MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

RESEARCH ARTICLE

Received: 11 Aug 2024

Revised: 20 Aug 2024

Accepted: 03 Sep 2024

*Corresponding author's e-mail: suvendukumarroy@gmail.com

Multivariate Studies On Diverse Rice (Oryza Sativa L.)

Genotypes For Agro-Morphological Characters Under Terai

Region Of West Bengal

Umamaheswar N¹, Kundu A², Roy S K^1*, Mandal R³, Sen S⁴, Hijam L¹, Chakraborty M¹, Das B⁵,

Barman R⁶, Vishnupriya S¹, Thapa B⁷, Maying B⁸ and Rout S⁹

1Department of Genetics and Plant Breeding, Uttar Banga Krishi Viswavidyalaya, Cooch Behar, West Bengal– 736165, India

2AICRN on Potential Crops, Uttar Banga Krishi Viswavidyalaya, Pundibari, Cooch Behar, West Bengal– 736165, India

3Regional Research Station, Terai Zone, Uttar Banga Krishi Viswavidyalaya, Cooch Behar, West Bengal– 736165, India

4AINP on Jute and Allied Fibres, Uttar Banga Krishi Vishwavidyalaya, Cooch Behar, West Bengal– 736165, India

5Department of Genetics and Plant Breeding, Uttar Banga Krishi Vishwavidyalaya, College of Agriculture (Extended Campus), Majhian,

Dakshin Dinajpur, West Bengal– 733133, India

6Regional Research Station (OAZ), Uttar Banga Krishi Vishwavidyalaya, Majhian, Dakshin Dinajpur, West Bengal– 733133, India

7Regional Research Station (Hill Zone), Uttar Banga Krishi Viswavidyalaya, Kalimpong, West Bengal– 734301, India

8College of Agriculture, Central Agricultural University, Pasighat Arunachal Pradesh– 791102, India

9Department of Genetics and Plant Breeding, Centurion University of Technology and Management, Paralakhemundi, Odisha– 761211,

India

ABSTRACT

A study was conducted to analyze trait variations among rice genotypes

in the Terai region of West Bengal and to select high-performing genotypes

based on specific characteristics. The study was conducted during the Kharif

(Aman) seasons of 2019 and 2020, focussing on 42 rice genotypes at Uttar

Banga Krishi Viswavidyalaya, Cooch Behar, West Bengal. The Mahalanobis

D² analysis revealed four distinct clusters, with significant variation observed

for grain length, plant height, and grain yield per plant. Additionally, the

maximum Mahalanobis D² distance was observed for Dudeswar, Baramshall,

and Khara. The Principal Component Analysis identified spikelet fertility,

grain yield per plant, filled grains per plant, and test weight as principal

discriminatory characteristics, with Dudeswar exhibiting the highest index

score of 2.49. It was followed by Geetanjali with a score of 1.92, according

to the Smith selection index. The significant characters identified through the

D² analysis and PCA, such as grain length, plant height, spikelet fertility and

others played a crucial role in revealing the diversity among the genotypes.

The maximum D² distances for specific genotypes, coupled with high index

scores, suggested a strong association with discriminatory characteristics

identified through the Smith selection index, emphasizing their importance in

genotypic classification and selection.

Keywords: D² statistic; Genetic diversity; PCA; Rice; Smith index; Terai region

INTRODUCTION

Rice serves as a staple food for about more

than three billion people (Zeigler and Adam, 2008).

In order to maintain self-sufficiency, it is essential

to develop new varieties or hybrids with high yield

potential and resilience in challenging conditions

(Papademetriou et al., 2000). There is a pressing

requirement for novel rice varieties with greater

genetic diversity, high yield, resilience to biotic and

abiotic stresses, and superior grain quality to meet the

needs of future consumers.

Genetic diversity, which refers to heritable variation

within and between populations,

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

plays a crucial role in determining breeding strategies

as it is a pre-requisite for the initiation of any crop-

breeding programme. It is crucial to comprehend the

available variability within the population being studied

(Nachimuthu et al., 2014). Due to their adaptability

to diverse environments, traditional rice varieties can

effectively cope with changing climates and serve as

repositories of genes resistant to pests and diseases.

Thus, employing traditional rice varieties would be the

optimal and sustainable approach to create climate-

smart rice varieties with resistance to significant

biotic and abiotic stresses. Assessing genetic diversity

is a critical element in developing effective crop

improvement breeding programs. This evaluation

also aids in establishing genetic relationships and

estimating genetic variability during germplasm

collection, by doing so, it helps to ensure that parental

combinations

segregating

populations

have

increased genetic variability, leading to the creation of

new recombination for the selection and incorporation

desirable

genes

into

superior

germplasm

(Thompson et al., 1998; Islam et al., 2012).

Multivariate analysis is one methodology for

measuring genetic distance estimates for a population

since it is important to recognise the useful variability

present in the population (Nachimuthu et al., 2014). It

is commonly used to summarize and characterize the

intrinsic diversity among genotypes.

Mahalanobis D² analysis of quantitative characters

is a powerful tool for measuring genetic divergence

among the material selected even from the same

geographic region as reported by Mahalanobis (1936)

followed by Rao (1952). A high level of genetic diversity

helps the plant breeder in selecting genotypes having

a desired specific character or a combination of

characters. D² statistic remains as the most effective

method for quantifying the degree of genetic diversity

among genotypes. Analyzing D² values, breeders can

identify groups with similar genetic characteristics and

also assess the genetic diversity within and between

groups or clusters, which is crucial for accurately

selecting parental lines, leading to more effective

exploration of heterosis, as emphasized by Murty and

Arunachalam (1966).

As variation occurs often in plants for yield and

yield-related characteristics (Maji et al., 2012),

Principal Component Analysis (PCA) identifies patterns

and reduces redundancy in datasets. According

to Anderson (1972) and Morrison (1978), PCA is

a powerful and well-known multivariate statistical

technique used for dimension reduction. It determines

the smallest number of components that can explain

the greatest amount of variability out of the total

variability. Principal components (PCs) are commonly

derived from either a covariance matrix or a correlation

matrix. When variables are measured in different

units, scale effects can change the composition of

derived components, emphasizing the importance of

standardizing the variables in these situations. The

primary advantage of PCA lies in its ability to quantify

the importance of each dimension in capturing the

variability of a dataset, as highlighted by Shoba et al.,

2019.

