Printer Friendly

MORPHOLOGICAL CHARACTERIZATION OF PAKISTANI MANGO (MangiferaindicaL.) VARIETIES USING PRINCIPAL COMPONENT ANALYSIS.

Byline: H. Mukhtar, M. Arif and F. Khan

ABSTRACT: The morphological data of mango varieties was provided by Shujabad mango research center. The point of present study was to evaluate diversity and grouping pattern of the mango varieties and traits. A total of 36 morphological characters were considered.Twenty fourqualitative traits fromtree(4), leaves (5), inflorescences (4), fruits and seeds (11) were included. Twelve quantitative traitsfrom leaf(2),flower and seed(10) were selected.Principal component analysis (PCA) was done by XLSTAT.Fifteen fruit traits and four leaf traits were used to index mango morphology.The first three principal components clarified 34%of variation and identified fruit/stone traits especially the fruit length, fruit thickness and fruit width, fiber length, total soluble salts (TSS), leaf colour, twisting of leaf blade as important traits that could be used to categorize mango varieties.

Fajri Kalan, Sindhri and Chaunsasufaidwere conspicuous due to highest fruit weight, stone thickness, fruit thickness and fruit length. The high morphological diversity within the mango varieties could prove to be helpful in their identification and categorization.

Key words: Principal component analysis, twisting, Quantitative Traits and Qualitative Traits.

INTRODUCTION

In Pakistan the area under Mango cultivation is 175 thousand hectares with a production of 1,784 thousand tonnes. Mango is the second major fruit crop of Pakistan which produces 8.5% of world's Mango(Govt. of Pakistan 2015). There is a dire need to characterize Pakistani mango varieties to compete in the global market. It is an important tool for improvement and breeding of mango in evolving new varieties. Morphological and bio-chemical description of mango is difficult and has never been addressed properly. (Rajwanaet.al.,2011).

Numerousmeasures for the documentation and characterization of mango varieties have been establishedon the basis of fruit morphological traits. Morphological characteristics are used as landmark to improve mango varieties(Jaramillo and Baena, 2000). Mango is a perennial crop having a long juvenile period;it is always very difficult to identify a cultivar at the initial stage of plant growth. If problem of proper identification issolved then it will make the mango improvement much easier. Morphological characters on the other hand, have great role for the identification of different cultivars (Joshietal., 2012). Fruit shape is a very prominent morphological character which influences the choice of consumers (Seyif and Rashidi, 2007). Fruit size and weight areimportant economical parameters too (Sinnott, 1932). A universally accepted procedure has been developed for characterization of mango varieties by the International Plant Genetic Resources Institute (IPGRI).

The IPGRI has a recognized and universal format of list of descriptors for mango that comprisesof the morphological traits of plant, flowers, leaves, seeds and fruits (Krishnapillai andWijeratnam, 2016).The morphological data of the mango varieties would be subjected to PCA in order to handle huge data properly. The target of present study was to provide a list of Morphological traits that could be used to identify mango varietiesat vegetative stage as well as to identify the mango fruit, depending upon the particular fruit characteristics.

MATERIALS AND METHODS

The morphological data of 46 Pakistani mango varieties was collected from the Shujabad Mango research center. Out of 36 morphological characters, 24 were qualitative while 12 were quantitative character. Qualitative characters included the Trees (tallness, vigour, branching and spreading), leaves (colour, shape, fragrance, tip shape and twisting of blade), inflorescence(length, branching, stalk colour and flower colour), fruits(beak, sinus, prescence of fiber, length of fiber, yield, peel colour and shape in cross section)traits.Quantitative characters were leaf (length and width), percent acidity, TSS, stone (length, thickness, width and average weight), fruit(length, width, thickness and average weight) traits,werecategorized and rated numerically (Table 1)

Table 1. Numerical Rating of the Qualitative Traits.

