BIOLOGY OF FLOWERING

BIOLOGY OF FLOWERING, EMBRYOLOGICAL AND CARYOLOGICAL PECULIARITIES

E.M. Kazimierska, T. Kazimierski

Biology of flowering and pollinating

Lupin is an entomophilous plant attracting insects by multi-colored flowers, nutritious pollen and fragrance. The flower of lupin is monoecious, zygomorphous and hermaphroditic. It has concrescent stamens, superior ovary and apocarpous gynaeceum (Дорофеев et al., 1990). Among lupin species of the eastern hemisphere there are strictly self-pollinated species (L. angustifolius L.) and self-pollinated species with facultative cross-pollination. Cross-pollination prevails in lupins of the western hemisphere. The main stem and the lateral branches of lupin terminate in inflorescences of the apical truss type, whereas a majority of other representatives of fam. Leguminosae bear flowers on leaf axils. The arrangement of inflorescences and their sequence of development also have peculiarities with different species. For example, in many perennial lupins peduncles are direct continuation of the stem, which means that their inflorescences are originally apical. However, there are also perennial species where trusses or even individual flowers are formed on leaf axils. With the annual forms of lupin, inflorescences are also apical, but in addition to accessory buds there are branches which, after their growth and development, would end in lateral inflorescences, often considerably exceeding the apical inflorescences. Annual forms usually have much more inflorescences than perennial ones, but they normally comply with the latter in length and number of flowers. The length of the flower truss is a quite reliable specific character. In some species the length of axial inflorescences reached 40-50 cm (Майсурян and Атабекова, 1974).

The flowers set in a normally developed inflorescence are veticillate or semi-verticillate. However, there are trusses where the strictly verticillate arrangement of flowers in the lowest part of the inflorescence goes up in a spiral. Hence, the arrangement of flowers in an inflorescence can be verticillate and, at the same time, semi-verticillate in some lupin species. The flower corolla of lupin is zygomorphic and papillionaceous. It consists of a vexillum (flag), wings, a keel and an ovary with a pistil and 10 stamens (Fig.17).

Fig. 17. Parts of the flower of lupin

1 – vexillum (flag);  2 - wings;  3 – keel (carina) consisted of two joined petals;  4 – staminate tube with the pistil and stamens.

The development of the flower goes on in cycles. In early phases the calyx forms a compressed ring consisting of five sepals. The vexillum covers the wings and keel (keel). Initially the stamens are joined together, but later on one of the stamens becomes slightly isolated from the other nine. The flowers of lupin do not contain honey, but they attract insects with their bright color, presence of pollen and secretion of a smelling liquid from the vexillum and other parts. The vexillum densely covers other parts of the corolla before the flower opens. In the middle, between two halves of the vexillum, there is a small vallecula, making it almost keeled. The vexillum has a rounded or oval contour; it is straightened in the middle during flowering, and becomes prone to strong curving at the edges. Claw of the vexillum is dense and fleshy. The cells of the epidermis in the vexillum are elongated from the outside, with straight walls. Specific features are inherent in the epidermis structure of the inner side of the vexillum. In the middle of the vexillum there are small nipple-shaped outgrowths emanating odoriferous liquid, which attracts insects. Towards the top of the vexillum, its cells are appreciably shorter and outgrowths are narrower, so there are two types of nipple-shaped outgrowths on its inner surface. Various species of lupin differ on the structure of epidermis and the size and structure of nipple-shaped outgrowths. These outgrowths are smaller and less numerous in completely non-odoriferous or insufficiently odoriferous species (Майсурян and Атабекова, 1974).

The flowers of some species have a strong insect-attracting smell. The intensity of this odor depends on the size and number of nipple-shaped outgrowths on the epidermis of the vexillum and corolla wings. There are well-advanced and closely placed nipple-shaped outgrowths on the epidermis of the vexillum and on the wings of odoriferous flowers (L. luteus), while in odorless flowers (L. albus) small obtuse nipple-shaped outgrowths are only at the basis of the vexillum on its inner side. Towards the top of the vexillum in odorless flowers, these outgrowths gradually decrease in size and finally disappear. The wings of lupin flowers are concave-convex, having narrow, normal or wide oval shape. They are usually linked at the top edge and loose on both sides. The outer part of the wings is always convex and forms an arch. In the closed hollow of the wings the keel is located. The epidermis on the outer side of the wings consists of rectangular cells with wavy-surfaced walls. Two types of nipple-shaped outgrowths can be found here: the first type is smaller, wide and blunt, while the second is larger and pointed. The inner side of the wings is covered by smooth epidermis and by blunt and pointed outgrowths. The size of cells markedly decreases towards the top part of the wings. The flower keel is falcate or falcate-curved, less often it is more or less straight, narrowed toward the top. The uppermost part of the keel accretes into a firm beak. There is a tiny aperture for pollen and the pistil’s stigma in it. The base of the keel is always open. The edges of the keel in some lupin species are ciliated. The keel is transparent, colorless or just a little hued. The keel’s edge can be colorless, slighly hued, or intensively colored, almost black with anthocyanin. The color of the keel’s edge is a simple and convenient distinctive feature of different lupin forms and varieties. It is taken into account by breeders and taxonomists. The color of the keel depends on the presence of chloroplasts, anthocyanin and other pigments. Besides, in all parts of the flowers it is possible to find rather large crystals, insoluble in hot water, alcohol or nitric acid (Майсурян and Атабекова, 1974).

There are a pistil and 10 stamens in the keel. In lupin flowers, all the stamens are joined at the bottom in the staminate tube, while their upper ends are loose. They are arranged in two circles. The anthers of five stamens of the outer circle located against the sepal are more elongated and much larger then the inner ones. The anthers of five inner stamens situated closer to the petals of the corolla are reniform; they are developed later than the outer ones. Pollen grains in the anthers of both groups have the same size, being rather large, and more or less triangular-obovate.

Fig. 18. The pollen of different lupin species

1 – L. angustifolius L.;  2 – L. cosentinii Guss.;  3  - L. mutabilis Sweet.;  4 – L. arboreus Sims.

The ovary is apical, with two or more ovules; the pistil is rounded, slightly curved, naked. The stigma consisting of friable elongated cells is capitate in the middle, and covered by numerous nipple-shaped outgrowths. It can be smooth enough (without hairs), or surrounded with a ring of firm hairs, longer at the side of the vexillum and shorter at opposite sides (in cross-pollinated species). The stigma of lupin is open, since a rather long and wide channel goes from it inside the plant. The formation of the pistil happens simultaneously with the development of the stamens and corolla. The calyx of the flower is bilabiate, the cut of the lips is deep, either reaching the base of the calyx, or less frequently ending halfway. The upper lip is usually shorter than the lower one, and has a deep incision or bilabiate shape. The lower lip is integral and pointed. Less frequently, both lips of the calyx are integral. The tube of the calyx is quite short, and more or less cylindrical. The color of the calyx in lupin is more often not the same. The base part of the lips has the same color as the floral shoot, while the top part of both lips has another color, closer to the color of the corolla. Thus, the calyx can be either all green, or similar to the corolla. The shape of the calyx is an important common descriptor and is widely used in classification of the species. The floral bract of lupin usually falls down early. Specific differences of floral bracts (their size and shape) are also very characteristic and serve as an excellent customary descriptor. In various lupin species, the shape of floral bracts can be obovate, rhombic or other. The texture of floral bracts is also variable, from paleaceous and transparent to rough-skinned and dense. Pubescence on floral bracts, as well as on the calyx, corresponds to the type of pubescence on the whole plant. The color of floral bracts can be extensively variable: cream, salad, green, green with anthocyanin and dark-anthocyanin, almost black.