In many breeding schemes, genotype selection is

carried out entirely based on grain yield, neglecting other

yield-determining characters in commercial breeding

programmes. The application of selection indices,

such as those initially suggested by Smith (1936)

and Hazel (1943) known as the classical index, in a

single study is an effective method for simultaneously

and effectively including multiple characters. Smith

(1936) contended that the genetic value could not

be accurately assessed through individual characters

alone, but rather it could be more effectively estimated

through a linear combination of observable phenotypic

values. Therefore, the application of a selection index

would optimize the genetic improvement for intricate

characteristics such as grain yield.

The purpose of the present study is to determine

genetic diversity among rice genotypes in terai region

of West Bengal based on D² analysis, PCA and Smith

selection Index. Best combinations of these yield

attributes and genotypes can be used as selection

criteria for creating high yielding rice genotypes based

on agro-morphological characters for future crop

improvement programmes.

MATERIALS AND METHODS

The field experiment was conducted at the

experimental farm of Uttar Banga Krishi Viswavidyalaya,

Pundibari, Cooch Behar, West Bengal, during Kharif,

2019 and 2020. The location receives high annual

rainfall (3200 mm). Moreover, there is a wide

distribution of rainfall coupled with high temperatures.

It is located at geographical coordinates at 26°34′19”

N latitude, 88°08′51” E longitude at an elevation of

113 meters above mean sea level (MSL). During the

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

crop growth period, the total amount of rainfall from

June to November, 2019 was 2100 mm and during the

second season i.e. June to November, 2020 was 3000

mm. The mean annual temperature is 30 °C and it

has sandy loam soil type. A total of forty-two diverse

rice genotypes were used for the study (Table 1.). The

genotypes collected from two states, namely West

Bengal and Andhra Pradesh, with wider adaptability in

areas of their recommendation were used. The field

trial was laid out in a Randomized Complete Block

Design (RCBD) with two replications with a spacing

of 20 cm × 15 cm row to row and plant to plant,

respectively. The recommended package of practices

was followed during the crop season to raise a good

crop in the main field. Twenty-eight days seedlings

were transplanted in the main field. The observations

were recorded on randomly selected five competitive

plants in the inner middle rows of each plot in all the

two replications for nine morphological characters

namely plant height (cm)- [PH], panicles plant^-1- [PPP],

filled grains spikelet^-1- [FGPP], spikelet fertility (%)-[SF],

grain length (mm)- [GL], grain breadth (mm)- [GB],

grain length: breadth ratio (mm)- [LBR], test weight (g)-

[TW] and grain yield plant^-1 (g)- [GYP].

The chi- square test indicated that the rice

genotypes

were

divergent

and

therefore

the

Mahalanobis D² (Mahalanobis 1928, 1936) analysis

was carried out. GENSTAT software was used for D²

analysis.

In order to classify the patterns of variation,

principal component analysis (PCA) was performed.

Those PCs with Eigen values greater than one were

selected as proposed by Jeffers (1967). Correlations

between the original characters and the respective

Principal Components (PCs) were calculated. The

mean data of the characters were used to perform

principal component analysis (PCA) using software

FactoMiner package (Lê et al., 2008) on a matrix of

nine morphological characters followed by visualization

by FactoExtra package in RStudio. GraphPad Prism 7

software (GraphPad Software Version 9.0, La Jolla

California USA) was used for the visualization plot of

proportion of variance.

Table 1. Details of the rice genotypes used in the experiment

Sl. No.

Code

Name of the Genotype

Sl. No.

Code

Name of the Genotype

Balam

22)

G22

Khalia Eulo

Baramshall

23)

G23

Kalonunia

Baskathi

24)

G24

Khara

Basmati

25)

G25

Lal Badsahbhog

Kharadhan

26)

G26

Patnai

Chamarmani

27)

G27

Sagar Sugandhi

Chamatkar

28)

G28

Tulsi Mukul

Dehradun Gandheswari

29)

G29

Nonabokra

Dudeswar

30)

G30

BPT 2295

G10

Gopalbhog

31)

G31

BPT 5204

G11

Indulshall

32)

G32

CR 910

G12

Jhara

33)

G33

Geetanjali

G13

JP 90

34)

G34

NL 44

G14

JP 120

35)

G35

NL 46

G15

Zugal

36)

G36

NLR 0106

G16

Kakri

37)

G37

NLR 3242

G17

Kalavati

38)

G38

MTU 1061

G18

Kalo Aush

39)

G39

NLR 20084

G19

Kamal

40)

G40

NLR 40058

G20

Kanakchur

41)

G41

NLR 145

G21

Kerala Sundari

42)

G42

BPT 2411

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

The weights obtained from the eigen values were

utilized in the construction of selection indices,

following the approach outlined by Smith (1936). The

weights assigned to each trait was determined using

the PCA loading values, with a scale ranging from 1 to

10. Notably, grain yield received the highest weightage

of 10 within this framework. Selection index for the

recorded data was computed using the software PB

Tools v. 1.4 (PB Tools, 2014).

RESULTS AND DISCUSSION

Mahalanobis D² statistic measured genetic

divergence by clustering the genotypes into four

clusters (Table 2.) based on nine morphological

characters. The significant difference indicates

the appropriateness of the use of D² statistics for

clustering the genotypes into different groups. In this

regard, Shanmugam et al. (2023), Singh et al. (2020),

Pavani et al. (2018), Singh et al. (2017), Karuppaiyan

et al. (2013), and Shanmugasundaram et al. (2000)

identified four clusters for various numbers of

genotypes in their study involving rice genotypes.

The genotypes were grouped into four clusters

and they contained a variable number of genotypes.

Cluster III contained the maximum number of 28 rice

genotypes (Balam, Baramshall, Baskathi, Basmati,

Kharadhan, Chamarmani, Chamatkar, Dehradun

Gandheswari, Dudeswar, Gopalbhog, Indulshall, Jhara,

JP 90, JP 120, Zugal, Kakri, Kalavati, Kalo Aush, Kamal,

Kanakchur, Kerala Sundari, Khalia Eulo, Kalonunia,

Khara, Lal Badsahbhog, Patnai, Sagar Sugandh and

Tulsi Mukul) followed by 10 genotypes in Cluster IV

(Nonabokra, BPT 2295, BPT 5204, Geetanjali, NLR

0106, NLR 3242, MTU 1061, NLR 20084, NLR 40058

and BPT 2411), two genotypes in Cluster I (NL 44 and

NL 46) as well as Cluster II (CR 910 and NLR 145).

