CHAPTER I

INTRODUCTION

1.1 Background Information

Brinjal (Solanum melongena L.), one of the most important

vegetable crops, belongs to family Solanaceae. In India, it is known by
baigan (HindI), Vangi (Marathi), Badanekai (Kannada), Katharikai (Tamil),
Vankai (Telagu) etc. Internationally, it is referred as Egg plant (England) or
Aubergine (France). India is regarded as the primary centre of
origin/diversity of brinjal (Bhaduri, 1951 and Vavilov, 1931) and shows
secondary diversity in South East Asia. This fact was confirmed by Isshiki
et al. (1994) based on the isozyme and morphological variation noticed in
large germplasm collection from India. The chromosome number of many
species of solanaceae under non tuberous group is fairly stable as 2x = 2n
= 24. There are 38 Asian species, which includes 22 Indian species. There
is a group of 5 related ones, namely S. melongena L., S. incum L., S.
xanthocarpum, S. indicum L. and S. maccani (Choudhury, 1976b). There
are three main botanical varieties under the species melongena
(Choudhury,1976 a).The common brinjal type which is large, round or egg-
shaped fruited forms belongs to group under var. esculentum. The long,
slender types are included under var. serpentinum and the dwarf brinjal
plants are put under var. depressum.

India is the second largest producer of brinjal in the world only

after China followed by Iran, Egypt, Indonesia, Japan, Spain, Italy,
Bangladesh & Pakistan where it is being grown extensively. It is a warm
season crop in tropical and subtropical region of India and hence it can be
grown in almost all parts of India, all the year round except in higher
altitudes. West Bengal, Orissa, Andhra Pradesh, Gujarat, Bihar, Madhya
Pradesh, Maharashtra, Chhattisgarh & Karnataka are the major brinjal
producing states in India which constitute 84.69 % production. Brinjal is
cultivated in India, over an area of 0.67 million ha, sharing 6.5 % to the total
vegetable area, with an average annual production of 12.40 million tones,
sharing 7.08 % production with 18.53 MT/ha productivity compared to 26.5

1
MT/ha of world. (Anon, 2017). In Maharashtra, it is grown on an area of
0.22 lakh ha with annual production of 4.33 lakh tonnes having productivity
19.78 MT/ha. Nagpur, Satara, Solapur, Parbhani, Pune Sangali, Bhandara,
Amrawati, Wardha Chandrapur, Latur, Nashik, Dhuley, Beed and
Aurangabad districts contribute more area and production to the state pool.

Brinjal has been a staple vegetable in our diet since ancient

times. Brinjal is highly productive and usually finds its place as poor man's
vegetable crop (Som and Maity, 2002). It is consumed as a cooked
vegetable in many ways and is liked by both poor and rich. Year round
availability, easy to cultivation, moderate to high yield and consumption in
varieties of ways i.e. salad, bhaji, stuffed brinjal, roasted, chatni, pickles
etc., has made brinjal the king of vegetables in India. Large numbers of
cultivars are grown throughout the country depending upon the consumer’s
preference for the colour, size, shape and the yield. Consumer’s preference
for shape and colour are specific which changes with region. It has also got
much potential as raw material in pickle making and dehydration industries
(Singh et al., 1963). Further, in recent years brinjal is being exported in the
form of products like baingan bhartha, chatni, pickles etc to Middle East
countries.

Brinjal fruits are rich source of minerals like calcium,

magnesium, potassium, iron, zinc and copper (Tomar and Kalda, 1998). It
is also a fair source of nutritive values (carbohydrates, proteins and fibres)
and it is used for medicinal purposes in curing diabetes, asthma, cholera,
bronchitis and diarrhea (Bose et al., 2002). It is reported to stimulate the
intrapeptic metabolism of blood cholesterol. The de-cholestrolizing action is
attributed to the presence of polyunsaturated fatty acids (lionleic and
linolenic) which are present in flesh and seeds of the fruit in higher amount
(65.1%). The presence of magnesium and potassium salts also helps in
decholesterolizing action. It is also known for the presence an alkaloid
‘solanine’ in roots and leaves. Dry fruit is reported to contain goitrogenic
principles. White Brinjals possesses several medicinal properties, which
have been known in India since antiquity.

2
 Brinjal is herbaceous annual with erect or semi spreading
habits. It is perennial plant but commercially cultivated as annual. It
develops into bushy plants with large, fuzzy leaves that grow to a height of
about 60 to 120 centimeters. The leaves are large, simple, lobed and
alternate on the stems. The stems, leaves and calyx of some cultivars are
spined. Inflorescence is often solitary but sometimes it constitutes a cluster
of 2-5 flowers. This character is dependent on the variety/hybrid. The
flowers are large, violet- colored and either solitary or in clusters of two or
more. There are five stamens which are free and inserted at the throat of
corolla. The fruit is pendent and fleshy berry borne singly or in clusters. The
shape of the fruit varies from ovoid, oblong, obovoid, or long cylindrical.
The colour of the mature fruit varies from mono-colored purple, purple
black, yellowish, white, green and variegated types of purple with white
stripes, green with light green/white stripes or even combination of three
colours. The number of seeds per fruit varies from few to many. The seed
color is white, light yellow, brownish yellow, brown to black brown for
different varieties. Anthers are cone shaped, free and with terminal
dehiscence. Ovary is hypogynous, bicarpellary, syncarpous and with basal
placentation.

In brinjal, heterostyly is a common feature. Four types of

flowers have been reported depending upon the length of style, viz., (i)
long-styled with large ovary, (ii) medium-styled with medium size ovary, (iii)
pseudoshort- styled with rudimentary ovary and (iv) true short-styled with
very rudimentary ovary. Long and medium styled flowers produces fruits
whereas pseudo-short and short-styled flowers do not set fruits. Further,
chances of cross pollination are more in long style flowers. All varieties
have flowers with different style length. The position of the stigma in
relation to stamens varies with the cultivars and can also vary in different
flowers of the same cultivars.

Brinjal is usually self pollinated, but the extent of cross-

pollination has been reported as high as 48% and hence it is classified as
often cross pollinated crop. The cone-like formation of anthers favors self
pollination but since the stigma ultimately project beyond the anthers, there

3
is an ample opportunity for cross pollination. The cross pollination depends
on the pollinating insects such as bumblebees (Bombus sp.), wild bees
(Exomalapis sp., Xylocopa sp., Anthophora sp.) and domestic bees (Apis
sp.)

1.2 Importance and need of study

The productivity of brinjal is considerably low in India. Among

the many reasons behind it, use of poor yielder cultivars and incidence of
pest and diseases are prominent. The present production is not in
commensurate with the demand by the burgeoning population at geometric
rate. So, brinjal deserves a deep contemplation for improvement for the
reasons cited above. It should be highly pragmatic by the fact that India
being the centre of origin and diversity of brinjal, it should pave the way to
bring about a kind of plant architecture, which could enhance its quality and
productivity without tossing the consumers’ requirements. There are
specific genotypes suited for specific preparation apart from the large
genetic variation observed with regard for traits like colour, shape and size
of fruits. In addition, variation is also noticed for traits like vegetative
growth, maturity and presence or absence of spines on leaves, stem and
fruit calyx among the indigenous material.

In India, consumer’s preference for the colour, shape and size

of fruits varies state wise as well as region wise. In Maharashtra, western
and central region mostly prefers round fruits having purple colour with
white stripes and spines on calyx; while eastern Vidharbha area mostly
prefers spineless round and green colored fruits. Though improved and
hybrid varieties of brinjal are being consumed, but due to consumer’s
preference towards habitual taste, still local cultivars are being cultivated.
Therefore there is need to improve the productivity and quality with pest
and disease resistance of such preferred local cultivars.

It can be achieved by using appropriate breeding methods

with considering the vast base of plant architecture. One of the best
methods employed is exploitation of hybrid vigour through hybridization.
For the first time, Bailey and Munson (1892) reported artificial hybridization
in brinjal. Nagai and Kida (1926) were probably the first to observe hybrid
4
vigour, hoping some commercial acceptance in crosses among some
Japanese varieties. Since then many public and private sectors have
developed various hybrids in India, but these hybrids lacked regional
preferences for colour, shape and presence or absence of spines and
lacked suitability to specific product preparations

Brinjal, due to hardy nature and comparatively large size of

flowers and large number of seeds in a single act of pollination, is always
the breeder’s choice for exploitation of heterosis. It is also due to highly
varied consumer acceptance from region to region which creates the
demand for high yielding F1 hybrids. In India, Pal and Singh (1946), Mishra
(1961), Sambandham (1962), Dhankar et at. (1980) and Chadha and Sidhu
(1982) have observed the hybrid vigour over their parents, which becomes
the potential tools for improvement in brinjal.

In genetic improvement, selection of suitable parents is

important for development of better hybrids. The per se performance of
parental lines provides clues, however, reliable information on magnitude of
heterosis, combining ability of parents for yield and its component
characters and gene action involved in the inheritance in different
characters are more helpful in selecting appropriate parents and desirable
cross combinations for commercial exploitation of hybrid vigour.

Different mating designs have been used by different workers

as an aid in the choice of parents and to understand their genetic worth.
Line x Tester analysis was suggested by Kempthorne (1957) to elucidate
the nature of gene action and combing ability of parents for different
characters. Line x Tester analysis is a useful technique for screening large
numbers of lines for identifying the best combiners. Similarly knowledge
about nature of gene action governing the expression of various traits helps
in determing the strategies to be adopted.

Estimation of genetic parameters is needed to understand the

genetic architecture of yield and yield contributing components. Information
about type of gene action, heterosis and combing ability for all the yield
contributing traits would be of immense help for a plant breeder to decide

5
about the proper breeding procedure to be adopted and the characters on
which the selection has to be made.

Considering the importance of heterosis and combing ability

studies in improvement of brinjal crop, the present investigation “Genetic
analysis of F1 hybrids in brinjal” was carried out.

