Cytogenetic Collection of Uzbekistan

alternate and adjacent orientations. The translocations were characterized by different frequencies of the multivalents at the MI of meiosis. Translocations with the large translocated chromosome segments greatest for tagging cotton chromosomes and developing new homozygous translocation lines. Electron microscopy of synaptonemal complexes at the pachytene stage revealed more cells with heterozygous chromosomal rearrangements than light microscopy at meiotic MI by a factor of 1.8 Those results indicate that the different frequencies of multivalents in heterozygotes for translocations at MI result from partial desynapsis and segregation of translocation multivalents into two “heteromorph ” bivalents, which cannot be distinguished from normal bivalents at meiotic MI by light microscopy. The analyses of tetrads were carried out in 201 translocations. Most of the translocations exhibited a high meiotic index when compared with the control However, 20.51% were characterized by a reduction in meiotic index and an increase in the percentage tetrads with micronuclei when compared with the control Six translocations were characterized with a low meiotic index 52.15±1.99 79.37±1.32%) and number of abnormal tetrads containing micronuclei 16.53±1.48 detected. The results suggested an irregular univalent chromosome centromere misdivision in the parental monosomics that led to a single chromosome arm missing and formed either telocentric or isochromosome in the case of an arm doubling. Our results demonstrated the rather rare occurrence of telo-and isochromosomes in the monosomic progenies studied, which showed univalent misdivision to be rare.


Introduction
The genetic improvement of cotton demands the use of cytogenetic stocks for molecular mapping of QTLs and introgression of beneficaila genes from wild and unadapted germplasms in Upland cotton. Development of the collection of cotton cytogenetic lines with translocations and chromosome deficiencies are necessary to fulfill these goals. During long-term studies the series of translocations and monosome stocks were developed in the USA, which provide chromosome identification and localization of marker genes on chromosomes. As the result, totally, 63 translocations were accumulated over a period of more than 20 years in USA [1][2]. The 62 heterozygous translocations were transferred to homozygous state and identified. Twenty reciprocal translocations of the cytogenetic tester set of G. hirsutum were selected and cytologically characterized [3,4]. This tester set of the translocations marks 25 of the 26 chromosomes of G. hirsutum, with chromosome 26 being identified by elimination.
Cultivated allotetraploid cotton, G. hirsutum (2n=52), is tolerant to the loss of individual chromosomes or their arms. During long-term investigations a big number of monosomic plants of different origin were isolated in the USA [5][6][7][8][9]. The majority of monosomes were found for chromosomes 2, 4 and 6 of the A-genome [10]. Unfortunately, the complete series of 26 monosomic lines in cotton have not been recovered yet. The monosomes for 15 of the 26 nonhomological chromosomes of G. hirsutum were identified [2]. Therefore the development of one or more deficiencies which would involve a part and/or all of these chromosomes has been high priority [11]. Use of the new molecular cytogenetic methods -meiotic fluorescence in situ hybridization (FISH) was identifed a new cotton monosome of chromosome 23 [12]. Recently, another new monosome for chromosome 21 in cotton was reported [13]. During the last years the monosomic stocks were used for chromosome assignment genetic and molecular markers to specific chromosomes [14][15][16][17][18][19][20][21].
Use of F 1 hypoaneuploid hybrids resulting from the crosses of G. hirsutum aneuploids (2n-1 or 2n-1/2) and G. barbadense L. species (2n) in molecular-genetic analyses has facilitated the localization of different molecular markers on specific cotton chromosomes [22][23][24][25]. However, some loci were not assigned using the aneuploids due to the lack of a full set of cotton aneuploids [e.g. 21,[25][26][27]. In the last decade chromosome-deficient stocks of G. hirsutum have been used for the development of chromosome substitution lines for G. barbadense, G. tomentosum and G. mustelinum chromosomes or chromosomes segment(s) [28][29].