The clustering pattern of genotypes showed that

the genotypes of different origins, collected from Uttar

Banga Krishi Viswavidyalaya (Majhian Campus and

Pundibari Campus) and Acharya N.G. Ranga Agricultural

University (Bapatla Campus and Tirupati Campus)

were clubbed in one cluster, whereas the genotypes

belonging to same origin were grouped in different

clusters indicating that the geographical distribution

need not always be considered to be the sole criterion

for genetic diversity. The genotypes included in the

same cluster were considered genetically similar

with respect to the aggregate effect of the characters

examined. From the pattern of clustering, it could

be inferred that sufficient divergence was present to

enable the formation of individual clusters.

In the present investigation, the inter-cluster and

intra-cluster distance was estimated among the nine

Table 2. Grouping of 42 rice genotypes into different clusters on the basis of D² analysis for nine

morphological characters (Combined over 2 years)

Cluster

No.

Total no. of

germplasm

accessions

Source

Name of germplasm accessions

(G34) and (G35)

(G32) and (G41)

III

(G1), (G2), (G3), (G4), (G5), (G6), (G7), (G8), (G9), (G10),

(G11), (G12), (G13), (G14), (G15), (G16), (G17), (G18),

(G19), (G20), (G21), (G22), (G23), (G24), (G25), (G26),

(G27) and (G28)

A1+A3+B

(G29), (G30), (G31), (G33), (G36), (G37), (G38), (G39),

(G40) and (G42)

A1- Acharya N.G. Ranga Agricultural University (Tirupati Campus), A2- Acharya N.G. Ranga Agricultural

University (Agricultural Research Station, Nellore), A3- Acharya N.G. Ranga Agricultural University (Bapatla), B-

Uttar Banga Krishi Viswavidyalaya (Majhian Campus and Pundibari Campus).

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

characters (Table 3). Members of cluster IV were more

dissimilar in morphological features and performance

than those of other clusters, as indicated by the

highest value of the intra-cluster distance (3222.974).

The maximum intra cluster distance was recorded in

cluster VI (3222.974) followed by cluster III (1978.974),

cluster II (145.187), and cluster I (100.023), indicating

greater genetic divergence between the genotypes in

these clusters. The maximum intra cluster distance in

cluster VI was because of wide genetic diversity among

its genotypes.

The maximum inter cluster distance was observed

between cluster III and I (10262.474) followed

by cluster IV and III (7248.518), cluster III and II

(6592.362), cluster IV and IV (3222.974), cluster IV

and I (2798.984), cluster II and I (2403.396), cluster

IV and II (2243.974) and cluster III and III (1978.974).

So, genotypes can be selected as parents between

cluster III and I because of maximum inter cluster

genetic distance.

The

larger

inter-cluster

distances

indicated

more diversity among the rice genotypes grouped

in different clusters with respect to the characters

considered for hybridization programme in rice. The

estimates of average intra and inter cluster distance

value of four clusters revealed that the genotypes

belonging to the same cluster (intra cluster) have less

genetic divergence as compared to genetic diversity

between the genotypes of different clusters (inter

cluster). When crossing is done between genotypes

belonging to the same cluster, no transgressive

segregants are expected from such combinations

because same cluster genotypes display the lowest

degree of divergence from one another. Therefore, a

hybridization programme should always be formulated

in such a way that parents belonging to different

clusters with maximum genetic distance can be

utilized to obtain desirable transgressive segregants.

The cluster means for various characters are

presented in Table 4 which showed that each cluster

had its own uniqueness that separated it from the

other clusters. Cluster I was characterized by highest

means for filled grains spikelet^-1 (1.965) and lowest

for grain length: breadth ratio (0.460). Cluster II,

consisting of only two genotypes, was characterized

by the highest value for plant height (2.023) and the

lowest for grain breadth (0.412). Cluster III had the

highest value for plant height (2.178) and lowest for

grain breadth (0.543). Cluster IV had the lowest value

for grain breadth (0.468) and highest for filled grains

spikelet^-1 (2.030). Cluster mean analysis indicated the

extent of diversity among different clusters, which can

be of practical value in rice breeding.

Table 3. Average intra (diagonal) and inter-cluster (off-diagonal) D² values of 42 rice genotypes

(Combined over 2 years)

Cluster

III

100.02

2403.40

10262.47

2798.98

145.19

6592.36

2243.97

III

1978.97

7248.52

3222.97

Table 4. Cluster means for nine characters of rice genotypes (Combined over 2 years)

Cluster

PPP

FGPP

LBR

GYP

Total

1.81

0.91

1.97

1.93

0.81

0.60

0.46

1.29

1.21

10.97

2.02

0.92

2.00

1.92

0.74

0.41

0.58

1.16

1.27

11.03

III

2.18

1.07

2.04

1.91

0.87

0.54

0.56

1.19

1.24

11.60

1.95

1.05

2.03

1.93

0.81

0.47

0.59

1.22

1.32

11.37

Population Mean

2.10

1.05

2.03

1.92

0.85

0.52

0.56

1.20

1.26

11.49

PC (%)

27.18

2.67

2.09

0.00

31.48

2.32

0.12

8.48

25.67

100.01

PH - Plant height (cm), PPP - Panicles plant^-1, FGPP - Filled grains spikelet^-1, SF - Spikelet fertility (%), GL - Grain

length (mm), GB - Grain breadth (mm), LBR - Grain length: breadth ratio (mm), TW - Test weight

(g) and GYP - Grain yield plant^-1 (g), PC – percent contribution.

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

The cluster means ranged from 1.81 (Cluster I) to

2.18 (Cluster III) for plant height (cm); 0.91 (Cluster I)

to 1.07 (Cluster III) for panicles plant^-1; 1.97 (Cluster

I) to 2.04 (Cluster III) for filled grains spikelet^-1; 1.91

(Cluster III ) to 1.93 (Cluster I and Cluster IV) for spikelet

fertility; 0.74 (Cluster II) to 0.81 (Cluster I) for grain

length; 0.41 (Cluster II) to 0.60 (Cluster I) for grain

breadth; 0.46 (Cluster I) to 0.59 (Cluster IV) for grain

length: breadth ratio; 1.16 (Cluster II) to 1.29 (Cluster

I) for test weight; 1.21 (Cluster I) to 1.32 (Cluster IV) for

grain yield plant^-1 in the present study.

