HELIA,
25
, Nr. 37, p.p. 109-118, (2002)
UDC 633.854.78:631.527
CORRELATION AND PATH ANALYSIS IN
SUNFLOWER
Chikkadevaiah
*
, Sujatha H.L. and Nandini
University of Agricultural Sciences, GKVK, Bangalore 560 065, India
Received: January 10, 2001
Accepted: September 18, 2002
SUMMARY
Fifty-one inbreds and three checks of sunflower were laid in a rand-
omized complete block design following recommended package of practices.
Correlation studies revealed that seed yield was positively and significantly
associated with seed volume weight, hull percentage, percent autogamy, days
to 50 percent flowering and oil yield. Oil content was positively associated with
plant height, number of leaves per plant, seed volume weight, hull percentage,
head diameter, percent autogamy, days to 50 percent flowering, 100-seed
weight, seed yield and oil yield. Path coefficient analysis for seed yield at phe-
notypic level showed that the direct effect was maximum for oil yield followed
by 100-seed weight. The maximum indirect effect for seed yield was minimum
through oil yield, while it was through yield per plant for oil.
Key words: character association, path analysis, sunflower
INTRODUCTION
The objectives of any plant breeder include selection from a natural population
or from one developed by him for one or several characters. Yield is a complex
character and is a function of several component characters and their interaction
with environment. It would be more meaningful if the structure of yield were probed
through breeding. It is necessary to measure the mutual relationship between vari-
ous plant characters and determine the component characters on which selection
can be based for genetic improvement in yield. Genotypic and phenotypic associa-
tion reveal the degree of association between different characters. Thus it helps to
base selection procedure to a required balance where two opposite desirable char-
acters affecting the principal characters are being selected. It also helps to improve
different characters simultaneously. Desai (1989), Patil et al. (1996b) and Abdelga-
wad et al. (1987) have reported positive associations of seed yield with various yield
components. Ibrahim (1985) and Srinivasa (1982) have reported positive associa-
tions of various yield components with oil content.
* Corresponding author, Fax: 0091-080-333-0277, e-mail: chikkadevaiah@yahoo.com
110 HELIA, 25, Nr. 37, p.p. 109-118, (2002)
The path analysis gives the cause and effect relationship. It splits the associa-
tion coefficient into measures of direct and indirect effects and determines the
direct and indirect contribution of various characters towards yield. It critically
breaks up different direct and indirect effects which finally make up association
coefficient. Hussain et al. (1995) and Sheriff et al. (1987) have reported direct
effects of different characters on seed yield.
MATERIAL AND METHODS
The experiment consisted of 51 inbreds and three checks (GAUSUF-7, TNAU-
SUF-15 and KBSH-1) of sunflower which were of diverse nature with respect to geo-
graphic distribution(Table 1).
Table 1: Sunflower inbreds and their country of origin
No Inbreds Country No Inbreds Country
1 RHA-348 USA 29 ARM-250B France
2 CMS-234B USA 30 CMS-852B USA
3 RHA-801 USA 31 7-1-B India
4 CMS-336B USA 32 CMS-302 USA
5 RHA-356 USA 33 RHA-180 USA
6 CMS-343B USA 34 CMS-597B Canada
7 RHA-274-1 USA 35 HAM-69 India
8 ARM-242B India 36 RHA-R-R-I USA
9 CMS-207B USA 37 CMS-234B-I USA
10 HAM-174 India 38 CMS-89B USA
11 ARM-244B India 39 RHA-MR-I USA
12 CMS-850B USA 40 86 B
3
France
13 ARM-246B India 41 CMS-851B USA
14 CMS-353B USA 42 RHA-272-I USA
15 ARM-243 India 43 RHA-587-II USA
16 MR-1 India 44 CMS-335B USA
17 ARM-245B India 45 RHA-272-II USA
18 CMS-351B USA 46 RR-I USSR
19 HAM-9R India 47 CMS-349B USA
20 ARM-247B India 48 CMS-339B USA
21 CMS-300B USA 49 CMS-62B France
22 HAM-196 India 50 338(C)-B USSR
23 ARM-248B India 51 ARM-248B India
24 CMS-350B USA
25 RHA-274-II Australia Checks
26 ARM-249B India 1 GAUSUF-15 India
27 CMS-843B USA 2 TNAUSUF-7 India
28 RHA-278 France 3 KBSH-1 India
HELIA, 25, Nr. 37, p.p. 109-118, (2002) 111
Table 2: Phenotypic correlation coefficients for 15 quantitative characters in sunflower
X
2
X
3
X
4
X
5
X
6
X
7
X
8
X
9
X
10
X
11
X
12
X
13
X
14
X
15
X
1
0.504** 0.103 -0.155 -0.088 0.042 0.659** -0.124 0.459** 0.289* 0.454** 0.199 0.546** 0.210 0.540**
X
2
-0.194 -0.236 -0.238 0.193 0.466** -0.008 0.476** 0.014 0.330* 0.184 0.356** 0.097 0.331*
X
3
-0.118 -0.098 0.034 -0.009 0.301* 0.114 0.087 -0.238 0.247 0.090 0.291* 0.176
X
4
0.166 -0.193 -0.293* 0.133 -0.124 0.021 -0.178 -0.203 -0.331* -0.148 -0.351**
X
5
-0.868** -0.071 0.079 -0.186 0.020 0.193 -0.110 0.085 0.133 0.097
X
6
0.095 -0.039 0.120 -0.029 -0.255 0.113 -0.115 -0.134 -0.119
X
7
-0.170 0.324* 0.037 0.583** 0.141 0.523** 0.124 0.511**
X
8
0.123 -0.093 -0.327* -0.228 -0.325* 0.100 -0.210
X
9
0.011 0.171 0.316* 0.352** 0.172 0.355**
X
10
0.105 0.033 0.076 0.179 0.125
X
11
0.059 0.677** 0.131 0.624**
X
l2
0.417** 0.208 0.383**
X
13
0.339* 0.948**
X
14
0.534**
X
1
=Plant height;
X
2
=Number of leaves per plant;
X
3
=Seed volume weight;
X
4
=Seed density;
X
5
=Hull percentage;
X
6
=Husk percentage;
X
7
=Head diameter;
X
8
=Percent autogamy;
X
9
=Stem girth;
X
10
=Days to 50% flowering;
X
11
=100-seed weight;
X
12
=Percent seed set;
X
13
=Yield per plant;
X
14
=Oil content;
X
15
=Oil yield
112 HELIA, 25, Nr. 37, p.p. 109-118, (2002)
It was laid out in a random complete block design during summer 2000 under
irrigated conditions with a spacing of 60 x 30 cm and two replications were
adopted. The observations were recorded on five randomly selected plants. Pheno-
typic and genotypic associations were computed as per Weber and Moorthy (1952).
