Study of genetic divergence for forage yield and biochemical traits in sorghum [Sorghum bicolor L. Moench] germplasm

Introduction

Sorghum is one of the most important and widely grown crops in the world having the area of 41.14 million hectare with the production of about 58.72 million tonnes globally whereas 5.00 million hectare and 4.50 million tonnes grain production in India (USDA Foreign Agricultural Services, 2019) [32] . Sorghum is known by various names in Africa, such that guinea-corn, dawa or sorgho in West Africa, durra in the Sudan, mshelia in Ethiopia and Eritrea, mtama in East Africa, kaffir corn in South Africa and amabele or mabele in several countries in Southern Africa. In the Indian sub-continent, it is known as jowar (Hindi), cholam (Tamil Nadu), jonna (Andhra Pradesh) and jola (Karnataka). Five basic races of cultivated sorghum are recognized as Bicolor, Guinea, Kafir, Durra and Caudatum (Harlan and De Wet, 1972) [5] . It has extensive variability of usage such as forage sorghum, grain sorghum and sweet sorghum, providing food, fodder, feed, fuel and fiber. The crop is mainly grown in tropical and subtropical areas because of its drought tolerance capacity, and quick growing habit, good palatability where agro-climatic conditions such as rainfall, temperature and soil are variable. Much of the crop is grown in the stress-prone and marginal areas of the semi-arid tropics, mainly on small holdings. In Northern Western India, it is grown for meeting the major fodder requirement of kharif and summer seasons.

Precise information on nature and degree of genetic variability helps the plant breeder in selecting the genetically diverse parents for the purposeful hybridization. (Arunachalam, 1981) [3] . Genetic improvement of yield especially in self-pollinated crops depends on nature and amount of genetic diversity (Joshi and Dhawan, 1966) [16] . Nutritionally, among the kharif fodders, sorghum is a crop par excellence with starch (63-68%), potential of high digestibility (50-60%), dry matter (20-35%), sugars (8-17%), crude protein (7.5-10.0%), calcium (0.53%), phosphorus (0.24%), and crude fiber (30-32%) (Sheoran et. al., 2000) [26] . Beside the higher content of carbohydrates, it has iron (Fe) and vitamin B 3 contents which are higher than maize and rice. It is a major staple food of many countries in Asia and Africa, sorghum is now a major feed crop in the United States, Argentina, Mexico, South Africa, and Australia (Miller and Kebede, 1984) [24] . Genetic diversity and relationship among different individuals is a prerequisite for any successful breeding programme. Genetic diversity among accessions provides opportunities for improvement of agronomic and nutritional quality traits in crops (Huang, 2004) [14] . It aids plant breeders to characterize and classify accessions into heterotic groups (Menz et al., 2004) [20] . Genetic diversity of plants determines their potential for improved efficiency and hence their use for breeding, which eventually may result in enhanced forage production. Genetic diversity explains the genetic differences between different populations within a species or between species. The parents having more genetic diversity result into higher heterotic expression in F 1 and greater amount of genetic variability in segregating populations (Shekhawat et al., 2001) [25] . One of the important approaches to sorghum breeding is hybridization and subsequent selection. Parents' choice is the first step in plant breeding program through hybridization. In order to obtain transgressive segregants, genetic diversity between parents is necessary (Joshi et al., 2004) [17] . The higher genetic diversity between parents, the higher heterosis in progeny can be observed (Joshi and Dhawan, 1966) [16] . Estimation of genetic diversity is one of appropriate tools for parental selection in sorghum hybridization programs. Appropriate selection of the parents is essential to be used in crossing nurseries to enhance the genetic recombination for potential yield increase. In view of the above, there is need to screen the diversity of sorghum germplasm based on yield and quality parameters to find out their suitability in different breeding programmes. There is a need to make genuine efforts to assess available diversity. Hence the present investigation was conducted to estimate the magnitude of genetic diversity present among the elite sorghum genotypes.