This implies that the grain yield plant^-1 showed a

consistent range across all the clusters like cluster IV

(1.322), cluster II (1.270), cluster III (1.241), and cluster

I (1.213). This uniformity in the range of grain yield

across the clusters suggest the presence of divergent

genotypes within each cluster, indicating a broad

spectrum of genetic diversity within the population.

Thus, the clustering pattern could be utilized in

selection of parents for crossing and deciding the best

cross combinations, which may generate the highest

possible variability for various characters.

The contribution to divergence (Figure 1..) has been

the maximum by grain length (31.48 %), followed by

plant height (27.18 %) and grain yield plant^-1 (25.67).

In contrast, remaining characters contributed very

little towards genetic divergence i.e., test weight (8.48

%), panicles plant^-1 (2.67), grain breadth (2.32), filled

grain panicle^-1 (2.09) and grain length: breadth ratio

(0.12). The interesting point in percentage contribution

is that spikelet fertility (0.00) showed no contribution

towards genetic divergence.

Using D² statistics, all the rice genotypes could be

distinguished from one another considering all the

characters collectively. The results suggested that

some genotypes performed better than others used in

this investigation over the years. A suitable crossing

programme may be justifiable to exploit genetic

divergence in characters such as grain length, plant

height, and grain yield plant^-1. This can be based on

the percentage contribution of these characters and

the identification of rice genotypes, namely Dudeswar,

Baramshall, and Khara, through intra-cluster distance,

inter-cluster distance, and the D² distance between

individual genotypes.

PCA was conducted using yield and yield attributes

on diverse rice genotypes. It has been proposed that

the genetic variation among the rice accessions, as

indicated by their agro-morphological characters,

should be considered to advance the improvement

programme (Shahidullah et al., 2009). Out of the nine,

only four principal components possessed more value

than 1.0 eigen value and showed about 75% of total

variability among the characters studied. Summary

of the contribution of the principal components to

variability are given in Table 5.

To choose the variable parents, the main

components with multiple eigen values demonstrated

greater variation among the rice genotypes. So,

these four principal components were given more

importance. The PC1 shared high proportion of total

variation of 25.86 % and the rest of the three principal

components viz., PC2, PC3 and PC4 contributed

18.48%, 17.16% and 15.16% of the total variance

respectively. Similar findings were also reported by Sar

and Kole (2023), Mushtaq el al. (2023), Shanmugam

et al. (2023), Pushpa et al. (2022), Akhtar et al.

(2022), Dhakal et al. (2020), Tuhina-Khatun (2015)

and Shoba et al. (2010).

Scree plot (Figure 2.) was plotted for illustrating the

percentage of variance to each principal component,

with PC1 exhibiting the highest variation (25.86%)

than the other principal components. Thus, selecting

characteristics based on PC1 would be effective.

Characters like grain length: breadth ratio, grain

length, and grain breadth had relatively longer vector

suggesting that the characters had relatively larger

effects on grain yield (Figure 3.). On the contrary,

Agric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

0.03

0.05

0.10

0.01

-0.10

0.03

0.06

0.10

nt height (cm), PPP - Panicles plant^-1, FGPP - Filled grains spikelet^-1, SF - Spikelet fertility (%), GL -

ngth (mm), GB - Grain breadth (mm), LBR - Grain length: breadth ratio (mm), TW - Test weight (g)

- Grain yield plant^-1 (g), MSI- Mean of Selected Individuals, MAI- Mean of all individuals, SDi:

n differential, EGG: Expected genetic gain.

Figure 1. The percent contribution of nine characters for combined over the years

Figure 1. The percent contribution of nine

characters for combined over the years

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

between two lines is acute angle (<90°) on PC1 but

grain breadth shows extreme negative correlation with

grain yield plant^-1 on PC1 as indicated by obtuse angle

(>90°). Characters namely grain length, plant height

and panicles plant^-1 displayed very close correlation

among themselves along with negligible positive

correlation while test weight had minimal negative

correlation with grain yield plant^-1 because they form

an acute angle (<90°) on PC2.

Percent contribution of variables on principal

components is given in Table 6. The findings

demonstrated that in PC1, spikelet fertility (0.460)

and grain yield plant^-1 (0.407) had the highest positive

values in PC1. While in PC2, grain length: breadth ratio

had the highest positive score (0.606), followed by

filled grains spikelet^-1 (0.386), grain length and plant

height (0.377). Surprisingly, only grain length: breadth

ratio had the positive value in PC3. In PC4, filled grains

spikelet^-1 (0.560)had highest positive value followed by

panicles plant^-1 (0.429). From these findings, specific

selection strategies can be formulated to enhance

characters such as grain yield. The agro-morphological

diversity and variability among rice genotypes are

pivotal for crop improvement (Seetharam et al., 2009).

Biplot is the merger of PCA score plot and loading

plot. The covariate effect of biplot based on correlation

among the characters is presented in Figure 5 and the

same explained 44% of the total variation and thus can

be considered as a good approximation, as far as the

effect of characters on yield as well as their similarities

were concerned. Because the first two principal

components (PC1 and PC2) contain the majority of

Table 6. Four principal components along with their factor loadings for combined over the years

Characters

Combined over the years

PC1

PC2

PC3

PC4

Plant height (cm)

-0.371

0.377

-0.151

0.247

Panicles plant^-1

-0.180

0.287

-0.059

0.429

Filled grains panicle^-1

0.232

0.386

-0.050

0.560

Spikelet fertility (%)

0.460

0.084

-0.038

0.243

Grain length (mm)

-0.424

0.377

-0.185

-0.280

Grain breadth (mm)

-0.418

-0.267

-0.469

0.155

Grain L/B ratio

0.040

0.606

0.321

-0.405

Test weight (g)

0.208

0.137

-0.615

-0.325

Grain yield (g plant^-1)

0.407

0.144

-0.483

-0.106

Madras Agric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

Figure 3. Variables of PCA

Figure 4. Biplot of PCA

Figure 4. Biplot of PCA

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

Table 7. Values of smith selection index for all the rice genotypes (Combined over the years)