1.3 Objectives of study

1. To estimate heterosis for yield and its component traits.

2. To estimate GCA and SCA effects for yield and its component traits.

3. To study the gene action of best cross combination for higher yield and
its component traits.

1.4 Hypothesis or assumptions

The assessment regarding “Genetic analysis of F1 hybrids in

brinjal” will help for understanding the inheritance pattern and other
genetical aspects of brinjal under selected genotypes and predicting
breeding procedure which will prove as a useful tool in brinjal improvement
for early maturity, higher yield and better quality. Heterosis results in
increase in earliness, higher yields, better quality fruits, etc. in F 1 hybrids
compared to parents. Existence of wider variability offer greater potential to
obscure hybrid vigour in this crop.

Combining ability analysis provides the information for

selection of desirable parents and cross combination for exploitation. In this
analysis, total genetic variation is partitioned into GCA and SCA effects to
verify the parents in terms of combining ability to combine in hybrid
combination. Combining ability is the ability to transmit the superior
performance to its progeny resulting hybrids.

1.5 Scope and limitation of the study

In India, mostly local types of brinjal are being cultivated all

over the country. In Maharashtra also a large number of local types are
being cultivated. During recent years some improved types have been
developed by Coordinated Centres and state Agricultural Universities from
wide spectrum of genetic variability in brinjal from their local types and

6
recommended for commercial cultivation. These improved varieties and
local types are lower in yield than the potential of this crop. This generates
need to improve these genotypes or to develop hybrids superior to these
types in yield and other characters. In recent years breeding for regional
specificity has acquired the importance in brinjal improvement. Several
studies in brinjal crop have revealed the evidence of non-allelic interactions
for economically important characters.

In brinjal, wide range of variability is available in fruit colour,

shape and size, which have created region-wise wide range of different
consumers preference and thus it offers great potential for exploitation of
hybrid vigour. Being an often cross pollination and large flower size, it
produces large number of seeds per fruit .Therefore the technology of
hybrid seed production and seed cost of F1 seeds, is comparatively
cheaper to other vegetable crops. Thus heterosis breeding will always have
great scope in brinjal improvement for early maturity, higher yield and
better quality which are the major advantages of brinjal. However,
limitations in hybridization work like dehiscence of pollen, heterostyly and
wide range of acceptance limits the exploitation of hybrids in brinjal.

Combining ability studies helps in identification of better

combing parents and the best specific cross combinations for their further
utilization in any breeding programme. However, the cross combination as
identified has to be tested on large scale before their exploitation at
commercial level.

7
 CHAPTER II

REVIEW OF LITERATURE

Brinjal breeder is primarily concerned with the improvement of

crop characters both quantitative and qualitative. Hence, adequate
knowledge of genetics of various traits of brinjal is very essential for
obtaining desired results in the succeeding generations. The magnitude of
such a success to be obtained lies on the selection from the base material
and creative manipulation.

Therefore, the study of heterosis, combining ability and gene

action is atmost important in the evaluation and selection of varieties and
deciding breeding method for the crop under study. Lot of work on
heterosis, combing ability and gene action in brinjal improvement has been
carried out.

Keeping in view the objectives of the present investigation,

relevant literature was reviewed and presented in the following headings.

2.1 Heterosis

2.2 Combining ability

2.3 Gene action

2.1 Heterosis

The term heterosis was first coined by Shull in 1914.

Heterosis may be defined as the F1 population obtained by crossing of two
homozygous inbreds of genetically unlike parental constitution showing
superiority or inferiority over both the parents in both quantitative and
qualitative traits. This increased productivity or superiority of the hybrids
over parents is known as hybrid vigour and the genetic cause of this
phenomenon is known as heterosis. It may be relative heterosis
(improvement over mid parent), heterobeltiosis (improvement over better
parent) and standard heterosis (superiority over commercial check).
Various theories has been put forth to explain the genetic basis of heterosis
such as favourable expression of heterozygosity and accumulation of
favorable dominant alleles in the F1 contributed by both male and female

8
parents. Moll and Stuber (1974), reported that partial to complete
dominance, over dominance and the epistasis are the three possible
genetic causes of heterosis. The exploitation of heterosis in cultivated
plants becomes a potential tool in crop improvements. At the National level,
first hybrid vigour appeared during 1933 in chilli at Indian Agricultural
Research Institute (IARI), New Delhi. The first commercial exploitation of
heterosis for high yielding hybrid in vegetables was in bottle gourd through
release of ‘Pusa Meghdoot’ and Pusa Manjari’.

The earliest recorded instances of artificial hybridization in

eggplant were evidently those carried out by Bailey and Munson in the
United States; however none of the hybrids exhibited heterosis but were
intermediate between the parents (Bailey and Munson, 1891). The first
positive report of heterosis in the eggplant came from Munson (1892).
Subsequently, Halsted (1901) reported that one of his cross was double the
size of the parents and also yielded more.

In the Philippines, Bayla (1918) hybridized some local

varieties of brinjal and found that the hybrids were more vigorous, stronger
and healthier than the respective parental lines.

In Japan, Nagai and Kida (1926) studied certain quantitative

characteristics in the brinjal hybrids and found that heterosis was
manifested in total yield and its traits. Tatesi (1927) also observed higher
productivity in certain crosses between Japanese brinjal varieties.

Kakizaki (1928), Schmidt (1935), Averjanova (1941) and)

Daskaloff (1941) also reported hybrid vigour for different characters in egg
plant.

In India, the first attempt to hybridize eggplant appeared to

have been made by Rao in 1934; however, in the cross between two wide
varieties, a high degree of partial sterility due to abortive pollen was
observed.

Pal and Singh (1946) reported that the majority of brinjal

hybrids exhibited heterosis with respect to seed germination, plant height,

9
plant spread, number of branches, early flowering, number of fruits per
plant, fruit size and yield.

Venkataramani (1946) reported that the hybrid eggplants

were taller, spread more, flowered earlier than the parents and yielded
more than either parent.

Odland and Noll (1948) experimented with sixteen hybrid

types and recorded that in every case the hybrids have out-yielded their
respective parents besides being early.

Capinpin and Alviar (1949) reported that hybrid seeds

exhibited higher germination percentage, the hybrid plants were superior to
the parental lines in early flowering and fruit setting, greater number of
fruits per plant, longer fruit, greater mean of equatorial diameter of fruits
and greater mean weight of the fruits.

Randhava et al. (1973) studied four varieties naming P-8,

Pusa purple cluster, Pusa purple long and S-1 to investigate the heterosis
over mid and better parents in brinjal. Maximum heterotic effect was
observed in the cross PPC x PPL in respect of fruit yield, number of fruits
per plant, fruit weight and number of primary branches, while the cross
PPC x S1 in the traits like fruit weight, yield per plant, number of primary
branches, number of fruit, fruit size and height of plant respectively.

During their studies, Singh and Nandpuri (1974) revealed

that, in a 7 x 7 egg plant diallel excluding reciprocals, F 1 hybrids exhibited
heterosis over the better parents for plant height, days to flowering, fruit
length, fruit width and yield per plant.

Vijay and Nath (1978) observed heterosis in F1s over better

parents for yield and days to flowering, whereas, Wadnerkar (1978)
reported the heterosis for yield in F1s over better parents up to 115.45 % in
cross SM-2 X HBR-112 and also observed significance heterosis in the
characters viz., plant height (28.5%), early maturity (33.33 %), length of
fruit (28.5 %), diameter of fruits (4.7%) average weight of fruit (14.28%) and
number of fruits per plant (9.52%) while studying seven parents and
hybrids in brinjal.

10
 Ram et al. (1981) revealed that, eight crosses had showed
significant positive heterosis over superior parent for plant height in brinjal.
Three crosses were better over superior parent and one cross over best
parent showed significant heterosis for plant spread.

Patil and Shinde (1984) studied heterosis in brinjal and

reported heterosis for fruit yield per plant as 53.3% followed by 46.7% for
fruits per cluster. Flowers per cluster, fruit girth, fruit per plant and days to
flower showed significant better heterosis.

Sidhu and Chadha (1985) reported that, the mean of hybrid

was higher than parents for all the characters other than height of plant,
number of branches, number of fruits and yield. The increase in yield over
their respective better parents in heterotic hybrid ranged from 7.42 to
45.71%. In all the characters, the best performing hybrid was better than
the top parent except weight of fruit, number of fruit and yield.

Sadawarte et al. (1993), found the presence of highly

significant heterosis over better parent for majority of the yield and yield
contributing characters while studying 11 X 11 diaallel crossing in brinjal.

Ponnuswami et al. (1994) reported high per se performance

of hybrids and parents for number of branches per plant, percentage of
long styled flower, fruit weight, fruit length and yield.

Singh and Krishnaprasad (1995) revealed that, five hybrids

over better parents and seven hybrids over the mid parents showed
positive and significant heterosis for yield per plant in brinjal. The better
parents and mid parent heterosis ranged from 3.09 to - 82.57 % and from
4.34 to -130.37% respectively. Significant values of heterosis over better
parents as well as mid parent were also observed for other attributes,
indicating their contribution to the heterosis for plant height.

Ingle and Patil (1997) recorded significant positive heterosis

over mid parent in 10 x 10 half diallel analysis in brinjal.

Patil (1998) observed highest range of crosses over their

respective mid parent value range from -8.64 to 14.62%, while -20.58 to
18.69 % over commercial check. None of the cross exhibited significant

11
positive or negative heterosis over mid parent or commercial check for
plant height. The expression of heterosis over mid parent in desired
direction was revealed in 46 crosses for number of branches per plant,
totally six cross combinations manifested significant negative heterosis
over commercial check. For days to 50% flowering, 25 crosses exhibited
significant positive heterosis over mid parent in fruit length; however with
respect to fruit diameter significant positive heterosis over mid parent was
recorded in seventeen crosses, for number of fruits per plant significant
positive heterosis in thirteen brinjal crosses over commercial check.

Prasath et al. (1998) reported that nine crosses were

significantly earlier in flowering than that of parents and for fruit weight,
crosses in generally showed positive heterosis. Only nine crosses showed
negative heterosis out of 30 F1 hybrids. Significant Positive heterosis in
brinjal hybrids viz., Arka Nidhi x MDU-1, 202 x Annamalai and 190 x
Annamalai over better parent for number of fruits per plant. However
significant heterobeltiosis for yield per plant in different F1s, ranged from
8.34 (SM6-6 x Annamali) to 101.86 % (H-0 x MDU-1) was recorded.