Materials and Methods

The initial research related to germplasm screening was carried out in the experimental area of Instructional Dairy Farm, Nagla, G.B. Pant University of Agriculture and Technology. Pantnagar, District U. S. Nagar, Uttarakhand during Kharif, 2018 and 2019. The experimental material for this experiment consisted of two hundreds and eighty diverse germplasm lines of sorghum along with six checks viz., SSG 59-3, Pant Chari-5, Pant Chari-6, CSV-21 F, CSH-22S, and CSV-24SS. The germplasm lines were evaluated in Augmented Block Design. The experiment was carried out in an Augmented Block Design (Federer, 1956, 1961, and Federer & Raghavrao, 1975) [10,11] with each block containing 35 test entries and 6 checks which were randomly allocated in 8 blocks. All genotypes were sown during Kharif 2018 and Kharif 2019 on 23 rd July 2018 and 27 th July 2019 respectively in single row of 5 meter length with a row spacing of 45 cm. All the recommended package of practices for sorghum was followed to raise a healthy crop. The observations were recorded on days to 50% flowering, days to maturity, number of leaves per plant, number of nodes, plant height (cm), leaf length (cm), leaf width (cm), leaf area (cm 2 ), flag leaf length (cm), flag leaf width (cm), stem girth (cm), internodal length (cm), panicle length (cm), panicle width (cm), leaf:stem ratio, 1000-grains weight (gm), grain yield per plant (gm), green fodder yield per plant (gm), dry fodder yield per plant (gm), foliar diseases zonate leaf spot and anthracnose (Thakur et al., 2010) [30] , shoot fly (Atherigona soccata) incidence (Dead hearts %), dry matter (%), brix %, HCN content (ppm) (Hogg and Ahlagreen, 1942) [13] and Gilchrist et al. (1967) [12] , protein content (%) (Jeckson, 1973) [15] , in-vitro dry matter disappearance (IVDMD) % (Erwin and Ellinston, 1959) [7] , neutral detergent fiber (Van Soest, 1991) [33] , acid detergent fiber (%) and cellulose (%) (Van Soest, 1991) [33] , acid detergent lignin (%), cellulose (%) and silica (%) (Van Soest, 1991) [33] . The data obtained from both years was pooled for diversity analysis. Hierarchical cluster analysis was performed on the basis of Euclidean distance between the genotypes. Euclidean distance was calculated by using the following method:

Euclidean distance: The Euclidean distance between i th and k th accession is:

Where, D ik = Euclidean distance between i th and k th accession A ij = performance of i th accession for j th character. A kj = performance of k th accession for j th character. m = number of accessions (I or k = 1, 2… m) n = number of characters (j =1, 2…n)

When the similarity matrix is computed from distance function, the hierarchical clustering method begins by finding the link between the two closest genotypes (Anderberg, 1973) [2] . The statistical analysis was performed by Indostat Hyderabad.

Results and Discussion

Knowledge about genetic diversity of parents in hybridization programme is essential as the crosses involving genetically diverse parents who are likely to produce not only high heterotic effects, but it also produce desirable transgressive segregants in the later segregating generations. The hierarchial cluster analysis discriminates genotypes in a different cluster on the basis of genetic diversity among the genotypes and thus enable breeder to select more genetically diverse parents for their crossing programme to recover desirable seggregants. The genotypes included in the same cluster may have different generations of time, different parental combinations or different generations of the same parental combinations. This proved that geographical diversity need not necessarily be related sown conditioned to the genetic diversity. The hierarchial cluster analysis had been found to be a potent tool in quantifying the degree of divergence in germplasm .This analysis provides a measurement of relative contribution of different components on diversity both in inter cluster and intra cluster level and genotypes drawn from widely divergent clusters are likely to produce heterotic combinations and wide variability in segregating generation (Rao, 1952) [22] . Among the different approaches of selecting parents, selection based on diversity has its own merit.