###TREE MORPHOLOGY

###1###TALLNESS

###Categories###Short###Medium###Tall

###Rating###1###2###3

###2###VIGOR

###Categories###Low###Moderate###Good###Well

###Rating###1###2###3###4

###3###BRANCHING

###Categories###Medium###Good###Well

###Rating###1###2###3

###4###SPREADING

###Categories###Semi###Spreading###Good###Well

###LEAF MORPHOLOGY

###1###COLOUR

###Categories###YG###LG###MG###DG

###Rating###1###2###3###4

###2###SHAPE

###Categories###Elliptical###Oblong

###Rating###1###2

###3###FRAGRANCE

###Categories###Absent###Present###Weak###Medium###Acute###Strong

###Rating###1###2###3###4###5###6

###4###TIP SHAPE

###Categories###Attenuate###Acuminate###Acute

###Rating###1###2###3

###5###TWISTING OF BLADE

###Categories###Absent###V.Weak###Weak###Slight###Medium###Present

###Rating###1###2###3###4###5###6

###Rating###1###2###3###4

###INFLORESCENCE MORPHOLOGY

1###LENGTH

###Categories###Short###Medium###Long###V.Long

###Rating###1###2###3###4

2###BRANCHING

###Categories###Less###Medium###Profuse

###Rating###1###2###3

3###STALK COLOUR

###Categories###GW###GY###LG###LP###MP###P###BP###DP###PR

###Rating###1###2###3###4###5###6###7###8###9

4###FLOWER COLOUR

###Categories###OW###GW###LG###G###LP###YP###P###MP###DP

###Rating###1###2###3###4###5###6###7###8###9

1###FRUIT MORPHOLOGY###FRUIT BEAK

###Categories###Absent###Almost###Weak to###Weak###Not###Medium to###Short to###Short###Very short###Medium but###Medium###Rounded###Present###Broadly###Prominent

###absent###absent###prominent###weak###medium###obtuse###pointed

###but pointed

###Rating###1###2###3###4###5###6###7###8###9###10###11###12###13###14###15

2###SINUS OF FRUIT

###Very###Slight to###Medium to

###Categories###Absent###Absent to weak###weak###Weak###Present###Slight###medium###strong###Medium###Strong

###Rating###1###2###3###4###5###6###7###8###9###10

3###SKIN THICKNESS

###Categories###Thin###Medium###Thick

###Rating###1###2###3

4###PRESCENCE OF FIBER

###Very###Medium to

###Categories###Absent###Few###Rare###rare###Low###Medium###high###High###Abundant

###Rating###1###2###3###4###5###6###7###8###9

5###LENGTH OF FIBER

###Medium

###Categories###Absent###Very short###Short###Medium###long###Long

###Rating###1###2###3###4###5###6

6###FRUIT YEILD

###Very

###Categories###Low###Medium###Good###good

###Rating###1###2###3###4

7###PEEL COLOUR

###Cat###L###P###Y###L###MY###C###C###G###Y###PIY###YGP###YR###PYL###Y###PYL###LEY###YRB###LY

###ego###Y###Y###E###Y###A###Y###G###PB###G###PT###PT###RB

###ries###Y###Y###L

###Rat###1###2###3###4###5###6###7###8###9###10###11###12###13###1###15###16###17###18

###ing###4

8###FRUIT SHAPE IN CROSS SECTION

###Cat###C.###O###OV.rou###OB.###OV.###O###OBL.###Oval###OBL###OBL.ob###OV.to###OB.to###BR.OB

###ego###V###nd###ovat###oblong###B###round###OBL.###. oval###long###OBL.###OBL.###L.

###ries###e###oval###round

###Rat###1###2###3###4###5###6###7###8###9###10###11###12###13

###ing

###FRUIT MORPHOLOGY

7###PEEL COLOR

###Categories###LEYRB###YGCP###LG###SG###BG###LGYRB###LGRB###MGRBB###GYLPB###GYPB###OYRPB###OYRB###BLGT###BYGG###PR

###Rating###19###20###21###22###23###24###25###26###27###28###29###30###31###32###33

8###Fruit Shape In Cross Section

###Categories###OV to###OBL.###OV.###OBL.###N. elliptic###BR.###OV. and

###IRR. Oval###flattened###Elliptic###elliptical###elliptic###N. elliptic

###Rating###14###15###16###17###18###19###20

Twelve quantitative characters were recorded includingleaf width, leaf length, fruit width, fruit length, fruit thickness, percent acidity, total soluble salt (TSS), average fruit weight, stone length, stone width, stone thickness and average stone weight. The length and width of leaf, fruit and stone was measured in centimeters. The weight of stone and fruit were measured by electric balance in grams.Total soluble salt was determined by using digital bench refractometer.Titrable aciditywas determined according to the protocol as described by Hortwitz(1960). Crude extract (10 ml) was taken in a beaker and titrated with 0.1NNaOH by two to three drops of phenolphthaleinwere added as an indicator. The percentage of citric acid was calculated as per following formula.