Kumar et al. (1999) noticed both negative and positive

heterosis in 12 F1s hybrids of brinjal.

During the year 2000, Anuroopa reported seventeen hybrids

in brinjal shown significant positive heterosis over mid parent, five hybrids
expressed significant heterobeltiosis. Significant positive heterosis over
standard check was shown by twenty hybrids for plant height, for fruit
weight significant positive heterosis in five brinjal crosses over mid parent
and only one hybrid showed positive significant heterosis for fruit diameter.
Among 32 brinjal hybrids, none of them exhibited significant and positive
heterobeltiosis for both the traits. Whereas, for fruit length, eleven crosses
had shown significant and positive heterosis over the mid parent ranging
from 15.05 to 33.99 per cent. The magnitude of heterosis for yield showed
significant positive heterosis over mid parent in 14 brinjal hybrids and none
of the crosses expressed significant positive heterosis over the best parent.

Bulgundi (2000) revealed that 15 crosses showed significant

plant height over mid parent, but 10 of them exhibited significant positive
12
heterosis. The average heterosis was in negative direction indicating that
brinjal hybrids flowered earlier than their parents. Among 30 crosses, 15
exhibited significant negative heterosis, for fruit weight, three hybrids out of
thirty showed significant positive heterosis, while, for fruit diameter
significant positive heterosis was recorded in 10 F1s and six crosses
showed significant negative heterosis. Significant positive heterosis for
number of fruits per plant in fourteen crosses, while equal number of
crosses exhibited significant negative heterosis over mid parent. An
appreciable amount of heterosis in F1s over mid parent value was prevalent
for yield. Fifteen crosses exhibited significant positive heterosis in brinjal.

In 2001, Chadha et al. observed both negative and positive

heterosis for number of branches in brinjal hybrids.

Bavage (2002) showed that the heterosis for plant height over
commercial check was observed in the range of -12.43 to 30.47% with four
crosses exhibiting significantly high vigour (positive heterosis) over
commercial hybrid Kalpataru, for number of branches per plant. Out of 28
crosses in brinjal, 26 over mid parents and 24 over better parents exhibited
heterotic effect, of which except three crosses, all had heterosis in desired
direction (positive heterosis). Per se performance in brinjal for days to 50 %
flowering ranged from 89 to 111 days and from 87.33 to 112.67 days for
parents and hybrids respectively. The magnitude of the heterosis for fruit
length over mid parent, better parent and commercial check ranged from -
11.90 to 40.69, -10.59 to 52.24 and -5.85 to 47.82% respectively, for fruit
diameter 23 crosses exhibited heterosis over mid parent, 16 over better
parent and 12 over commercial check significantly, while, for the number of
fruit per plant, 8 crosses over mid parent, 6 crosses over better parent and
13 crosses over commercial check had significant positive heterosis. None
of the crosses exhibited significant heterosis over mid parent and better
parent for yield. However three brinjal crosses have exhibited high positive
heterobeltosis.

Das and Barua (2003) reported that, crosses JC-2 x JC-4,

JC-1 x JC-2, JC-2 x JC-6 and JC-4 x JC-6 exhibited significant heterosis for
yield and most of the yield contributing characters, JC-2 xJC-4 was found

13
to be most heterotic cross for yield, fruit weight, fruit per plant, days to 50%
flowering and days to flowering.

Kanthaswami et al. (2003) observed -54.72 to 57.18%

heterosis of fruit borer infestation over better parents in his study on 28 F 1
hybrids

During studies on double crossing in brinjal, Karaganni (2003)

observed significant positive heterosis for plant height over better parent.
For number of fruits per plant, six eggplant crosses had significant positive
heterosis in desirable direction and five DC F1s were negative over mid
parent. All 15 DC F1 had showed significant positive heterosis over local
check.

Pratibha et al. (2004), during the studies on heterosis in

brinjal, observed significant positive heterosis over the better parent for
fruit length, maximum heterosis for fruit diameter in cross T-3 x PB-45
(96.8%) followed by PB-33 x PB-54 (92.2%) and T-3 x KS-331 (88.7%)
over Pant Samrat (SP2). None of these crosses showed positive heterosis
over Pant Rituraj (SP1). Maximum Positive and significant heterosis for the
brinjal cross Pant Samrat x White Brinjal (349.6%, 68.6%) followed by Pant
Samrat x KS-331 (259.8, 34.9%) over both the standard parents (SP1 and
SP2) for number of fruits per plant. Standard heterosis for yield per plant
varied from 49.1 to 27.8 % over SP1 and -39.8 to 51.3% over SP2 and the
cross DBL-11 x PB-33 (27.8%, 51.3%) showed maximum standard
heterosis followed by Pant Samrat x KS-331 (26.7%, 50.1%) and PB-33 x
PB-30 (23.3%, 45.9%) in brinjal.

Singh et al. (2004) found both negative and positive heterosis

for number of branches per plant over better and mid parents in eggplant.

Ashwani and Khandelwal (2005) noticed that the cross

combination P1 x P4 have recorded significantly high heterosis for days to
first flowering in brinjal.

Bavage et al. (2005) reported significant negative heterosis

for days to 50 per cent flowering whereas; heterosis ranged from -10.26 to
0.63% over better parent in F1 brinjal hybrids.

14
 Prabhu et al. (2005) observed the maximum positive
heterosis in the hybrid EP12 x MDU-1 over better parent and standard
check for plant height, number of branches per plant and number of fruits
per plant was high in the parents EP65 and APAU Bagmathi in brinjal. The
hybrid EP12 x MDU1 showed significantly higher mean values and the
hybrid EP65 x Pusa Uttam showed superiority based on mean performance
as well as standard parent heterosis. The significant high mean values and
heterosis for fruit weight over better parent were found in the brinjal hybrids
EP12 x MOU1 and MDU-1 x Surya. Negative heterosis over better and
standard parents was observed for fruit weight. Three brinjal hybrids
namely EP65 x Pusa Uttam, EP12 x MDU-1 and MDU-1 x Surya exhibited
negative heterosis over better and standard parents for fruit weight. Three
brinjal hybrids namely EP65 x Pusa Uttam, EP12 x MDU-1 and EP104 x
APAU Bagmathi, exhibited heterosis over standard parent Co-2. EP65 x
Pusa Uttam showed significant heterosis over better parent. He also
observed the heterosis with the range -27.23 to 1.82 per cent, -21.76 to
9.08 per cent of heterobeltiosis and -53.97 to 34.82 per cent of standard
heterosis in fruit borer infestation.

Singh and Maurya (2005) revealed that, most of the eggplant

crosses manifested significant heterosis over better parent, mid parent and
check parent for fruit yield per plant.

Ajjappalavara (2006) noticed higher positive heterosis for

growth parameter in the cross Melavanki spine x DWD-3 over mid and
better parent. Whereas, over standard check nine crosses showed
negative heterosis, minimum days for 50 per cent flowering was in DWD-3
among parents, whereas, DWD-1 x DWD-3 was recorded minimum days
for flowering among brinjal hybrids, highly significant heterosis over mid
parent for number of branches per plant by Kuduchi x DWD-2, whereas,
the cross DWD-1 x Malapur showed significant heterosis over standard
check, eleven brinjal crosses were in positive direction of which the cross
DWD-1 x Rabakavi was showed highest and significant heterosis over mid
parent for fruit length. For fruit diameter, eleven, ten and two brinjal crosses

15
were in positive heterosis over mid parent, better parent and standard
check respectively.

Prakash et al. (2006) studied on heterosis in brinjal and

reported that, significant heterobeltiosis in desired positive direction
observed by seven hybrids. All the crosses exhibited significantly positive
heterosis over check for fruit weight and yield.

In 2007, Shafeeq et al., while studying on heterosis in brinjal,

observed significantly higher number of branches at peak flowering in 10
crosses among 24 hybrids. The crosses involving Mullu Badane and Arka
Sheel parents exhibited significantly higher number of branches at peak
flowering. Per se performance of days to 50 per cent flowering, ranged
from 43 days (Arka Shirish) to 56 days (Arka Nidhi) in parents and 36.50
(Arka Nidhi x Malapur local) to 49.50 days (Arka Shirish x Green round) in
hybrids. Among the 24 crosses, 22 crosses expressed significant
superiority for fruit length while, all crosses showed positive heterosis over
commercial check. Whereas, significant heterosis over commercial check
was recorded in 18 hybrids with all showing positive heterosis for fruit
diameter.

Suneetha et al. (2008) noticed that, fruit yield per plant had
high heterobeltiosis to an extent of 114.43 per cent and standard heterosis
of 38.77 per cent. High heterobeltiosis and standard heterosis (>60 per
cent) were also noticed for all the quality traits in brinjal.

Das et al. (2009) reported positive relative heterosis and

heterobeltiosis for plant height and primary branches per plant in brinjal. In
case of 50% flowering, where negative heterosis was desirable, all the
hybrids showed significant relative heterosis and ten hybrids showed
negative heteobeltiosis. In case of fruit length and fruit girth, both positive
and negative heterosis was observed.

During studies on heterosis in brinjal, Shanmugapriya et al.

(2009) observed five hybrids namely, L7 (White Brinjal) x T1 (Annamalai),
L4 (Mullukathri) x T1 (Annakalai), L3 (Kunnam) x T1 (Annakalai), L2
(Putheri) x T1 (Annamalai) and L7 (White Brinjal) x T2 (PLR 1) recorded

16
more than 50 per cent heterobeltiosis for fruit yield and number of fruits per
plant, more than 20 per cent for number of branches per plant and more or
less 10 per cent negative heterobeltiosis for days to first flowering and plant
height.

Chowdhary et al. (2010) carried out a study to estimate

magnitude of heterosis over better parent and standard check for some
important characters in 15 crosses, resulting from a half diallel mating
design of 6 inbred lines of brinjal with F1 standard check Tarapuri.
Significant levels of heterosis were observed for all the traits studied.
Promising hybrids exhibited significant positive heterosis for fruit yield,
magnitude of which ranged 9.6 to 74.89% and 8.52 to 72.60% over better
parent and standard check respectively. Some of the promising hybrids
showed desirable negative heterosis for earliness, positive heterosis for
increased fruit number and yield.