Materials and methods
Inbred cotton lines L-458, L-461, L-500 and L-501 (Gossypium hirsutum L.) from the Genetic Collection of the National University of Uzbekistan were used for producing cytogenetic stock series. Three types of radiations were applied -combined treatment of colchicine and gamma of irradiation translocation plants were obtained in the M 1 , M 2 and M 3 generations (Tables 2). They included exchanges between two different pairs chromosomes, translocations among three pair chromosomes and complex exchanges. Three types of the irradiation treatment were characterized by differences of the frequency and spectrum of the translocations. The treatment of seeds with combined colchicines and gamma-rays resulted 25 disomic plants with two or more translocations in the M 1 generation, with two, or even four multivalent configurations per PMCs in different plants. The unaffected plants from M 2 generation, a further 22 disomic plant with a number multivalents were detected. Progeny from different branches and bolls of one and the same parents was kept separate. Some of the parents were chimeras and as it was expected and subsequently confirmed, that different bolls from one and the same parent would give different result. From the meiosis, M 3 progenies were only recovered in which there was just one multivalent per PMC. Therefore, 23 translocations obtained in the M 3 generation were listed in the Table 1. In comparison to the single irradiation the combined irradiation treatment was most effective for producing the highest number of single as well as complex translocations in a number of meiocytes. Irradiation of seeds by thermal neutrons induced 31 translocations in the M 1 , 31 plants with interchanges in the M 2 and 10 -in M 3 generation (Table 2). Only four plants were recorded with complex translocations per PMC in the M 1 generation (Figure 1). Other translocations involved to two nonhomologous chromosomes ( Figure 2) and four -of three chromosomes. Moreover, the concurrent appearance of the two chromosomal aberrations causing and inducing chromosomal deficiencies and rearrangements, was also detected. The highest number of the chromosome changes occurred from the thermal neutron and unique chromosomal aberrations such as complex of the interchanges on PMS, multiple translocations involving up to three nonhomological chromosomes. Moreover, such rare mutations induced have not been observed earlier in experiments with other treatments. Such rare mutations were cluster fruiting habit for translocation (Tr18- Figure  10 C, F), reduced stigma (Mo62- Figure 18 C), cytoplasmic mutation virescent simultaneous with translocation (Tr28), unique desinaptic plant, pollen semisterility in homozygous stock (Tr21). It is also notable that one of the translocation plant 1475/30 4 had also a clear phenotypic character with cytoplasmatic mutation of yellow-green color of the leaf. This finding indicates the simultaneous appearance of chlorophyll deficiency and the chromosome translocation in a single plant. These specific mutations were not frequent, but they were specific for the thermal neutron irradiation and very important for cotton genetics as new genetic markers.  Table 2). All translocations were involving of two nonhomologous chromosomes formed quadrivalent associations ( Figure 3). Three translocations were involving of three chromosomes formed hexavalent associations. There were no PMCs with more than one multivalent among translocation plants after pollen irradiation. The greatest number of translocation plants with medium and high frequency (58.14%) was found in experiments with irradiation at 20 and 25 Gy. In comparison to the other irradiation the gamma-irradiation of pollen was the most effective for producing more different deficiencies.
Comparison of the mean frequency of multivalent per cell between translocations obtained in the M 1 , M 2 and M 3 generations after irradiation suggests that the highest number (26,42%) of interchanges with a high average number of multivalent at the MI of meiosis occurred in the M 1 generation. M 2 and M 3 progenies had more normal karyotypes than that seen in the M 1 plants and translocations with a high frequency of multivalent were uncommen in next generations. Those translocations, integrated into this collection from homozygous translocation lines.
Among 235 translocations, 224 involved two chromosomes but only 11 involved three nonhomologous chromosomes. Different types of multivalent configurations were found with alternate and adjacent orientations. The translocations were characterized by different frequencies of the multivalents at the MI of meiosis. Translocations with the large translocated chromosome segments are of greatest interest for tagging cotton chromosomes and developing new homozygous translocation lines. Electron microscopy of synaptonemal complexes at the pachytene stage revealed more cells with heterozygous chromosomal rearrangements than light microscopy at meiotic MI by a factor of 1.8 [44]. Those results indicate that the different frequencies of multivalents in heterozygotes for translocations at MI result from partial desynapsis and segregation of translocation multivalents into two "heteromorph " bivalents, which cannot be distinguished from normal bivalents at meiotic MI by light microscopy. The distribution of pollen fertility for 189 translocations is summarized in Figure 4. Pollen fertility was estimated by acetocarmine staining. Although the acetocarmine-based pollen fertility considered relatively insensitive method, it is widely used for preliminary screening of pollen quality in plants. The    In 19 translocations, pollen fertility varied between flowers of one and the same plant and 11 translocations were characterized by pollen sterility. On the whole, the high frequency of abortive pollen grains in flowers was typical for 48.5% of the translocations studied.
