Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

Relationship of Seed Traits on Initial Progeny Growth Performance and

Divergence Studies in Madhuca latifolia Macb. for Further Use in

Tree Improvement

B.N. Divakara
1*

1
Institute of Wood Science and Technology, Mellashwaram, Bangalore, India

Date Received: 18-01-2013 Date Accepted: 03-07-2014

Abstract

Evaluation of 23 genotypes of Madhuca latifolia was carried out based on relationship of seed

traits with initial progeny growth performance and divergence studies as a scope for further breeding

programme. Variability studies revealed that, more than 12 accessions recorded above average for 100-

seed weight (247.5±49.2), oil content (43.8±3.7) and volume index (346.0±97.7). The maximum values

observed in studied CPTs were as follows: seed length (39.1 mm) in CPT-15 genotype, seed breadth

(19.2 mm) in CPT–8 and CPT–9, aspect ratio (2.2) in CPT-6 and CPT-15, 2D surface area (501.4 and

491.6 mm
2
) in CPT-9 and CPT-3 respectively. CPT–16 recorded maximum for 100 seed weight (282.4

g) and oil content (51.2%). However, maximum volume index was recorded by CPT–3 (578.3 cm
3
)

followed by CPT–16 (496.0 cm
3
). The phenotypic and genotypic coefficients of variations are close to

each other for all traits, except volume index that exhibited striking difference between PCV (40.0%)

and GCV (19.9%) indicating that for most traits genetic control was quite high. Trait oil content and

100 seed weight expressed high heritability (93.5%, 93.0%) accompanied with moderate genetic

advance (17.2%, 15.6%), indicating that, heritability is due to additive gene effects and selection may

be effective. At genotypic level 100 seed weight registered positive significant correlation with plant

height (0.73), oil content with volume index (0.71). Hence seeds with large breadth, high seed weight

and oil content may be selected for producing better progenies. Since traits viz. 100 seed weight and oil

content are under strong genetic control, improvement in these characters can bring improvement in

volume index. On the basis of the divergence, the 23 genotypes studied were grouped into 5 clusters,

indicating wide diversity. The clustering pattern shows that geographical diversity is not necessarily

related to genetic diversity. The genotypes in cluster IV and V were most heterogeneous and can be

best used for within group hybridization. Cluster means indicated crosses involving under cluster II and

V and cluster II and I may result in substantial segregates and further selection for overall improvement

of species.

Keywords: Madhuca latifolia, heritability, genetic advance, genetic divergence, correlation

1. Introduction
Madhuca latifolia Macb. (Syn M. indica J.F. Gmel; Bassia latifolia Roxb.) of family

Sapotaceae, vernacularly and commonly known as mahua and Indian Butter Tree is a large, branched

deciduous tree indigenous to the Indian subcontinent. It is predominantly grown in dry tropical and

* Correspondence: bndsira@gmail.com

Tel: +91 8022190198

ISSN 2235-9370 Print/ISSN 2235-9362 Online © University of Sri Jayewardenepura

sub-tropical regions of Indo-Pakistan subcontinent. The species is scattered in deciduous forests and

dry sal plain forests. Plant grows upto 18 m height with short bole, grey to black bark and round

crown. It is extensively cultivated in villages of northern India and Deccan peninsula for sweet fleshy

corolla and ripe fruits which is used as a major source of industrial alcohol as well as country liquor,

portable spirit and vinegar. Seeds contain a valuable fatty oil ranging from 38% to 57%, known as

mahua oil or butter of romance. Mahua oil is used as edible oil, manufacture of margarine, soap,

glycerin, lubricating grease and medicine. Seed cake is used as biofertilisers, organic manure, biocide

and fish meal. Its leaves used for fodder and green manure while bark gives tannin (CSIR, 1998). Due

to presence of similar properties to that of diesel, Mahua oil has gained the importance as bio-diesel

and is emerging as a viable alternative to fossil fuel. Importantly, the successful adoption of bio-fuels is

reliant on the supply of feedstock from non-food crops with the capacity to grow on marginal land not

destined to be used for the cultivation of food crops (Hill et al., 2006). Meeting both of these criteria

and thus, can form the basis of a highly promising, profitable, and self-sustaining platform for small-

scale entrepreneurship and self-employment in rural areas, ensuring optimum utilisation of wasteland

resources and unemployed manpower. Although M. latifolia is well known as oil yielding tree having

wide adaptability and plethora of uses, little attempt has been directed to improve it as a crop plant

because of long gestation period and slow growing nature. The wide gap in potential and actual yield is

due to the use of locally available wild material. No systematic breeding program for breeding superior

high yielding genotypes has been initiated. M. latifolia being open pollinated crop (Anemophillic),

provide ample scope for genetic improvement through selection of superior trees with genetic variation

in seed morphology and oil content along with initial progeny performance which later can be of great

potential and may have greater impact than the conventional breeding. Hence, the challenging task, as

of today is to screen the naturally available M. latifolia genetic resources to select the best planting

material with high oil content for higher productivity. The available information in literature does not

provide a complete understanding of geographical variation and its influence on improvement of seed

quality and quantity. The information on the genetic structure and diversity relationship of candidate

plus trees provide a basis for planning and conducting future collections and efficient utilization of

genetic resources to realize the potentiality for maximizing seed and oil yield. Keeping precarious

scenario in view, an effort was made to investigate the relationship of seed traits with initial progeny

growth performance for early selection and divergence studies to understand the diversity among 23

accessions for assessment and creation of diverse lines in M. latifolia for further use in tree

improvement.