Flowering. Most of Lupinus spp. are cross-pollinated. However, there is a species which is almost exclusively self-pollinated (L. angustifolius). Within the genus Lupinus L. it is possible to observe in different species all transitional pollination types – from cross-pollination to self-pollination. Reports on frequent visiting of lupin flowers by bees and on cross-pollination opportunities are found in the works of Fruwirth (1919, 1927). One of the conditions enabling cross-pollination is the property of pollen to be separated from the flowers during the contact with insects. The mechanism of this process has been known for a long time and is called the technology of the swinging pump (Szafer, 1968). The essence of this process is that anthers push out in the keel’s edge constantly falling pollen during growth. A newly opened flower is usually abundantly covered with plenty of pollen and, as a result, is readily visited by insect. As a rule, the insects sit on the vexillum and wings and by their weight stir the wings and the adjoining keel. Consequently, the anthers hidden in it become straightened and start working as a pump pushing out pollen upon the abdomen and legs of an insect. Detailed information on the pollination of lupin by insects was published by Maissurjan and Atabiekova (1974). This work presents consistent description of the pollination process in L. ornatusDougl. It goes on in the following sequence. Prior to flowering, under the dense closed petals of flower buds there are two circles of stamens: the stamens with longer anthers are colored more richly in yellow-orange color, and the stamens with reniform anthers are cream-green. One of the stamens of the lower circle is isolated from the other four. The stigma of the pistil develops a little earlier that the stamens. In the beginning of flowering process, the distance between the upper and lower circles of anthers becomes markedly reduced. The elongated anthers in the upper circle are the first to be opened lengthwise. The reniform anthers of the lower circle turn yellow; on their surface very sparse and hardly noticeable fractures appear. Afterwards, the stamens of the lower circle of anthers also extend to the top level. Then the anthers of the upper circle start to crack completely lengthwise. At this phase of development, a small quantity if pollen occurs on the keel’s edge. The extreme stamen of the lower circle, being slightly isolated from the others, develops markedly faster than the other four. This stamen is intermediate between the upper and lower stamens. Then, the lower circle outruns the upper one in its development. In this period the keel is filled with pollen. The flowering phase is terminated when all ten stamens completely wither. By that time, the ovary turns into a pod, and a light stain at the base of the flag acquires bright color. More accurate comparative analysis showed a number of essential differences in the manner of pollination between different species of lupin. For example, in L. mutabilis Sweet pollination happens in an opened flower. In L. nanus Dougl., some setback in flowering is from time to time observed, when the wings of the corolla do not fall from top to bottom, but rising above the keel straighten out to a part of the inflorescence axis. Thus, the transfer of androecium and gynaecium can go in opposite directions. Long-blossoming inflorescence at L. elegans is of an unusual kind. Drooping of the floral shoot occurs during flowering, and it carries away the wings with the keel, pistil and stamens. Fertilization comes to pass in the lowered flower; then the ovary turns into a pod and begins to grow and develop vigorously. With its new size, the pod does not find sufficient place in the keel and is gradually put forward beyond the borders of the keel. Then the floral shoot straightens and becomes a fruit spur. L. angustifolius is reported by many authors to be a self-pollinated species. This is justified, in particular, by the absence of pubescence on the stigma (Fig.19). As a result, pollen can freely penetrate the stigma of the pistil.

Fig. 19. The stigmas of pistils at different species of lupin

1 – Lupinus angustifolius L;   2 – Lupinus mutabilis Sweet.

Fruwirth (1925) marked that artificial isolation had no negative effect on the seed yield of this species. This opinion is shared by other researchers: Hallquist, 1921; Sypniewski, 1930; Шарапов, 1937; Mackiewicz, 1958. Roemer (1924) also attributed L. angustifolius to self-pollinated crops, because cross-pollination happens in this species extremely seldom, approximately in one case out of one thousand. However, that situation can change depending on the growing conditions and especially on the availability of bees. Sengbusch (1931) also expressed that classification of narrow-leafed lupin as a self-pollinated species is more justified than with yellow lupin. Flowering and fertilization in self-pollinated species, for example in L. angustifolius, occur only in closed flowers and on the earliest phases of their development. L. luteus is peculiar for the presence of not only self-pollination but also cross-pollination habit. Pollination occurs in this species also on very early phases of flower development. Thus, the arrangement of flowers in this species is verticillate, while in L. angustifolius and L. albus it is alternate. However, most part of lupins, especially from the American continent, is cross-pollinated. There are grounds to believe that the polymorphic genus of Lupinus L. includes all possible biological forms: from a self-pollinated up to mandatory cross-pollinated species. The flowers of lupin are pollinated by bees gathering pollen with their abdomen (Megalichidae). They have a special brush on the abdomen for collecting pollen. This group of insects includes many wild bees living alone (Wojciechowska, 1976, 1993). Honeybee and bumblebees also take part in the pollination process of yellow lupin. There are very interesting observations concerning the variation of color in the corolla during the flowering phase of L. pilosus Murr. and many other species of lupin (Wojciechowska, 1976, 1993). These data gain special significance, since bees and bumblebees have poorly sensitivity to the red color. Bumblebees visiting an inflorescence of lupin will penetrate only into such flowers that posses a white spot at the base of the flag. Bumblebees never pay any attention to the flowers with a red spot on the flag. As the inflorescence of lupin is a truss, flowering proceeds in the ascending order. The bottom flowers begin to blossom first, and then come the middle ones, and the last to open is the uppermost flower. According to Maissurjan and Atabiekova (1974), bumblebees come down on the lower flowers of a young blossoming inflorescence, then on the flowers in the middle of a more mature inflorescence, and lastly on the top of the inflorescence when the flowering phase comes to an end. But the data of our researches (T. Kazimierski, unpublished) evidence that bees also react to the color of flowers. In the experiments with yellow lupin having citric-yellow and orange flowers, it has been established that bees regularly come down on the flowers whose color is within certain limits (from citric to orange). If any plants with different colors were included in the experiment, the bees were attracted only by that color, from which their visiting had started, and avoided other colors. Bumblebees consecutively visited all flowers of yellow lupin. Alternate blossoming of flowers within the limits of one truss (from the bottom toward the top) as well as on the main stem and lateral branches also attests to the cross-pollination habit of lupin. The lateral branches begin to grow quickly after the appearance of pods on the main stem and spreading of the already existing pods over the main stem. Long-blossoming flowers on the lateral branches attract insects. Nipple-shaped outgrowths and hollows in the epidermis on different organs of the flower have recently been identified as another tool of adaptation to cross-pollination. The legs of insects seek them for support against sliding, and at the same time collect the pollen. Furthermore, these nipple-shaped outgrowths produce an odor attractive to insects. There are also a number of features hampering the process of self-pollination. There are such morphological obstacles as dichogamy, i.e. non-simultaneous ripening of male and female organs. Besides, the stigma of the pistil in many species is set much higher than the anthers. One more impediment is a collar of dense hairs covering the stigma. Such hairs were observed by us in many cross-pollinated species: L. mutabilis Sweet (Fig. 19 and 20), L. polluphullys Lindl, L. nannus Doudl and others, whereas in L. angustifolius L. (Fig.191) very short hairs or their absence do not create any obstacle for self-pollination. The castration of flowers will be carried out before artificial hybridization of lupin forms. The castration of flowers at lupin will carry out when the flower buds achieve half of normal size. The bottom flower buds are most suitable for castration. Usually are castrated 3-4 flower buds on one flower truss, others buds are deleted together with floral shoot.