Note: the remaining 19 translocations were not included in the histogram due to their varied pollen fertility level in different flowers and 4 plants were complex interchanges. This broad variability of pollen fertility in the plants with interchromosomal exchanges hampers using this trait as a marker of heterozygosity for exchanges in cotton, in contrast to species of Pisum, Zea, Sorghum, and Petunia. Heterozygosity for translocations in these species is always accompanied by half-sterile pollen because of equal probabilities of ring-shaped and zigzag-shaped quadrivalent orientation. In addition, the detection of complete male and female sterility in some cotton translocants suggested that exactly the translocations were responsible for their sterility so far as these translocation plants did not produce any seed sets from self-pollination and intercrossing. Apparently, short translocated segments involved vital chromosome domains, whose rearrangements induced to abortive gametes.
Different techniques have been employed for isolating plants homozygous for the translocations [45]. Common methods have been used in maize and barley [46]. To identify homozygous, self the plants with normal pollen crossed to a standard normal line. If the normal being tested were homozygous for the translocation, all the F 1 hybrid plants should be partially sterile. A technique for quick isolation of translocation homozygotes that not require of the analysis pollen fertility into account was also worked out. This character has been varied in the cotton translocation heterozygotes from high up to pollen sterility and cannot be used the marker characteristic. Such techniques allowed us to isolate the number of different translocation homozygotes in progeny of one and the same parent containing two or more interchanges per PMC.
Translocations have been confirmed as homozygous after cytogenetic studies from selfpollination progenies of the heterozygotes according to the scheme ( Figure 5). Heterozygotes were selfed and progeny from each plant was examined at the metaphase I of meiosis. The plant that exhibited normal pairing were backcrossed to the control line L-458 in order to identify the plants homozygous for the translocations and the F 1 BC 1 progeny was examined at the metaphase I of meiosis for the presence of multivalents. F 1 BC 1 plants with multivalents pointed out homozygous for the translocation under consideration. As a result 33 new homozygous translocation stocks were isolated among those, 13 new translocation stocks of cotton (Tr1-Tr11, Tr25 and Tr26) were obtained in the combined treatment of seeds with colchicine and γ-rays hybrid progeny L-500 x L-461, one stock (Tr12) -in the pollen γirradiation hybrid progeny L-461 x L-501, while the others-from irradiation of seeds by thermal neutrons highly inbred line L-458.
In the progeny of the translocation heterozygotes, the deviations were found from the 1:2:3 ratio with a deficit of different types of plants. The latter can be explained by time limitation of examining plants or low viability by some types of progeny. So, it was not possible to establish one translocation (1020 1 -9) in the homozygous condition because only heterozygotes and normal plants detected in progeny. Probably, it can be attributed to localization of the breakpoints in the region of the chromosome, which cannot be reconstructed.
As it was discussed earlier for 28 of the 34 translocation lines in barley, homozygous plants were available, although one translocation -C 951 was not identified in the homozygous condition [47].
Translocation lines from our collection were numbered from Tr1 to Tr33 in the order of their detection. Thirty one translocations were simple reciprocal interchanges, involving only two nonhomologous chromosomes, whereas the two remaining (Tr2 and Tr20) were interchanges involving three non-homologous chromosomes. Subgenome assignment of the translocation was carried out with using hybrid DD-subgenome (F 1 G. thurberi x G. raimondii). Three types of modal configurations are expected at metaphase I of meiosis in the triploid hybrids [48]. Translocations being involving two A-subgenome chromosomes will have not A homoeologues in the triploid hybrid and showed modal chromosome configurations -13 (DD) bivalents and 13 (A) univalents; translocations being involving two DD-subgenome chromosomes -11 (DD) bivalents and 13 (A) univalents and one (DDDD) quadrivalent; translocations being involving AD-subgenome chromosomes -12 (DD) bivalents and 12 (A) univalents and one (ADD) trivalent. Tests involved many translocation lines but results were obtained for 5 lines only. Translocation lines Tr1, Tr7, Tr8 and Tr16 had AA-subgenome location translocated chromosomes because their triploid hybrids shown modal configurations 13 DD-bivalents and 13-A univalents ( Figure 6). Translocation line Tr2 is interchange which involves three non-homologous chromosomes. Their triploid hybrid characterized 11 univalents and 12 bivalents and one quadrivalent that pointed out on two A-subgenome and one D-subgenome location chromosomes.