2. Material and Methods

An extensive wild germplasm exploration survey was conducted to identify the high yielding

Candidate Plus Trees (CPTs) of M. latifolia at fruiting stage from different predominant naturalized

locations in Jharkhand, India (Table 1, Figure 1). The selection was made on phenotypic assessment of

characters of economic importance viz yield potential, crown spread, total height, girth at breast height,

age of the tree, free from pest and diseases, seed size and seed weight. A total of 23 CPTs

(morphologically superior trees) covering a latitude and longitudinal range between 22
0
00

N to 24
0
50

N and 83
0
30

E to 87
0
00

E, respectively, were selected (Table 1). From each CPTs, 2 Kg of mature

capsules were collected during June-July, 2005. Quantitative characterization was carried out (6 seed

and 3 progeny characters) at Forest Research Centre, Institute of Forest Productivity Mandar, Ranchi

District during 2005-07.

Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

2.1 Site characteristics

The study area under Forest Research Centre (latitude: 23
0
27′ 40″ N, longitude: 85

0
05′ 56″ E, altitude

2,320 ft msl approx.) has a semi-arid type of climate. Mean annual rainfall is 1231.6 mm with mean

number of rainy days 73.6. Annual minimum and maximum temperature is recorded as 17.7
0
C and

30.2
0
C respectively with lowest temperature in January and highest temperature in May. Soils of the

study area is characterized by pH (5.7), EC (35), organic carbon (0.33%), nitrogen (105 ppm),

phosphorus (11 ppm) and potassium (74 ppm).

2.2 Seed characters

Samples of 300 seeds were randomly collected from each CPT to make three replicates with 100 seeds

in each. Measurement of morphological traits viz. seed length, seed breadth, aspect ratio and 2D

surface area, was carried out using Image analyzer (Leica Quantimet called QWin 500). Seeds were

spread on a glass platform of macro-viewer for each replication and images were captured using charge

coupled device (CCD) camera. These images were analysed using Quantimet 500+ or Qwin software.

The Qwin identifies the object based on given specification for seed colour and calibrates captured

images to actual scale. Different two dimensional (2D) measurements of the detected images were

measured (Table 2).

Figure 1: Distribution of Madhuca latifolia Candidate Plus Trees in

Jharkhand, India mapped using Microsoft Encarta 2005

Note: Details of number representation is in table 1.

Table 1: Locational details of Madhuca latifolia Candidate Plus Trees (CPTs) selected in Jharkhand, India.
CPTs District Location/Village Latitude Longitude Alt,

Age,

Seed

yield,

kg yr
-1

Crown

area

(m
2
)

CPT-1 Lohardaga Chingri Nawatoli 23° 26' 19 N 84° 18' 25 E 590 95 10.0 80 300 294.19

CPT-2 Simdega Pharsabera Kusumtoli 22° 39' 17 N 84° 31' 56 E 410 65 13.6 90 100 191.21

CPT-3 Gumla Samra Velluwa toil 23° 03' 03 N 84° 52' 28 E 520 80 15.3 53 120 123.75

CPT-4 Simdega Piosokra 22° 36' 06 N 84° 43' 01 E 370 80 17.5 119 180 331.81

CPT-5 Latehar Hotwag 23° 47' 28 N 84° 27' 08 E 340 70 15.3 70 120 247.55

CPT-6 Daltonganj Sarja Polpol 23° 54' 17 N 84° 31' 21 E 290 45 13.8 70 70 166.34

CPT-7 Daltonganj Akhra 23° 55' 59 N 84° 10' 23 E 250 100 14.4 110 200 481.30

CPT-8 Garhwa Singakala 23° 55' 18 N 83° 44' 15 E 385 100 13.6 110 200 452.57

CPT-9 Garhwa Garbandh 24° 21' 24 N 83° 28' 39 E 510 60 10.3 85 150 166.34

CPT-10 Daltonganj Simarbar batawa 24° 27' 39 N 84° 13' 04 E 215 70 12.0 74 100 247.55

CPT-11 Chatra Pitiz 24° 15' 19 N 85° 06' 50 E 375 40 10.3 55 30 133.81

CPT-12 Latehar Nachna 23° 51' 54 N 84° 49' 44 E 540 120 9.8 103 200 216.51

CPT-13 Ranchi Piska Mahua Toli 23° 28' 03 N 85° 26' 24 E 610 50 14.0 60 150 203.67

CPT-14 Hazaribag Ahoanhe moraha 23° 49' 37 N 85° 24' 55 E 510 40 10.2 54 70 166.34

CPT-15 Koderma Banpok 24° 34' 36 N 85° 42' 47 E 390 80 10.2 70 100 125.73

CPT-16 Giridih Madhgopali 23° 56' 56 N 86° 00' 44 E 340 70 11.0 78 100 212.62

CPT-17 Chaibasa Khas Jamda 22° 09' 08 N 85° 29 54 E 550 70 12.8 80 100 229.75

CPT-18 Saraikela Palobera 22° 15' 52 N 85° 59' 14 E 300 45 15.4 65 60 243.38

CPT-19 Jamshedpur Golkatta 22° 39' 02 N 86° 27' 09 E 190 50 12.7 52 70 128.73

CPT-20 Chaibasa Junko 22° 45' 19 N 85° 26' 25 E 450 50 17.7 70 100 308.03

CPT-21 Ranchi PaniMahua Chamma 23° 32' 27 N 85° 06' 17 E 650 50 14.0 60 150 203.67

