CONTENTS

PARASHURAM and JAYARAME
MATERIALS AND METHODS
Heterosis for yield and its attributing traits
Four lines were crossed to four testers in a
line x tester mating design (Kempthorne, 1957) in
kharif-2009, to generate hybrids by following hot water
treatment (Rao and Rao, 1962) for emasculation then
contact method for crossing (Ayyangar and Warrior,
1934). Crossed seeds along with their parents were
sown in nursery during rabi-2009. The hybrids were
first identified in the nursery using purple plant
pigmentation as marker. All the eight parents (4 lines
and 4 testers) together with 16 crosses were evaluated
during rabi-2009. The material was grown in a single
row of 3 m length with a spacing of 30 x 10 cm in a
randomized block design and replicated twice. The
data were recorded on plant height, number of fingers
per ear, number of productive tillers per plant, finger
length, finger width, culm width, peduncle length,
days to 50 per cent flowering, days to maturity, straw
yield per plant, grain yield per plant and test weight
on five randomly selected plants. The analysis of
Randomised Block Design was carried out based on
the methods described by Panse and Sukhatme
(1967) and significance of heterosis was tested using
simple‘t’ test at five per cent and one per cent level
of significance. The magnitude of heterosis was
estimated over mid parent and better parent (Singh
and Chaudhary, 1985).
Intra Spikelet Competition for Seed Size
The material for this study comprised of
mature spike/fingers of three varieties viz., HR 911,
PR 2O2 and GPU 28. These varieties were grown
both in field and pots in green house. Each spike
was divided into three portions. The florets from 1/3 rd
position of top, 1/3 rd of middle and 1/3 rd of bottom
position of spike of each variety were taken
separately. Approximately 100 spikelets from each
portion were selected in all three varieties. In each
finger there are about 70 spikelets, each spikelet
having five to seven complete flowers. However, the
experimental varieties having six florets from each
spikelet from top to bottom of the spike. The positions
of florets in each spikelet were numbered (only six
florets hence numbered from one to six) and the
seeds of particular position were taken out which were
numbered from one to six to a particular bowl which
were also numbered as one to six. Thousand grain
weights of separately collected floret wise seeds of
three different positions was weighed and compared.
Analysis of variance for three factorial Completely
Randomized Design for floret wise seed weight of
three portions of three different varieties was
constructed. The varieties were considered as main
factors. Sub factors were portions (3), because each
variety has been divided into three portions and subsub
factors were 6, because each spikelet consists
of six florets, for which competition between florets
was studied.
RESULTS AND DISCUSSION
Heterosis for yield and related traits
The analysis of variance revealed significant
differences among the parents as well as crosses
for all the traits. Higher level of significance in the
variance of parents vs hybrids for all the characters
clearly indicated the existence of significant level of
average heterosis in the hybrids (Table 2). Non
additive gene action was noticed for all the characters
studied. The results support the findings of
Tamilcovane (1994), Madhusudhan et al. (1995) and
Patel (1994).
Among 16 hybrids developed, per cent
heterosis over mid parent and better parent was
negatively significant in most of the hybrids except
in crosses, GE 6216 x GPU 69 and GE 6216 x GE
5095 for days to 50 per cent flowering and except
GE 6216 x GPU 69 and GE 4906 x GE 5095 for days
to maturity suggesting the involvement of dominant
gene action with negative effects (Table 3). The earlier
reports suggest that early types can be obtained from
crosses which record negative heterosis. Similar
results were obtained by Konstantinov and Linnik
(1985) and Ramesh (1990) in proso millet. Significant
level of heterosis over mid parent and better parent
have been observed for the characters like plant
height, number of productive tillers per plant, straw
yield per plant and grain yield per plant.
From this study, it can be concluded that
the non additive gene action favouring hybridization
to some extent and the crosses, GE 4596 x L 5 and
GE 4596 x GPU 69 are the best crosses for grain
yield and most of the yield contributing characters.
34