Dharwad et al. (2011) studied on heterosis in brinjal and

observed that IC-112 x IC-909, IC-112 x IC-997, IC-112 x IC-996, IC-111 x
IC-996 and IC-111 x IC-136 crosses exhibited high heterosis for yield and
yield components viz., number of branches per plant, fruit weight and
number of fruits per plant while studying heterosis in brinjal.

Murthy et al. (2011) studied heterosis in brinjal and reported

that among the 12 hybrids,7, 4 and 1 hybrid exhibited significantly positive
heterosis for fruit yield per plant over mid parent, better parent and
commercial check respectively. Maximum positive significant heterosis was
observed in cross Arka Shirish x WCGR (84.93%).The hybrid Mattugula x
WCGR exhibited highly significant heterosis for number fruits/cluster and
number and number of flowers per cluster while the same hybrid Mattugula
x WCGR showed significant heterosis for average fruit weight.

Biswas et al. (2013) evaluated sixty hybrids along with ten

parents for heat tolerance and earliness during summer season. Highest
standard heterosis was shown by IBWL x PPC (46.86%) followed by GL x
PPL (46.13%) for total yield per plant.

17
 Makani et al. (2013), in his study on brinjal, estimated the
magnitude of heterosis for yield and its eleven yield components. He found
appreciable heterosis in 28 F1’s hybrid over mid, better and standard
parent for all the traits studied in desirable direction. F 1 hybrid AB-07-08 x
GP-180 (136.39%) followed by AB-07-08 x KS-331 (102.20%) and NBD-18
x AB-07-08 (97.63%) were observed significant heterosis over mid parent
while the maximum heterobeltiosis for fruit yield per plant was exhibited by
the hybrid AB-07-08 x GP-180 (125.78%).In case of standard heterosis,,
significant and positive heterosis over standard check GBL-1 for fruit yield
per plant was observed in hybrid GBL-1 x KS-331 (50.41%).

Reddy and Patel (2014) conducted a heterosis studies using

line x tester mating design comprising of 5 lines and 4 testers, observed
high heterosis response in most of the hybrids of brinjal. The maximum
heterosis for fruit yield per plant and number of fruits per clusters were
observed by cross AB-8/5 x GJB-2 (103.59 %). Whereas, hybrid JRB-6/7 x
GJB-2 showed significant heterosis (17.53%) for average fruit weight.
Some of the promising hybrid showed desirable heterosis for earliness,
number of flowers per cluster and phenol content.

2.2 Combining ability

The selection of most appropriate parents for hybridization

programme could be made on the basis of their ability to transmit desirable
trait to their crosses (Sprague and Tatum, 1942). The combing ability was
powerful tool to discriminate for SCA and GCA. It helps in choosing
appropriate parents and also provides the nature and magnitude of gene
action to decide breeding action and methodology for the crop.

With this method, the resulting total genetic variance is

partitioned into effect of general and specific combing abilities. The general
combing effect is defined as the average performance of the line in a series
of crosses, which can be recognized as a measure of additive gene action
and specific combining ability as a deviation from expectation on the basis
of average performance of a line in a series of crosses, which can be
recognized as a measure of dominance.(i.e. deviation from expectation on
the basis of average performance of lines involved in the hybrid
18
combination) and is a measure of non additive gene action. SCA effect help
to sort out superior cross combinations and or could be utilised further in
breeding programme to exploit transgressive segregants.

Brief review of work pertaining to combining ability is

presented below.

Griffing (1956) presented a model to show that variance for

gca involved mostly additive genetic effects whereas, sca resulted from
dominance (intra-allelic) and epistatic (inter-allelic) components of genetic
variance.

Kempthorne (1957) discussed gca and sca variances in terms

of covariance of half and full sibs in a random mating population.

Where,

σ2gca = Cov (H.S.) and

σ2sca = Cov (F.S.) - Cov (H.S.)

Where, Cov (H.S.) = covariance between half sib

Cov (H.S.) = covariance between full sib

Iqbal et al. (1986) done combining ability analysis for fruit

yield and its components in a diallel set involving four varieties of eggplant.
Mean squares due to general combining ability (GCA) were highly
significant for all the characters, while mean squares due to specific
combining ability (SCA) effects for number of fruits per plant, fruit length,
fruit yield per plant and plant height were highly significant. Both GCA and
SCA effects contributed significantly to fruit yield and its components. SCA
effects were relatively of greater importance than GCA effect.

Singh and Kumar (1988) studied diallel cross involving 5 long-

fruited brinjal varieties and revealed that ARU1 and Sel-5 were the best
general combiners for yield. The cross Pusa Purple Cluster x Sel-5 was the
best specific combination for yield.

Chadha and Hedge (1989) derived combining ability from

data on 8 characters in 9 brinjal (Solanum melongena) lines and their F1
hybrids. H4 was a good general combiner for yield and other characters.

19
The crosses Pusa Purple Cluster x Pusa Kranti, Pusa Purple Cluster x
Punjab Chamkila and Sultanpur X H4 were good specific combinations for
yield.

Patil and Shinde (1989) studied general and specific

combining ability variances and effects for six characters studied involving
seven parents of eggplant in all possible combinations excluding
reciprocals. General combining ability (gca) and specific combining ability
(sca) variances were highly significant for all characters under study. The
gca variances was higher for all the characters, suggesting the
predominance of additive gene action. The gca effects indicated that none
of the parents was a good general combiner for all characters, suggesting
that separate parent will have to be used for improvement of different traits
studied. The use of varieties Pusa Purple Long and Dorly is recommended
for breeding programme, as they have high yielding performance. The
predictability ratio is near unity for fruits/plant, followed by leaf area and
days to flowering, suggesting greater importance of additive genetic
variance for these characters.

Joshi and Chadha (1991) conducted study on 7 X 7 diallel

cross involving homozygous aubergine lines (3 round-fruited (PTR, P8,
BR11), 1 small oval-fruited (SM17-4) and 3 long-fruited (7-3, PPC, PPL)
during spring 1987. The resulting 21 hybrids, and the 7 parents, were
grown during autumn 1987 and spring 1988. The best combiners for yield
and its components were recorded. Results suggested that it was not
possible to improve fruit weight and fruit number simultaneously. The
crosses PTR X PPL, PTR X SM17-4, PTR X PPC and BR11 X SM17-4
exhibited highly significant SCA for total yield in autumn 1987. It was
concluded that the crosses 7-3 X PPL and 7-3 X PPC would be of use in
breeding.

Kale et al. (1992) while studying 11 x 11 diallel cross set in

brinjal found that variance due to SCA was higher than their respective
GCA variance components for the characters plant height, spread days to
first harvest, yield/plant (kg), no. of fruits per plant and weight of fruits.

20
 During the genetic studies in brinjal, Sadawarte et al. (1993)
reported significant gca effect for yield and yield contributing characters.
Cross combinations viz., Jalgaon local x HBR-112, Pusa purple round x
SM-2, Pusa purple cluster x Banaras giant white and Jambhala doria x
HBR-112 with maximum sca effect viz. 13.33, 10.83, 8.48 and 7.25
respectively for yield per plant in environment 3.

Padmnabhan and Jagdish (1996) studied combining ability

analysis for yield and fruit characters in eight-parent dialllel cross without
reciprocals of round fruited brinjal and their 28 F1 hybrids and found both
additive and non-additive gene actions for the all characters, except for
yield where non additive gene action alone was exhibited. Among the
parents, Cluster White was found to be good general combiner for early
flowering, dwarf plant type and fruits per plant and the variety, Hyderabad
Local for early flowering and the number of secondary branches. The
crosses Manjari Gota X Mettavanga, Black Beauty X Srikakulam and
Cluster White X Mettavanga were superior in yield per plant.

During their studies on combining ability analysis in brinjal,

Kumar et al. (1996) evaluated the F1 hybrids of a three lines x four testers
and SM-6 was found good general combiner for days to 50 per cent
flowering and additive gene effects were significant for fruit weight in all the
crosses and were higher in magnitude than dominance gene effects.

In 1998, Patil observed the lines, SL (6.10) and P (-5.25)

were found to possess significantly positive and negative gca effects
respectively. Out of the 10 testers, only two showed significant gca effects
in opposite direction for plant height. Highly significant gca effects for two
lines, one each in positive desirable direction (RL) and negative non-
desirable direction (P) The number of crosses which displayed significant
positive and negative sca effects was seven and nine respectively for
number of branches per plant, for days to 50% flowering, significant sca
effect for 14 crosses, among which, the proportion of crosses showing
negative sca was equal to the crosses exhibiting positive sca effect. 23
crosses registered significant positive sca effects and none of the crosses
excelled P x CP2 (15.41) for fruit weight. The parental combination involved

21
in this cross was negative x non-significant gca effect. For fruit length 24
crosses gave significantly positive sca effect and the rest were negative in
specific effect. The majority of the significant positive sca effect crosses
involved at least one of the parents with significant positive gca effect
specially in fruit diameter. The proportion of crosses with significant
negative sca effects were twelve, among the 12 crosses there were eleven
crosses which were at par with each other excepting P x MK (-0.41), the
higher value (-0.78) being displayed by KL x CP1 for yield per plant.

Bulgundi (2000) noticed significant positive gca effects in two

lines and one tester out of 11 parents of brinjal, while two lines and one
tester revealed significant negative gca effects in plant height. Whereas,
positive and negative sca effects in six hybrids, out of 12 hybrids,
manifesting significant sca effects for number of branches per plant. The
cross MG x W-8, which was early in flowering, had parents combination of
negative x significant positive gca effect. Significant positive sca effect for
fruit weight in 11 crosses and they ranged from 5.99 to 24.17. Six crosses
showed significant negative sca effect with highest being noticed in RL x R
cross (-23.40). Whereas, out of total 14 crosses displayed significant sca
effects, seven crosses each exhibited significant positive and negative sca
effects for fruit length, four lines expressed significant gca effect, two each
in positive desirable direction (ML and Mg) and negative undesirable
direction (P and KR) for fruit diameter and the tester C1 was the best
positive general combiner (0.33), among the give testers for same trait. 24
hybrids displayed significant sca effects for number of fruits per plant. Ten
hybrids showed significant positive and negative sca effects for yield per
plant and Hybrids RL x TL and MG x W-8 had parent combination of
significant positive x significant positive gca effects.