Path coefficient analysis was carried out as suggested by Wright (1921) and illus-
trated by Dewey and Lu (1959). The characters which exhibited significant associa-
tions with seed yield and oil yield were considered for path analysis viz., plant
height, number of leaves per plant, seed volume weight, hull percentage, head diam-
eter, stem girth, days to 50 percent flowering, 100-seed weight, percent seed set, oil
content and oil yield.
RESULTS AND DISCUSSION
Phenotypic association
It is evident from the results (Table 2) that seed yield was positively and signifi-
cantly associated with plant height (Narayana, 1998), number of leaves per plant
(Satisha et al., 1995), head diameter (Singh et al., 1998), stem girth, (Punia et al.,
1994), 100-seed weight (Singh et al., 1990), percent seed set (Singh et al., 1998;
Narayana et al., 1998) and oil content (Jhagirdhar, 1986). In general, seed yield
was positively associated with all other characters except with seed density, husk
percentage and percent autogamy (Figure 1).
Since most of the associations are in the desirable direction, these aid in easy
selection of inbreds with desirable characters. On the other hand, oil content was
positively associated with plant height (Jhagirdhar, 1986; Abdelgawad, 1987)
number of leaves per plant (Ibrahim, 1985; Abdelgawad et al., 1987), seed volume
X
1
= Plant height (cm)
X
2
=Number of leaves per plant
X
3
=Seed volume weight (g/100 cm
3
X
4
=Seed density (g/cc)
X
5
=Hull percentage (%)
X
6
=Husk percentage (%)
X
7
=Head diameter (cm)
X
8
=Percent autogamy (%)
X
9
=Stem girth (cm)
X
10
=Days to 50% flowering
X
11
=100-seed weight (g)
X
12
=Percent seed set (%)
X
14
=Oil content (%)
X
15
=Oil yield (%)
Figure 1: Phenotypic correlation coefficient of 14 quantitative characters with seed yield in
sunflower
HELIA,
25
, Nr. 37, p.p. 109-118, (2002)
113
weight (Lakshmanaih, 1980; Shivaraju, 1984), hull percentage, head diameter
(Lakshmanaiah, 1980) days to 50 percent flowering (Ibrahim, 1985), 100-seed
weight (Abdelgawad et al., 1987), seed yield and oil yield. Among these positive
associations, the associations with seed volume weight, percent seed set, yield per
plant and oil yield, while oil content were negatively and non-significantly associated
with seed density and husk percentage (Figure 2).
Lack of strong association between seed volume weight, hull percentage, days to
50 percent flowering and seed yield gives ample scope to select desired high seed
volume weight with high yielding, higher hull percentage with high yielding and
early flowering with high yielding types. Fortunately, seed yield and oil content were
reported to be desirably and negatively associated with husk percentage. One of the
interesting features in sunflower reflected by the present study is that the associa-
tions among all yield components were positive which would be encouraging for
rapid improvement of seed yield and oil content. Though seed density was nega-
tively associated with most of the characters including seed yield and oil content, its
lower magnitude would not affect much in the improvement of seed yield and oil
content.
Path analysis
Path analysis was carried out in respect of seed yield and oil yield (Tables 3 and 4).
Path analysis for seed yield
The direct effect for seed yield was maximum through oil yield followed by 100-
seed weight. The maximum indirect effect for any character was through oil yield
which itself was the major contributor.
F
igure 2: Phenotypic correlation coefficient of 14 quantitative characters with oil content in
sunflower
X
1
=Plant height (cm)
X
2
=Number of leaves per plant
X
3
=Seed volume weight (g/100 cm
3
X
4
=Seed density (g/cc)
X
5
=Hull percentage (%)
X
6
=Husk percentage (%)
X
7
=Head diameter (cm)
X
8
=Percent autogamy (%)
X
9
=Stem girth (cm)
X
10
=Days to 50% flowering
X
11
=100-seed weight (g)
X
12
=Percent seed set (%)
X
13
=Yield per plant(g)
X
15
=Oil yield (%)