Titrable Acidity = volume of 1.2N NaOH x factor (0.0064) x 100 / Volume of sample used

The qualitative ratings and quantitative readings were used to prepare the Excel input file of XLSTAT (2012) to execute the principal component analysis(PCA). The PCA based on correlation matrix, was performed to evaluate diversity and grouping pattern of the germplasm and other traits evaluated. The data file was selected and the option of analysing data was selected through which principal component analysis was done. For observation / variable table the data input file was selected except the column of varieties. Pearson (n) PCA type was selected for observation labels the column of varieties was selected. The detailed PCA was generated including the summary statistics, Pearson correlation matrix, eigen value, eigen vectors, scree plot, factor loading, PCA biplot, percent contribution of observations and variables. The criterion of the significance of the eigenvalues, was used to select the statistically significant principal components Kaiser (1960).

Only those principal components (PCs)which haveeigenvalues >one were consideredas significant PCs.

RESULTS AND DISCUSSION

Thecorrelations of the first three significant principal componentsi.ePC1, PC2 and PC3 with the variableswere observed in mango varieties. Only these first three principal components that exhibited the eigenvalues >one were considered as significant in accordance to (table 1). In a study Hair et.al.,(1998) suggested that eigen value >one was significant The first PC accounted for 15.25% cumulativevariability, second 9.69% and third 9.12%. According to Gueiet. al., (2005), the seprincipal components were the most imperative in reflecting the variation patterns among the differentvarieties and the related characters were most important in differentiation.

Hence, the first three components were extracted to explain the variability which existed among the 46 mango varieties Principal component analysis, as a data reduction tool played an important role in identifying the traits, responsible for differentiation among mango varieties as has been reported by Marbohet al.,(2015). The first principal component (PC1) was highlyassociated with ten of the original variablesPC1 was highly subjective to characteristics of the fruit morphology.PC1 increased with increase in leaf fragrance (LF,0.298), peel colour (PEC,0.276), fiber length (FBRLN,0.451),yield (Y,0.214) leaf width (LFW,0.291), total soluble salts (TSS,0.331), average weight of stone(AVWT,0.377), fruit width (FRWD,0.376), fruit thickness(FRTH,0.555) and average weight of fruit(AVGWT,0.640). Furthermore, PC1 was highly correlated with AVGWT (0.640).

The second principal component (PC2) increased with the increase in five of the original variables including twisting of blade (TW,0.223), color of inflorescence stalk (SKCLR,0.383), color of flowers (FLCLR,0.419), stone length(STL,0.298), stone width (STW,0.262),and fruitlength (FRLG,0.324). PC2 was highly correlated withFLCLR(0.419). Third principal component (PC3) wascorrelated with three of the original variables.andincreased with the increase in leaf color (LC,0.163), leaf shape (LSH,0.256) and presence of fiber (FBR,0.233) while it was highly correlated with FBR (0.233).The conclusion of present study are in accordance with those of Malik et.al.,(2012) and Shrestha et.al.,(2012) who narrated that fruit weight, fruit length, fruit diameter, fruit rind thickness, TSS, leaf length and leaf width as important variables with the highest provenance to the variation contributed by the principal components.

The variables and observations were projected on a bipolt on the basis of first two PCs(figure 1 and 2). The net variation of the biplot is illustrated by PC1 and PC2. PC1 (15.26%) represented FRWD, FRTHK, AVGWT, PEC. According to Jintanawonget.al.,(1992)fruit color was the highlyattractivecharacter for commercial recognition of a variety. While PC2 (9.70) represented FRLG, STW, STL, STTHK. In first quadrant of biplot, the positive value of PC1 andPC2 indicated that, the varietiesi.eFajrikalan,Shindri and Chaunsasufaid showed diversityin the qualitative character of fruitsthe highest values of FRWD, FRTHK, AVGWT and FRLG, STW, STL, STTHK) were seen in second quadrant,the positive values for PC2 showed that LSH had highest values linked with the local varietiesi.eGhulam Muhammad wala, Zardaluand Golden ball.The varieties in second quadrant showed diversity in leaf shape and inflorescence color.

In a study Toilliet.al.,(2013) reported thatthe color of fresh leaf, tree tallness, type of leaf margins, circumference of stem, strength of fragrance and length of leaf blade had strong correlation.In third quadrant,the negative values for PC1 and PC2 showed the lowest values of FRSK, LC, TSS, TW in local varietiesi.eBadia manna seyed, Intikhab, Almas, Rohaan, Aalishan, HaiderShah wala,Neelum and Burma surkha. These varieties showed diversity in leaf and tree characters.In fourth quadrant,the negative value of PC2 showed lowest values for SKCAV, linked with Badiamunaseyed. The positive value of PC1 indicated highest values of LF, IFL, IF W, AVWT, FBRLN, FBRrelated with Yakta and ChaunsasummarBahisht. The varieties in the first and fourth quadrant were famous commercial varieties.Commercially renowned mangos must have low fiber content, higher values for fruit length, width, thickness and weight as has been reported by (Human and Rheeder, 2004).