Identification of translocation homozygotes from our collection was carried out using double translocation heterozygotes obtained after intercrossing translocation plants [1]. Modal chromosome configurations at metaphase I of meiosis -21 bivalents (II)+1 quadrivalent (IV)+1 hexavalent (VI) and 22 bivalents (II)+2 quadrivalents (IV) showed the involvement the different chromosomes, modal chromosome configurations -22 bivalents (II)+1 oktavalent (VIII) and 23 bivalents (II)+1 hexavalent (VI) indicated the chromosome in common and modal chromosome configurations -26 bivalents (II) indicated the involvement of the same arms of the two homological chromosomes. It is important to note that translocation lines Tr2 and Tr20 showed multivalents in hybrids because of involving the translocations among three non-homologous chromosomes. With respect to the crosses between translocation lines Tr7, Tr8 and Tr9 in which the M 1 parent plant was formed of several quadrivalents in the same meiocyte, three types were present in homozygote giving rise to the three translocation lines. The data indicated that Tr7 and Tr8 involved the same two non-homological chromosomes, but Tr7 and Tr8 on the one hand and Tr9 on the other hand are different chromosomes. The differences were found between translocation stocks both in the number of hybrids and the number of common chromosomes in the translocations in our investigation. Thus, the translocation lines Tr7, Tr8, Tr14, Tr16 and Tr27 showed more common chromosomes, whereas other lines-Tr1, Tr18, Tr20, Tr21, Tr23 and Tr26 showed uncommon ones. From these data it is observed that chromosomes in the translocations Tr7, Tr8, Tr14, Tr16 and Tr27 were more frequently involved in chromosome translocations, but chromosomes from the translocations lines-Tr1, Tr18, Tr20, Tr21, Tr23 and Tr26 were less frequently involved in translocations ( Figure 7). On the base of rare occurrence the chromosomes in common in the hybrid involved Tr1 and Tr20 translocation stocks (two and one, respectively) it was shown that these stocks have translocated chromosomes that were rare involving into interchanges. Thus, when Tr1 stock was crossed with Tr2 and Tr20 the ring consisting of 8 chromosomes was observed in both hybrids showing that Tr1 stock has interchanged chromosomes in common with Tr2 and Tr20 stocks. However, the crosses between Tr2 and Tr20 stock detected two rings consisting of six chromosomes that pointed out the involvement unique chromosome pair in Tr1 and Tr20. Preliminary chromosome numeration in the interchanges pointed out the involvement about 50% chromosome set into reciprocal translocations in 27 studied translocation stock from our collection.     (Table 3).  Seven of the monosomic plants had simultaneously independent chromosome interchanges so far as these shown both quadrivalent and univalent in MI meiosis. More than 70% of M 1 primary monosomics (25 of 34), were induced by high doses of pollen irradiation (20 -25 Gy) ( Table 3). The number of monosomics detected declined in subsequent generations (24 and 7, respectively), and one M 3 monosomic (Mo54) also displayed heterozygous translocation ( Figure 11). A specific feature of the pollination with irradiated pollen of cotton was a lot of genomic mutations such as chromosome deficiencies, chromosome arm deficiencies (22.51%) in comparison with the neutron irradiation (16.85%). The latter resulted from elimination of whole chromosomes, chromosome arms, or even the complete paternal genome and yielded monosomic, monotelodisomic, and haploid plants.  Figure 12).
An addition to traditional radiation-induced cotton monosomics, we used the desynaptic effect which have been found to be a useful source of aneuploidy in other crops. and an indicator of meiotic stability, which proposed by Love [49] for evaluation of meiosis in  wheat. We observed that pollen fertility was varied among the flowers on the same plant (from 2.61±1.27% to 91.81±1.18% in 179/2 desynaptic plant; Table 4). One unique desynaptic plant-356/8 (row number/plant number in M 1 generation in field) was observed from the desynaptic progenies studies. This plant produced monosomics in high frequency with a small size of univalents and strong phenotypic differences, suggesting monosomy for different chromosomes of cotton genome. In previous we identified two new monosomics (Mo30 and Mo67) using progeny of translocation plants [33]. Those results suggested association of univalent with two homeologous chromosomes. Such trivalents formed by pairing of homologous chromosomes were also found in other monosomic plants [50]. Moreover, two of them (Mo56 and Mo61) were also characterized with additional univalents. The other 12 cotton primary monosomics showed quadrivalent associations with different frequencies suggesting heterozygosity for their translocation. Analysis of the sizes of monosomes revealed medium univalent size in 44 monosomics (Fig. 13 C, D); whereas there were 22 monosomics with large univalents (Fig. 13  A, B). The number of monosomics having small univalents was slightly higher (27); moreover, among these, 6 monosomics with very small univalents were detected (Fig. 13 E, F). Therefore, according to a preliminary assignment of monosomes considered on the basis of their sizes to the subgenomes, 22 large monosomes can be assigned to the A t -genome and 27 monosomes of small sizes to the D t -genome.