Distribution of genotypes into different clusters:

The clustering pattern of genotypes on the basis of Hierarchial cluster analysis has been presented in Table 1. The genotypes were grouped into XI distinct non-overlapping clusters suggesting considerable amount of genetic diversity prsesent in the experimental material. The cluster pattern of the genotypes showed non-parallelism between geographic and genetic diversity (Singh et al., 2009) [27] . The genotypes were grouped into XI distinct non-overlapping clusters. The highest number of genotypes were grouped into cluster-VIII (75) followed by cluster-II (39), cluster-IV (35), cluster-VII (35), cluster-III (26), cluster-I (21), cluster-IX (20), cluster-VI ( 16), cluster-X (15), cluster-V (3) whereas cluster-XI exhibited only single genotype.

Cluster-V: This cluster consisted of only three genotypes viz., CHS-22-SS, CSV-24-SS and CSV-19. This cluster mean had high cluster mean for leaf width, leaf area, flag leaf width, internodal length, leaf: stem ratio, 1000-grains weight, green fodder yield per plant, dry fodder yield per plant, dry matter percent, total soluble solids, in-vitro dry matter disappearance, acid detergent fiber and hemicelluloses content.

Cluster-VI: This cluster was marked with sixteen genotypes viz., EJ-3, RAJ-9-1, EJ-42, C-43, RAJ-15, IS-313, Pant Chari-6, EJ-19, EJ-26, EJ-27, EJ-40, IS-23992, EJ-30, IS-4925 and IS-33096. This cluster had high cluster mean for inter-nodal length, leaf: stem ratio, dry matter percent and total soluble solids. Cluster-X: This cluster was marked with fifteen genotypes viz., CO (FS)-29, IS-18850, SSG-21, IS-30117, IS-3353, E-159, IS-14357, IS-12735, IS-13566, IS-18927, IS-18844, IS-18933, SMC-8, SMC-13 and IS-28313. This cluster mean had high cluster mean for days to flowering, days to maturity, number of leaves, number of nodes, leaf length, flag leaf length, green fodder yield per plant, dry fodder yield per plant, protein percent and lignin content.

Cluster-VII:

Cluster-XI: This cluster had only single genotype IS-14241. This cluster mean had highest cluster mean for days to flowering, days to maturity, number of leaves, number of nodes and leaf length. The pattern of distribution of genotypes in different cluster exhibited that geographical diversity was not related to genetic diversity as genotypes of same geographical region were grouped into different clusters and vice-versa (Kumar et al., 2009) [19] .

Average intra and inter cluster distances: The intracluster and inter-cluster distances were calculated to determine the genetic relationship between members of different clusters and among the individuals within a cluster. The intra-cluster and inter-cluster distances has been represented in Table 2. Inter-cluster distance is the main criterion for the selection of genotypes (Khare et al., 2015) [18] . The genotypes belonging to those clusters having maximum inter-cluster distance are genetically more divergent and hybridization between these genotypes of different clusters is likely to produce wide range of variability with desirable individuals in segregating generations.

Intra-cluster distance:

The maximum intra-cluster distance was exhibited by cluster-II (58.202) followed by cluster-III (54.572), cluster-VII (49.145), cluster-X (48.898), cluster-VI (47.994), cluster-IX (45.248), cluster-I (44.766), cluster-IV (43.784), cluster-VIII (43.133), cluster-V (32.891) whereas minimum intra-cluster distance was exhibited by cluster-XI (0.000).

Inter-cluster distance:

The genotypes belonging to those clusters having maximum inter-cluster distance are genetically more divergent and hybridization between these genotypes of different clusters is likely to produce wide variability with desirable individuals. The highest intercluster distance was observed between clusters-V and XI (341.437) suggested distant relationship between members of these two clusters and upon crossing the members of these two clusters will give more genetic diversity in segregating generation followed by clusters-VI and XI (307. whereas the lowest inter-cluster distance was observed between clusters-VIII and IX (53.27) suggested a closer relationship between these two clusters and low degree of genetic diversity among the genotypes. Presence of substantial genetic diversity among the genotypes screened in the present study indicated that this material may serve as a good source for selecting the diverse parents for hybridization programme. In order to increase the possibility of isolating good trangressive segregants in the segregating generations it would be logical to attempt crosses between the diverse genotypes belonging to clusters separated by large inter-cluster distances.