Chaudhary et al. (2000) conducted combining ability studies

on brinjal to isolate desirable parents and F1 cross combinations for yield
and physiological growth parameters. Both additive and non-additive types
of gene effects were observed to be operative for various characters.
Parental lines SM6-6, Punjab Barsati, Hisar Shyamal and Arka Nidhi were
observed to be good general combiners for yield and other characters. The

22
mean squares for SCA were greater than those for GCA in most of the
characters.The superior performance of F1 hybrids showing high SCA was
largely due to epistatic interaction. Punjab Barsati x SM6-7, SM6-6 x Hisar
Shyamal and Arka Nidhi x Punjab Barsati were the best specific crosses for
fruit yield per plant and different growth parameters.

During the year 2002, Bavage observed the positive sca

effects for thirteen crosses in brinjal, out of which two crosses were
significant. All the crosses with negative sca effects were non-significant
except G3 x G4 for plant height and five hybrids exhibited significant
positive sca effect and three crosses exhibited significant negative sca
effect for number of branches per plant. Also reported that the sca effect
were significant for eight crosses, out of which three crosses had useful
negative sca effects and 5 crosses had positive specific combining effects
for days to 50% flowering, only 8 crosses have exhibited favourable
significant positive sca effects against 11 crosses with significant negative
sca effect for fruit weight and for fruit length only two crosses exhibited
significant positive sca effect. While none of the parents and hybrids
exhibited significant gca and sca effect respectively for fruit diameter. For
number of fruits per plant ten and thirteen hybrids exhibited significantly
positive and negative sca effects respectively. The sca effects were
recorded significant for twenty four crosses, out of which thirteen were
positive and eleven were negative with respect to yield per plant.

While carrying out studies on double crosses in brinjal,

Karaganni (2003) noticed a good amount of mean performance highly
significant gca effects coupled with moderately high GCA variance and
SCA variance were manifested by parent SCHN2 for plant height, for fruit
weight high significant gca effect (4.22) comparatively high GCA variance
(17.77) and moderately high SCA variance and significant negative gca
effect for fruit diameter and also for number of fruits per plant high mean
performance, high significant positive gca effect, high GCA variance and
comparatively low SCA variance. However, SCHN3 and SCHN5 reported
high mean performance, high GCA variance, high SCA variance and

23
significant negative gca effect proved to be a potential specific combiner
useful to transgressive breeding for yield per plant.

Singh and Maurya (2003) estimated Combining ability effects

for different characters in a line x tester crossing programme of brinjal
comprising 27 hybrids produced by crossing of nine lines and three testers.
Parents and hybrids/crosses differed significantly for gca and sca effects,
respectively. The parents JB-15, DBL-24, KS-352, Green round and Green
long were found good general combiners for most of the characters
including fruit yield per plant. High sca effects were expressed for the cross
combinations of DBL-24 x Pant Samrat, Aruna x Pant Rituraj, Punjab
Barsati x Ramnagar Giant and DBL-24 x Pant Rituraj and were good
specific combiners for yield and most yield attributing characters.

Singh et al. (2003) revealed that the variance due to

interaction between parents and hybrids was found highly significant for
branches per plant in brinjal. The GCA / SCA variance ratio being less than
unity revealed predominance of non-additive components of variance. He
also reported the crosses exhibited sca effects indicating the presence of
dominance and epistatic (non-additive) type of gene action for yield per
plant.

During the year 2003, Vinodkumar and Pathania observed

highly significant sca effects in the cross Arka Keshav x Pusa Purple
Cluster in respect to plant height, highly significant positive gca effects for
fruit length were recorded by the lines Arka Keshav, Pusa Anupam and
Punjab Barsati for fruit length in brinjal. The cross Arka Keshav x Pusa
Purple cluster exhibited best sca effects.

Biswajit et al. (2004) reported significant negative sca effects

for days to first flower and significant positive effect for fruit length in brinjal.
The cross combination PB-62 x T-3 for fruit diameter and PB-62 x T-3 for
yield per plant with highly significant sca effect.

Ashwani and Khandelwal (2005) noticed that the cross

combination P7 x P8 have recorded significantly high sca effects for plant
height, plant spread and fruit yield per plant in brinjal. The cross

24
combination P1 x P4 have recorded significantly high sca effects for days to
first flowering, for days to 75 per cent flowering, while, the cross
combination P6 x P9 for fruit length. The cross P1 x P8 have recorded
significantly high sca effects for fruit width, P2 x P7 for number of fruits per
plant indicating the importance of both additive and non-additive effects for
these characters.

During their study on brinjal, Premalakshme et al. (2005)

reported best GCA of parents producing high SCA hybrid is an indication of
the role of additive x additive gene interaction. The hybrid had significant
and positive heterosis over the best parent. Positive heterosis with positive
sca effect indicated the role of additive gene action for fruits per plant and
also reported the number of fruits per plant has a considerable influence on
total yield of fruits in tomato. Parent P6 and P1 had the maximum number
of fruits P1 (58.38 and 47.90) had positive gca effect while P6 had
significant negative gca effect.

Bendale et al. (2005) studied twenty eight F1 hybrids of

aubergine and their parents (CHES-309, SM-6-6, Arka Keshav, BB-64, BB-
60-C, DPL-B-4, DPL-B-5 and Bandhtiware, were subjected to combining
ability for plant height, leaf area per plant, number of branches per plant,
number of days to 50% flowering, number of days to first picking, and fruit
yield per plant. CHES-309 was a good general combiner for yield per plant,
plant height, and number of branches per plant. DPL-B-5 was a good
general combiner for plant height, leaf area per plant, and yield per plant.
CHES -309 x BD-60-C had the highest significant positive specific
combining ability (SCA) effect for fruit yield per plant. Arka Keshav x BB-60-
C exhibited the highest significant SCA effect for yield per plant, plant
height, number of days to first flowering, number of days to 50% flowering,
and number of days to first picking. SM-6-6 x Bandhtiware showed the
greatest SCA effect in the desired direction for fruit yield per plant, leaf area
per plant, number of days to first flowering and number of days to 50%
flowering.

Biradar et al. (2005) studied nine aubergine cultivars (Pragati,

Pusa Kranti, Black Beauty, Manjari Gota, Green Round, Sheetal Local,

25
Udgir Local, Jalgaon Green and Vidarbha Local) were crossed in all
possible combinations following a half diallel design. The resulting 36 F1
hybrids and the 9 parents were evaluated for combining ability studies.
Data were recorded for days to 50% flowering, days to first fruit picking,
fruit length, fruit girth, number of fruits per plant, fruit size index, fruit yield
per plant, average fruit weight and plant height. Mean sum of squares due
to general (GCA) and specific combining ability (SCA) were significant for
most characters, indicating the existence of a good amount of variability in
the parents. Black Beauty was the best general combiner, producing
significant GCA effects for fruit length, fruit size index, number of fruits per
plant, fruit yield per plant and average fruit weight. Another parent, Pragati,
also produced significant GCA effects for fruit girth, fruit yield per plant,
number of fruits per plant and fruit size index. Pusa Kranti x Manjari Gota,
Black Beauty x Sheetal Local, Black Beauty x Green Round and Pragati x
Manjari Gota were considered as the best cross combinations, recording
highly significant SCA effects for more than 5 characters.

While studying the combining ability analysis in brinjal, Bisht

et al. (2006) reported significant general combining ability effects for the
height. He identified the line Uttara as the promising combiner for branches
per plant and good specific combinations were Punjab Barsati x Green
Long Cluster and Punjab Sadabahar x DBL-24, Uttara x Green Long
Cluster had highest specific combining ability for fruit weight and were the
good specific combiners for this trait. The hybrids JB-7 x Nessppe,
Nessppe x Uttara, Punjab Barsati x Punjab Sadabahar were found to be
good specific combiners for fruit length. Whereas, three crosses (Uttara x
Small Long White, Pujab Barsati x Small Long White, Nessppe x Small
Long White) exhibited high specific combining ability effects for fruit
diameter. The parents Uttara, Green Long Cluster and Nessppe proved to
be good general combiners for number of fruits per plant and also observed
the best three crosses with high specific combining ability effects and
hybrid Uttar x Green Long Cluster had highest specific combining ability
effects for total fruit yield per plant.

26
 While studying on combining ability effects in Brinjal,
Praneetha and Veeragavatham (2007) revealed that parent EP-65
registered the lowest significant gca effect both for shoot borer infestation
(-1.34) and fruit borer infestation (-5.26)

In 2007, Shafeeq et al. reported that the line which

contributed significant positive GCA effects for number of flowers per
inflorescence was Budihal local (0.31) Lines Arka Nidhi (-0.41), Arka
Shirish (-0.09) and Mullu Badane (-0.21) exhibited negative gca effects. For
days to 50% flowering, four of the six lines exhibited significant gca effects.
Suneetha et al. (2008) revealed that among all the hybrids, 22
crosses had exhibited significant and desirable sca effects for fruit yield per
plant in brinjal.

Shanmugapriya et al. (2009) revealed that the lines which

recorded high fruit yield per plant, EP 378, White Brinjal and Pusa
Sadabahar were found good general combiners for days to first flowering,
number of branches per plant, plant height, number of fruits per plant, fruit
weight, fruit length, fruit girth and fruit yield per plant. The tester which
showed high fruit yield per plant viz., Hissar Pragathi was good general
combiner for all the eight traits studied.