Material
It is known that only three chromosome pairs of G. hirsutum have long arms that are two or three times the length of the short arms [51]. Monosomes of medium sizes demand special translocation tests with subgenome and chromosome number assignment of the translocated chromosomes. The analysis of subgenome assignment of unidentified monosomes of medium sizes showed the A t subgenome location [52] and significant deviation from the expected 1:1 ratio of the A t -subgenome monosome number to the D t -subgenome ones. This observation implied that preferential loss of the A t -subgenome chromosome was caused by specific genetic regulation system of chromosome disjunction and was not due to size of monosomes [52]. In our experiments, we detected a nearly 2 : 1 ratio of the A t -to the D t -subgenome monosomes for all that monosomes of medium sizes to be from A t -genome. The ratio observed us is not significantly different from the ratio given by Myles and Endrizzi [52]. This confirms a greater tolerance of G. hirsutum to loss of the large A t -genome chromosome than the small D t -genome chromosomes.
Analysis of the tetrads was carried out for 87 primary monosomics of our collection. Most of the monosomics (73 or 83.91%) had high meiotic index (more than 90%) than that of the control plants (95.11±0.46%). That indicates regular univalent chromosome disjunction. Fourteen of the monosomics (16.09%) were characterized with lowering of meiotic index from 89.33% (Mo74) to 68.32% (Mo4). Moreover, 10 of the monosomics had a smaller reduction of meiotic index (to 80%) compared others two monosomics (Mo4 and Mo16). These monosomics were induced in M 1 generation by pollen gamma-irradiation in doses of 20 and 25 Gy, leading to strong meiotic index reduction (to 68.32±1.10% and 76.07±0.93%, respectively). We also observe an increase of percentage of tetrads with micronuclei (to 6.87±0.60% and 21.56±0.90% respectively) in comparison with the control line (1.42±0.25%).Two other monosomics (Mo88 and Mo90), selected from M 3 generation treated by thermal neutrons and pollen with gamma-rays were characterized with different meiotic  index in various buds. Variation limits were also observed for the number of tetrads with micronuclei.
Meiotic index decrease in 6 monosomics (Mo16, Mo28, Mo52, Mo74, Mo88 and Mo90) could be explained because of the presence of a additional univalents at meiotic metaphase I. In contrast, meiotic index decrease in 4 monosomics (Mo8, Mo21, Mo23 and Mo57) was connected with simultaneous translocation heterozygosity that led to chromosome disjunction disturbances and the production of tetrads with micronuclei. However, meiotic index decrease in 3 monosomics with the modal chromosome pairing (Mo4, Mo34 and Mo37) and increase of number of tetrads with micronuclei In Mo4 (to 6.87±0.60%) directly demonstrated disturbances in monosome disjunction and imbalanced gamete formation. Therefore, the low frequency of tetrads with micronuclei in cotton monosomic argues for stability of monosomes, which seldom undergo irregular division (misdivision) of univalent centromeres [43]. This is confirmed by the fact that we found only 9 monotelodisomics and one isochromosome in more than 1000 cytologically examined plants of the progeny of various monosomics from our cytogenetical collection. These results are in contrast with earlier data on the degree of chromosome lagging in wheat monosomics, where the frequency of tetrads with micronuclei varied from 34.1 to 65.2% [53].
Pollen fertility after acetocarmine staining was studied in 93 primary cotton monosomics, isolated mainly from different types of irradiation. High pollen fertility was detected only in 30 plants with chromosome deficiencies that pointed out probable early haplo-deficient microspore abortion prior to mature pollen stage. Remaining monosomics were characterized with pollen fertility decrease. Thus, 17 monosomics had small lowering of pollen fertility (to 70%), 11 -semisterile pollen (to 40%) and 15 -strong pollen fertility reduction (to 5%) (Fig. 14). Pollen sterility was established in 6 monosomics (Mo5, Mo7, Mo10, Mo44, Mo45 and Mo47) derived from M 1 generation after irradiation of pollen and in two monosomics (Mo57 and Mo74) isolated after thermal neutron seed irradiation. Monosomics Mo5 and Mo44 did not produce any seeds from self-pollination and intercrossing that suggest their complete sterility. In 11 monosomics pollen fertility was varied among different flowers on the same plant; moreover, the variation limits were strongly differed. Reproduced monosomics from the 3 other families (Mo22, Mo39 and Mo46) also had pollen fertility variation in different flowers on the same plant from semisterile or low to reduced pollen fertility.