CPT-22 Dhanbad Kokra 23° 59' 50 N 86° 30' 45 E 240 90 14.3 97 100 274.76

CPT-23 Latehar Chandwa 23° 41' 42 N 84° 43' 36 E 520 80 13.8 106 120 335.05

2.3 The progenies

Seeds of all the CPTs were pre-treated by soaking in cold water for 24 hours. Pretreated seeds

of each CPTs were directly sown in polythene bags of dimension 30×12×10 cm filled with potting

mixture of soil, sand and farmyard manure (2:1:1) in three replicates of 100 seeds each, during July

2005. Samples of six one-year-old seedlings were planted with spacing of 3.5×3.5 m in the field with

(pit size 45×45× 45 cm) for field evaluation at Forest Research Centre in July 2006. Randomised

Complete Block Design (RCBD) with three replicates was used. After 30 months of sowing (Juvenile

stage), observations were recorded on the trial for plant height (m) and collar diameter (cm) were

measured at bimonthly intervals over a period of 18 months. The data recorded at 18 months after

sowing (MAS) alone was considered for variability, correlation and diversity studies. The results are

given in the (Table 3).

Table 2: Methodology for measuring seed and progeny traits of Madhuca latifolia.

Sl. No. Traits Method

1 Seed length (mm) Length of the seed at longest side.

2 Seed breadth (mm) Length of the seed at shortest side.

3 Aspect ratio Length was divided by breadth.

4 2D surface area (mm
2
) 2D surface area of the seed in the direction of measurement.

5 100 – seed weight (g) Weight of 100 seeds weighed on electronic balance was measured in

grams.

6 Oil content (%) Estimated using soxhlet apparatus following the procedure of

Sadasivam and Manickam (1992).

7 Plant height (cm) Length of the plant from ground level to tip.

8 Collar diameter (cm) Stem diameter near the ground level.

9 Volume index (cm
3
) [Collar diameter (cm)]

2
× Plant height (cm)] (Manavalan 1990).

Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

2.4 Data analysis

The seed parameters and progeny measurements were analysed using Analysis of variance

(ANOVA) and Duncan Multiple Range Test (DMRT) to understand the significance of differences

between the seeds and progenies of CPTs (Gomez and Gomez, 1984). The phenotypic variation for

each trait was partitioned into components based on genetic (hereditary) and non-genetic

(environmental) factors and estimated using the following formula (Johanson et al., 1955):

Vp = MSG/r; Vg = (MSG – MSE)/r; Ve = MSE (1)

MSG, MSE and r are the mean squares of CPTs, mean squares of error and number of

replications, respectively.

The phenotypic variance (Vp) is the total variance among phenotypes when grown over the

range of environments of interest, the genotypic variance (Vg) is the part of the phenotypic variance

that can be attributed to genotypic differences among the phenotypes, and the error variance (Ve) is part

of the phenotypic variance due to environmental effects. To compare the variation among traits,

phenotypic (PCV) and genotypic (GCV) coefficients of variation were computed according to the

method suggested by Burton, (1952)

PCV = (√Vp/X) × 100; GCV = (√Vg/X) × 100 (2)

Vp, Vg and X are the phenotypic variance, genotypic variance and grand mean for each pod and

seed-related trait, respectively.

Broad sense heritability (h
2
B) was calculated according to Allard (1999) as the ratio of the

genotypic variance (Vg) to the phenotypic variance (Vp). Genetic advance (GA) expected and GA as

per cent of the mean assuming selection of the superior 5% of the genotypes were estimated in

accordance with Johanson et al., (1955), as in the following equation.

GA = K·h
2
B·√Vp; GA (as % of the mean) = (GA/X) × 100 (3)

K is the selection differential (2.06 for selecting 5% of the genotypes).

Phenotypic (rp) and genotypic (rg) correlations were further computed to examine inter-character

relationships among seed and seedling traits following Goulden (1952) as given below.

rp = Covp (x1, x2)/[Vp(x1)·Vp(x2)]
½
(4)

rg = Covg (x1, x2)/[Vg(x1)·Vg(x2)]
½

(5)

Covp and Covg are phenotypic and genotypic covariances for any two traits x1 and x2,

respectively, and Vp and Vg are the respective phenotypic and genotypic variances for those traits. The

genetic diversity was estimated using the Mahalanobis D2 statistics (Mahalanobis, 1936). Tracing D
2

as a generalised distance, the criterion used by Tocher as described by Rao (1952) was applied for

determining the clustration group. Average intra and inter cluster distances were determined using

GENRES version 3.11, 1994 Pascal Intl. Software and suggested by Singh and Chaudhary (1977).

3. Results
Mean parent seed and progeny growth characters of twenty-three tested CPTs of M. latifolia are

presented in Table 3. There is a significant difference among all seed and progenies traits. Variability

studies revealed that, more than twelve accessions recorded are above average for 100-seed weight

(247.5±49.2), oil content (43.8±3.7) and volume index (346.0±97.7). Maximum seed length (39.1 mm)

was observed in CPT-15 genotype, while seed breadth (19.2 mm) was maximum in CPT–8 and CPT–9,

Aspect ratio (2.2) was highest in CPT-6 and CPT-15. 2D surface area (501.4 and 491.6 mm
2
) was

highest in CPT-9 and CPT-3 respectively. CPT–16 recorded the maximum for 100 seed weight (282.4

g) and oil content (51.2%). However, maximum volume index was recorded in CPT–3 (578.3 cm
3
)

followed by CPT–16 (496.0 cm
3
). Lowest seed weight (for 100 seed), oil content and volume index

was expressed in genotypes CPT–4 (216.0 g) lowest oil content in CPT–21 (37.7 %) and lowest

volume index in CPT–12 (176.5 cm
3
) respectively. Genotype CPT–19 and CPT–23 recorder lowest for

seed length (26.8 mm), aspect ratio (1.6) and seed breadth (15.5 mm), 2D surface area (321.5 mm
2
).