Cluster mean for different characters: Cluster means were calculated for all the yield and quality traits along with some biochemical traits which exhibited considerable differences among the clusters. The mean performance of the clusters was used to select genetically diverse and agronomically superior genotypes under present study. The highest cluster mean for days to flowering was exhibited by cluster-XI (83.600), cluster-X (77.600), cluster-IX (73.100), cluster-II (69.900), cluster-III (67.000), cluster-VII (63.700), cluster-V (62.900), cluster-VIII (62.000), cluster-I (61.400), cluster-VI (59.000) whereas lowest cluster mean for days to flowering was exhibited by cluster-IV (57.700). The maximum cluster mean for days to maturity was observed in cluster-XI (145.000) followed by cluster-X (140.000), cluster-IX (137.000), cluster-II (133.000), cluster-V (132.000), cluster-III (130.000), cluster-VII (129.000), cluster-I (126.000), cluster-VIII (126.000), cluster-VI (125.000) whereas minimum by cluster-IV (120.000). The highest cluster mean for number of leaves was exhibited by cluster-XI (20.000), cluster-X (18.000), cluster-IX (17.000), cluster-V (17.000), cluster-IV (16.000), cluster-VIII (16.000), cluster-II (16.000), cluster-III (16.000), cluster-I (16.000), cluster-VII (16.000) whereas lowest cluster mean for number of leaves was exhibited by cluster-VI (12.000). The maximum cluster mean for number of nodes was observed in cluster-XI (18.100) followed by cluster-V (17.400), cluster-X (16.500), cluster-IX (16.100), cluster-IV (15.100), cluster-II (14.900), cluster-VIII (14.800), cluster-III (14.300), cluster-I (14.300), cluster-VII (14.100) whereas minimum by cluster-VI (11.800). The highest cluster mean for plant height was exhibited by cluster-X (409.000), cluster-IX (393.000), cluster-VIII (390.000), cluster-I (378.000), cluster-XI (366.000), cluster-II (362.000), cluster-III (350.000), cluster-IV (350.000), cluster-VII (333.000), cluster-VI (291.000) whereas lowest cluster mean for plant height was exhibited by cluster-V (290.000). The maximum cluster mean for leaf length was observed in cluster-XI (98.430) followed by cluster-X (92.540), cluster-IV (92.260), cluster-II (90.180), cluster-III (83.960), cluster-V (83.760), cluster-I (82.720), cluster-IX (81.010), cluster-VIII (80.630), cluster-VII (72.870) whereas minimum by cluster-VI (69.250). The highest cluster mean for leaf width was exhibited by cluster-V (11.200), cluster-IV (9.160), cluster-III (8.890), cluster-II (8.780), cluster-IX (7.970), cluster-VII (7.490), cluster-VIII (7.290), cluster-I (7.220), cluster-VI (7.160), cluster-X (5.080) whereas lowest cluster mean for leaf width was exhibited by cluster-XI (4.590). The maximum cluster mean for leaf area was observed in cluster-V (732.000) followed by cluster-IV (549.000), cluster-II (531.000), cluster-III (501.000), cluster-IX (436.000), cluster-VIII (398.000), cluster-I (397.000), cluster-VII (368.000), cluster-VI (331.000), cluster-XI (327.000) whereas minimum by cluster-X (304.000). The highest cluster mean for flag leaf length was exhibited by cluster-XI (51.100), cluster-IV (46.800), cluster-X (46.500), cluster-II (45.400), cluster-III (42.900), cluster-V (42.700), cluster-I (41.500), cluster-IX (40.900), cluster-VIII (40.700), cluster-VII (36.900) whereas lowest cluster mean for flag leaf length was exhibited by cluster-VI (34.600). The maximum cluster mean for flag