Fruit traits were best for studying mango diversity (Galvez-Lopezetal.,2010). In future, mango varieties with superior fruit traits must be used in breeding efforts to produce improved hybrids and new cultivars.

According to PCA, peel colour, total soluble salt, average weight of stone average weight of fruit, fruit thickness, fruit length and fruit width were morphological traits used to differentiate between different mango varieties were found to bein line with the findings of. Marbohet.al.,(2015).

Table 1: Eigen values of the three significant Principal components

Values###PC1###PC2###PC3

Eigen values###5.493###3.491###3.284

Variability (%)###15.259###9.698###9.123

Cumulative %###15.259###24.957###34.080

Table 2: Correlation of the three Significant PCs with the original variables.

Variables###PC1###PC2###PC3

Leaf colour###0.018###0.071###0.163

Twisting of blade###0.044###0.223###0.152

Leaf fragrance###0.298###0.151###0.078

Leaf shape###0.240###0.046###0.256

Stalk colour###0.016###0.383###0.125

Flower colour###0.052###0.419###0.037

Peel colour###0.276###0.063###0.181

Prescence of Fiber###0.026###0.089###0.233

Fiber Length###0.451###0.044###0.165

Yeild###0.214###0.013###0.122

Leaf width###0.291###0.069###0.032

Total soluble salts###0.331###0.019###0.028

Avg. wt. per stone###0.377###0.025###0.224

Stone length###0.175###0.298###0.287

Stone width###0.082###0.262###0.019

Fruit length###0.033###0.324###0.288

Fruit width###0.376###0.037###0.062

Fruit thickness###0.555###0.040###0.070

Avg. wt. per fruit###0.640###0.024###0.133

Traits ABBREVATIONS###ABBREVATIONS

TT###tree talness###SPL###Shah pasand

TV###tree vigour###ZRL###Zardalu

TB###tree branching###HSWL###Haider shah wala

TS###tree spreading###SARL###Saroli

LC###leaf colour###BPL###Bagan pali

TW###twisting of blade###YKL###Yakta

TPS###shape of tip###CSBL###Chaunsa (SammarBahisht)

LF###leaf fragrance###SNDL###Sindhri

LSH###leaf shape###MLDL###Malda late

INFL###inflorescene length###KCHL###Kala chaunsa

INFB###inflorscence branching###CHSL###Chaunsasafaid

SKCLR###stalk colour###RETLL###Retaul late

FL CLR###flower colour###SANGL###Sanglakhi

FRSP###fruit shape in cross section###SOBL###Sobe de ting

PEC###colour of peel###POHL###Pohi lot

SZLN###size of lenticles###TAIL###Taimuria

FRSK###fruit skin###CHRPL###Chaunsa (Rampuri)

SK CAV###Stalk cavity###SABL###Saleh bhai

F BK###fruit beak###NEEL###Neelum

FR SI###fruit sinus###ZAFL###Zafran

SKTHK###skin thickness###BURL###Burma surkha

FBR###prescence of fiber###BMSL###Badiamunasyed

FBR LN###fiber length###ALL###Almas

Y###Yield###INTL###Intikhab

LF L###leaf length###LANL###Langra

LFW###leaf width###ANL###Anmole

AC###Acidity###GHUL###Ghulab-e-khas

TSS###total solid salt###BML###Bara mashi

AV WT###avgwtpr stone###DUL###Dusehri

ST L###stone lenth###ARL###Anwar retaul

ST W###stone width###FKL###Fajrikalan

ST THK###stone thickness###TPL###Totapari

FR LG###fruit length###LMWL###Langramayewala

FR WD###fruit width###PNL###Pan

FR THK###fruit thickness###LML###Lab-e-Mashooq

AVG WT###avgwtpr fruit###LAL###Lahotia

JWL###Joiyawala

KWL###Kachnaliwala

DKL###Dusehrikalan

GMWL###Ghulam Muhammad wala

GL###Golden

GBL###Golden ball

AML###Aminia

ALH###Aalishan

ROH###Rohan

HH###Hasaan

Conclusion: Principal Component Analysis reduced the dimensionality of the data. The first three significant Principal components were highly correlatedto the traits, such as fruit weight, fruit thickness, fruit width, Flower colour and prescence of fiber, twisting of leaf blade, leaf colour and leaf shape which considerably differentiated the 46 mango varietiesin the groups under study.

