sa in maize

RELATIVE STABILITY ANALYSIS OF PUBLIC AND PRIVATE

BRED HYBRIDS OF MAIZE (Zea mays L.)

Thesis submitted to the University of Agricultural Sciences, Dharwad in partial fulfilment of the requirements for the

Degree of

MASTER OF SCIENCE (AGRICULTURE)

IN

GENETICS AND PLANT BREEDING

By

NAGABHUSHAN

DEPARTMENT OF GENETICS AND PLANT BREEDING COLLEGE OF AGRICULTURE, DHARWAD

UNIVERSITY OF AGRICULTURAL SCIENCES, DHARWAD - 580 005

AUGUST, 2008

ADVISORY COMMITTEE

Dharwad (M.C. WALI) AUGUST, 2008 MAJOR ADVISOR Approved by:

Chairman : ______________________ (M.C. WALI)

Members : 1.____________________ (P.M. SALIMATH) 2.____________________ (R.C. JAGADEESHA) 3.____________________ (B.S. LINGARAJU)

C O N T E N T S

Sl. No. Chapter Particulars Page No.

CERTIFICATE

ACKNOWLEDGEMENT

LIST OF TABLES

LIST OF FIGURES

LIST OF APPENDICES

1 INTRODUCTION

2 REVIEW OF LITERATURE

2.1 Genetic variability, heritability and genetic advance

2.2 Character association and path coefficient analysis

2.3 Importance of G × E interactions, stability models and stability parameters

2.4 G × E interaction and stability analysis in maize

3 MATERIAL AND METHODS

3.1 Selection of material

3.2 Locations

3.3 Experimental layout

3.4 Collection of data

3.5 Observations recorded

3.6 Statistical analysis

4 EXPERIMENTAL RESULTS

4.1 Mean performance of hybrids in different locations



4.4 Stability analysis

5 DISCUSSION

5.1 Mean performance of genotypes in different locations




6 SUMMARY AND CONCLUSIONS

REFERENCES

APPENDICES

LIST OF TABLES

Table No.

Title Page No.

1 Character–wise chronological report on variability, heritability and genetic advance (as % of mean) in maize

2 Importance of genotype × environment interactions in maize and other crops

3a Stability models

3b Stability parameters

4 Mean, range and environmental index for traits in maize at different locations

5 Genetic variability, heritability and genetic advance parameters for grain yield and its components traits in maize

6 Phenotypic correlations among different quantitative traits in maize at Dharwad

7 Phenotypic correlations among different quantitative traits in maize at Arabhavi

8 Phenotypic correlations among different quantitative traits in maize at Bheemarayanagudi

9 Direct (diagonal) and indirect effects of grain yield component traits on grain yield per plant at phenotypic level in maize at Dharwad

10 Direct (diagonal) and indirect effects of grain yield component traits on grain yield per plant at phenotypic level in maize at Arabhavi

11 Direct (diagonal) and indirect effects of grain yield component traits on grain yield per plant at phenotypic level in maize at Bheemarayanagudi

12 Analysis of variance for grain yield and other quantitative characters at three locations in maize

13 Pooled analysis of variance for stability analysis (Eberhart and Russell, 1966) for twelve quantitative traits in maize over three locations

14 Stability parameters for days to 50 per cent tasseling and days to 50 per cent silking

15 Stability parameters for plant height and ear height

16 Stability parameters for cob length and cob girth

17 Stability parameters for number of kernel rows per cob and number of kernels per row

18 Stability parameters for 100- grain weight and grain yield per plant

19 Stability parameters for grain yield per hectare and fodder yield per hectare

20 Stable genotypes identified for each characters over locations

21 Stability of hybrids for different characters

LIST OF FIGURES

Figure No.

Title Page No.

1 Relationship between the regression coefficient (bi) and cob length (cm)

2 Relationship between the regression coefficient (bi) and cob girth (cm)

3 Relationship between the regression coefficient (bi) and number of kernel rows per cob

4 Relationship between the regression coefficient (bi) and number of kernels per row

5 Relationship between the regression coefficient (bi) and 100-grain weight (g)

6 Relationship between the regression coefficient (bi) and grain yield (q/ha)

7 Relationship between the regression coefficient (bi) and fodder yield (t/ha)

LIST OF APPENDICES

Appendix No.

Title Page No.