Genotype

PPP

FGPP

LBR

GYP

Smith

index

1. Balam

2.22

1.26

2.01

1.93

0.90

0.60

0.53

1.25

1.29

0.63

2. Baramshall

2.12

0.91

2.28

1.94

0.90

0.51

0.62

1.16

1.19

1.36

3. Baskathi

2.22

1.10

2.09

1.94

0.95

0.49

0.69

1.16

1.19

1.29

4. Basmati

2.22

1.07

1.92

1.91

0.94

0.51

0.65

1.35

1.37

1.09

5. Kharadhan

2.16

1.17

2.16

1.93

0.86

0.53

0.56

1.13

1.37

1.38

6. Chamarmani

2.20

0.92

1.86

1.92

0.91

0.61

0.52

1.13

1.18

-1.69

7. Chamatkar

2.18

0.95

2.06

1.89

0.89

0.46

0.67

1.16

1.20

0.07

8. Dehradun Gandheswari

2.20

1.11

2.02

1.91

0.97

0.59

0.60

1.11

1.15

-0.34

9. Dudeswar

2.18

1.16

2.28

1.93

0.85

0.49

0.59

1.17

1.39

2.49

10. Gopalbhog

2.18

0.89

2.06

1.91

0.84

0.59

0.48

1.27

1.42

0.35

11. Indulshall

2.11

1.13

1.83

1.86

0.89

0.60

0.52

1.2

1.18

-2.32

12. Jhara

2.19

1.31

2.07

1.90

0.92

0.61

0.52

1.17

-0.30

13. JP 90

2.19

1.09

2.06

1.90

0.84

0.45

0.65

1.13

1.14

-0.32

14. JP 120

2.18

1.14

1.93

1.88

0.9

0.59

0.53

1.26

1.22

-0.99

15. Zugal

2.20

0.94

2.10

1.94

0.82

0.59

0.47

1.34

1.38

0.85

16. Kakri

2.26

1.08

2.29

1.94

0.85

0.51

0.58

1.20

1.22

1.59

17. Kalavati

2.19

0.89

1.92

1.93

0.83

0.53

0.54

1.26

-0.51

18. Kalo Aush

2.15

1.17

1.91

1.87

0.86

0.54

1.11

1.14

-2.05

19. Kamal

2.15

1.14

2.01

1.90

0.90

0.49

0.63

1.17

1.21

0.12

20. Kanakchur

2.10

1.03

1.98

1.93

0.76

0.60

0.42

1.19

1.39

-0.56

21. Kerala Sundari

2.17

1.04

2.17

1.92

0.84

0.61

0.47

1.14

1.15

-0.77

22. Khalia Eulo

2.18

0.91

2.00

1.89

0.94

0.52

0.64

1.18

1.16

-0.66

23. Kalonunia

2.20

1.29

2.08

1.94

0.79

0.47

0.55

1.17

0.26

24. Khara

2.20

1.10

2.07

1.90

0.85

0.50

0.58

1.28

1.30

0.62

25. Lal Badsahbhog

2.12

1.04

2.04

1.88

0.73

0.51

0.48

1.09

1.30

-1.32

26. Patnai

2.23

0.93

1.99

1.90

0.95

0.52

0.65

1.36

1.34

1.01

27. Sagar Sugandhi

2.13

0.95

1.85

1.90

0.95

0.56

0.61

1.16

1.17

-1.36

28. Tulsi Mukul

2.16

1.12

2.07

1.90

0.78

0.60

0.43

0.97

1.07

-2.41

29. Nonabokra

1.87

0.86

1.90

1.92

0.85

0.50

0.58

1.01

1.16

-1.93

30. BPT 2295

1.98

0.97

2.19

1.95

0.79

0.45

0.58

1.23

1.40

1.89

31. BPT 5204

1.93

1.01

2.02

1.95

0.79

0.45

0.59

1.18

1.34

0.75

32. CR 910

2.04

0.91

1.97

1.92

0.73

0.41

0.58

1.13

1.28

-0.63

33. Geetanjali

1.93

1.13

2.04

1.91

1.02

0.58

0.64

1.26

1.42

1.92

34. NL 44

1.82

0.86

1.95

1.92

0.82

0.59

0.46

1.29

1.22

-1.51

35. NL 46

1.80

0.96

1.98

1.93

0.8

0.59

0.45

1.29

1.21

-1.19

36. NLR 0106

1.91

1.03

1.98

1.94

0.72

0.38

0.61

1.20

1.42

1.03

37. NLR 3242

2.02

1.10

2.09

1.92

0.79

0.51

0.52

1.37

1.42

1.30

38. MTU 1061

2.04

1.07

1.99

1.92

0.84

0.52

0.55

1.28

1.31

0.25

39. NLR 20084

1.96

1.15

2.11

1.91

0.78

0.43

0.60

1.32

1.43

1.74

40. NLR 40058

1.85

1.06

1.99

1.94

0.78

0.43

0.60

1.17

1.16

-0.36

41. NLR 145

2.01

0.94

2.02

1.92

0.75

0.42

0.59

1.19

1.25

-0.18

42. BPT 2411

2.03

1.12

2.02

1.91

0.79

0.44

0.60

1.14

1.17

-0.57

PH - Plant height (cm), PPP - Panicles plant^-1, FGPP - Filled grains spikelet^-1, SF - Spikelet fertility (%), GL - Grain

length (mm), GB - Grain breadth (mm), LBR - Grain length: breadth ratio (mm), TW - Test weight (g) and GYP - Grain

yield plant^-1 (g).

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

Table 8. Ranking of the 42 rice genotypes based on the Smith index for each year and combined

over the years basis

Genotype

2019

2020

Combined over the

years

S.I.

Rank

S.I.

Rank

S.I.