The reproduction of the monosomic plants was studied in the self-pollination and outcrossed progenies under the field and greenhouse conditions. Comparative analysis of the cotton monosomics produced both in the field and greenhouse revealed distinct morphological differences in comparison with disomic sibs. As a result, monosomics were reproduced in 18 generations under field condition. However, we did not analyze most of the progenies and determine exact transmission frequency in the field due to limited space, time and cost. Thirteen of 18 reproduced later under greenhouse condition whereas five monosomics (Mo22, Mo36, Mo39, Mo46 and Mo53) plants did not produce daughter monosomics.
The progenies of 81 different monosomics were studied in the greenhouse. All monosomic families strongly differed in number of plants studied, and in only 18 families were all progenies cytologically examined for monosome transmission rate ( Significant variability in transmission rates could be explained by differences in the viability of haplo-deficient gametes involving specific chromosomes. Theoretically, after selfing monosomics must produce progenies with 2n, 2n-1 and 2n-2 chromosome number in the ratio 1:2:1, but, in fact, cotton nullisomic gametes with 2n-2 are nonviable whereas n and n-1 gametes form in unequal frequencies because of lower haplo-deficient gamete viability and their incompetitiveness in comparison with normal gametes. Thus, all the differences in detected transmission rates involve deficiencies in various chromosomes of the cotton genome. Nevertheless, transmission rate similarities in some monosomes in our collection could indicate identities that should be explored. Note: the remaining 11 monosomics were not included in the histogram due to their varied pollen fertility level in different flowers.    A similar effect detected in daughter monosomics, confirmed the genetic determination of such variation and suggested possible chromosome localization of the gene(s) for male gametophyte viability in the deficient chromosomes. It is known that the majority of cotton chromosome deficiencies are not transmissible via pollen due to non-functionality of chromatindeficient pollen [54]. Besides, Kakani et al. [19] reported that gene(s) responsible for pollen spine development were located on long arm of chromosome 12 using the advanced technique of confocal laser scanning microscopy and substitution lines.
A study of the morphology of cotton monosomic plants revealed the specific influence of monosomy on many characters that were differentiated them from disomic sibs. Such characters were thin stem, feeble leafing, small leaves, short internodes, crooked sympodia, small flowers and bolls, as well as deformed and obligospermous bolls.  The most variability was observed for the character "presence/absence of nectarines" where in 15 monosomic lines not all bracts had nectarines, and Mo66 lacked any external nectarines. Nectarines of different sizes within a single flower were presented in 6 monosomic lines (Mo9, Mo27, Mo31, Mo39, Mo84 and Mo89) ( Figure 17). Monosomy had an influence on the stigma structure and sizes in a flower. Thus, there were shorter stigmata in 3 lines (Mo17, Mo19 and Mo28) and a broad "reverting" stigma in Mo39.
A new phenotypic marker for cotton monosomy -"reduced" stigma was detected in Mo62. Analysis of Mo62 progeny revealed the presence of reduced stigma only in monosomic cytotypes whereas disomic ones had normal stigma as did the control (Figure 18). This trait it makes possible to distinguish cytotypes within the progeny without cytological analysis. However, stigma reduction rate was varied in different flowers within the same plant: a little reduction stigma (to 7-9 mm); medium reduction (stigma to 2-6 mm), and strong reduction (stigma to 1 mm). Moreover, as a rule, strongly reduced stigmas were located inside the staminate column. Besides flowers with reduced stigma, there were flowers in which the stigma was closed inside the stylar tissue. A dependence of stigma reduction rates related to the seasons of a year was also established.
All daughter monosomics of Mo62 were fertile both as male and female but had lower seed number per a boll (22.30±1.83) and lower seed set (76.90±2.47 %) in comparison with the parental line L-458 (34.40±0.62 and 89.81±1.55, respectively). A monosome of G. hirsutum with a strong reduction of stigma but still fertile, has not been described. Thus the monosome in Mo62 for the chromosome of cotton genome could be new.