Sao and Mehta (2010) conducted genetical studies on fruit

yield per plant and its attributing traits along with quality traits like total
soluble solids (TSS) and rind thickness following line x tester mating design
comprising of 8 lines and 6 testers in brinjal. The analysis revealed that all
the parents were found good general combiners for most of the characters.
However, line IGBO 65 and tester KS 327 were found best combiners for
fruit yield per plant. The ratio of gca variance/sca variance were observed
less than unity for all the characters which revealed the predominance of
non additive gene action. The hybrid, IGBL 70 x PPL was found best on the
basis of specific combining ability

Dharwad et al. (2011) noticed that parents MG, IC-112, IC-

111 and IC-136 had high gca effect for characters such as days to
flowering number of fruits per plant, fruit weight and yield per plant and
were found to be good general combiners in brinjal.
27
 Rai and Asati (2011) revealed that RCMB-10 and BB-64 were
best general combiners for major yield contributing characters of brinjal
viz., plant height, number of primary branches per plant, fruit weight and
fruit yield. However, the estimates of specific combining ability showed the
highest desirable sca effects in crosses RCMB-3 x BB-64 for plant height,
number of branches and number of fruits, RCMB-10 x RCMB-4 for fruit
breadth, fruit weight and yield, RCMB-10 x RCMB-3 for percent fruit set
and RCMB-1 x BB-64 for fruit length.

Sane et al. (2011) studied combining ability analysis of brinjal

in a 8 x 8 diallel, excluding reciprocals for fruit yield and yield contributing
characters. Non additive gene action was noticed to be preponderant for all
the traits studied. Perusal of the gca effects revealed parents MF and PLK-
1 to be good combiner for the fruit yield. Hence, these parents may be used
in breeding programmes aimed at the development of high yielding hybrids.
Four crosses had desirable sca effects in addition to high per se
performance for the trait. The hybrid MF x PLK-1, involving both good
combiners for fruit yield per plant had recorded maximum fruit yield, in
addition to desirable sca effects for fruit yield and early harvest. The
hybrids Krishna Kathi x PLK-1, MF x IAB-87, MF x Krishna Kathi, JB-19 x
PLK-1 and RHRB-7-7-6-1 x IAB-87 were also found promising for yield and
yield traits studied.

Mishra et al. (2013) estimated combining ability effects in

brinjal for seven different characters by crossing twelve lines and three
testers in line and tester mating design. Significant differences were
observed among the parents and hybrids for gca and sca effects
respectively. Analysis of gca effects for parent revealed that the lines
Punjab Neelam, DBSR-31, Ramnagar Giant, BR-SPS-14, ABSR-2 and
Pant Rituraj were identified by as most promising parents for inclusion in
hybridization programme with the aim to improving fruit yield as well as
other important yield contributing characters. Crosses exhibiting high per se
performance and significant positive sca effects for fruit yield and some
other important characters were Panjab Sanjyog x Black Beauty, Arka
Nidhi x Dudhiya, DBSR-31 x Pant Rituraj, Ramnagar Giant x Dudhiya ,BR-

28
SPS-14 x Pant Rituraj, Pussa Uttam x Dudhiya, ASRB-2 x Dudhiya and
ASRB-2 x Black Beauty. These crosses may exploit in the breeding
programme for obtaining trangressive segregants towards developing
hybrid varieties.

Combining ability for yield and its components in a diallel

crosses was analyzed by Patel et al. (2013), involving 7 genotypes of egg
plant. The gca and sca effects were significant for all the characters,
indicating the importance of both additive and non-additive genetic
components. The genotype R-1 proved as best general combiner for days
to 50% flowering and earliness,P-3 for plant height (tall stature),C-2 proved
as best general combiner for average fruit weight, fruit diameter, fruit size
and yield per plant, whereas I-9 for fruit length. Among the crosses C-2 x P-
3, P-1 x P-2, P-2 x I-9 and P-3 x R-1 were good specific combiners for yield
per plant.

Chawdhary and Didel (2014) studied 10 x 10 half diallel cross

of brinjal and found gca and sca effects significant for all characters under
study. Among the parents, based on fruit yield DBSR-91, was found to be
the best combiner and showed gca effects only for plant height and cruide
protein content besides fruit yield. Pusa purple round x Pusa Kranti
followed by DBR-31 x Pusa Kranti, HLB-12 x DBR-31, Hissar Shyamal x
Pusa Purple Roun and HLB-25 x DBSR-91were the best crosses on basis
of sca effects of fruit yield.

2.3 Gene action

The genetic improvement of a population depends largely

upon the nature and magnitude of additive and non-additive type of gene
actions. Heterosis is the results of the certain types of gene effects viz.,
additive, dominance and epistasis (additive x additive, additive x
dominance and dominance x dominance) of these additive types of gene
effects contribute to additive genetic variability. Therefore, higher the
contribution of additive type of gene effects to the manifestation of
heterosis, greater would be the retention of vigour in subsequent
segregating generations, whereas heterosis contributed in presence of non
additive gene effects i.e. dominance and dominance x dominance type of

29
epistatic gene effects may indicate possibility of exploiting such F1 hybrid
commercially.

Inference on gene action can also be made from combining

ability analysis besides, getting information on outstanding parents and
crosses. The choice of an appropriate breeding method for improvement of
quantitative characters also depends largely on gene action. But the effect
of individual gene cannot be measured. Non heritable agencies also
influence the phenotypic expression of these characters. Therefore the
effects of individual gene must be considered along with suitable statistical
procedure to obtain genetic information.

Fisher et al. (1932) used gene model to describe gene action

of many numbers of genes on a given characters. Mather (1949) also
developed gene models to study and assess the relative importance of
additive and dominance gene effects, but they assumed epistatic effects to
be negligible. Anderson and Kempthorne (1954) in their gene model
partitioned genotype value into additive, dominance and epistatic effects.
Mather (1949) and Hayman and Mather (1955) proposed a gene model to
assess the contribution of gene interactions to continuous variation based
on the theory developed by Fisher et al.(1932). Hayman (1958) describe a
general procedure to estimate parameters referring to additive (d),
dominance(h), additive x additive (i), additive x dominance (j), and
dominance x dominance effects.

The literature pertaining to gene action on various quantitative

characters in brinjal are reviewed below.

Swami (1970), observed dominance of purple colour over

green, elongated fruit over round and clustered fruits over non-clustered in
brinjal.

Randhawa and Sukhija (1973) observed dominance of

spininess over non-spineness, dominance of purple leaf colour over green.

While studing the genetic architecture of six characters of

brinjal, Mittal et al. (1976), reported that additive component was significant

30
only for all characters, whereas dominance variance was significant only for
plant height, fruit diameter and fruit volume.

Singh et al. (1976) registered additive and dominance

variance were highly significant for plant height, branches per plant, days to
first flower and fruit length, additive genetic variance for fruit diameter and
reported dominance were highly significant for fruit yield per plant.

Sidhu et al. (1980) noted that additive gene action effect was
significant for most of the characters, whereas dominant gene action was
significant for yield, of fruit, number of flowers and days to flowering. Over
dominance was reported for all the characters except number of fruits,
weight of fruit and girth of fruit.

Kandasamy et al. (1981) studied diallel analysis of forty five

brinjal hybrids and ten parents are observed dominance for plant height
and days to first fruit set, partial dominance for fruit per plant, over
dominance for fruit index and yield. Almost all the characters were
governed by non-additive gene action except for number of fruits per plant.

Verma (1986) observed the importance of both additive and

non-additive components, of genetic variance expression for days to 50%
flowering and fruit length.

During their studies on Genetics in brinjal, Mankar et al.

(1994) reported the presence of epistasis gene action for number of fruit
per plant in 26 crosses and for fruit yield per plant in 14 crosses. Whereas,
non additive gene action in the inheritance of fruit weight and fruit yield was
mainly govern by dominance effect.

Suneetha et al. (2000) observed the pre-ponderance of non-

dditive gene action for all the traits studied in brinjal. Among all the hybrids,
22 crosses had exhibited significant and desirable sca effects for fruit yield
per plant.

Vaghasiya et al. (2000) studied six generations (P1, P2, F1, F2,
BC1 and BC2) of brinjal (Solanum melongena L.) crosses KS 233 x PLR1
and Green Round x PLR1 were grown during kharif 1996-97 and evaluated
for 6 yield components. Results showed that in KS 233 x PLR1, additive as

31
well as non-additive gene effects were important for fruit weight and days to
first picking, while only non-additive effects were important for fruit yield per
plant, fruits per plant, fruit girth and plant height. In Green Round x PLR1,
all characters were under the control of both additive and non-additive gene
effects.

In 2005, Biradar et al. noticed the additive gene action for fruit
girth and average weight. Whereas during the year 2005, Suneetha
observed non-additive gene action was pre-ponderant for yield and yield
contributing characters and for infestation of shoot and fruit borer, the study
also revealed significant and desirable effects for several hybrids.

Shanmugapriya et al. (2009) reported that the ratio of

GCA/SCA variance was less than unity and showing high SCA variance for
all the characters studied. This indicated the preponderance of non-additive
gene action in the inheritance and improvement of the traits of Interest.

Dharwad et al. (2011) studied gene action in brinjal and

revealed gene action showed predominant non-additive gene action for all
traits under investigation.

Rai and Asati. (2011) observed that, gene action analysis

revealed preponderance of both additive and non-additive genes for yield
and its contributing characters in brinjal.

Chowdhary and Didel (2014) studied gene action from 10 x

10 half diallel crosses of brinjal. The analysis of variance for combining
ability selected revealed that mean squares due to gca and sca variance
were significant for all the characters which indicated that all the characters
were controlled by both additive and non- additive gene effects. However
the σ gca / σ sca ratio being less than unity for all the characters except
fruit diameter and plant height indicating that the non-additive gene action
was more important in the inheritance.

32
 CHAPTER III

MATERIAL AND METHODS

The present investigation "Genetic analysis of F1 hybrids in
brinjal" involving the Line X Tester analysis, was conducted on the
experimental field of Main Garden, Department of Horticulture, Dr.
Panjabrao Deshmukh Krishi Vidyapeeth, Akola. The experiment was
conducted during kharif and summer of 2014 -2015. The details of the
materials used and methods followed for the studies are described in this
chapter.

3.1 Soil

The soil was medium black with clay soil, well leveled and
uniform in topography with appropriate drainage.

3.2 Climate

Akola comes under tropical belt and is situated at attitude

307.4 meters above mean sea level. The geographical situation is 20.42°N
latitude, 77.02°E longitude. The mean annual precipitation recorded during
2012-2013 was 699.5 mm, which almost received from south-east
monsoon.