Fig. 20. The pistil of Lupinus mutabilis Sweet.

Microsporogenesis

Differentiation of stamens occurs in anthophorous burgeons of lupin at early stages of their formation. A young anther consists from multiserial row of homogeneous meristematic cells covered by epidermis. Maheshwari (1950) and a majority of other authors have no doubts that archespores are formed as a result of the division of subepidermal cells. The archesporeal cells make up by division the primary parietal layer from the outside and primary sporogenous layer from the inside. Further, the cells of the parietal layer being divided by periclinal and anticlinal partition form a number of concentrated layers of anther walls. The primary sporogenous cells either directly or by a series of consecutive divisions become the parent cells of spores. However Maissurjan and Atabiekova (1974) consider that the cells of epidermis divide only in the anticlinal direction with further development of an anther, whereas meristematic cells divide in the radial one. Subsequently, subepidermal cells of the meristem, being also divided only by cross partitions, form the outer and inner layers of parietal tissue, whereas other meristematic cells, being divided in a usual manner, turn into a multi-layered archesporium. As a result of further differentiation of the anther, the inner cells of two-layered parietal tissue, being divided only longitudinally, form the tapetal layer, and the outer cells of this tissue, being divided transversely, are transformed into the cells of the middle and fibrillar layers. Thus, according to Maissurjan and Atabiekova (1974), the walls of the anthers of all lupin species consist of four layers: epidermal, fibrillar, middle and tapetal. The structure of anther walls is quite unusual, for it is the subepidermal cells (not the cells of the archesporium) that are the source of anther wall construction. The archesporial cells of the anther differ from the cells of parietal tissue in size, dimensions of nuclei and density of cytoplasm. They form a multicellular row of sporogenous cells that, being divided in various directions, expand and transform into the parent cells of microspores. It has been shown by comparative researches on meiosis in the anthers of lupin that the transition from dormancy to prophase proceeds in the nucleus gradually. The increase observable in the nucleus is seen more clearly during the last period of the prophase. The duration of the prophase is the longest in the process of microsporogenesis and most favorable for observation. Other phases are so short that sometimes all of them can be observed on one and the same preparation. The process of meiosis in cells at different species precedes more ore less equally if to take into account different number of chromosomes. At study of prophase of the first division was not found out of any qualitative differences in passage of this process at various species of lupin. The comparative supervision have revealed a strict sequence in passage of phases of leptonema, pachinema, diplonema and diakinesis, differing among themselves at separate species only in a quantitative sense, as at them cells are different number of chromosomes. The process of meiosis proceeds as follows. The tapetal cells in anthers concern to secretory type and they filled by rather rich cytoplasm in young anther.

Fig. 21. Microsporogenesis at  L. hartwegii Lindl. (a-f) and L. pubescens Dougl. (g-l)

a – diacinesis;  b – metaphase first;  c  and  d – anaphase first;  e – prophase second;  f – anaphase second.

g – diacinesis;  h - metaphase first;  i - anaphase first;  j – metaphase second;  k - anaphase second;  l – tetrad of microspores.

The cytoplasm in cells of tapetum gradually vacuolated, and the nucleuses begin to degenerate to time of the termination of meiosis divisions in parent cells of pollen. By then, when the microspores begin already to be separated one from another, the cells of tapetum also undergo rather significant destruction, though them wall up to some degree are still kept. The division of parent cells of microspore occurs on simul type, peculiar to all groups of dicotyledonous plants. The isolation of microspores among themselves comes at lupin immediately. Mature pollen is two-celled. The repeated division of generative cell comes already in growing pollineferous tube. Some interspecific hybrids of lupin can produce as a tetrad and also octad of microspores, which are formed from two nuclear cells of pollen (Kazimierska, 1970).

Such type of cells differs from the typical by size and form. They usually lengthened and with constriction. They have two nucleuses in interkinesis located in opposite territories of a cell. Two reduced sets of chromosomes are formed in them in time of leptonema and diakinesis. The set of chromosomes in binucleate cells is the same as well as in one nuclear. Two expressive blocks with spindles of karyokinesis are formed during the first metaphase from chromosome complexes. The first nuclear tetrad is formed after meiotic division. Four groups of chromosomes with 24 chromosomes in each are visible during the stage of the second prophase in the cell of pollen, and in the cell there are already four blocks and four spindles of karyokinesis during the second metaphase. There are already eight regularly placed groups of chromosomes in the cell during the second telophase. The normal octad of viable microspores appears after the division of such cell. It is possible to explain the genesis of binucleate cells of pollen by the presence of two different genes, one of which regulates the division of the nucleus and the other controls the cytoplasm. The activity of these genes goes on synchronously and regularly in homozygous forms. Karyokinesis is always followed by cytokinesis, therefore the cells of pollen are uninucleate. For heterozygous organisms, the activation of gene-induced karyokinesis is possible with inactivity of the gene causing cytokinesis. As a result, binucleate cells of pollen appear. When identical sporogenous cells initiate formation of uninucleate and binucleate cells of pollen under similar conditions, it is the evidence of heterozygosis of the initial pollen cells. Another hypothesis explaining this phenomenon is as follows. There is only one gene regulating division of the nucleus and cytoplasm. However, archesporious cells are divided with different speed, and a part of them remains in the binucleate state. However, uninucleate cells are typical. The tapetal cells are uninucleate and uninucleolar in the anthers of lupin, with chromocenters in the nucleus and with a large vacuole at the anther’s wall. The cells of the tapetum increase their volume and change their shape to the flattened one during microsporogenesis.

Fig. 22.  Meiosis in binuclear mother cells  of pollen

a – prophase;  b – diacinesis; c and d - metaphase first;  e – tetrad of nuclei  after first division;  f – metaphase second; g – anaphase second;  h – telophase second;  i – octad of nuclei;  j – octad of microspores.

The tapetum plays an important role in the cell and has become the object of many researches. Pakendorf (1970), comparing the development of a male gametophyte in fruit-bearing and fruitless plants of L. mutabilis, marked that that fruitlessness is connected with the absence of disintegration with the tapetal tissue. The phenomenon of disintegration with the tapetal tissue is necessary for correct development and is caused by activation of the gene responsible for programmed death of this tissue in the required period of ontogenesis. There are breaches in the development of the gametophyte when this process is discontinued. This phenomenon was confirmed by discovering the gene of “Programmed cell death – PCD” (Havel and Durzan, 1996; Jones and Dangl, 1996; Charzyńska, 1997). By its shape, the pollen of lupin resembles loaf of bread (Mackewicz, 1958). Hackbart (1939,1943) defines the shape of pollen as dumplings. It does not essentially differ in different species of lupin (Fruwirth, 1919). However, our researches have shown that pollen has some variation in size among the Mediterranean species. Rough-seeded species of Mediterranean origin have larger pollen (Tab.14). Between the American species no morphological differences have been found in pollen. Pollen of four lupin species was examined under microscope, and the results are presented on Fig.18. Physiology of pollen germination in lupin still needs thorough investigation. Very few references are available on this problem. However, it has been confirmed that the process of pollen germination to a great extent depends on genetic interrelation between male and female gametophytes, as well as on the effect of environment. Cytological researches have proven that the quality of pollen is determined by the conditions of its formation and by the environments in which it comes after the opening of the anther (Майсурян and Атабекова, 1974). Valuable findings were made in the course of generalized studies of germination of the pollen tube in intra- and interspecific hybrids of L. albus, L. angustifolius and L. mutabilis (Przyborowski et al. 1996). The pollen of lupin can be germinated not only on the stigma, but also on artificial substances and in water. It is not substrate-selective, and its division or germination may well be attained on agar at a concentration of sugar from 1% to 50%. However, there is specific variation in the rate and duration of growth of the pollen tube, depending on the concentration of sugar. The most favorable concentration for L. albus is 15 %, and for L. luteus it is 10%. (Атабекова, 1955; Атабекова and Лин Цзян-син,1962).