Table 3: Mean performance of selected genotypes of Madhuca latifolia for seed and progeny growth traits.

Genotype

Seed traits Progeny traits

Seed

length,

Seed

breadth,

Aspect

ratio

2D surface

area, mm
2

100 seed

weight, g

Oil

content,

Height,

Collar

diameter,

Volume

index,

cm
3

CPT-1 33.45
bcd

16.97
efg

1.97
cde

408.48
de

235.93
ij
44.03

ghi
60.50

bcd
2.03

bcdef
253.61

CPT-2 31.55
def

16.75
fg

1.88
defg

385.18
efg

271.67
b
39.32

klm
54.67

d
1.99

cdef
235.94

CPT-3 37.99
a
18.11

bc
2.04

bc
491.57

a
264.27

bc
50.30

ab
81.17

ab
2.74

a
578.30

CPT-4 30.45
efg

15.56
hi
1.95

cdef
355.43

fgh
216.00

l
40.38

kl
61.67

bcd
1.93

ef
235.93

CPT-5 34.39
bcd

17.98
cd

1.91
cdef

461.59
abc

286.27
a
49.18

bc
64.33

abcd
2.20

abcdef
310.97

bcd

CPT-6 38.77
a
16.81

efg
2.21

a
478.49

ab
249.72

efgh
40.85

k
68.50

abcd
2.21

abcdef
344.32

abcd

CPT-7 32.17
cdef

17.64
cde

1.82
fgh

417.48
cde

221.58
kl
40.25

kl
60.50

bcd
2.32

abcde
352.01

abcd

CPT-8 33.69
bcd

19.19
a
1.75

gh
465.87

ab
260.20

cd
45.45

efg
70.67

abcd
2.58

ab
449.69

abc

CPT-9 35.38
b
19.17

a
1.85

efgh
501.43

a
245.96

fgh
38.68

lm
71.83

abcd
2.28

abcde
385.56

abcd

CPT-10 27.59
hi
18.18

bc
1.56

j
358.18

fgh
227.42

jk
45.92

def
60.17

bcd
2.28

abcde
335.69

bcd

CPT-11 33.65
bcd

16.33
gh

2.00
bcd

400.78
def

242.31
ghi

44.05
ghi

66.17
abcd

2.35
abcde

387.45
abcd

CPT-12 33.31
bcde

15.48
hi
2.12

ab
381.31

efg
257.81

cde
46.12

de
62.50

abcd
1.68

f
176.45d

CPT-13 29.84
fgh

15.73
hi
1.90

cdefg
344.37

gh
254.89

cdef
47.48

cd
62.83

abcd
2.50

abcd
415.24

abcd

CPT-14 32.25
cdef

18.38
abc

1.76
gh

436.97
bcd

240.18
hi
42.58

ij
72.17

abcd
2.41

abcde
423.82

abc

CPT-15 39.12
a
16.94

efg
2.21

a
488.35

a
261.59

cd
41.02

jk
68.83

abcd
2.27

abcde
354.35

abcd

CPT-16 32.86
bcde

18.08
bc

1.82
fgh

438.88
bcd

282.36
a
51.15

a
79.50

abc
2.53

abc
496.00

CPT-17 34.45
bcd

15.78
hi
2.12

ab
400.69

def
219.41

kl
39.02

lm
58.33

cd
1.97

def
251.05

CPT-18 32.23
cdef

16.96
efg

1.90
cdefg

407.17
de

225.96
k
43.82

ghi
55.50

d
2.08

bcdef
255.19

CPT-19 26.84
i
16.81

efg
1.60

ij
334.32

h
235.75

ij
45.05

efgh
64.83

abcd
2.43

abcde
379.78

abcd

CPT-20 34.74
bc

17.25
def

2.02
bcd

440.61
bcd

250.25
efg

44.18
fghi

56.83
d
2.07

bcdef
268.84

bcd

CPT-21 32.27
cdef

18.84
ab

1.71
hi
459.23

abc
226.55

jk
37.65

m
58.83

cd
2.00

cdef
252.19

CPT-22 32.26
cdef

16.66
fg

1.94
cdef

399.02
def

264.78
bc

48.05
c
70.50

abcd
2.24

abcde
356.74

abcd

CPT-23 28.41
ghi

15.46
i
1.84

efgh
321.52

h
252.11

def
43.35

hi
83.67

a
2.34

abcde
458.01

abc

MEAN 32.94 17.18 1.91 416.39 247.52 43.82 65.85 2.24 345.96

SEM 0.89 0.27 0.04 14.08 3.09 0.58 8.90 0.23 98.07

CD (5%) 2.59 0.77 0.13 40.96 9.00 1.68 25.89 0.66 285.38

Trait means not followed by the same superscript letter are significantly different at p = 0.05.