leaf width was observed in cluster-V (9.000) followed by cluster-III (5.000), cluster-IV (5.000), cluster-II (4.000), cluster-IX (4.000), cluster-VII (4.000), cluster-VI (4.000), cluster-VIII (4.000), cluster-I (4.000), cluster-XI (3.000) whereas minimum by cluster-X (2.000). The highest cluster mean for stem girth was exhibited by cluster-IX (2.658), cluster-III (26.636), cluster-IV (2.595), cluster-VII (2.539), cluster-II (2.471), cluster-I (2.465), cluster-VIII (2.438), cluster-V (2.303), cluster-VI (2.217), cluster-X (2.020) whereas lowest cluster mean for stem girth was exhibited by cluster-XI (2.017). The maximum cluster mean for inter-nodal length was observed in cluster-V (36.000) followed by cluster-VI (31.000), cluster-VIII (30.500), cluster-I (29.700), cluster-II (28.200), cluster-IV (27.000), cluster-IX (27.000), cluster-VII (26.900), cluster-X (26.600), cluster-III (26.300) whereas minimum by cluster-XI (19.200). The highest cluster mean for panicle length was exhibited by cluster-XI (37.000), cluster-VIII (27.000), cluster-I (27.000), cluster-X (26.000), cluster-IV (24.000), cluster-IX (23.000), cluster-II (22.000), cluster-V (20.000), cluster-III (19.000), cluster-VI (17.000) whereas lowest cluster mean for panicle length was exhibited by cluster-VII (14.000). The maximum cluster mean for panicle width was observed in cluster-XI (20.000) followed by cluster-I (15.000), cluster-X (15.000), cluster-VIII (15.000), cluster-V (12.000), cluster-IX (11.000), cluster-II (10.000), cluster-IV (10.000), cluster-III (8.000), cluster-VI (7.000) whereas minimum by cluster-VII (7.000). The highest cluster mean for leaf:stem ratio was exhibited by cluster-V (0.440), cluster-VI (0.38), cluster-II (0.330), cluster-I (0.320), cluster-VIII (0.310), cluster-VII (0.310), cluster-III (0.300), cluster-IX (0.300), cluster-X (0.300), cluster-XI (0.300) whereas lowest cluster mean for leaf:stem ratio was exhibited by cluster-IV (0.290). The maximum cluster mean for 1000-grains weight was observed in cluster-V (40.000) followed by cluster-III (28.000), cluster-IV (27.000), cluster-IX (27.000), cluster-II (22.000), cluster-VIII (21.000), cluster-VII (20.000), cluster-I (20.000), cluster-VI (15.000), cluster-XI (13.000) whereas minimum by cluster-X (12.000). The highest cluster mean for grain yield per plant was exhibited by cluster-III (108.100), cluster-IV (107.800), cluster-IX (106.400), cluster-V (103.400), cluster-II (87.890), cluster-VIII (83.240), cluster-VII (79.440), cluster-I (78.210), cluster-VI (57.600), cluster-XI (48.340) whereas lowest cluster mean for grain yield per plant was exhibited by cluster-X (47.830). The maximum cluster mean for green fodder yield per plant was observed in cluster-XI (940.000) followed by cluster-V (445.000), cluster-X (378.000), cluster-IV (366.000), cluster-II (344.000), cluster-III (336.000), cluster-IX (335.000), cluster-VIII (301.000), cluster-VII (301.000), cluster-VI (260.000) whereas minimum by cluster-I (259.000). The highest cluster mean for dry fodder yield per plant was exhibited by cluster-XI (426.000), cluster-V (182.000), cluster-X (140.000), cluster-IV (133.000), cluster-III (122.000), cluster-II (119.000), cluster-IX (118.000), cluster-VII (111.000), cluster-VIII (107.000), cluster-VI (99.300) whereas lowest cluster mean for dry fodder yield per plant was exhibited by cluster-I (95.100). The maximum cluster mean for dry matter percent was observed in