Significant specific variability is also observed in the group of lupins from the American continent. The least favorable concentration is reported for the annual species L. barkeri: the optimum conditions for pollen germination lay between 3% and 5% of sugar. According to our data, this species is the earliest of all known ones, so its pollination happens early, before the hottest days of the vegetation period. The optimum level of pollen germination is between 5% and 10% for perennial L. polyphyllus. The greatest activity is observed in the pollen of L. mutabilis, which is capable to germinate on sugar solutions of most diverse concentration. As shown by our research, the pollen grain of L. mutabilis normally grows at a concentration of sugar from 5% to 25%. Undemanding nature of the pollen of L. mutabilis as regards germination conditions favorably distinguishes this species from the other species of the western hemisphere. Natural and artificial crossings of this species with others lupins of the western hemisphere have been successful, obviously, owing to this biological property of pollen (Атабекова, 1955; Атабекова and Лин Цзян-син, 1962; Майсурян and Атабекова, 1974).

The pollen of lupin retains its fertilization ability for about 50-115 days (Жуковский, 1982). The observations by this author were partially proved to be true by our researches as well. L. hispanicus produced hybrid seeds when this species had been pollinated by the pollen of L. luteus preserved for 30 days at +100C. The greatest percent of germinated pollen was identified with earlier pollination of the flowers of yellow and narrow-leafed lupin (Brodowska, 1970). However, such interrelation was not found in white lupin. There were attempts to measure the effect of weather conditions on the efficiency of pollen germination. Such interrelation was not proven by most of the researches. However, there are reports concerning essential significance of air temperature and humidity for this process (Lesin, 1961; Rausch, 1964; Tomaszewski, 1966). Low temperatures (less than 8-12 degrees) and high temperatures (above 36-40 degrees) produce negative effect on the process of pollen germination (Brodowska, 1970). The stigma has the greatest susceptibility to pollen on the second day after castration. The percent of fructification is considerably reduced by the fifth day. Pollen tubes begin to germinate in 15-30 minutes and inpour through all columellae in 1-5 hours (Дорофеев et al.1990). The best way is to carry out pollination of flowers past 24-36 hours after castration; it is desirable to do in the morning. There are various ways of pollination (Майсурян, Атабекова, 1974). The pollen is taken from completely open flowers. Fertilization of embryo sacs occurs approximately in 32 hours after application of pollen on the stigma (Дорофеев et al. 1990). Pollen tubes of lupin grow straight, instead of the spiral habit of many other plants. Variability in the shape of pollen tubes was not revealed in different species (Mackewicz, 1958). This author found out that pollen germination was weaker when there was too much pollen on the stigma, than with small quantities.

Our research has confirmed this assumption. In the process of growth, pollen tubes pass from autotrophic to heterotrophic nutrition, and when they are in plenty they lack hormones and nutritive elements. The length of pollen tubes in vitro varied within a wide range. The results of the researches of different authors on this problem are essentially at variance. It may be exemplified by the materials presented in the following table (Tab.15) and published by Mackewicz (1958). The difference in results can be explained by distinction in genotypes of investigated plants, terms and years of researches. Sometimes happens that from one pollen grain grow two pollen pipes or one ramified pipe. However, such anomalies are rather rare. The length of pollen pipes depends also from what inflorescence (main or lateral) and from what place of inflorescence (bottom part, middle or top) the pollen are taken from (Mackewicz, 1958). The most valuable pollen usually is from the bottom part of main inflorescence though is present specificity on species. The best conditions of its development, presence of nutritious substances and hormones explain this phenomenon. But the data showing of the best viability of pollen with lateral inflorescence are published also (Caliński, 1953).

Megasporogenesis

The first researches on megasporogenesis and, in particular, on the structure of the embryo sac and embryo of lupin date back to the 18th century (Hofmeister, 1858). He managed to provide a correct description of the structure of the embryo sac, specify its components, and trace the development of the embryo, despite the imperfection of the used methods and techniques. It is necessary to take into account that this research on Lupinus hirsutus L. and Lupinus cruckshanksii Gray. is not reliable enough. According to Hofmeister (1858), before fertilization a rather massive integument surrounds in a spindle-like manner the slightly bent nucellus whose middle part is completely superseded by the embryo sac located a little bit sideways under the micropyle. The pollen tube will penetrate in the embryo sac or between the nucellus and integument, thus reaching the lateral surface of the embryo sac, or pierce directly the tissue of the nucellus. In the author’s opinion, the pollen tube forms numerous short outgrowths and all possible curvatures during germination. The first divisions of the fertilized ovicell occur in lupin by formation of a cross partition. Later there will be subsequent and repeated divisions in the final cells of the embryo. In Hofmeister’s opinion, this area of research is rather complicated due to the properties of a young embryo whose cells are very sensitive because of the absence of a solid cell boundary. He also described the development of nutritive tissue, nowadays called endosperm, which started its formation simultaneously with the first division of the fertilized ovum. There are large nuclei with large nucleoli on all inner surface the embryo sac quickly growing in length and in width and in the thick layer of protoplasm connected to the walls of the embryo sac. Hofmeister observed a set of chlorophyll grains in the bottom part behind the sac in the cells of epidermis, as well as in the cells of the embryo. According to his observations, the embryo sac expands mainly in the top part, and forms short and long protrusions during the described process. The works on lupin embryology by other authors (Hegelmaier, 1880 and Strasburger, 1880) were published after Hofmeister’s research. Hegelmaier (1880) studied L. varius Gaertn, L. mutabilis Sweet, L. polyphyllus Lindl. and L. luteus L. with one or two integuments. Double cover was found only in L. luteus L. Hegelmaier ardently debated the conclusions of Hofmeister (1858) and stated that in addition to a solid external cell layer in lupin there was also an inner integument consisting of two-celled layers. On the basis of this character, Hegelmaier subdivided lupins into two different embryo groups. Further, he marked that had not observed the suspensor described by Hofmeister. Embryological research of Strasburger (1880) also confirmed the presence of two integuments in L. luteus postulated in the work of Hegelmaier. As to the problem of the suspensor, he shared the viewpoint of Hofmeister. The suspensor was not noticed in many lupin species because, in his opinion. it was placed very deep in the embryo and broke up early. Thus, first Hofmeister (1858), then Hegelmaier (1880) and Strasburger (1880) published more or less detailed observations on the embryological development of some lupin species.