Acknowledgement: We are grateful to Mr. Abdul Ghaffar Grewal (Senior Horticulturist at Shujabad Mango Research Center) for providing the morphological data and Higher Education Commission of Pakistan for funds.

REFERENCES

Galvez-Lopez, D., M. Salvador-Figueroa, M.L. Adriano-Anaya and N. Mayek-Perez (2010).Morphological characterization of native mangos from Chiapas, Mexico.Subtrop.Plant. Sci. J.62: 18-26.

Govt. of Pakistan(2015).Agriculture Statistics of Pakistan.Ministry of food, agriculture and livestock.Economic, trade and investment wing. Islamabad, Pakistan.

Guei R.G., K.A. Sanni, F.J. Abamu and I. Fawole (2005).Genetic diversity of rice (Oryza sativa L.).Agron.Afri. 5: 17- 28

Hair, J., R. Anderson, R. Tatham and W. Black (1998). Multivariate Data Analysis, 5th Edition. Prentice-Hall Inc., New Jersey

Horwitz, W. (1960).Official and Tentative Methods of Analysis, 9th Edition.Association of Official Agricultural Chemists, Washington, D. C., pp.314-320.

Human, C. F. and S. Rheeder (2004). Mango breeding: results and successes. Acta.Horti. 645: 331-335.

Jaramillo, S and M. Baena (2000).Material de apoyo a la capacitacionenconservacion ex situ de recursosfitogenUticos.InstitutoInternacional de RecursosFitogenUticos, Cali (Colombia).

Jintanawong, S., H. Hiranpradit and S. Chandraparnik(1992).Quality standardization of mango (MangiferaindicaL.).Acta.Hortic.321: 705-707.

Joshi, R., M. Kundu and C.P. Singh (2012). Morphological characters: Efficient tool for identification on different mango cultivars.Environ.Eco.31: 385-388.

Kaiser, H.F. (1960). The application of electronic computers to factor analysis. Educ. Psychol. Meas. 20: 141-151.

Krishnapillai, N and R.S.W. Wijeratnam (2016). Morphometric analysis of mango varieties in Sri Lanka Aust. J. Crop Sci.10(6): 784-792.

Malik, S. K., M. R. Rohini, S. Kumar, R. Choudhary, D. Pal and R. Chaudhury, (2012).Assessment of genetic diversity in sweet orange [Citrus sinensis (L.)Osbeck] cultivars of India using morphological and RAPD markers.AGR. RES. 1: 317-24.

Marboh, E.S., A.K. Singh, A.K Dubey and J. Prakash (2015).Analysis of genetic variability among citrus (Citrus spp) genotypes using morphological traits.Indian. J. Agr. Sci.85(2). 203-11

Rajwana, I.A., I.A. Khan, A.U. Malik, B.A. Saleem, A.S. Khan, K. Ziaf, R. Anwar and M. Amin (2011). Morphological and biochemical markers for varietal characterization and quality assessment of potential indigenous mango (Mangiferaindica) germplasm.Int. J. Agric. Biol. 13: 151-158.

Seyfi, K, and M. Rashidi (2007).Effect of drip irrigation and plastic mulch on crop yield and yield components of cantaloupe.Int. J. Agric. Biol. 2: 247-249.

Shrestha, R. L., D. D. Dhakal, D. M. Gautum, K.P. Paudyal and S. Shrestha (2012). Study of fruit diversity and selection of elite acid lime (Citrus aurantifoliaSwingle) genotypes in Nepal. Am. J. Plant. Sci. 3: 1098-1104

Sinnott, E.W., (1932). Shape changes during fruit development in Cucurbita and their importance in the study of shape inheritance.Am. Nat. 301-309.

Toilli, M.E., F.K. Rimberia, A.B. Nyende, U. Mutwiwa, J. Kaluli, and D. Sila (2013). Assessing morphological diversity of mango germplasm from the upper Athi river (uar) region of eastern kenya. Inscientific conference proceedings.602-612.
COPYRIGHT 2016 Asianet-Pakistan
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Pakistan Journal of Science
Geographic Code:9PAKI
Date:Sep 30, 2016
Words:3633
Previous Article:AN ECONOMETRIC ESTIMATION OF POST-HARVEST LOSSES OF CUT-FLOWERS IN PUNJAB, PAKISTAN.
Next Article:GEOGRAPHIC INFORMATION SYSTEM (GIS) BASED HIGHWAY ASSET MANAGEMENT SYSTEM FOR MOTORWAYS: CASE STUDY OF MAJOR PAKISTAN'S MOTORWAYS.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters