recognized that for the time being, in spite of all attempts, efficient sources of resistance to anthracnose have not been found in these two species (L. albus and L. luteus). As to L. angustifolius, cvs. Illyarrie, Chittick, etc. from Australia already carry the gene An ensuring resistance to this disease (Gladstones, 1977).

Virus diseases

Bean yellow mosaic virus (Phaseolus virus 2, Smith.) is the most harmful disease of lupins. Yellow lupin (L. luteus L.) is most susceptible to this disease among all studied species. Testing of 670 yellow lupin accessions helped to identify 50 samples with low (0-10%) susceptibility and 80 with medium (10-20 %) susceptibility.
Among them, noteworthy accessions are k.k.-2153, 2146, 1892, 1656, 1363, 2539, 2650, 2653, 2657, 2667, 2668, 2707, which under natural spreading of the virus infection during a number of years showed low percentage of infected plants. However, these samples were affected to 76-100% under the conditions of artificial infection with both strains of BYMV. More or less resistant under such conditions were k-1834 and k-1844 from the U.S. and k-2551 from Belarus.
White lupin was rather resistant to BYMV in natural environments. Affection of such accessions as k.k. - 201, 240,1785, 249, 738, 306, 490, 1494, 529, 532, 2237 changed from1,1 up to 29,6 %; k.k. - 1724, 2239, 2056, 2057, 1675, 2061, 249 and 2295, in general, showed no symptoms of this disease. However, these accessions were severely injured by artificial infection. Accessions of the small-seeded American species were much more resistant. Such species as L. elegans H.B.K., L. nanus Dougl., L. ornathus Dougl., L. succulentus Dougl. did not shown any symptoms of this disease in natural conditions during the whole period of observations. Signs of the disease were not marked for 3 years in L. succulenthus Dougl. (k - 193) and L. elegans H.B.K. (k.k. - 348, 262, 466), while k-348 and k-193 were resistant during 6 years of monitoring in the conditions of Belarus. Two accessions of L. elegans H.B.K. (k-2100 and k-348) manifested high resistance to two strains of BYMV under artificial infection as well (Курлович et al., 1989). Repeated selection among the non-affected plants of the sample E-875 and cv. Fakel (L. luteus) helped to identify weakly affected forms under artificial infection conditions, which were included in the collection of VIR (k.k. - 2933-2939). The accessions from the Ukraine (k.k. - 2651, 2736, 2847) appeared to have complex resistance to Fusarium wilt and viruses.

Resistance to Drought

Lupin is usually cultivated on the low-fertile sandy soils that lose moisture quickly. In view of this, plants frequently suffer from the lack of moisture, especially at early stages of development when the root system is yet underdeveloped. Thus, there is an urgent need to search for drought resistant material. Laboratory evaluation of accessions on this parameter was based on the principle of seed germination in saccharose solutions. As a result, significant inter- and intraspecific variation was observed in this trait. It appeared that that among the cultivated species yellow lupin accessions showed higher relative drought resistance. The accessions of white and narrow-leafed lupins were characterized, on the whole, by lower drought resistance, although some forms resistant to drought were also identified among them (Курлович and Чернышова, 1986, 1989).
Accessions of L. luteus L. with high resistance to drought: K –1350 from Bryansk Province (Russia), cv. Volynsky (Ukraine), the material that arrived from the Byelorussian Institute of Arable Farming (k.k.-2306, 2311, 2312, 2319, 2321, 2322, 2324, 2325, 2326, 2485, 2489, 2519, 2652) and from the Byelorussian Agricultural Academy (k-2211 and k-2520). Besides, high drought resistance was found in cvs. Szybkoedny wezesny and Aga, accessions k-2587, WTD 2025 (k-2582), WTD 2026 (k-2583) from Poland, and also in one sample from Portugal (k - 2271).
Accessions of L. albus L. with high resistance to drought: Cv. Start from Russia, k - 1540 from Poland, k - 2192 from Australia.
Accessions of L. angustifolius L. with high resistance to drought:LA-V (k - 2166) from Poland, cv. Bryansky 35 and samples k - 2500 and k- 2501 from Russia.

The scheme and technique of breeding process

The scheme of breeding process on lupin is similar with the generally accepted breeding practice with other crops. But it has also some differences ensuing from the features of this crop, directions and problems that should be solved in order to obtain the needed cultivars. This scheme includes the following basic links (Таранухо, 1979):

· Nursery of basic material (collection of species, cultivars, lines, wild forms and other initial materials of different genesis); 
· Nursery of hybrids and mutants (first and second generation, spontaneous hybrids and mutants); 
· Selection nurseries of the first and second year, selected families from hybrids and mutants of the second and subsequent breeds; 
· Monitoring nursery (test of the best constant numbers);
· Trial test of the best accessions and their preliminary reproduction; · Competitive test of the most valuable accessions; · State trials and production line testing; 
· Zoning, organization of primary seed production, and approval of new cultivars for agricultural production. 

While implementing the intended program, it is necessary to fulfill rather carefully all the requirements set for the techniques of field and laboratory research, phenological observations, analysis of growth dynamics, definition of a crop pattern, registration of grain and green matter productivity, analysis of biochemical structure of seed and vegetative organs of plants, definition of resistance to shattering, diseases and pests and other useful traits. Correctly matched pairs may be crossed in hothouses in the winter/spring period to accelerate the breeding process, while the obtained hybrid is to be grown in spring under customary conditions of thinned sowing with the purpose of increasing the multiplication factor.

In the subsequent breeds, it is also expedient to use different facilities of artificial protected ground which make it possible to grow and obtain 2-3 harvests a year. Nowadays, most of the breeders are equipped with climate chambers and greenhouses. Proper organization of such activities helps to reduce considerably the whole process of creating new cultivars from 10-12 down to 5-6 years.