Rank

1. Balam

0.76

0.68

0.63

2. Baramshall

1.34

0.64

1.36

3. Baskathi

0.80

0.54

1.29

4. Basmati

0.36

0.70

1.09

5. Kharadhan

1.07

0.30

1.38

6. Chamarmani

-1.39

0.07

-1.69

7. Chamatkar

-0.70

-0.03

0.07

8. Dehradun Gandheswari

-0.77

0.47

-0.34

9. Dudeswar

2.40

0.37

2.49

10. Gopalbhog

1.04

0.32

0.35

11. Indulshall

-2.13

-0.48

-2.32

12. Jhara

0.19

0.34

-0.30

13. JP 90

-0.62

-0.43

-0.32

14. JP 120

-1.00

0.20

-0.99

15. Zugal

1.65

0.72

0.85

16. Kakri

1.97

0.54

1.59

17. Kalavati

-0.19

0.03

-0.51

18. Kalo Aush

-2.45

-0.55

-2.05

19. Kamal

-0.19

-0.09

0.12

20. Kanakchur

-0.15

-0.05

-0.56

21. Kerala Sundari

0.20

0.13

-0.77

22. Khalia Eulo

-1.05

0.10

-0.66

23. Kalonunia

0.37

-0.16

0.26

24. Khara

0.35

0.27

0.62

25. Lal Badsahbhog

-1.95

-0.69

-1.32

26. Patnai

0.51

0.82

1.01

27. Sagar Sugandhi

-1.74

0.03

-1.36

28. Tulsi Mukul

-1.78

-0.65

-2.41

29. Nonabokra

-2.21

-0.97

-1.93

30. BPT 2295

1.81

0.10

1.89

31. BPT 5204

0.10

-0.16

0.75

32. CR 910

-0.90

-0.94

-0.63

33. Geetanjali

1.30

0.88

1.92

34. NL 44

0.02

-0.40

-1.51

35. NL 46

0.32

-0.30

-1.19

36. NLR 0106

0.32

-0.63

1.03

37. NLR 3242

1.77

0.25

1.30

38. MTU 1061

0.46

0.06

0.25

39. NLR 20084

1.70

-0.09

1.74

40. NLR 40058

-0.51

-0.63

-0.36

41. NLR 145

-0.23

-0.66

-0.18

42. BPT 2411

-0.84

-0.65

-0.57

S.I - Smith index

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

Table 9. Ranking of the best 20% rice genotypes on the basis of Smith index

Year

Genotypes

Mean values for the characters

Smith

index

Rank

PPP

FGPP

LBR

2019

2.18

1.16

2.28

1.93

0.85

0.49

0.60

1.21

1.46

2.4

G16

2.26

1.10

2.30

1.94

0.85

0.51

0.58

1.23

1.24

1.97

G30

1.99

0.95

2.20

1.95

0.79

0.46

0.58

1.26

1.43

1.81

G37

2.02

1.13

2.13

1.92

0.79

0.52

1.37

1.45

1.77

G39

1.96

1.19

2.17

1.90

0.78

0.45

0.59

1.35

1.47

1.70

G15

2.20

0.95

2.11

1.94

0.82

0.60

0.47

1.38

1.39

1.65

2.12

0.94

2.29

1.94

0.9

0.52

0.60

1.17

1.20

1.34

G33

1.94

1.14

2.05

1.91

1.02

0.58

0.64

1.28

1.45

1.30

2.17

1.20

2.18

1.92

0.86

0.54

0.55

1.16

1.38

1.07

MSI

2.09

1.08

2.19

1.93

0.85

0.52

0.57

1.27

1.39

MAI

2.10

1.06

2.05

1.91

0.85

0.53

0.56

1.22

1.29

SDi

-0.01

0.02

0.14

0.02

0.01

-0.10

0.01

0.05

0.10

EGG

0.02

0.04

0.11

0.02

0.01

-0.10

0.01

0.07

0.10

2020

G33

1.93

1.12

2.03

1.92

1.02

0.57

0.65

1.24

1.38

0.88

G26

2.23

0.91

1.98

1.91

0.95

0.51

0.65

1.34

1.31

0.82

G15

2.20

0.93

2.09

1.94

0.81

0.58

0.47

1.30

1.36

0.72

2.21

1.06

1.91

0.94

0.51

0.65

1.34

0.70

2.21

1.25

1.99

1.93

0.90

0.59

0.53

1.23

1.28

0.68

2.11

0.88

2.28

1.95

0.90

0.50

0.63

1.15

1.18

0.64

G16

2.26

1.06

2.28

1.94

0.85

0.50

0.58

1.17

1.20

0.54

2.22

1.08

2.06

1.94

0.95

0.48

0.68

1.14

1.17

0.54

2.19

1.09

2.01

1.92

0.97

0.58

0.61

1.10

1.14

0.47

MSI

2.17

1.04

2.07

1.93

0.92

0.54

0.61

1.22

1.26

MAI

2.10

1.03

2.01

1.92

0.85

0.51

0.57

1.17

1.23

SDi

0.08

0.01

0.06

0.01

0.07

0.02

0.04

0.05

0.03

EGG

0.11

0.04

0.07

0.06

0.03

0.02

0.07

0.05

Combined over the years

2.18

1.16

2.28

1.93

0.85

0.49

0.59

1.17

1.39

2.49

G33

1.93

1.13

2.04

1.91

1.02

0.58

0.64

1.26

1.42

1.92

G30

1.98

0.97

2.19

1.95

0.79

0.45

0.58

1.23

1.4

1.89

G39

1.96

1.15

2.11

1.91

0.78

0.43

0.6

1.32

1.43

1.74

G16

2.26

1.08

2.29

1.94

0.85

0.51

0.58

1.2

1.22

1.59

2.16

1.17

2.16

1.93

0.86

0.53

0.56

1.13

1.37

1.38

2.12

0.91

2.28

1.94

0.9

0.51

0.62

1.16

1.19

1.36

G37

2.02

1.1

2.09

1.92

0.79

0.51

0.52

1.37

1.42

1.3

2.22

1.1

2.09

1.94

0.95

0.49

0.69

1.16

1.19

1.29

MSI

2.09

1.08

2.17

1.93

0.86

0.5

0.6

1.22

1.34

MAI

2.10

1.05

2.03

1.92

0.85

0.52

0.56

1.2

1.26

SDi

-0.10

0.04

0.14

0.02

-0.10

0.03

0.02

0.08

EGG

0.03

0.05

0.10

0.01

-0.10

0.03

0.06

0.10

PH - Plant height (cm), PPP - Panicles plant^-1, FGPP - Filled grains spikelet^-1, SF - Spikelet fertility (%), GL - Grain

length (mm), GB - Grain breadth (mm), LBR - Grain length: breadth ratio (mm), TW - Test weight (g) and GYP - Grain

yield plant^-1 (g), MSI- Mean of Selected Individuals, MAI- Mean of all individuals, SDi: Selection

differential, EGG: Expected genetic gain.