cluster-V (42.500) followed by cluster-III (38.200), cluster-VI (37.500), cluster-I (37.500), cluster-IV (37.500), cluster-X (37.100), cluster-IX (37.100), cluster-VII (36.800), cluster-VIII (36.100), cluster-II (34.600) whereas minimum by cluster-XI (34.400). The highest cluster mean for total soluble solids was exhibited by cluster-V (13.000), cluster-II (9.000), cluster-VI (9.000), cluster-IV (8.000), cluster-IX (7.000), cluster-VIII (7.000), cluster-XI (7.000), cluster-III (6.000), cluster-X (6.000), cluster-I (5.000) whereas lowest cluster mean for total soluble solids was exhibited by cluster-VII (5.000). The maximum cluster mean for hydrocyanic content was observed in cluster-XI (110.000) followed by cluster-I (94.48), cluster-IV (93.320), cluster-VII (92.370), cluster-VIII (87.500), cluster-VI (87.230), cluster-III (87.170), cluster-X (85.470), cluster-II (84.200), cluster-IX (83.260) whereas minimum by cluster-V (72.930). The highest cluster mean for protein content was exhibited by cluster-XI (16.200), cluster-III (12.900), cluster-X (12.400), cluster-VIII (11.800), cluster-II (11.700), cluster-I (11.100), cluster-VII (10.900), cluster-IV (10.800), cluster-IX (10.600), cluster-VI (9.640) whereas lowest cluster mean for protein content was exhibited by cluster-V (7.170). The maximum cluster mean for in-vitro dry matter disappearance per cent (IVDMD) was observed in cluster-III (59.000) followed by cluster-V (59.000), cluster-IV (58.000), cluster-VI (57.000), cluster-IX (57.000), cluster-X (56.000), cluster-XI (56.000), cluster-VIII (55.000), cluster-VII (54.000), cluster-I (54.000) whereas minimum by cluster-II (53.000). The highest cluster mean for neutral detergent fiber was exhibited by cluster-XI (60.000), cluster-IX (57.000), cluster-IV (56.000), cluster-VIII (56.000), cluster-VI (56.000), cluster-VII (56.000), cluster-X (55.000), cluster-II (55.000), cluster-III (55.000), cluster-I (54.000) whereas lowest cluster mean for neutral detergent fiber was exhibited by cluster-V (52.000). The maximum cluster mean for acid detergent fiber was observed in cluster-III (38.800) followed by cluster-I (38.600), cluster-V (37.500), cluster-X (36.800), cluster-II (36.300), cluster-IV (35.100), cluster-VII (35.000), cluster-VIII (34.700), cluster-VI (34.400), cluster-IX (33.200) whereas minimum by cluster-XI (32.300). The highest cluster mean for cellulose content was exhibited by cluster-III (31.000), cluster-I (30.000), cluster-X (30.000), cluster-II (30.000), cluster-IV (29.000), cluster-VIII (29.000), cluster-VII (29.000), cluster-VI (29.000), cluster-V (28.000), cluster-IX (28.000) whereas lowest cluster mean for cellulose content was exhibited by cluster-XI (28.000). The maximum cluster mean for lignin content was observed in cluster-I (6.890) followed by cluster-III (6.275), cluster-X (5.684), cluster-II (5.585), cluster-VI (5.566), cluster-VII (5.389), cluster-IV (5.056), cluster-VIII (5.011), cluster-V (5.005), cluster-XI (4.931) whereas minimum by cluster-IX (4.917). The highest cluster mean for silica content was exhibited by cluster-I (2.640), cluster-II (2.420), cluster-III (2.370), cluster-V (2.130), cluster-X (2.070), cluster-IV (2.040), cluster-VII (2.010), cluster-VIII (1.920), cluster-VI (1.860), cluster-IX (1.710) whereas lowest cluster mean for silica content was exhibited by cluster-XI (1.230). The maximum cluster mean for hemicelluloses content was observed in cluster-XI (29.300) followed by cluster-IX (26.500), cluster-V (24.800), cluster-VIII (23.600), cluster-IV (23.600), cluster-VII (23.500), cluster-VII (23.100), cluster-X (21.200), cluster-II (29.900), cluster-III (17.900) whereas minimum by cluster-I (16.700). The highest cluster mean for anthracnose was exhibited by cluster-VII (53.700), cluster-III (29.500), cluster-XI (27.700), cluster-I (24.800), cluster-IV (22.500), cluster-IX (20.000), cluster-VI (18.400), cluster-VIII (18.400), cluster-X (15.300), cluster-II (13.300) whereas lowest cluster mean for anthracnose was exhibited by cluster-V (8.160). The maximum cluster mean for zonate leaf spot was observed in cluster-II (12.000) followed by cluster-III (9.000), cluster-IV (8.000), cluster-VII (8.000), cluster-V (8.000), cluster-I (7.000), cluster-IV (7.000), cluster-IX (7.000), cluster-VI (5.000), cluster-X (4.000) whereas minimum by cluster-XI (2.000). The highest cluster mean for shoot fly incidence was exhibited by cluster-VII (38.620), cluster-IX (32.350), cluster-III (30.380), cluster-II (21.070), cluster-V (17.570), cluster-IV (17.450), cluster-VI (16.780), cluster-VIII (14.260), cluster-I (10.710), cluster-X (9.718) whereas lowest cluster mean for shoot fly incidence was exhibited by cluster-XI (4.906). Classification of the germplasm in to divergent groups based on inter cluster distances, per se performance and selection of parents from diverse clusters was reported in several studies [ (Damor et al., 2017) [4] , (Ahalawat et al., 2018) [1] and (Rohila et al., 2022)] [24] . Crosses suggesting parents belonging to most divergent clusters would be expected to manifest maximum heterosis and also wide variability of genetic architecture [(Thant et al., 2020) [31] and (Deep et al., 2020) [6] . These results of our present study areare somewhat in accordance with the findings of, [28,1,24,23] . cluster-VIII (75) whereas cluster-XI exhibited only single genotype. The maximum intra-cluster distance was exhibited by cluster-II (58.202) whereas minimum intracluster distance was exhibited by cluster-XI (0.000). The clusters with high intra-cluster distances suggested that genotypes in these clusters were more genetic diverse than the genotypes in other clusters with low intra-cluster distances. Low intra-cluster distance suggested a closer relationship between these two clusters and low degree of genetic diversity among the genotypes whereas high intra cluster distance represented high amount of genetic diversity among members of same cluster. The highest inter-cluster distance was observed between clusters-V and XI (341.437) suggested distant relationship between members of these two clusters and upon crossing the members of these two clusters will give more genetic diversity in segregating generation whereas the lowest inter-cluster distance was observed between clusters-VIII and IX (53.27) suggested a closer relationship between these two clusters and low degree of genetic diversity among the genotypes. Presence of substantial genetic diversity among the genotypes screened in the present study indicated that this material may serve as a good source for selecting the diverse parents for hybridization programme. In order to increase the possibility of isolating good trangressive segregants in the segregating generations it would be logical to attempt crosses between the diverse genotypes belonging to clusters separated by large inter-cluster distances.