The phenotypic and genotypic coefficients of variations are close to each other for all traits,

except for volume index that exhibited striking difference between PCV (40.0%) and GCV (19.9%)

indicating that for most traits, genetic control was quite high (Table 4). All the seed traits showed high

heritability while progeny growth characters showed low to moderate heritability. Trait oil content and

100 seed weight expressed high heritability (93.5%, 93.0%) accompanied with moderate genetic

Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

advance (17.5%, 15.6%) indicating that, heritability is due to additive gene effects and thus selection

may be effective. Trait volume index expressed moderate heritability (24.7%) accompanied with high

genetic advance (20.4%). The low heritability is being exhibited due to high environment effects, since

trait has high genetic advance, selection will be effective. Similar trend was observed in Casuarina

equisitifolia relationship of cone and seed traits on progeny growth performance (Mahadevan et al.

1999).

Table 4: Genetic estimates of parent seed and progeny growth traits.
Seed traits Variance Coefficient of variation (%)

Heritability

(%)

GA (%) of

mean Genotypic Phenotypic Genotypic Phenotypic

Seed length 9.17 11.54 9.19 10.31 79.5 16.89

Seed breadth 1.29 1.50 6.61 7.13 85.9 12.61

Aspect ratio 0.03 0.03 8.70 9.58 82.4 16.26

2D surface area 2571.64 3165.93 12.18 13.51 81.2 22.61

100 seed weight 379.11 407.76 7.87 8.16 93.0 15.63

Oil content 14.24 15.24 8.61 8.91 93.5 17.15

Height 25.17 143.95 7.62 18.22 17.5 6.56

Collar diameter 0.03 0.11 8.22 14.93 30.3 9.33

Volume index 4742.53 19168.55 19.91 40.02 24.7 20.40

Of the 72 (36 genotypic and 36 phenotypic) correlations, 13 genotypic and seven phenotypic

combinations were significant at 1% along with four genotypic and one phenotypic combinations was

significant at 5% (Table 5). Skinner et al. (1999) suggested only those correlation coefficients, which

are greater than 0.70 or smaller than -0.70 as biologically meaningful so that 50% of the variation in

one trait is predicted by the other. The seed trait pair showing such high correlation was 11 and all were

positive. At genotypic level 100 seed weight registered positive significant association with height

(0.73) and oil content with volume index (0.71) at 30 MAS.

In the present investigation, attempts were made to assess the genetic diversity among the

twenty-three CPT’s candidate plus trees based on seed and progeny traits using Mahalanobis D
2

statistics. On the basis of D
2
values for all possible 253 pairs of populations, twenty-three genotypes

were grouped into five clusters. Cluster III had maximum number of genotypes which is six and cluster

IV and V followed accommodating five genotypes each. Cluster I and II had four and three genotypes

respectively (Table 6). The clustering pattern showed that geographical diversity is not necessarily

related to genetic diversity. Intra-cluster distance D values ranged between 5.59 in cluster II to 9.56 in

cluster V having three and five genotypes respectively (Table 7). Cluster V with 9.56 followed by

cluster IV with 9.19 intra-cluster distance were the most diverse because the genotypes used for

breeding programme were from different locations. The divergence within the cluster indicates the

divergence among the genotypes in the same cluster. Contrarily cluster II showed the minimum intra

genetic distance (5.59) between them revealing that these genotypes were somewhat similar in genetic

constitution and hybridization amongst these groups not showing sufficient variability. Inter-cluster

distance ranged from 8.73 between III and V to 14.25 between II and V (Table 7). The highest inter-

cluster distance 14.25 was followed by 13.14 between cluster I and II. Inter-cluster divergence suggests

the distance (divergence) between the genotypes of different clusters. The tendency of genotypes from

diverse eco-geographic regions to group together in the same cluster or scattered distributions of

genotypes of same geographic origin in different clusters have been observed in the present study.

Table 5: Correlation coefficient matrix of seed and progeny growth traits of Madhuca latifolia.

Seed traits Seed

breadth

Aspect

ratio

surface

area

100 seed

weight

Oil

content

Height Collar

diameter

Volume

index

Seed length G 0.197 0.765
**

0.850
**

0.329 -0.094 0.389 0.054 0.172

P 0.241 0.757
**

0.859
**

0.275 -0.041 0.027 0.009 0.015

Seed breadth G -0.482
*
0.673

**
0.148 0.034 0.351 0.600 0.550

P -0.428
*
0.683

**
0.130 0.036 0.052 0.266 0.194

Aspect ratio G 0.319 0.201 -0.106 0.055 -0.427
*
-0.282

P 0.342 0.159 -0.064 -0.041 -0.205 -0.156

2D surface area G 0.323 -0.085 0.469
*
0.324 0.390

P 0.268 -0.044 0.041 0.106 0.080

100 seed weight G 0.604
**

0.733
**

0.405
*
0.518

P 0.573
**

0.296 0.238 0.270

Oil content G 0.680
**

0.652
**

0.705
**

P 0.273 0.328 0.330

Height G 0.732
**

0.861
**

P 0.643
**

0.833
**

Collar diameter G 0.978
**

P 0.947
**

*
significant at p = 0.05,

**
significant at p = 0.01

Table 6: Clustering of Madhuca latifolia genotypes using Tocher’s method.
Clusters Number of

accessions

Accessions (CPTs)

I 4 CPT-1, CPT-2, CPT-6, CPT-15

II 3 CPT-3, CPT-5, CPT-16

III 6 CPT-4, CPT-7, CPT-8, CPT-9, CPT-11, CPT-20

IV 5 CPT-10, CPT-12, CPT-13, CPT-14, CPT-22

V 5 CPT-17, CPT-18, CPT-19, CPT-21, CPT-23

Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

Table 7: Average intra and inter-cluster distance and D
2
values.