These researches were followed by the publication of Cuignard (Cuignard, 1881) who supplemented and revised the data of the previous authors. Cuignard paid a lot of attention especially to the structure of the embryo in lupin. According to Cuignard (1881), the ovicell of L. polyphyllus, surrounded by the thinnest wall, becomes elongated after fertilization. In the lower part of this ovicell there is a nucleus. It is divided so quickly that is difficult to catch the moment of formation of an octonucleate embryo. Six of these nuclei set in pairs one above the other make the suspensor, while two other smaller nuclei participate in the formation of an embryo. Then, the outside cell of the embryo being cross-divided forms a body of three cell layers. Less frequently, each of the two cells of the embryo, being divided is a longitudinal manner, form the tetrad. By that time, according to Cuignard’s observations, the cells of the suspensor start to slide one upon the other. As a result, their pairing is broken. The suspensor’s cells before the appearance of an embryo are represented by large round nuclei surrounded by a rich layer of protoplasm. Owing to the expansion of the embryo cavity, they are separated from each other and are set in a row. Despite their separate position and the presence of surrounding endosperm nuclei, they may easily be distinguished by their large size, while the nuclei of endosperm retain the same size. The position of the embryo in the embryo cavity varies in the process of its growth. The embryo, initially found in the mycropyle area of the embryo sac, soon appears in the bent part of the cavity under the isolated cells of the suspension bearing. Cuignard’s observations were completely proven by a series of subsequent researches. Studying the embryology of lupins is complicated by the fast rate of division at the initial stages of the fecundated ovicell in the embryo sac, as Soueges (1948a) noticed.

Nevertheless, he marked that the earlier researches by Cuignard had produced correct data describing the structure of the embryo sac, the origin of nutritive tissue and the suspensor’s bearing, as well as the role of the elements of its composition. His other work (Soueges, 1948b) presented the evidence of consecutive segmentation of the embryonic sphere in the favorite species of many authors – Lupinus pollyphyllus Lindl. The researches of Romanienko (Романенко, 1939), Markova (Маркова, 1944) and Palamarchuk (Паламарчук, 1959) were dedicated to the same problem. In their works one can find a detailed development of the embryo sac, endosperm and suspensor in the process of fructification. Jaranowski (1962a and 1962b) described the fertilization and formation of the embryo in the Mediterranean species — L. albus, L. angustifolius and L. luteus. The author made up interesting tables presenting the rates of fertilization and development of the embryo and seed. According to him, the inner integument consists of two layers of cells only, whereas the outer integument consists of several layers. The outer integument, developing faster than the inner one, soon surround the nucellus from above forming a channel, which is called the exostome. The inner integument also surrounds the nucellus, and makes only an insignificant part of the micropyle channel — the endostome. In the opinion of this researcher, these formations set in one line, as a rule, produce one straight and long common channel (for L. albus — 250 microns; for L. luteus — 230 microns, and for L. angustifolius — 150 microns). Rapid growth of the outer integument in the basis of the ovule promotes of its declination. Further Jaranowski marked that thr nucellus in lupin is powerful. There is a group of cells forming a cap close to the apical end of the nucellus, as a result of periclinal divisions. It is visible more clearly in L. luteus and L. albus. The referred author considered this tissue to be epistatic.

The nucellar cells go on forming hypostatic tissue at the opposite sides of the ovule, which in the process of development of the embryo sac is gradually destroyed. During fertilization the embryo is separated from the inner integument by nucellar tissue. The cells of the inner integument are differentiated neither in the integumental tapetum, nor in the endothelium, neither during this process nor afterwards. Maissurjan and Atabiekova (1974) challenged the task of detailed embryological study of various lupin species belonging to two remote geographical areas. Amongst the representatives of the Mediterranean group (eastern hemisphere) they investigated L. angustifolius L., L. linifolius Roth., L. pilosus Murr., L. albus L. and L. luteus L., and from the American group (western hemisphere), L. subcarnosus Hook., L. nanus DougL, L. barkeri Lindl., L. mutabilis Sweet, L. hartwegii Lindl., L. elegans H. B. K. and L. ornatus Dougl.

The results of these studies were described in detail in a separate section of the monograph “Lupin” (Майсурян and Атабекова, 1974). The ovules of all known lupin species representing the Mediterranean group and the unique American species L. subcarnosus were reported to have two integuments. The ovules were monochlamydeous in all other American species. Their observations showed that the outer and inner integuments were formed almost simultaneously in dichlamydeous lupins. The sporogenic cell is easy distinguished from other cells of the nucellus by the size of the nucleus and the density of the cytoplasm. While the primary cells of the inner integument pass the stages of prophase and metaphase, the primary cells of the outer integument are in the state of prophase. Further, the outer integument surpasses the inner one in the speed of cell formation. Finally, being made only from two layers of cells, it does not reach the top of the nucellus. Therefore, the micropyle is formed apparently only from one multilayered outer integument. The border between the inner and outer integuments is almost indistinct. The integuments are sound and multilayered at all known American monochlamydeous species of lupin. The ovule of lupin, looking in the beginning like a straight cone, transforms into a bent one in the process of ontogenesis. One apical subepidermal cell increases in size and becomes direct by an archesporial cell in the young ovule. Then, this cell being divided by cross-partition forms the primary sporogenous and parietal cells. Finally, being divided again crosswise and lengthwise, it forms three layers of parietal tissue. The primary sporogenous cell being increased in size without subsequent division directly turns into the parent cell of megaspore. It gradually turns from round to more stretched in the course of the leptonema phase. The mycropile cell can divide during the secondary meiosis at right angle to the axis of the first division; therefore, the T-shaped arrangement of a tetrad of megaspores is being formed.

The processes going on during the formation of archesporial cells in L.polluphullys are best of all described, in our opinion, in Guignard’s publication (1981). According to him the large axial subepidermal cell, being divided in the middle, forms apical and subapical cells in the ovules of L. polluphullys. Then, the apical cell is several times divided in the horizontal direction. Besides, new cells are being formed at the expense of crosswise division. In the lateral layers of the nucellus, the cells at that time start to be divided in the tangential direction. According to Guignard’s data, the subapical cell, being extended, remains in most cases undivided. Crosswise division sometimes can be observed in it, but it would soon disappear after the parent cell is formed. The described process, according to Guignard (1881, 1882), proceeds in L. luteus in a little different way. The apical cell forms only two cells located above it. Vertical subdivision happens seldom, and the subapical cell is divided always crosswise. Thus, what makes an infrequent exception in L. polyphyllus is constantly observed with L. luteus. The subapical cell is divided repeatedly in other species, for example in L. pilosus. The process of megasporogenesis can be presented at L. pubescenc Dougl. as follows on our opinion. The archesporial cell is formed from one of apical cells, which begins to be increased in the sizes and become oblong (Fig. 23). The parent cell of pollen divides by the cross wall during the first meiosis (Fig. 23b). The second section is carried out by division of cells from the side of micropyle under a direct corner concerning the first division; therefore, the T -figurative tetrad is formed. However, the megaspores have more often described by Guignard (1881, 1882) linear arrangement (Fig. 23c). The integument surrounds all nucellus in the phase of megaspore tetrad. The parent cell of embryonal bag is formed usually from the first megaspore from the side of chalaza. Guignard describes as exception exclusive cases of formation of the parent cell at L. nanus not from last halaziferous cell but from the penultimate cell of megaspore tetrad. The chalaziferous microdispute, as is specified, will be transformed to one-nuclear parent cell of embryonal bag. Three spores located above surround it (Fig. 23с). One-nuclear embryonal bag will be transformed in eight nuclear because of subsequent meiosis (Fig. 23d and e). Mature embryonal bag (Fig. 23g) is bent in an midle part and extended in micropilar and chalariferous point. It fills all space of nucellus. The antipodes disappear earlier, before unification of nucleuses and formation of the secondary nucleus of embryonal bag. The egg cell and two synergides are in micropilar point of mature embryonal bag. The secondary nucleus of embryonal bag settles down at the bottom of egg cell.

