Genetic Analysis in mango (Mangifera indica L.) based on fruit characteristics of 400 genotypes

Mango (Mangifera indica ..) is originated from the Indo-Burma region and genus Mangifera has more than 60 species world-wide, the highest diversity being found in the Malayan Peninsula, Borneo and Sumatra (Bompard,1993). Mukherjee (1953) opines that mango has been under cultivation for at least 4000 years with over 1000 varieties in cultivation. Almost all these are selections made from open-pollinated seedlings and selection by man from seedlings of unknown parentage has played the most significant role in the development of new mango cultivars (Singh, 1963). Mango is a premier fruit crop in India as well as some other countries in the tropical world with respect to being its eminent place in nutritional security and employment and income generation. The present scenario and expected future need of mangoes necessitate bringing improvement in mango with respect to the productivity not only per tree but also per unit area of land. A need has therefore, arisen to develop high yielding varieties of dwarf plant type, high fruit quality and resistant to biotic and abiotic stresses. Studies have been made for understanding diversity in the genus Mangifera and the possibility of its use in improvement of mango through introduction and selection of promising varieties for commercial cultivation and making further improvement in the existing varieties through inter-specific and inter-varietal hybridization and induction of useful mutations.

Mukherjee (1948) has described 72 mango varieties from Bengal, Bihar and Uttar Pradesh.

Simultaneously, Naik and Gangolly (1950) have described 335 varieties of South India. Apart from the fruit characters, they have also laid great stress on the vegetative characters. The list of 1595 cultivars of mango in world (Pandey, 1998) was revised with the names of 1663 cultivars by Pandey and Dinesh (2010). The updated list now contains the names of 1682 cultivars. There are seven centres of diversity exists in India which includes (i) humid subtropical region (Manipur, Tripura, Mizoram and south Assam), (ii) Chhota Nagpur Plateau (trijunction of Madhya Pradesh, Odisha and Bihar), (iii) Santhal Paragana,(iv) Southern Madhya Pradesh (tribal area) adjoining Odisha and Andhra Pradesh, (v) Dhar Plateau of Madhya Pradesh adjoining South Rajasthan and Gujarat, (vi) humid tropical southern peninsular India and (vii) Andaman & Nicobar group of islands (Yadav and Rajan, 1993). A list of the names of cultivars available in the world as probable gene sources for dwarf-ness, fruit size, red peel colour, high pulp content, high content of total soluble solids, long shelf- life of fruit, regularity in fruit bearing, earliness and lateness in fruit maturity and good processing quality have been mentioned by Pandey and Dinesh (2010). Dinesh and Vasugi (2002) catalogued 151 cultivars of mango including M. zeylanica. Another lot of 223 varieties of mango were catalogued by Dinesh et al., (2012) using Bioversity International Descriptors and they developed barcodes for these varieties through molecular characterization. There is a great variation in fruit weight in mango. Pandey and Dinesh (2010) categorized mango on the basis of fruit weight as very small (99g and below), small (100-149g), medium large (150-299g), large (300-500g) and very large (more than 500g) fruited varieties. Out of 61 varieties Out of 61 varieties of mango registered in U.S.A., 19 varieties have been reported to bear large to very large fruits (Brooks and Olmo, 1972). The evaluation of genetic variability with in a cultivated crop has important consequences in plant breeding and germplasm management. The yield and its contributing traits improvement in this crop can be achieved through selection of superior genotypes with desirable traits existing in nature. Mahalnobis (D2) statistics which is based on the multivariate analysis of quantitative trait is powerful tool for measuring genetic divergence among the population. Therefore, an attempt has been made to study the variability of fruit characteristics among 400 mango germplasm.

The experimental materials consists of 400 mango genotypes belonging to different geographical regions of the world were evaluated for over the three years (2014-17) at ICAR-Indian Institute of Horticultural Research, Bengaluru. The experimental material comprised of 400 mango germplasm belonging to different geographical regions and evaluated in the year 2015, 2016 and 2017. The fruit characteristics such as fruit weight (g), fruit length (cm), fruit diameter (cm), fruit thickness (cm), pulp recovery (%) and TSS (ÚBrix) have been recorded by using the standard procedure. Thirty qualitative characteristics have been used to catalogue 400 varieties as per the standard descriptors given by the Bioversity International (IPGRI, 1997). At the first instance, statistical tools such as ANOVA and F-test were used to evaluate the significant difference (p<0.05) among the varieties/hybrid individually for all the traits. Least Significant Difference (LSD) was computed as a Post-hoc test (Cochran and Cox, 1957). SAS GLM was used to develop suitable codes for the statistical analysis (SAS V 9.3 2012).