References

1. Ahalawat NK, Arya VK, Kumar P, Singh SK. Genetic divergence in forage sorghum (Sorghum bicolor L. Moench). Journal of Applied and Natural Science. 2018;10(1):439-444.
2. Arunachalam V. Genetic distances in plant breeding. Indian Journal of Genetics and Plant Breeding. 1981;4:226-236.
3. Damor HI, Parmar HP, Gohil DP, Patel AA. Genetic variability, character association, path coefficient in forage sorghum (Sorghum bicolour L. Moench). Green Farming. 2017;9(2):218-233.
4. De Wet JMJ, Harlan JR. Origin and evolution of guinea sorghums. East Africa Agriculture Journal. 1972;38:114-119.
5. Deep H, Chakraborty I, Arya S, Lamba P, Pahuja SK, Tokas J. Genetic diversity analysis by D2 clustering of fodder yield and its related traits in forage sorghum. Pantnagar Journal of Research. 2020;18(3).
6. Erwin ES, Elliston NG. Rapid method of determining digestibility of concentrates and roughages in cattle. Journal of Animal Science. 1959;18:1518.
7. Federer WT. Augmented Designs. The Hawaiian Planters' Record. 1956;55(2):191-208.
8. Federer WT. Augmented designs with one-way elimination of heterogeneity. Biometrics. 1961;17(3):447-473.
9. Federer WT, Nair RC, Raghavarao D. Some augmented row-column designs. Biometrics. 1975;31:361-373.
10. Gilchrist DG, Lueschen WE, Hittle CN. Revised method for the preparation of strands in the sodium picrate assay of HCN. Crop Science. 1967;7:267-268.
11. Hogg PG, Ahlgren HC. A rapid method of determining hydrocyanic acid content of single plant of sudan grass. Journal of the American Society of Agronomy. 1942;43:199-200.
12. Joshi AB, Dhawan NL. Genetic improvement of yield with special reference to self fertilizing crops. Indian Journal of Genetics and Plant Breeding. 1966;26:101113.
13. Joshi BK, Mudwari A, Bhatta MR, Ferrara GO. Genetic diversity in Nepalese wheat cultivars based on agromorphological traits and coefficients of parentage. Nepal Agriculture Research Journal. 2001;5:7-17.
14. Khare M, Rangare NR, Singh RP. Evaluation of genetic diversity in Mexican wheat (Triticum aestivum L.) genotypes for quantitative and qualitative traits. International Journal of Plant Protection. 2015;8(1):7780.
15. Kumar B, Lal GM, Ruchi, Upadhyay A. Genetic variability, diversity and association of quantitative traits with grain yield in bread wheat (Triticum aestivum L.). Asian Journal of Agricultural Sciences. 2009;1(1):4-6.
16. Menz MA, Klein RR, Unruh NC, Rooney WL, Klein PE, Mullet JE. Genetic diversity of public inbreds of sorghum determined by mapped AFLP and SSR markers. Crop Science. 2004;44(4):1236-1244.
17. Miller FR, Kebede Y. Genetic contributions to yield gains in sorghum, 1950 to 1980. In: Fehr WR, editor. Genetic Contributions to Yield in Five Major Crop Plants. Crop Science Society of America Special Publication 7. Madison, WI: Crop Science Society of America; c1984. p. 1-14.
18. Rathod VU, Kalpande HV, Rathod PS. Genetic divergence studies in induced mutants of Rabi sorghum (Sorghum bicolor L. Moench). The Pharma Innovation Journal. 2023;12(1):1770-1772.
19. Rohila N, Arya S, Pahuja SK, Kumari P, Panchta R. Genetic diversity among forage sorghum (Sorghum bicolor L. Moench) genotypes. Forage Research. 2022;48(1):28-30.
20. Shekhawat US, Vijay P, Singhania DL. Genetic divergence in barley (Hordeum vulgare L.). Indian Journal of Agricultural Research. 2001;35(2):121-123.
21. Sheoran RK, Govila OP, Singh B. Genetic architecture of yield and yield contributing traits in pearl millet. Annals of Agricultural Research. 2000;21:443-445.
22. Singh D, Singh SK, Singh KN. Diversity of salt resistance in a large germplasm collection of bread wheat (Triticum aestivum L.). Crop Improvement. 2009;36(1):9-12.
23. Tesfaye K. Genetic diversity study of sorghum (Sorghum bicolor L. Moench) genotypes, Ethiopia. Acta Universitatis Sapientiae Agriculture and Environment. 2017;9:44-54.
24. Thakur RP, Rao VP, Wu BM, Subbarao KV, Mathur K, Tailor HC, et al. Genetic resistance to foliar anthracnose in sorghum and pathogenic variability in Colletotrichum graminicola. Indian Phytopathology. 2010;60(1):13.
25. Thant SM, Kumari P, Guar A, Pahuja SK. Genetic diversity study for identification of dual purpose sorghum. Forage Research. 2020;46(2):148-151.
26. Van Soest PJ, Robertson JB, Lewis BA. Methods for dietary fiber, neutral-detergent fiber and non starch polysaccharides in relation to animal nutrition. Journal of Dairy Science. 1991;74:3583-3597.