Fig. 23. Macrosporogenesis at L. pubescens Dougl.

a – archesporium;  b – diad of macrospores;  c –tetrad of macrospores with three macrospores growing from micropylar side and turning to parent ceel  of embrional bag and one chalaziferous macrospre; d – two nuclear embrional bag;   e – four nuclear embrional bag;  f –  eight nuclearembrional bag;  g – mature embrional bag.      

The embryonal bag is formed at lupin for a type of Polygonum. However some researchers (Guignard, 1881; Романенко, 1937; Маркова, 1944) working with L. polyphyllus, have described of Allium-type of development of embryonal bag. The attempt of explanation of this problem can be found in the work of Palamarchuk (Паламарчук, 1959). Author shown, that developing chalazal megaspore at L. polyphyllus, passing stages of two and four nuclear embryonal bag, forms eight-nuclear bag of the normal Polygonum-type. Our investigations in area of megasporogenesis and macrosporogenesis of interspecific hybrids of lupin decline us however up to deep proviso of the contradictions in the work of the above-stated researcher We also have found out cases of development of embryonal bag for a type of Allium in the second-generation (F2) of hybrid L. hartwegii x L. pubescens. Some researchers treat such phenomenon as marginal, as deviations in ontogenesis. Others, for example Bergman (1957), describe the embryonal bags both mono and tetra sporiforous at the same plant of Leontodon hispidus var.hyosercides, and embryonal bags at Antenaria carpatica are formed as a minimum of three different types. Romanów (1957, 1960) mentions, that in the plants of species Tulipa analyzed by him, which have tetra sporiforous embryonal bags, he also found individual infringements and deviations in ontogenesis. He approves that such cases are essentially important for understanding of mutual connections and types of development. It is necessary to add, that at species breed by apomixis or vegetative, the type of development of embional bag has a supporting role, as at such forms usually are not fastened seeds. Urbańska (1956) has found out at Homogzne alpina, that embryonal bag can be developed from each of maccrospores. Kordium (1965) has observed within the limits of one plant of Bupleurum rotundifoliun development of embryonal bags for Polygonum, Allium and Adoxa types.

Gerasimova-Nawaszina (1954) opens development and evolution of embryonal bag from the point of view of the most general laws of life cycle of cells. She believes that the angiospermous plants have polyphyletic origin, and embryonal bag such as Polygonum as well as its other types has polyphyletic origin. In opinion of the author, the essential factors of evolution are the conditions of external environment. Partially agreeing with such statement, we however believe, that necessary to take into account also genetic factors of evolution, such as mutation and hybridization. Each of these factors influences on genetic structure of a plant. There cannot be exceptions from this rule, and an embryonal bag cannot be as independent structure in a plant, with other inheritance of characters than at other organs. The hybridization is one of the factors causing variability, because of which in progeny at hybrids a new characters (transgressions) can appear. Hence, explaining a type of development of embryonal bag as the genetic characters can be admitted, that the new types can be formed as the result of recombination of genes of the parental forms. The occurrence in hybrids of the forms with different types of embryonal bags can only testify to the large meaning of the hybridization factor in evolution of the plants, and cannot be treated as a deviation in ontogenesis.

The passage of megasporogenesis and macrospotogenesis resulting in formation of the certain type of embryonal bag is as well as other characters by result of long process of evolution. Accepted by the part of embryologists the point of view about that that the type of development of embryonal bag is the conservative character and not subject of changes, is not correct. In our opinion, this position is in contradiction with the facts of occurrence of different types of development of the embryonal bags in limits of concrete variety, plant or even separate flower. Because of study of interspecific hybrids is revealed by us before not known bisporiferous type of embryonal bag formed from two chalazal microspores and above which were two micropyleous microspores. Such bisporiferous embryonal bags were not described up to us, and we have offered for them the name pseudo-bisporiferous (Kazimierska, 1970). Described embryonal bags of pseudo- bisporiferous type are found out by us also at hybrids L. varius L. x L. pilosus Murr. (Kazimierska, 1970). Such sort of supervision declines us to opinion of Modiiewskij (1953a and 1953b), that the development of embryonal bag is characterized by high plasticity at the angiospermous plants, and in particular at lupin. We consider widely popular opinion that embryonal bag such as Polygonum is initial, and it is possible to receive all other known types on its basis. The materials of literature and the own researches show that bi-sporiferous and tetra-sporiferous bags can be formed at the forms, which usually have mono-sporiforous embryonal bags. Their formation is carried out by means of disappearance of cell walls at dividing microspores, or walls does not occur at all. In last case is formed two or four-nuclear oocytes, transformed in subsequent in two or four-nuclear embryonal bag.

Fertilization. Many embryologists encountered great difficulties while defining the number of nuclei and cells in embryo sacs of lupins, for it is not at all easy to count them up. Obviously, the difficulty in calculating the compound elements of the embryo sac is connected with the unstable existence of antipodes which are not easily perceptible in lupin. Among all analyzed preparations it was possible to observe antipodes only in the embryo sac of L. polyphyllus (Атабекова and Лин Цзян-син, 1962; Майсурян and Атабекова, 1974). Thus, a mature embryo sac does not contain an antipode up to the time of fertilization, which is characteristic of all known species of both Mediterranean and American origin. The ovicell always differs from two synergids adjoining with it. Pollen tubes in lupin penetrate into the ovule during fertilization only through a micropyle. It is always possible to find a significant quantity of germinating pollen on the stigmata of lupin plants and other representatives of fam. Leguminosae.

Mature pollen always consists of two cells. The second division of the generative cell goes on concurrently with the germination of the pollen tube. The shape of the ovicell is not always stable before fertilization. This is connected probably with the thinness of its walls. The spermia of lupin have helminthoid shape. The cavity of the embryo sac considerably extends lengthwise, and acquires the curved shape typical for all known lupin species during fertilization. By that time the synergides are usually degenerated, and only the ovicell and central nucleus retain their integrity in the embryo sac. Pollens germinate and penetrate into the columella of the pistil between its tissue and the nipple of the stigma. Then, the pollen tube takes root in the cavity of the embryo sac, passes either directly through the ovicell or between the synergides and ovicell, sometimes destroying one of the synergides. The process of fertilization takes place 50 hours after the pollen has been cleared from the anthers under natural conditions, or 30 hours after the artificial pollination. The fusion of the male cell with the central nucleus during fertilization occurs earlier than the fusion of the second male cell with the ovicell. Moreover, sometimes it was possible to observe in L. albus two free nuclei of endosperm in the cavity of the embryo sac. It usually happens at the moment of the fusion of spermium with the ovicell (Майсурян and Атабекова, 1974).