The analysis of variance for 6 quantitative traits showed significant differences among the 400 genotypes of mango which indicates the existence of genetic diversity. Means for different qualitative characters of 400 accessions of mango are given as Supplementary Table S1 (Available online). The maximum fruit weight of 1404.27g in variety Sora followed by 1280.67g in variety Amini where as the minimum fruit weight was of 29,51g in variety Kana Appe followed by 39.31g in variety Halasage. The fruit length was maximum (22.03 cm) in var. Sora followed 16.77cm in hybrid 7/15 where as the minimum fruit length of 4.33cm was recorded in var. Pacharasi. The maximum fruit diameter 13.50 cm was recorded in var. Dorgani Kavi followed by 11.33 cm in var. Maharaja Pasand where as minimum fruit diameter of 3.33 cm was recorded in var. Dodderi Jeerige. The maximum fruit thickness of 11.40cm was recorded in var. Dorgani Kavi followed by 9.87 in var. Amini whereas the minimum thickness was recorded (2,80cm) in Haldotta Appe. The highest pulp recovery (89.67%) was recorded in var. Manoranjan followed by 88.20 % in var. Lahara whereas the lowest pulp recovery was recorded in var. Halasage. The highest TSS (ÚBrix) of 31.00 was recorded in var. Dattatreya local followed by 30.67 in var.K-0-7 where as the lowest was recorded in var. Halasage (5.5%)

The cluster mean analysis (Table 1) reveals that there is a huge variation among the clusters. Highest fruit weight of 1404.27 g was recorded in cluster 10 followed by cluster 15 with 1280.67g. Whereas the lowest fruit weight was recorded in cluster 16 (30.94g). The highest fruit length (22.03 cm) was recorded in cluster 10 followed by 17.80 cm in cluster 14. Similarly, the fruit diameter was highest (12.18 cm) in cluster 10 followed by 12.03 cluster 4. The fruit thickness was highest (10.60 cm) in cluster 15 followed by cluster 4 with 9.96 cm. The pulp recovery was maximum (87.16%) in cluster-14 followed by cluster 4 and 18 with 79.28 and 78.41 %, respectively.

Table 1
Cluster mean for quantitative traits of 400 mango genotypes

The maximum inter cluster (D2) value was obtained between cluster 10 and cluster 11. These clusters may be used for hybridization programme due to wide variability and possibility of transgressive sergeants. The minimum cluster distance was obtained in cluster 2 and cluster 5 which indicates that the accessions belonging to such clusters are relatively close. The selection of parents from genetically close clusters may be due to narrow genetic base and inbreeding depression.

The pattern of distribution of accessions in different clusters indicating the existence genetic diversity which is related to geographical distribution. These 400 genotypes were grouped in to 18 clusters as presented in Table 2 which is apparent that cluster 1 (28 accessions), cluster 2(63 accessions), cluster 3(50 accessions), cluster 4 (3 accessions), cluster 5 (56 accessions), cluster 6 (20 accessions), cluster 7 (1 accessions), cluster 8 (34 accessions), cluster 9 (7 accessions), cluster 10 (1 accession), cluster11 (24 accessions), cluster 12 (6 accessions), cluster 13 (53 accessions) cluster14 (1 accession), cluster15 (1 accession), cluster16 (3 accessions), cluster17 (12 accessions) and cluster 18 (10 accessions).The clusters such as 7, 10, 14 and 15 had 1 accession in each with hybrid 7/15, Sora, Tella Gulabi and Amini, respectively. All these accessions possessed the fruit weight ranged from 1015.20g to 1280.67g which are meant for pickle making.

Table 2
Distribution mango accessions in various clusters

Estimates of phenotypic variance and genotypic variance had only a narrow difference for all six characters studied indicating that these characters are not much influenced by environmental factors (Table 3). This also suggests the presence of sufficient genetic variability which can be exploited by adopting clonal selection or selection of chance seedlings. The maximum PCV was recorded for fruit weight followed by fruit length and fruit diameter. This indicates the better scope for phenotypic selection of these traits for improvement.

Heritability and genetic advance for fruit characters varied considerably. The high heritability (0.84 to 0.94) and high estimate of genetic advance recorded for fruit weight, fruit length, fruit diameter, fruit thickness, pulp recovery and TSS. High heritability indicates the effectiveness of selection through phenotypic performance but it does not mean a high genetic gain. However, high heritability associated with high genetic advance proves more useful for efficient improvement of a character through simple selection. In the present study, high heritability estimates with high genetic advance as per cent over mean was observed for the all the fruit traits studied indicating the possible role of additive gene action, whereas moderate heritability with low genetic advance as per cent over mean indicating the non-additive gene action. In the present investigation, the estimates of genotypic correlations in general were higher than phenotypic correlations, indicating the presence of inherent association between various characters.