3.3 Experimental Material

The experimental material for this study comprised of eleven

inbred parental genotypes which were collected from University
Department of Horticulture, Dr.Panjabrao Deshmukh Krishi Vidyapeeth,
Akola, All India Co-ordinated Research Project on Vegetable crops and
Mahatma Phule Krishi Vidyapeeth,Rahuri based on their diversity for
various traits and popular local cultivars from Bhandara, Wadsa
(Gadchiroli) and Akola district. From these eleven genotypes, four were
used as lines and seven as testers and twenty eight crosses were evolved
in a line x tester design. One commercial hybrid Phule Arjun was used
as check. The list of parents, and check along with sources are given in
Table 1.

33
Table 1. List of parents, crosses and check

Sr. Parents Code Source

No.
Lines
1 Aruna P1 Dr. PDKV, Akola
2 Bhandara Local P2 Local collection of Bhandara
(Green round) District
3 Wadsa local (White P3 Local collection of Wadsa, Dist.
Long) Gadchiroli
4 Bhatai local (Dark P4 Local collection of Bhandara
Purple Round) District
Testers
1 Manjari Gota P5 MPKV, Rahuri
2 Chandur Local P6 Local collection from Chanduri area
of Akola district

3 Ruchira P7 MPKV, Rahuri

4 Krishna Kathi P8 MPKV, Rahuri
5 White Round P9 Local collection of Bhandara
District
6 DBSR-52 P10 MPKV, Rahuri
7 Brinjal White P11 MPKV, Rahuri
Check
1 Phule Arjun MPKV, Rahuri

3.4 Production of F1 hybrid seeds

The seeds of genotypes used as parents were sown during

kharif 2013 to constitute a crossing block. The genotypes were crossed by
using P1 to P4 as female parents and P5 to P11 as male parents in line x
tester design, to obtain the seeds of F1 hybrids. The healthy flower buds
which were supposed to open on next day were selected for emasculation
and pollination. The selected buds were emasculated by hand using
forceps in the evening hours between 4.00 pm to 6.00 pm. Emasculated

34
flowers were covered with butter paper bags to avoid contamination by
foreign pollen.

Pollination of emasculated flowers was done on following day

morning during anthesis (7.30 am to 10.30 am). Well opened flowers with
dehisced anthers were collected from the male parents, butter paper bag
was removed carefully and the stigma was touched with dehisced anthers
of male flowers. The female flower was covered with white colour butter
paper bag immediately for easy identification and further avoiding the
contamination from other pollens. The pedicel of each pollinated flowers
was tied with label, bearing information of female and male parents and
date of crossing for identification. The butter paper bags were removed by
6 -7 days after pollination to encourage better fruit development. The
parents were also selfed simultaneously to obtain pure seeds of each
genotype. The extracted seeds of fully matured fruits were thoroughly
washed to remove the mucilaginous and suitably dried in shed. Sufficient
selfed and crossed seeds were obtained to carry out the experiment.

3.5 Experimental Details

1. Name of crop : Brinjal

2. Experimental Design : Randomized Block Design (RBD)
3. No. of Replication : Three (3)
4 Spacing : 75 X 60 cm
5. Number of treatments : Eleven parents, twenty eight
Crosses and one Check
(11 parents + 28 F1 + 1 check=40)
6. Mating Design : Line x Tester
7. Season : Kharif 2014 and Summer 2015

Cultivation of experimental crop, brinjal, was conducted as

per the recommended package of practices of Dr. PDKV, Akola

35
3.6 Observations recorded

Five randomly selected plants were tagged in each entry to

record the observations and the average from these five plants was worked
out for statistical analysis. Following observations were recorded in this
experiment.

3.6.1 Plant height (cm)

Height of the plant from ground level to the top of the plant
was measured in centimeter and the average was calculated.

3.6.2 Number of branches per plant

Total number of primary with secondary branches were

counted at final harvest stage and the average was calculated.

3.6.3 Plant spread (cm)

East- west and north-south spread of each observational

plant was recorded by fixing the marks at extremes. Spread points at east-
west and north-south was measured with the help of meter scale .Finally,
average spread per plant was computed.

3.6.4 Days to first flowering

Days taken for first flowering was recorded from the date of
transplanting to the appearance of first flowering in the observational plants
in each treatment and replication.

3.6.5 Days to 50% flowering

Numbers of days to first flower appearance in 50 per cent

plants of each replicated treatment from transplanting were recorded.

3.6.6 Number of flowers per branch

Numbers of flowers per branch were counted on the

observational plants and the average was worked out.

3.6.7 Number of fruits per branch

Numbers of fruits harvested per branch of observational plant

at each harvesting were recorded and the average was worked out.

36
3.6.8 Number of fruits/plant
Numbers of fruits harvested per observational plant at each
harvesting were recorded and the average was worked out.
3.6.9 Fruit set (%)
Total numbers of fruits set were counted per selected plant
and the average as per cent to the number of flowers per plant were
calculated.
3.6.10 Fruit length (cm)
Length of five fruits at marketable stage was measured
individually from the base of calyx to tip of fruit in cm using vernier caliper
and average was calculated at each harvest.
3.6.11 Fruit diameter (cm)
The diameter of fruit was meaasured by using vernier caliper
at the widest point of the fruit. Average of five fruits diameter was worked
out in centimeter.
3.6.12 Fruit weight (g)
Five marketable size fruits selected at randomly from five
observational plants per plot and fresh weight was recorded in grams and
the average was computed.
3.6.13 Yield per plant (kg)
Weight of the fruits from every harvest from each of the
observational five plants in each entry was added and the average was
calculated and the expressed in kilograms.
3.6.14 Yield per plot (kg)
Weight of the fruits from different pickings from each plot in
each entry was added and the average was calculated and the expressed
in kilograms.
3.6.15 Yield per hectare (q)

The average yield was worked out on the basis of yield per
plot throughout the harvesting period under each treatment and it was
totaled and then average yield per hectare per treatment was worked out.

37
3.6.16 Number of seeds / fruit

Numbers of seeds of fully mature and ripe fruits per five

selected plant were counted and the average was worked out for analysis.

3.6.17 Infestation of shoot and fruit borer (%)

The number of infested fruits was recorded from the every

harvesting and mean per cent infested fruits per plant was worked out as
follows.

Number of infected fruits

Per cent fruit infestation = ___________________ x 100
Total number of fruits

3.6.18 Disease incidence (%)

The number of plants effected by diseases like bacterial wilt

and other major diseases in each entry in the field were recorded and
expressed as per cent. The angular transferred values were used for the
analysis. They were scored as per the scale given by Mew and Ho (1976).

3.6.19 Leaf chlorophyll content (%)

Leaves of observational plants were detached at first

harvesting stage and average chlorophyll content was calculated by using
chlorophyll meter in percent.

3.7 Statistical analysis

The data was subjected to the following statistical and

biometrical analysis.

3.7.1 Analysis of variance for the experimental design

The first step in the line x tester analysis is to perform

analysis of variance to test the significance of differences among
genotypes. This was carried out separately for each character at each
season as suggested by Fisher and Yates (1974) and Panse and
Sukhatme (1985). In order to test the significance of females, males and
crosses separately further partitioning of treatment sum of squares was
done for each character.

38
Table 1.1. Analysis of variance for parents and hybrids for individual
season

Source of Sum of F Value

D. F. M.S. S.
variation square calculated

Replications r-1 SSr Mr Mr/Me

Parents p-1 SSp Mp Mp/Me
Females f-1 SSf Mf Mf/ Me
Males m-1 SSm Mm Mm/ Me
Female vs Males 1 SSf Vs SSm Mf/ Mm Mf vs Mm/Me
Crosses c-1 SSc Mc Mc/ Me
Parents Vs crosses 1 SSp/c Mp/Mc Mp vs Mc/ Me
Error (r-1) (g-1) SSe Me

Where, r = number of replications

p= number of parents

f = number of females

m = number of males

c = number of crosses

df = degrees of freedom

Me = Error mean sum of squares

The significance of mean squares was tested against error

variance by using F-test. The standard error of difference for comparing
any two progeny means were computed by using error variance of
respective characters as follows:

2 MS e
S.E. (d) 
r

The critical difference was computed by multiplying the

standard error of difference with ‘t’ test value for (r-1) (g-1) at error degree
of freedom at p = 0.05 (5%) and p = 0.01 (1%).

39
3.7.2 Pooled analysis of variance:

The genotypes were tested over two seasons and pooled

analysis was carried out as given below.

Table 1.2. Analysis of variances for pooled over two seasons for
parents and crosses

F value
Sources of variation d. f. MSS
calculated
Seasons (n-1)
Replications over seasons (n-1) (r-1)
Parents p-1
Females f-1
Males m-1
Female vs Males 1
Crosses c-1
Parents vs Crosses 1
Parents x Seasons (p-1) (n-1)
Females x Seasons (f-1) (n-1)
Males x Seasons (m-1) (n-1)
(Females vs Males) x Seasons 1
Crosses x Seasons (c-1) (n-1)
(Parents vs crosses) x Seasons 1
Pooled error n(r-1) (g-1)

Where,

g, n and r stand for genotypes, seasons and replications

respectively.

The standard error and critical differences between two

means were calculated as follows.

S. E. of means for individual seasons = √ 2 e / r

40
C. D. = √ 2 x 2 e x t value
r
(‘t’ value at 5% and 1% level of probability for error degree of freedom).

3.7.3 Estimation of average heterosis, heterobeltiosis and standard

heterosis

A) Estimation of heterosis

Heterosis is the superiority(+ve) or inferiority(-ve) of F1 hybrid

over both of its parents in terms of yield or some other characters. It is
expressed as percent. The Heterosis over the mid parent (MP), better
parent (BP) and standard check (CC) was estimated for all biometrical
characters under study for each season and pooled data as given below.