The wall of the zygote is consolidated after fertilization, and the zygote acquires oblong or pear-shaped form.

Endosperm. Division of the nucleus of endosperm always precedes the division of the zygote (Jaranowski, 1962a, 1962b). When the zygote is in the phase of dormancy or metaphase it is already possible to find two (or less frequently four) nuclei of endosperm in the embryo sac. Division of nuclei occurs synchronously with the beginning of endosperm formation; afterwards it may happen on different stages of mitosis. Some differences were observed in the character of endosperm distribution in the cavity of the embryo sac in various species of lupin (Hegelmaier, 1880). Variation of the size of endosperm nuclei was traced while studying the endosperm of L. subcarnosus, including the nuclei of the suspensor’s cells containing in the mycropile part of the embryo sac. These cells are usually much smaller than the nuclei of halazal part (Майсурян and Атабекова, 1974). Therefore, the circumference of a nucleus of endosperm in the mycropile part of the embryo sac is only 7.7 microns in a 50-celled embryo, whereas in halazal parts it makes 9.7 microns. These dimensions equal to 10.64 and 20.20 microns respectively in a 104-celled embryo. The nuclei of the middle part of the embryo sac have an intermediate size. The endosperm surrounds the embryo and fills the mycropile part of the embryo sac bag after the beginning of the spherical phase in the body of the embryo. Besides the cellular endosperm, the embryo sac contains free nuclei, usually being much larger than the nuclei of the cellular endosperm. The cellular endosperm is formed much earlier in the Mediterranean species, than in the American ones (Майсурян and Атабекова, 1974). Endosperm gradually disintegrates in the process of development of an embryo. By the time of seed ripening, only one- or two-layered endosperm is retained, which surrounds the generated embryo.

Suspensor. The first division of the zygote, as a result of which a two-celled embryo is formed, happens in a crosswise or slightly angled direction, thus forming a proembryo of two cells: the top cell, called apical, and the lower basic cell. The apical cell is characterized by rich cytoplasm and significant quantity of plastids and mitochondria. It gives a start to the embryo. The suspensor is developed from the lower cell. It serves to keep the embryo fixed in a certain location to the wall of the embryo sac. It also performs the functions of a mediator in regulating the flow of nutrients. Besides, it is capable of intensifying metabolism at the early stages of the embryo’s development. Documentary data on the structure of different types of the suspensor in leguminous crops were presented by Guignard (1982). According to him, it is, as a rule, absent in the plants of the subfamily Mimosaceae. This organ is rudimental in the genera Soja and Trifolium, and consists of 3-4 cells. In Pisum, the suspensor consists of two pairs of cells located crosswise, and these cells have many nuclei. The suspensor is well discerned in the genus Lupinus. Part of the suspensor’s cells in the mycropile part of the embryo sac are separated and have unfixed location within this genus. Jaranowski (1962a and 1962b) described the development of the suspensor in three species: L. albus, L. angustifolius and L. luteus. He reported that the zygote formed 4-celled body by crosswise divisions. The embryo originated from the apical cell, and the suspensor was formed as a result of divisions from the basic and two middle cells. The species described by this author are characterized by a long and massive suspensor. It plays the role of an absorbent of nutrients and supports the embryo in certain situations. It is also possible to regard the suspensor of lupin as a rudimentary body (Майсурян and Атабекова, 1974), since it is reduced during the growth and development of an embryo. The number of cells in the suspensor is determined in the 4-8-celled phase of the embryo in all the Mediterranean and one American (L. subcarnosus) species, while in all other American species it happens in the 4-celled phase of the embryo. Characteristic of the genus Lupinus is a two-celled suspensor. The number of cell pairs in the suspensor is constant and can be used for the purposes of systematics: there are about 37 pairs in L. pilosus, 33 pairs in L. albus, about 15 pairs in L. luteus, 12 pairs in L. angustifolius, about 50 pairs in L. subcarnosus, 6 pairs in L. mutabilis, 5 pairs in L. albococcineus, 4 pairs in L. hartwegii and 3 pairs in L. elegans. The suspensor’s cells are separated and collapse after the coming of the 4-celled phase of the embryo body in quite a few species of the American group; in L. hartwegii and L. elegans it occurs before the beginning of the spherical phase of the embryo body, while in the species of the Mediterranean group and at L. subcarnosus it happens after the coming of this phase. The evolution of the genus Lupinus proceeded from longer suspensors, with late separation of cells, to shorter ones, with early separated cells. The Mediterranean species and L. subcarnosus are considered the most primitive as regards the structure of the suspensor and the rate of dissociation and destruction of its cells. L. hartwegii and L. elegans are also primitive with respect to these traits, as compared with other American species of lupin.

The suspensor’s cells are quite abundant of chlorophyll in all known Mediterranean species (plus one American species, L. subcarnosus). It is gradually destroyed in the process of seed maturing. Chlorophyll is absent in the representatives of the American group of lupins (Майсурян and Атабекова, 1974).

The Mediterranean group is more primitive than the American one (except for L. subcarnosus) as far as the presence and distribution of chlorophyll in the suspensor’s cells are concerned.

Development of the embryo. The zygote is increased in size soon after the fusion of gametes, the wall of the cell becomes clearer, and the large vacuole stands disconnected in its bottom part. It is usually possible to find two nuclei of the endosperm in the phase of dormancy or metaphase in this period in the embryo sac. The first division of the zygote is made by a crosswise or slightly slanted partition, and leads to forming a pro-embryo from two cells. Then, the cells are divided lengthwise forming a 4-celled pro-embryo, or so-called first tetrad (Soueges, 1948). Two top cells of the four-celled pro-embryo are divided crosswise a little bit later, originating two apical cells and two middle cells. Sometimes, not all cells of the first tetrad, dividing simultaneously, produce a six- or eight-celled pro-embryo. The period of formation of the eight-celled pro-embryo is an important stage in the development of an embryo, corresponding to the period of its substantial growth with a long delay of segmentation (Soueges, 1948a). Two finer apical cells give a start to the embryo, and the suspensor is formed from three pairs of larger cells. The nuclei of apical cells are increased and there comes their first division at the time of formation of the suspensor. In some days it already becomes possible to observe a small four- or eight-celled embryo and a huge braided suspensor.

The embryo is frequently equipped with haustoria, which in due tine will collapse. Afterwards, the eight cells of the proembryo are cross-divided, thus forming a 16-celled embryo. Each tissue in it consists of four cells (quadrants). The shape of the embryo changes to this end, becoming more or less obovate. The top quadrant represents the epiphyses, from which bud leaves and the growing point of cotyledons will be formed. One of the middle quadrants produces cotyledons, while the other forms a root and a hypocotyl. The bottom quadrant represents the hypophysis. The embryo root and haustorium of the embryo are developed from it. The cells of hypophysis being repeatedly divided form buldge in the bottom part of the embryo, from which the embryo’s haustorium will later grow (Soueges, 1948a and 1948b; Атабекова, 1965; Майсурян and Атабекова, 1974).

Thus, the embryo of lupin develops according to the pattern of Lupineae, characteristic of this tribe. A distinctive character for this pattern is a certain type of the suspensor, consisting from a certain number of cell pairs. The arrangement of ovicells in embryo sacs and the arrangement of embryos correspond to the ovule structure types of two geographical groups of lupin. This fact can be useful for taxonomic classification of the genus Lupinus L.