Clusters I II III IV V

I 8.056

(64.906)

13.144

(172.769)

9.092

(82.669)

11.406

(130.104)

10.394

(108.041)

II 5.591

(31.255)

12.674

(160.618)

10.322

(106.540)

14.247

(202.974)

III 8.702

(75.725)

9.739

(94.845)

8.725

(76.128)

IV 9.192

(84.499)

10.079

(101.582)

V 9.559

(91.366)

*
Figures given in the parenthesis are D

2
values.

Table 8: Cluster wise mean values of six seed traits and three progeny traits.
Traits\Clusters I II III IV V Percent

contribution

Seed length (cm) 35.7 35.1 33.4 31.1 30.8 5.1

Seed breadth (cm) 16.9 18.1 17.5 16.9 16.8 11.1

Aspect Ratio 2.1 1.9 1.9 1.9 1.8 0.0

2D surface area (mm
2
) 440.1 464.0 430.3 384.0 384.6 0.0

100 seed weight (g) 254.7 277.6 239.4 249.0 232.0 19.8

Oil content (%) 41.3 50.2 42.2 46.0 41.8 19.4

Plant height (cm) 63.1 75.0 64.6 65.6 64.3 4.0

Collar diameter (cm) 2.1 2.5 2.3 2.2 2.2 7.1

Volume index (cm
3
) 297.1 461.8 346.6 341.6 319.2 33.6

Cluster means indicated a wide range of variation for all the seed and progeny traits (Table 8). The best

cluster for seed breadth of 18.1 mm, 2D surface area of 464.0 mm
2
, 100 seed weight of 277.6 g, oil

content of 50.2%, plant height of 75 cm, collar diameter of 2.5 cm and volume index of 461.8 cm
3
was

for cluster II. Maximum seed length (35.7 mm) and aspect ratio (2.1) was recorded by cluster I. Cluster

V recorded minimum for seed length (30.8 mm), seed breadth (16.8 mm) and aspect ratio (1.8) and 100

seed weight (232.0 g) while cluster I recorded minimum for oil content (41.3%), plant height (63.1 cm)

collar diameter (2.1 cm) and volume index (297.1 cm
3
). Cluster II has genotype (CPT–3, CPT–5 ,

CPT–16) containing high volume index (578.3 cm
3
), 100 seed weight (286.3 g), oil content (51.2%)

and cluster I has genotype (CPT-15) containing high seed length (39.1 mm) and aspect ratio (2.2). Thus

it may be suggested that crosses involving under cluster II and I may result in substantial segregates

and further selection for overall improvement of the species.

4. Discussion
Seed weight, depends on reserve food material, which is produced as a result of double

fertilization (endosperm) and is dominated by the maternal traits and is also influenced by the nutrient

availability at the time of seed setting and environmental factors (Allen, 1960; Johnsen et al., 1989).

Embryo development and its physiological function are contributed by the maternal as well as by

paternal (pollen grain) traits in the species. The occurrence of M. latifolia over a wide range of habitats

with diverse geo-climatic conditions was expected to be reflected in the genetic constitution of its

populations. In the present study, the seeds from various CPTs exhibited significant seed trait

variability (Table 3) which could be attributed to population isolations that inturn influence gene flow.

Significant variability of seed characters like; seed size and weight was observed in seeds of the

selected plus trees (Bagchi and Sharma, 1989) and among various provenances of Santalum album

(Veerendra et al., 1999). Genetic control of seed size traits has been observed in several tree species

(Loha et al., 2006).

Though the selection of superior trees was carried out intensively and clonal superiority over

seed raised plants was established (Kumar, 1995), genetic superiority per se needs to be determined.

Genetic estimates could considerd as useful tools in predicting the amount of gain expected in short

period of time. The variation among genotypes is commonly used as an estimate of total genetic

variation and to calculate the degree of genetic control for a particular trait (Foster and Shaw, 1988).

Marginal difference between PCV and GCV and high estimates of heritability for all pod and seed

traits studied revealed the presence of heritable nature in variability (Table 4). The magnitude of the

error variance was relatively lower than the genotypic variance for all traits except for height, collar

diameter and volume index (data not given). Higher magnitude of genotypic variance over error

variance on one hand and close values of phenotypic and genotypic variances on the other hand for all

the seed traits, indicated considerable scope for selection. Relatively high value of genotypic variance

that resulted in high estimates of heritability which contributed to the high genetic gains expected in oil

content. Gains from tree breeding programs depend on the type and extent of genetic variability. In the

present study the genotypic coefficient of variation and the genetic gain were found to be

comparatively higher for an important trait such as volume index, 100-seed weight and oil content. The

high estimates of heritability combined with high genetic advance suggests that population means for

volume index may be changed considerably by selecting the superior 5% of the genotypes. High

heritability for growth parameters have been reported in Tectona grandis (Gera et al., 2001) and

accompanied by high genetic advance in Prosopis cineraria (Solanki et al., 1984).