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

the variance, a biplot was constructed in the present

study to explore the relationship among the 45 rice

genotypes based on their observed yield and yield

components. The top right corner of the biplot between

PC1 and PC2 revealed a group of genotypes including

Dudeswar, NLR 20084, BPT 2295, Kakri, Baramshall,

and NLR 0106. These genotypes displayed positive

values for both principal components and characters

such as filled grains spikelet^-1, grain yield plant^-1,

spikelet fertility, and grain length: breadth ratio (mm),

indicating their occupation of the same quadrant and

influence on grain yield. Similar findings are observed

by Dhakal et al. (2020) for filled grains spikelet^-1 and

grain yield plant^-1, and Shanmugam et al. (2023) for

grain length: breadth ratio and spikelet fertility.

In numerous breeding programmes, genotypes

are typically selected based solely on grain yield.

However, a plant’s economic value is contingent upon

its various characteristics. Therefore, it is crucial for

plant breeders to concurrently consider the selection

of multiple characteristics to optimize the economic

value of a plant. The use of a selection index

(Smith, 1936) aids in computing these characteristics

to facilitate the development of an optimal genotype.

The calculated index scores for all forty-two

genotypes grown under terai region varied from -2.41

(Tulsi Mukul) to 2.49 (Dehradun Gandheswari) in

combined over year basis. The top nine genotypes were

chosen based on their high selection index scores, with

Dudeswar, an optimal genotype, achieving the highest

rank under the study. Similar results are reported by

Pavithra et al. (2020) under drought environment

and Venmuhil et al. (2020) for various characteristics

in rice. Sabouri et al. (2008) and Habib et al. (2007)

reported greater genetic improvements through

selection based on multiple characters compared to

selection based on a single trait in rice.

The values of smith selection index for all the

genotypes are given for combined over the years is

given in Table 7, list of the genotypes with rank based

on smith index value for each year and combined

over the years is given in Table 8 and the best nine

genotypes based on smith index for each year and

combined over the years is given in Table 9.

CONCLUSION

Based on the findings of the present study, it

is evident that the D² analysis has successfully

categorized the rice genotypes into four distinct

clusters, each exhibiting significant variation in grain

length, plant height, and grain yield per plant. This

suggests that these genotypes, representing different

clusters, hold potential as donors for enhancing various

agronomic characters through hybridization programs.

By incorporating these beneficial characters into

modern high-yielding varieties, it becomes possible

to develop new varieties with improved adaptability

and resilience. Furthermore, the principal component

analysis highlights spikelet fertility, grain yield per plant,

filled grains per plant, and test weight as the principal

discriminatory characteristics. This emphasizes their

importance in influencing the overall performance

and suitability of the rice genotypes. Additionally, the

findings from the smith selection index for multiple

characters has identified that genotypes, Dudeswar

and Geetanjali are found to be the best performing

genotypes, as evidenced by their highest index scores.

Furthermore, utilizing indexing through PCA-selected

characters could enhance the dependability of the

selection process. These conclusions provide valuable

insights and implications for future research and

agricultural applications, particularly in the selection

and breeding of rice varieties. The identified genotypes

and their associated characters can serve as valuable

resources for the development of improved rice

varieties, ultimately contributing to the advancement

of the agricultural sector.

ACKNOWLEDGMENT

The authors acknowledge the Dean, Faculty of

Agriculture and Director of Research, Uttar Banga

Krishi Viswavidyalaya, Cooch Behar, West Bengal for

providing all the facilities required to carry out this

study.

Funding and Acknowledgement:

No external funding was received to carry out this

research.

Ethics Statement:

There was no human participants and or/or animal

included in this research.

Consent for publication:

All the authors agree to publish the content.

Competing interests:

The authors declare that there is no conflict of

interest for publishing this content.

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

Authors contribution

N. Umamaheswar: Conducted the field experiment

along with collection and analysis of data and drafting

manuscript; A. Kundu: Co-supervisor, guided in

planning and implementation of research work and

data analysis; S.K. Roy: Supervisor, guided in planning

and implementation of research work; S. Sen: Helped

in implementation of research work; L. Hijam: Guided

in planning and implementation of research work; M.

Chakraborty: Guided in layout of the field experiment

and writing the manuscript; B. Das: Helped in collection

of rice genotypes from different parts of West Bengal;

R. Barman: Helped in collection of rice genotypes from

different parts of West Bengal and Andhra Pradesh;

Vishnupriya S: Helped in data collection; B. Thapa:

Helped in data compilation; S. Rout: Helped in writing

the manuscript B. Maying: Helped in writing the

manuscript.

REFERENCES

Akhtar, R., Iqbal, A. and T. Dasgupta. 2022. Genetic

diversity analysis of aromatic rice (Oryza sativa

L.) germplasm based on agro-morphological

characterization. Oryza, 59(2): 141-149. https://

doi.org/10.35709/ory.2022.59.2

Dhakal, A., Pokhrel, A., Sharma S. and A. Poudel. 2020.

Multivariate analysis of phenotypic diversity of

rice (Oryza sativa L.) landraces from Lamjung

and Tanahun Districts, Nepal. Int. J. Agron., 1-8.

https://doi.org/10.1155/2020/886 7961

Habib, S. H., Iftekharuddaula, K. M., Bashar, M. K., Akter,

K. and M. K. Hossain. 2007. Genetic variation,

correlation and selection indices in advanced

breeding lines of rice (Oryza sativa L.).

Bangladesh J. Plant Breed. Genet., 20(1): 25-

32. https://doi.org/10.3329/bjpbg.v20i1.17024

Hazel, L.N. 1943. The Genetic Basis for Constructing

Selection Indexes. Genetics, 28(6): 476-90.

https://doi.org/10.1093/genetics/28.6.476

Jeffer, J. N. 1967. Two case studies in the application

principal

component

analysis.

Journal

of the Royal Statistical Society: Series C

(Applied Statistics), 16(3): 225-36. https://doi.

org/10.2307/2985919

Kang, M. S. 2015. Efficient SAS Programs for Computing

Path Coefficients and Index Weights for Selection

Indices. Journal Crop Improvement., 29: 6-22.

https://doi.org/10. 1080/15427528.2014.959628

Lakshmi, V.I., Sreedhar, M., Vanisri, S., Anantha, M. S.,

Rao L. L. S. and C. Gireesh. 2019. Multivariate

analysis and selection criteria for identification

of African rice (Oryza glaberrima L.) for genetic

improvement of indica rice cultivars. Plant

Genetic Resources, 17(6): 499-505. https://doi.

org/10.1017/S147926211900 0327

Mahalanobis, P. C. 1936. On the generalized distance in

statistics, in Proceedings of the National Institute

of Science of India, Calcutta, India, 2: 49-55.

http://hdl. handle.net/10263/6765.