In the species of Mediterranean origin and in L. subcarnosus, the embryo is located at the mycropile side of the embryo cavity. In other species of American origin, the embryo is situated near the chalazal side. It moves there with the help of plasmatic rods. The first type of the arrangement of the embryo is considered as more primitive (Майсурян and Атабекова, 1974).

The embryo develops in lupin from a baffle apical cell formed by the preceding division of the zygote. Such development of an embryo is typical for all family of Fabaceae Lindl. The division of the zygote, the shape of which resembles a platen or pear and has oblong symmetry, goes on transversely at first. Every subsequent division is right-angled to the previous one that is a longitudinal division is followed by a transverse one. As the cells are increasing in number, they begin to divide in different directions, and the speed of their division is always directly proportional to the density of cytoplasm. Cells with rich cytoplasm are divided faster, even when they are smaller. Division usually goes faster in the zygote’s apical part with dense cytoplasm. Besides, more cells are formed in this part than in the chalazal part. Thus, cells are smaller-sized in the chalazal part (Soueges, 1948a and 1948b). This author suggested that the first group of cells should be called micromer, and the other centromer. All such divisions result at first in the development of a non-differentiated and many-celled body called by some authors a pro-embryo (Soueges, 1948a and 1948b) and by others an embryo (Poddubnaja-Arnoldi, 1976). Many of the authors regarded 2- and 4-celled formations as pro-embryos, which in the subsequent phases of development would be transformed into embryos. Jakowlew (1973) offered the following criterion for differentiation between the phases of an embryo and pro-embryo. The phase of a pro-embryo proceeds from the first division of the zygote up to the moment when embryoderma occurs in the embryo, and the embryo phase lasts from the isolation of embryoderma to the formation of a fully completed embryo. However, this criterion is conventional enough, as it is very difficult to catch the beginning and the end of the specified moments. Jaranowski (1962a and 1962b), who analyzed the development of embryos in L. albus, L. angustifolius and L. luteus believed that their development is not typical and is difficult for classification. The ovules of lupin correspond to the Cariophyllad Type, Medicago Variation according to the system of Johansen (1945).

Microsporogenesis and megasporogenesis at plants of yellow lupin infected by virus diseases

Lupin is the important source of protein at geographical latitude of Poland, Belarus and Russia. However, its not sufficient stability to virus diseases constrains its wide distribution. It is expedient to present materials on character of a defeat and deformation of cells, tissues and flowers of lupin by this pathogen from the cytological and embryological position in this connection. Bean yellow mosaic virus essentially deforms the structure of all parts of flower at lupin (Fig. 24). Such deformations are very similar to action of the mutagenic factors. The majority of the infected flowers prematurely withers and falls down. Frequently frutificated young beans fall down also. It is possible only sporadically to receive seeds from the infected plants (Kazimierski, 1961c).

Fig. 24.  Flowers of  L. luteus L.  from healthy plant and  from plant infected  by virus.

Flowers:  a – from normal (healthy) plant;  b – from infected plant (deformed);

        Vexilums:  a1 – from normal plant (from the left) and a number from infected plant;

  Keels: a2  – from normal plant (from the left) and a number from infected plant;

 Wings: a3 – from normal plant (from the left) and a number from infected plant.

There are infringements in conjugation of chromosomes and in their division at infected plants. Therefore are formed damaged chromosomes and chromatides, the pollen have every possible deviations in development. In author’s researches is established that the pollen from the infected plants badly sprouts or does not sprout on artificial environments. The short pollen pipes are formed at sprouted pollen. (Kazimierska and Kazimierski, 1965).

Fig. 26.  Meiosis and microsporogenesis at yellow lupin plant infected by virus

a – M I with three polar cariocinetical spindle;  b – prophase with 25 chromosomes in one and 27 chromosomes in the second nucleus;  c – M II with chromosomes outside the plate;  d – T II with three chromosomes in the cytoplasm;  e – A II in one nucleus, in the second chromosomes remained on the phase;  f – tetrad of microspores;  g – diad of microspores  with two nuclei in each; h – diad of microspores, one three –nucleate; i – triads  of microspores, on bi-nucliate.

The essential infringements are marked also in process of megasporogenesis. The embryonal bag develops not for the type of Polygonum at the infected plants. The nucleuses are degenerated or are situated in two nuclear embryonal bag. A part of nucleuses degenerate completely at the further development, the infringements in development of synergids also take place. The division of nucleuses at antipodes can take place in chalazal part, which are usual die and disappear at healthy plants. The embryonal bags are exposed to complete degradation very often; the nucleuses disappear and remain only tapetum (Kazimierski and Kazimierska, 1965).

Karyological characteristic of species

The genus Lupinus L. is investigated karyologically poorly. The researches are made only about number of chromosomes at separate species. It is connected with difficulties of cytological researches of lupin having a big quantity of small chromosomes (Figs. 27 and 28), which studying and calculating is complicated by possibility of their segmentation (Maissurjan and Atabiekova, 1974). The species of lupin concerning to different subgenera form two completely original karyological groups with different quantity of chromosomes (Tab. 16) and do not reshape of interspecific hybrids. The diploid number of chromosomes varies from 32 up to 52 at species of Eastern hemisphere (subgen. Lupinus). It is impossible to establish base number of chromosomes for these species. On the literary data (Hackbarth, 1957a; Carstairs et all, 1992), aneuploidal variability is characteristic for them. It was possible to cross among themselves yellow lupin (L.luteus) with both subspecies of Spanish lupin (ssp.hispanicus and ssp.bicolor) to the present time (Lamberts, 1955, 1958; Kazimierski and Kazimierska, 1970; Maissurjan and Atabiekova, 1974; Swęcicki, 1988). More viable and valuable in the breeding relation progeny turns out from 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). The results of interspecific crossings among the species of section Scabrispermae (L.cosentinii, L.digitatus, L.atlanticus, L.pilosus, L.princei) are described also (Roy and Gladstones, 1985, 1988; Carstairs et all, 1992). It was possible to receive sterile hybrids F1 in a number of crossings, and sometimes were fastened impractical seed at hybrids F1. More successful result in this area is described by Buirchell (1994) and Atkins et al. (1998).

Fig.27. Chromosomes from flower buds of lupins (haploid number):

1, 2 – L. angustifolius; 3, 4 – L. pilosus; 5 – L. albus;  6 – L. luteus;  7 – L. subcarnosus; 8 – L. nanus;

 9 – L. barkeri.

Among lupins of Western hemisphere the species with 2n=48 prevail. The diploid set of chromosomes makes 36 at L.subcarnosus. 96-chromosomal species are described also (Maissurjan and Atabekova, 1974). The basic number is 6 or 12 for species of this hemisphere. Many species with 2n=48 are easily crossed among themselves and give fertile or enough fertile progeny.

Fig. 28. Chromosomes from tips of radixes at lupins (diploid number):

1, 2 – L. angustifolius; 3 – L. albus;  4 – L. luteus; 5 – L. subcarnosus; 6 –L. barkeri; 7 – L. mutabilis;

 8 – L. hartwegii; 9 – L. elegans; 10 – L. ornatus; 11 – L. affinis.

Besides, many from the recognized nowadays species of Western hemisphere are the latent hybrids (Maissurjan and Atabekova, 1974). In the literature are most full described the following hybrids: 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).