A total of 30 characters have been used to group the 400 genotypes as per standard descriptor of the Bioversity International (Table 4) which is evident that there are gradations of variations due to the heterozygosity nature of the crop.

Table 3
Genetic parameters for fruit characteristics of 400 mango accessions

Table 4
Grouping of 400 germplasm of mango based on qualitative traits

The analysis of variance for 6 quantitative traits exhibited the significant differences among the 400 genotypes of mango which indicates the existence of genetic diversity within the Mangifera indica. The maximum fruit weight of 1404.27g in var. Sora followed by 1280.67g in var. Amini whereas the minimum fruit weight was of 29,51g in var. Kana Appe followed by 39.31g in var. Halasage. The results are similar as reported by Pandey and Dinesh (2010). While selecting the parents for hybridization programme, the quantitative characters such as fruit weight, fruit length, fruit thickness, fruit diameter, and pulp recovery should be considered as all of them have high heritability. Transgressive segregation was observed for fruit size in the progeny (Iyer and Subramanyam, 1987) and it appeared to be governed by additive genes (Sharma and Majumder, 1988a). Detailed study conducted by Prabhuram (1998) based on heritability, expected genetic advance and total genetic variance gave useful information on inheritance of different fruit characters. Fruit weight was influenced considerably by the environmental conditions. Equal proportion of the additive and non-additive components of the total genetic variance influenced its genotype. Rajan et al., (2009) found high degree of broad sense heritability in mango varieties for the length and weight of fruit, peel weight and length and weight of stone. The fruit length was maximum (22.03 cm) in var. Sora followed 16.77cm in hybrid 7/15 where as the minimum fruit length of 4.33cm was recorded in var. Pacharasi. Fruit length shows transgressive segregation in either side of the parental limits and therefore, suggested a polygenic control of this character. Also, this character was influenced by environment to a considerable extent (Pandey, 2012). The maximum fruit thickness of 11.40cm was recorded in var. Dorgani Kavi followed by 9.87 in var. Amini whereas the minimum thickness was recorded (2,80cm) in Haldotta Appe. Fruit thickness showed polygenic control and was highly influenced by environmental factors. For breadth of fruit the genotypic contribution appeared to be comparatively lesser than in most of the traits. The peel colour and attractiveness are considered as important qualitative traits in mango, only 10% of the germplasm possess excellent attractiveness out of 400 germplasm screened. A high frequency of hybrids with red peel or burgundy blush can be recovered from crosses where one of the parents has an intense red blush (Brittell et al., 2004). The highest pulp recovery (89.67%) was recorded in var. Manoranjan followed by 88.20 % in var. Lahara whereas the lowest pulp recovery was recorded in var. Halasage. The highest TSS (ÚBrix) of 31.00 was recorded in var. Dattatreya local followed by 30.67 in var.K-0-7 where as the lowest was recorded in var. Halasage (5.5%). Sharma and Majumder (1988) stated that total Beta carotenoid pigments and T.S.S. content in these two hybrids exceeded the better parent Dashehari suggesting the gene action showing transgressive segregation for this trait. On the other hand, light yellow colour of pulp appeared dominant over orange yellow in the progenies of Alphonso x Neelum cross (Iyer, 1991). According to Prabhuram (1998), total soluble solids (T.S.S.) content in some hybrids transgressed either of the parents Amrapali and Sensation, which suggested a polygenic control for this trait and that it was influenced considerably by environmental factors.

Bretell et al., (2004) observed that many important fruit quality aspects such as fruit weight, fruit shape, ground skin colour, fruit width and pulp depth have high heritability, and can therefore be readily selected in a breeding programme. For non-ordered traits scored in discrete categories (blush colour, bloom, lenticel colour, embryo type and flavour), an estimate was made of data consistency from multiple scores for individual hybrids at different times and locations. Relatively high consistency value was recorded for fruit flavour, and in combinations involving Kensington Pride. The analysis of blush colour and fruit flavour in twelve families of hybrids has confirmed that these characters have a strong genetic component, the high frequency of hybrids with red or burgundy blush can be recovered from crosses where one parent has an intense red blush colour. Singh et al., (2004) observed wide magnitude of phenotypic coefficient of variation with high genetic advance for yield per plant and fruit weight in 31 chance seedlings of mango.

The analysis of variance for 6 quantitative traits showed significant differences among the 400 genotypes of mango which indicates the existence of genetic diversity. The maximum inter cluster (D2) value was obtained between cluster 10 and cluster 11. These clusters may be used for hybridization programme due to wide variability and possibility of transgressive sergeants. Estimates of phenotypic variance and genotypic variance had only a narrow difference for all six characters studied indicating that these characters are not much influenced by environmental factors. This also suggests the presence of sufficient genetic variability which can be exploited by adopting clonal selection or selection of chance seedlings.