1) Average heterosis

It is estimated as percent over mid parent as follows

F1 - MP
Average heterosis (H1) = x 100
MP

P1 + P 2
MP =
2
Where,

F1 = Mean performance of cross

P1 = Mean performance of parent 1

P2 = Mean performance of parent 2

MP = Mean performance of both the parents

2) Heterobeltiosis

It is estimated as percent over better parent as follows,

F1 - BP
Heterobeltiosis (H2) = x 100
BP

41
Where,

F1 = Mean performance of cross

BP = Mean performance of better parent

3) Standard heterosis

It is estimated as percent over standard commercial check

variety as follows,

F1 - CC
Standard heterosis (H3) = x 100
CC
Where,

F1 = Mean performance of cross

CC = Mean performance of check

The significance of Heterosis was tested by least significant

differences i.e. Standard error of difference for heterosis effect was
calculated by the following formulae.

For average heterosis,

S. E. (Diff.) H1 = √3 MSe / 2r

For heterobeltiosis and useful heterosis,

S. E. (Diff.) H2 = √2 MSe / r

Where,

MSe = Error mean squares

r = Number of replications

In case of pooled estimate all formulae remain unchanged

except that ‘r’ is replaced by 2r in the dominators.

The critical difference was computed by multiplying the

standard error of difference with‘t’ test value for error degree of freedom at
p = 0.05 (5%) and p = 0.01 (1%).

42
 3.7.4 Analysis of variance for combining ability (Line x Tester
analysis)

ANOVA for combining ability was based on the methodology

given by Kempthorne (1957).

3.7.4.1 Combining ability analysis for single season

The model used to study the general and specific combining

ability was

Yijk = µ + gi + gj + sij + eijk

Where,

Yijk = value of the ijkth observation

µ = an effect of common to all hybrids in all replications

gi = gca effects of the ith parents

gj = gca effects of the jth parents

sij = specific effect of the progeny of i x j

eijk = random error effect associated with ijkth observation

i = number of female parents, f

j = number of male parents, m

k = number of replications, r.

Table 1.3. Analysis of Variance for combining ability

Sources of Expectations of Mean squares

D. F. MSS
variation Component Co-variance
Replications r-1
Females σ 2 e + r σ 2 fm σ 2 e + r (cov. F.S. – 2 cov.
f-1 M1
(lines) + rm σ 2 f H. S.)+ rm (cov. H. S.)
σ 2 e + r σ 2 fm σ 2 e + r (cov. F.S. – 2 cov.
Males (tester) m-1 M2
+ rf σ 2 m H. S.)+ rf (cov. H. S.)
Females x σ 2 e + r cov(F.S) – 2 cov.
(f-1) (m-1) M3 σ 2 e + r σ 2 fm
Males (H.S)
Error (r-1) (fm-1) M4 σ2 e σ2 e

43
Where,

f = Number of females (Lines)

m = Number of males (Testers)

r = Number of replications

σ 2 e = genetic variance among individuals from same mating

σ2 f = progeny variance arising from differences among female parents

σ 2 m = progeny variance arising from differences among male parents

σ 2 fm = progeny variance arising from interaction of the contribution of

female and male parents

Cov. H. S. = co-variance between half sib

Cov. F. S. = co-variance between full sib

Estimation of variance components was done as follows by

solving the simultaneous equations of σ 2 f, σ 2 m and σ 2 fm based on the
expectations as shown in above ANOVA table.

σ2 f =M1 – M3 / rm

σ 2 m = M2 – M3 / rf

σ 2 fm = M3 – M4 / r

Estimates of Cov. H.S. and Cov. F. S. can be calculated as

Cov. H. S. = m σ2 f + f σ2 m
------------------------
f+m

Cov. F. S. = σ 2 fm + 2 Cov. H. S.

These can also be calculated from the expectations of mean squares as

M1 + M2 – 2 Ms
Cov. H.S. = -----------------------
r (m + f)

Cov F.S.= M1 + M2 + M3 – 3M4 σ r cov. H.S.– r (f + m) cov.H.S.

---------------------------- + ----------------------------------
3r 3r

44
3.7.4.2 Estimation of general and specific combining ability effects

X… f m r
i. Population mean µ = ------ …… where, X … = ∑ ∑ ∑ Xijk
rmf i=1 j=1 k=1

ii. General combining ability effects of female parents

Xi X… m r
gi = ---- - ------ …… where, Xi … = ∑ ∑ Xijk
rm rmf j=1 k=1

iii. General combining ability effects of male parents

Xj X… f r
gi = ---- - ------ …… where, Xi … = ∑ ∑ Xijk
rf rmf j=1 k=1

iv. Specific combining ability effect of crosses

Xij Xi . Xi x j X…
S σ ij = ----- - ------- - ---------- + ----------
r rm rf rmf
r
Where, Xij = ∑ x Xijk
k =1
Where,

X … - Total of all hybrid

Xi . - Total of ith female parent over all males

X.j - Total of jth male parent over all females

Xij - Total of ijth combinations over replications.

3.7.4.3 Standard error for testing combining ability effects

Standard error of effect was calculated as square root of the

variance of effects.

SE (gca female) = (M4 / mr) 1/2

SE (gca male) = (M4 / fr) 1/2

SE (sca crosses) = (M4 / r) 1/2

45
Where,
M4 = MSS due to error.
Standard error for testing combining ability effects in pooled
analysis were calculated as.
SE (gca female) = (M5 / mrs) ½
SE (gca male) = (M6 / frs) ½
SE (sca crosses) = (M7 / rs) ½
Where,
M5 = MSS due to female x season
M6 = MSS due to male x season
M7 = MSS due to female x male x season
s = number of seasons
3.7.4.4 Estimation of GCA and SCA variance

From the estimates of Cov.(H. S.) and Cov. (F.S.) variance

due to general combining ability and specific combining ability were
calculated as follows:

M1 + M2 – 2M3
σ2 gca = Cov. H. S. = -------------------------------------

r (f + m)

σ 2 sca = Cov. F. S. – 2 cov. H. S.

M3 – M4
= -----------------
r

The critical differences were calculated by multiplying the

standard error of difference with respective `t’ table value at 5% and 1%
level at error degrees of freedom.

3.7.4.5 Pooled analysis of combining ability (L X T)

The material was grown in two seasons. In order to have an

idea of genotype x season interactions as the estimate from one
environment may be biased and thus would not represent a true picture of
genetic architecture.

46
 The model used for the analysis was,

Yijkr = µ + fi + mj + (fm)ij + sk + (fs) ik + (ms) jk + (fms) ijk + eijkr

Where,
Yijkr = value of ijkrth observation
µ = an effect common to all hybrids in all replications
fi = gca effect of ith parent
mj = gca effect of jth parent
(fm) ij = specific effect of progeny i x j
Sk = season effect in kth replication
(fs) ik = female x season interaction in kth replication
(ms) jk = male x season interaction in kth replication
(fms) ijk = interaction of season with specific effects of the progeny
eijkr = random error effect associate with eijkr observation
The analysis of variance for treatment x season interaction
was based on the procedure developed by Kempthorne (1957) related to
design II of Comstock and Robinson (1952). The estimates of variance
were obtained by equating mean squares to expectations and solving for
the component.

Table 1.4. Pooled ANOVA for combining ability (L X T)

Source D. F. MSS Expectation

σ 2e + r σ 2fms + rm σ 2fs + rf
Season (s-1) M1
σ ms + + rfm σ 2s
2

σ 2e + r σ 2fms + rs σ 2fm + rm
Females (f-1) M2
σ 2fs + + rms σ 2f
σ 2e + r σ 2fms + rs σ 2fm + rf
Males (m-1) M3
σ 2ms + + rfs σ 2m
Females x (f-1)
M4 σ 2e + r σ 2fms + rm σ 2fs
Seasons (s-1)
(m-1)
Males x Seasons M5 σ 2e + r σ 2fms + rf σ 2ms
(s-1)
(f-1)
Female x male M6 σ 2e + r σ 2fms + rs σ 2fm
(m-1)
(Females x (f-1)
Males) x (m-1) M7 σ 2e + r σ 2fms
Seasons (s-1)
Error s (fm-1) (r-1) M8 σ 2e

47
Where,

f = number of female parents

m = number of male parents

σ 2e = genetic variance among individuals from the same mating

σ 2m = the variance of male effects

σ 2f = the variance of female effects

σ 2fm = the variance due to interaction between female effect and male

σ 2fs = the variance due to interaction between female effects and season

σ 2ms = the variance due to interaction between male effects and seasons

σ 2fms = the variance due to interaction between female, male and season.

The test of significance for females x males x season

interaction is F = M7 / M8 and for female x male interaction is F = M5 / M7.
The variance σ 2m and σ 2f will be tested against σ 2ms and σ 2fs
respectively.

Estimation of variance components was done as follows for σ

2m, σ 2f, σ 2fm, σ 2s, σ 2fs, σ 2ms and σ 2fms by equating mean sum of
squares to expectations and sorting out for the appropriate components.

σ 2f = (M2 – M4 – M6 + M7) / rms

σ 2m = (M3 – M6 – M5 + M7) / rfs

σ 2fm = (M6 – M7) / rs

σ 2s = (M1 – M4 – M5 + M7) / rfm

σ 2fs = (M4 – M7) / rm

σ 2ms = (M5 – M7) / rf

σ 2fms = (M7 – M8) / r

The estimates of σ 2 gca and σ 2 sca were based on

covariance of full sibs and half sibs.

(r σ 2 m + m σ 2 f)
Cov. H. S. = -----------------------------
(f + m)

48
 M2 + M3 – 2M6 – M4 – 2 M5 + 2M7
= ---------------------------------------------------
Rs (f + m)

Cov. H. S. = σ 2 fm + 2 cov. (H. S.)

From the above equations variance due to general and

specific combining ability were estimated as

σ 2 gca = Cov. H. S.

σ 2 sca = σ 2 fm = Cov. F. S. – 2 Cov. H. S.

3.7.5 Gene action

The ratio between σ 2 gca and σ 2 sca if greater than one (>1)
indicates the additive gene action and less than one (<1) indicates non
additive gene action in governing the respective characters.

3.7.6 Proportional contribution of females, males and their

interactions

a) Contribution of females

Sum of squares due to females

= ---------------------------------------------- x 100
Sum of squares due to crosses

a) Contribution of males
Sum of squares due to males
= ---------------------------------------------- x 100
Sum of squares due to crosses

a) Contribution of females x males

Sum of squares due to f x m
= ---------------------------------------------- x 100
Sum of squares due to crosses

49