Maji, A. T. and A. A. Shaibu. 2012. Application of principal

component

analysis

for

rice

germplasm

characterization and evaluation. J. Plant Breed.

Crop Sci., 4(6): 87-93. http://doi.org/10.5897/

JPBCS11.093

Morrison, D.E. 1978. Multivariate Statistical Methods (2nd

ed. 4th Print, McGraw Hill Kogakusta Ltd. https://

lib.ugent.be/catalog/rug01:000004122

Mushtaq, D. and Kumar, B. 2022. Multivariate analysis

among advanced breeding lines of rice (Oryza

sativa L.) Under sub-tropical ecology of Jammu

region of Jammu and Kashmir. Electronic Journal

of Plant Breeding, 13(4): 1387-1394. https://doi.

org/10.37992/2022.1304.179

Nachimuthu, V. V., Robin, S., Sudhakar, D., Raveendran,

M., Rajeswari, S. and S. Manonmani. 2014.

Evaluation of rice genetic diversity and variability

in a population panel by principal component

analysis. Indian J. Sci. Technol., 7(10): 1555-

1562.

https://dx.doi.org/10.17485/ijst/2014/

v7i10.14

Papademetriou, M. K., Dent, F. J. and E. M. Herath.

2000. Rice production in the Asia-Pacific region:

issues and perspectives. Bridging the rice

yield gap in the Asia-Pacific region. Food and

agriculture organization of the United Nations.

Regional office for Asia and the pacific, Bangkok,

Thailand,

220:

4-25.

https://www.fao.org/3/

x6905e/x6905e.pdf

Pavani, M., Sundaram, R. M., Ramesham M. S., Kishor

P. B. K. and K. B. Kemparaju. 2018. Prediction

of heterosis in rice based on divergence of

morphological and molecular markers. J. Genet.,

97: 1263-79. https://doi.org/10.1007/s12 041-

018-1023-8

Pushpa, R., Sassikumar, D., Suresh, R., and Iyanar, K.

(2022). Evaluation of nutritional and grain quality

diversity in rice (Oryza sativa L.) germplasm

MadrasAgric.J.,2024; https://doi.org/10.29321/MAJ.10.500017

111|7-9|

based on multivariate analysis. Electronic

Journal of Plant Breeding, 13(4), 1187-1197.

https://doi.org/10.37992/2022.1304.171

R Studio and Inc. shiny. 2013. Web Application Framework

for R. R package version 0.5.0. https://doi.

org/10.32614/CRAN.package.shiny

Sar, P. and P. C. Kole. 2023. Principal component

and cluster analyses for assessing agro-

morphological diversity in rice. Oryza, 60(1): 117-

24. https://doi.org/10.35709/ ory.2023.60.1.2

Seetharam, K., Thirumeni, S. and Paramasvum, K. 2009.

Estimation of genetic diversity in rice (Oryza

sativa L.) genotypes using SSR markers and

morphological characters. African J. Biotechnol.,

2050-2059.

https://doi.org/10.5897/

AJB2009.000-9275

Shahidullah, S. M., Hanafi, M. M., Ashrafuzzaman,

M., Ismail, M. R. and M. A. Salam. 2009.

Phenological characters and genetic divergence

in aromatic rices. African J. Biotechnol., 8: 3199-

3207. https://doi.org/10.5897/AJB09.686

Shanmugam, A., Suresh, R., Ramanathan, A., Anandhi,

P. and D. Sassikumar. 2023. Unravelling genetic

diversity of South Indian rice landraces based

on yield and its components. Electronic Journal

of Plant Breeding, 14(1): 160-9. https://doi.org/

10.37992/2023.1401.007

Shoba, D., Vijayan, R., Robin, S., Manivannan, N.,

Iyanar, K., Arunachalam, P., Nadarajan, N.,

Pillai, M. A. and S. Geetha. 2019. Assessment

of genetic diversity in aromatic rice (Oryza

sativa L.) germplasm using PCA and cluster

analysis. Electron. J. Plant Breed., 10(3): 1095-

104. https://doi.org/10.5958/0975-928X.2019.00

140.6

Singh, M., Singh, S. K., Vennela, P. R., Singh, D. K. and

D. Kumar. 2017. Genetic divergence studies

for drought tolerance in rice (Oryza sativa

L.) using morphological traits and molecular

markers. Oryza, 54(4): 385-91. https://doi.

org/10.5958/2249-5266.2017.00053.4

Singh, S. K., Pandey, V., Mounika, K., Singh, D. K.,

Khaire, A. R., Habde, S., & Majhi, P. K. (2020).

Study of genetic divergence in rice (Oryza

sativa L.) genotypes with high grain zinc using

Mahalanobis’ D2 analysis. Electronic Journal

of Plant Breeding, 11(02), 367-372. https://doi.

org/10.37992/2020.1102.065

Smith, H. F. 1936. A Discriminant Function for Plant

Selection. Annals Eugen., 7, 240-50. https://doi.

org/10.1111/j.1469-1809.1936.tb02143.x

Sparks, D.N. 1973. Euclidean cluster analysis. Journal

of the Royal Statistical Society: Series C

(Applied Statistics), 22(1): 126-30. https://doi.

org/10.2307/2346321

Tuhina-Khatun, M., Hanafi, M. M., Rafii Yusop, M., Wong,

M. Y., Salleh, F. M. and J. Ferdous. 2015. Genetic

variation, heritability, and diversity analysis of

upland rice (Oryza sativa L.) genotypes based

on quantitative traits. Biomed Res. Int., 1-7.

http://dx.doi.org/10.1155/2015/290861

Venmuhil, R., Sassikumar, D., Vanniarajan, C. and R.

Indirani. 2020. Selection indices for improving the

selection efficiency of rice genotypes using grain

quality traits. Electron. J. Plant Breed., 11(2):

543-549.

https://doi.org/10.37992/2020.1102.0

Zeigler, R.S., and B. Adam. 2008. The relevance of rice.

Rice, 1: 3-10. https://doi.org /10.1007/ s12284-

008-9001-z