As variation among genotypes is used for estimation of genetic variation and genetic gain, co-

variance estimates between traits can be used to estimate genetic correlations between the traits (Foster,

1986). In genetic improvement of M. latifolia clear understanding of the relationships among different

seed and progeny traits is very essential. Correlation establishes the extent and cause of association

between seed traits and its attributes so that these components may form additional criteria for selection

in breeding program. Correlated quantitative traits are of a major interest in an improvement program,

as the improvement of one character may cause simultaneous changes in the other character. Similar

correlation trend was seen in Jatropha curcas (Kaushik et al., 2007a) and Pongamia pinnata (Kaushik

et al., 2007b) seed traits. Hence 100 seed weight may be included among the criteria for selection of

plus trees. The genotypic correlation is an estimated value, whereas, phenotypic correlation is a derived

value from the genotype and environmental interaction (Chaturvedi and Pandey 2004). The genotypic

correlation is, therefore, a more reliable estimate for examining the degree of relationship between

character pairs. The present study indicates the correlation of 100 seed weight and oil content to growth

of M. latifolia at juvenile stage. Positive correlation between seed weight and seedling height in Pinus

spp. has been observed but it disappeared with the growing age of the seedlings (Righter, 1945).

However correlation between seed weight and height till 15 years was observed in Pinus taeda

(Robinson and Van Buijtenen, 1979). Khalil (1981) stated that, significant positive correlation and

regression between 1,000 seed weight and height at four years in Picea glauca which is not mere

carryover of the initial effect in the first year growth but appears to be genetically correlated and

recommended that seed weight may be included among the criteria for selection of plus trees.

Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

Genetic diversity in plant species is a gift to mankind as it forms the basis for selection and

further improvement. In Jatropha curcas Kaushik et al. (2007a) attempted to analyse the diversity

among the 24 CPT’s from Haryana of India. In the present study clustering pattern showed that

geographical diversity is not necessarily related to genetic diversity (Table 6). In rice it is reported that

this kind of genetic diversity might be due to differential adoption, selection criteria, selection pressure

and environment (Vivekananda and Subramanian, 1993). This indicated that genetic drift produces

greater diversity than the geographic diversity (Singh et al., 1996). Absence of any relationship

between genetic diversity and geographical distribution is in accordance with the findings of Kaushik et

al. (2007a) and Gohil and Pandya, (2008) in J. curcas. The contribution of individual characters to the

diversity has been studied. The trait volume index contributed maximum for 33.6% for genetic

diversity and rank total of 85. The character contributing maximum diversity can be given more

emphasis for the purpose of fixing priority of parents in hybridization program. Cluster V and cluster

IV with high intra-cluster distance were the most diverse and the divergence within the cluster indicates

the divergence among the genotypes in the same cluster (Table 7). Hence, best suited for within group

is the hybridisation. Cluster means indicated crosses involving under cluster II and V and cluster II and

I may result in substantial segregates and further selection for overall improvement of species. In

general, the cluster II and I, V exhibited high and low mean values respectively for most of the

characters (Table 8). It is also suggested that for creating variability and developing the best selection a

large number of divergent lines, instead of few should be used in the hybridization. Earlier studies, in

crop plant had indicated that inter-mating of divergent groups would lead to greater opportunity for

crossing over which would release latent variation by breaking up predominantly repulsion linkage

(Thoday, 1960) and utilization of diverse parents in breeding was also stressed by (Singh et al., 1981).

5. Conclusions
CPT-3, CPT-5 and CPT-16 found to be superior for volume index, 100 seed weight and oil

content respectively; hence seeds of these CPT may be given importance for massive afforestation

programme. The traits 100 seed weight and oil content were highly correlated with growth (volume

index) of the tree. In addition, seed breadth expressed correlation with volume index. Hence

identification of good CPTs may be graded based on seed weight and/or size and/or shape is

advantageous. Since traits viz. 100 seed weight and oil content are under strong genetic control,

improvement in these characters can bring improvement in volume index. The divergence among the

genotypes in cluster V and cluster IV as indicated by intra-cluster distance can be best used for within

group hybridization. Inter-cluster distance suggested that crosses involving under cluster II & V and

cluster II and I may result in substantial segregates and further selection for overall improvement of the

species.

As tree improvement is a continuous programme, progenies of all CPTs will further be

evaluated for total yield and oil content in future to check the pattern of character association and

selection of elite material. The present study can however serve as a pointer to be compared with the

results to be obtained at later stages especially seed and oil yield and also to establish the correlations.

This study would perspectively determine whether genetic analysis at early stage is reliable. If reliable,

genetic assessment for other population can also be carried out with suitable correlation factors the

extent of relationship can be determined

Acknowledgements

The author is grateful to National Bank for Agriculture and Rural Development (NABARD),

Mumbai for financial assistance in the form of Research and Development grants and The Director,

Institute of Forest Productivity (ICFRE), Ranchi for providing the necessary facilities. Sincere thanks

are due to CCF (Research) and field staff of Jharkhand forest department for their cooperation in

survey and identification and collection of clones.

Reference

Allard, R.W. 1999. Principles of Plant Breeding, (2
nd

edition) John Wiley & Sons, NY, pp.254.

Allen, G.S. 1960. Factors affecting the viability and germination behaviour of coniferous seed. IV.

Stratification period and incubation temperature, Pseudostuga menziesii (Mirb.) Franco.

Forestry Chronicle, 36: 18-19.

Bagchi, S.K. and Sharma, V.P. 1989. Biometrical studies on seed characters of Santalum album L.

Silvae Genet, 38: 152-153.

Burton, G.W. 1952. Quantitative inheritance of grass. Proceedings of 6
th

International Grassland

Congress held at Pennsylvania State College. Pa. U.S., 1: 74-83.

Chaturvedi, O.P. and Pandey, N. 2004. Correlation and path analysis studies between biomass and

other characters in Bombax ceiba L. Silvae Genetica, 53: 5-6.