The most important changes due to monosomy concerned sizes and shapes of bolls as majority of them formed smaller round bolls almost ranging from spherical to elongated bolls with beaks or without beaks compared to control plants. Many of the bolls of monosomics were ribbed or deformed due to a number of abortive ovules and immature seeds ( Figure 19). As a result, the number of seeds per boll and seed set were lower in all monosomic lines (from 9.50±1.62 in Mo13 to 32.61±3.99 % in Mo76) in comparison with the parental line (34.40±0.62 and 89.81±1.55, respectively). Mo4 was characterized with variation of boll sizes within the same monosomic plant and also the fruit occurred in clusters. Flowers and fruit clusters were also observed in Mo19. Mo66 was distinguished by a large broad beak at the top of an ovoid boll (Figure 19 D). Thus, it was shown that an individual chromosome deficiency had a specific influence in plant morphology and that some of them had unique marker characters. However, the clear similarity both morphological and cytogenetic features in some monosomics of our  collection suggested probable redundancy for the same monosomic chromosomes among the plants.
Many small chromosomes presence in karyotype analysis of tetraploid cotton G. hirsutum and absence of distinctive morphological markers for the chromosomes make it impossible to distinguish and identify chromosomes in karyologic analysis. Therefore, we identified monosomes to be specific chromosomes of the cotton genome using the known translocation test on hybrids of monosomics with translocation lines from the Uzbek Cytogenetic Collection ( Table 6). Analysis of hybrid chromosome pairing was used to reveal monosomic translocation F 1 hybrids and to study "critical configurations". The recently developed 28 translocation lines (Tr1-Tr28) from our collection were used for monosome identification according to the method described previously [6].  in PMCs of the F 1 monosomic hybrid plants (Fig. 20). We also identified four monosome pairs    Tr1  Tr3  Tr5  Tr8  Tr11  Tr12  Tr16   Mo3  -+  -4   Mo7  ----7 Mo10 Mo36 3 Mo38 ---9 Mo39 Mo89 -5 Note:+associated,-independent Table 6. Cytological test for indentification of the monosomes with the help of translocation lines.
We had isolated 4 monosomics (Mo70 -Mo73) from the progeny of the same desynaptic plant and proposed possible monosomy for different nonhomologous chromosomes of the cotton genome. Indirect confirmation was available with the detection of monosome Mo73 homology and one of the chromosomes involved into interchanges in the line Tr3 whereas the other three monosomes from the progeny of the same desynaptic plant (Mo70, Mo71 and Mo72) did not has any chromosomes in common in the Tr3 interchange. Another monosome (Mo85), isolated from the other desynaptic progeny, showed homology with a chromosome involved in an interchange with Tr1. This test revealed that the chromosomes of Tr1 were rarely involved in translocations. Translocation line Tr1 had common chromosomes only with two lines -Tr2 and Tr20 with multiple interchanges [41]. This verified our assumption that new or rare monosomes would occur in progenies of desynaptic forms of cotton [33]. Translocation tests involving other 24 monosomic lines have not yet revealed any homology of the monosomes and the chromosomes involved in interchanges because they showed detections of chromosome pairing with 23 bivalents plus one univalent plus one quadrivalent ( Figure 21). However they did demonstrate the differences in the studying level of the lines as well as depended on transmission rates of the monosomics in hybrid progenies. There is an evidence of the comparative rareness of other monosomes from our collection.

Monotelodisomics, monoisodisomics and haploids in cotton G. hirsutum L.
Other type of chromatin deficiency namely monotelodisimics are characterized with absence of a chromosome arms. As a result 25 normal bivalents and one heteromorphic bivalent with arm deficiency formed in meiosis. There are 11 monotelodisomics following pollen and seed irradiation and 9 from different monosomic progenies in our collection at present time.
Misdivision of a chromosome via centromere region followed by irradiation produced telocentric chromosome formation. Centromere inactivation caused loss of a chromosome arm. Telosome pair was detected as two different size univalents with various frequencies in the monotelodisomics (to 1.85±0.15 in average per cell). A high frequency of heteromorphic bivalents (to 0.95±0.05 in average per cell) was registered in monotelodisomics ( Figure 22). Among the plants with arm deficiencies three had also a translocation. Monotelodisomics with the translocation had a telocentric chromosome as univalent in the majority of PMCs. As it is known, telocentrics for short arms are less often paired with normal homologous chromosomes [55]. Meiotic index was high (from 81.86±0.85 to 99.56±0.16), but the pollen fertility reduced in the monotelodisomics.
Isochromosome formation was also connected with damaging action of radiation to centromere chromosome regions. As a result of centromere inactivation instable telocentric formed its arm developed on 180 0 , gave an isochromosome. Three monoisodisomics our collection were differed with isochromosome pairing at metaphase I of meiosis. If one of them had heteromorphic in most PMS (to 0.96±0.04) in other plants the isochromosome was often as univalent (to 0.91±0.22 an average on PMS). In spite of high Mi (to 98.55±0.25) in two monoisodicomics pollen fertility was reduced.
Deficiencies for one chromosome arm occurred in the progenies of 9 monosomics. Thus, in four monosomic progenies (Mo2, Mo19, Mo34 and Mo61) that differed with respect to monosome transmission rates, monotelodisomics were produced due to univalent instability and resulted in misdivision. In the progenies of Mo6, Mo21, Mo22, Mo49, Mo54 and Mo68 daugher monosomics failed to produce, but monotelodisimics (from the progenies of Mo6, Mo21, Mo22, Mo49 and Mo68) and a monoisodisomic plant (from the progeny of Mo54) were detected. The results suggested an irregular univalent chromosome centromere misdivision in the parental monosomics that led to a single chromosome arm missing and formed either telocentric or isochromosome in the case of an arm doubling. Our results demonstrated the rather rare occurrence of telo-and isochromosomes in the monosomic progenies studied, which showed univalent misdivision to be rare.   Different mechanisms are proposed to explain the emergence of haploids during pollen irradiation. One of them suggests that haploids result from female parthenogenesis induced by pseudofertilization with irradiated pollen. According to another mechanism, fertilization occurs before zygotization, but the damaged paternal genome is eliminated early in development [56]. Some authors believe that haploids of G. hirsutum are completely sterile, wheareas haploids G. barbadense are fertile and produce seeds after pollination with normal pollen [57]. Moreover, a line of the latter species is known that frequently produced haploids of the androgenous and matroclinous types.

Storage and propagation cytogenetical collection of cotton
Seeds Cytogenetical Collection of cotton maintained under room conditions (20-25 0 C). There is no facility available for cold storage of seeds. They are placed in to parchment paper bags. Each bag has catalogue number and year of collection. Bags are stored in special metal boxes (30 x 11 cm) and boxes are placed in wooden-cases. Monosomic and translocation plants and of their hybrids are grown at the greenhouse conditions in soil. All data collected are stored as a hard copy catalogue book that is being conversed to electronic format.

Location, maintenance and funding
The Cytogenetical Collection of cotton currently stored in the National University of Uzbekictan at Tashkent. It is funded by Committee for Coordination of Science and Technology Development (CCSTD) under the Cabinet Ministry of Uzbekistan.

Conclusions
In conclusion we studied new Cotton Cytogenetic Collection adapted to the Central Asian condition in contrast Cytogenetic Collection from USA using different types of seed and pollen irradiation. We propose the presence of unique cotton aberrations involved chromosomes for absent chromosomes in American collection. The results suggested a detection of "reduced" stigma as a useful phenotypic marker for cotton monosomics which makes it possible to distinguish different cytotypes without cytological analyses. The results demonstrated of new unique desynaptic cotton plants in which progeny produced monosomics with high frequency. We observed the very occurrence of univalents misdivision probably owing to monosome stability in the unique genetic background. Our cotton monosomic lines are unique and should be a valuable cytogenetic tool not only for chromosome assignment of new marker genes and genome enrichment with new chromosome deficient plant, but also for a development of new cotton chromosome substitution lines and germplasm introgression.
Alternatively, the creation of chromosome substitution lines through crossing of each of the new monosomics with G.barbadense genotype (Pima 3-79) is in progress. This will serve as a foundation to apply molecular marker (e g., SSPs) for the identification of our monosomics in hybrids with chromosome substitutions for a given monosome. At the same time, our monosomic cotton lines with initial cytogenetic characteristics, which developed using single genome background, should be useful germplasm for cotton researchers to use as material for future breeding genetic, cytogenetic and molecular-genetic investigation of cotton genome.
In future we plan to identify the chromosome deficiencies by molecular markers (SSR) to map of cotton genome. Also we will continue identification monosomic lines of our cytogenetic collection using a well-defined tester-set of translocation lines of the USA Cytogenetic Collection, kindly provided by Dr. D.M. Stelly, Texas A&M University, USA, under USDA germplasm exchange program.