CSIR. 1998. Wealth of India: A dictionary of Indian raw material and industrial products. Vol. VI (LM)

Council of Scientific and Industrial products, New Delhi, pp. 207-216.

Foster, G.S. 1986. Provenance variation of eastern cottonwood in the lower Mississippi valley. Silvae

Genet, 35: 32-38.

Foster, G.S. and Shaw, D.V. 1988. Using clonal replicates to explore genetic variation in a perennial

plant species. Theor. Appl. Genet, 76: 788-794.

Gera, M., Gera, N. and Sharma, S. 2001. Estimation of variability in growth characters of forty clones

of Tectona grandis. Indian Forester, 127: 639–43.

Gohil, R.H. and Pandya, J.B. 2008. Genetic diversity assessment in physic nut (Jatropha curcas L.)

International Journal of Plant Production, 2(4): 321-326.

Gomez, A.K. and Gomez, A.A. 1984. Statistical procedure for agricultural research. John Wiley and

sons, Inc.

Goulden, C.H. 1952. Some distance properties of latent root and vector methods used in multivariate

analysis. Biometrica, 53: 325-338.

Hill, J., Nelson, E. and Tilman, D. 2006. Environmental, economic and energetic costs and benefits of

biodiesel and ethanol biofuels. Proceedings of National Academic of Science, USA, 103:

11206–11210.

Johanson, H.W., Robinson, H.F. and Comstock, R.E. 1955. Estimate of genetic and environmental

variability in soyabeans. Agron. J, 47: 314-318.

Johnsen, O., Dietrichson, J. and Skaret, G. 1989. Phenotypic changes in progenies of northern clones of

Picea abies (L.) Karst. grown in a southern seed orchard. III. Climate changes and growth in a

progeny trial. Scand. J. For. Res, 4: 343–350.

Kaushik, N., Kumar, K., Kumar, S., Kaushik, N. and Roy, S. 2007a. Genetic variability and divergence

studies in seed traits and oil content of Jatropha (Jatropha curcas L.) accessions. Biomass

Bioenergy, 31: 497-502.

Kaushik, N., Kumar, S., Kumar, K., Beniwal, R. S., Kaushik, N. and Roy, S. 2007b. Genetic variability

and association studies in pod and seed traits of Pongamia pinnata (L.) Pierre in Haryana, India.

Genet Resour Crop Evol, 54: 1827-1832.

Khalil, M.A.K. 1981. Correlation of Juvenile height growth with cone morphology and seed weight in

white spruce. Silvae Genetica, 30(6): 179-181.

Divakara /Journal of Tropical Forestry and Environment Vol. 4. No 02 (2014) 11-23

Kumar, A. 1995. Genetic improvement of Casuarina equisetifolia. PhD thesis, Forest Research

Institute (Deemed University), Dehra Dun, India.

Loha, A., Tigabu, M., Teketay, D., Lundkvist, K. and Fries, A. 2006. Provenance variation in seed

morphometric traits, germination and seedling growth of Cordia africana Lam. New Forests,

32: 71-86.

Mahadevan, N.P., Sivakumar, V. and Gurudev Singh, B. 1999. Relationship of cone and seed traits on

progeny growth performance in Casuarina equisetifolia Forst. & Forst. Silvae. Genet, 48(6):

273-277.

Mahalonobis, P.C. 1936. On the generalized distance in statistics. Proceeding of National Institute of

Sciences, India, 2: 49-55.

Manavalan, A. 1990. Seedlind vigour and bioproductivity in woody biomass Species. Ph.D. Thesis,

Madurai Kamaraj University, Madurai, India.

Rao, C.R. 1952. Advanced statistical methods in biomatrics research. John Wiley and Sons, Inc.,

NewYork, 260pp.

Righter, F.I. 1945. Pinus: the relationship of seed size and seedling size to inherent vigor. Journal of

Forestry, 43: 131-137.

Robinson, J.F. and Van Buijtenen, J.P. 1979. Correlation of seed weight and nursery bed traits with 5,

10 and 15 year volume in loblolly pine progeny test. Forest Science, 35(4): 591-596.

Singh, A.K., Singh, S.B. and Singh, S.M. 1996. Genetic divergence in scented and fine genotypes of

rice (Oryza sativa L.) Ann Agric Res, 17: 163-166.

Singh, R.K. and Chaudhary, B.D. 1977. Biometrical methods in quantitative genetic Analysis. Kalyani

Publication, New Delhi, pp.596.

Singh, Y.P., Kumar, A. and Chauhan, B.P.S. 1981. Genetic divergence in pearl millet. Indian Journal

of genetics, 41: 186-190.

Skinner, D.Z., Bauchan, G.R., Auricht, G. and Hughes, S. 1999. A method for the efficient

management and utilization of large germplasm collections. Crop Science, 39: 1237-1242.

Solanki, K.R., Muthana, K.D., Jindal, S.K. and Arora, G.D. 1984. Variability, heritability and

correlation for growth parameters in Prosopis cineraria. Journal of Tree Science, 3: 86-8.

Thoday, J.M. 1960. Effect of disruptive selection. III. Coupling and repulsion. Heredity, 14: 35-49.

Veerendra, H.C.S., Ramalakshmi, S. and Mallesha, B.B. 1999. Variation in seed characterstics in

provenances of sandal (Santalum album L.). Indian Forester, 125: 308-312.

Vivekananda, P. and Subramaninan, S. 1993. Genetic divergence in rainfed rice. Oryza, 39: 60-62.