I Monthly meteorological data during crop growth period (2007-08) at Main Agricultural Research Station, University of Agricultural Sciences, Dharwad

II Monthly meteorological data during crop growth period (2007-08) at ARS, Arabhavi (Gokak)

III Monthly meteorological data during crop growth period (2007-08) at ARS, Bheemarayanagudi

1. INTRODUCTION

Maize (Zea mays L.) is an important cereal crop belongs to the tribe Maydeae, of the grass family, poaceae. The plant is native to South America. Zea mays is the only species in the genus Zea with chromosome number 2n=20. It has two close relatives, among the American Maydeae viz., genus Tripsacum (gamagrass) and Teosinte (Euchlena) which is regarded as the closest relative, and five among Asiatic Maydeae viz., Chionachne sclerachne, Coix trilobachne and polytoca.

The suitability of maize to diverse environments is unmatched by any crop as the expansion of maize to new areas and environment still continues, as it has a range of plasticity. It is grown from latitude 58°N to 40°S, from sea level to higher than 3000 m altitude and in areas receiving yearly rainfall of 250 to 5000 mm (Downsell et al., 1996). Most of the area under this crop is however in the warmer parts of temperate regions and in humid subtropical climate. Highest production is in area having the warmest month isotherms from 21

0 C to 27

0 C and a frost free season of 120 to180 days.

Maize is widely cultivated crop throughout the world. The United States produces almost half of the world’s production; other top producing countries are China, Brazil, France, Indonesia and South Africa.

In world, maize grown with a production 590 million tonnes and productivity 4229 kg per ha and occupies an area of 139 million ha. Maize is known as king of cereals because of its high production potential and wider adaptability. About 30% of world production is used for direct human consumption and as an industrial input, while 70% is used as animal feed. In India, maize is fourth most important cereal crop next to rice, wheat and sorghum. India ranks fourth in maize production with 14.93 million tonnes grown on an area of 7.32 million hectares with productivity of 1904 kg per hectare during 2007.

Karnataka is one of the major maize producing states in the country. During 2007, at total of 9.30 lakh hectares of maize was grown with a production of 16.2 lakh tonnes and average productivity of 2950 kg per ha was realized with almost cent per cent adoption of modern maize production technology. Karnataka state ranked first in yield realization till 1996-97, now lagging behind to Andhra Pradesh. The average productivity of state is much higher than the national productivity. Area under maize is increasing rapidly in the state, because of better environment. Thus there is a greater scope to increase maize productivity to a global level (Anon., 2007).

Corn is increasingly used as an animal feed and fodder crop for both green forage and silage. About 50 to 60 per cent of maize production in India is consumed as food, 30 to 35 per cent goes to poultry, piggery and fish meal, 10 to 12 per cent in wet milling industry e. g. in starch and oil and about 3 per cent in dry milling for traditional requirements like Dalia, Sattu and other food products such as corn bread corn chips.

However, now a day’s corn grain is also a key raw material used for making starch, glucose and oil. Corn starch is used to make sweeteners, as well as such items as disposable forks and spoons. Corn starch is widely used in industrial purposes such as coating for paper and paper products and wallboards for buildings. The pharmaceutical industry also uses corn starch to make pills and other similar products. Recently, high fructose corn syrup has also been made from corn starch. This is mostly used in manufacture of coal and other drinks. In some countries, alcohol made from corn is blended in fuel for gasoline-powered vehicles to reduce emission of pollutants.

It is known fact that the genotypes performing well under a particular environment may or may not perform well over other environments due to genotype-environment interactions (G X E). A genotype with low G X E interactions will have high yielding cultivar, if care is not taken to select for both yield and stability of performance, one may end-up with a high yielding genotype that is suitable only for a particular environment. It therefore necessitates the development of variety with wider adaptability.

Allard and Bradshaw (1964) suggested that the selection of genotypes should be based on least interactions with environments. They further opined that, heterozygous and heterogeneous populations offer the least opportunity to produce varieties which show small

genotype-environment interactions. They used the term “individual buffer” for genotypes where the individual members of a population are well buffered such that each member of the population are well adapted to a wide range of environments. Three-way cross hybrids have both individual and population buffering, while single cross hybrid have only the individual buffering and inbred have no buffering capacity. In India, presently three-way cross hybrids, single cross hybrids and open-pollinated varieties are cultivated over a wide range of environments.

Maize being a cross pollinated crop offer wide scope for the development of hybrids and varieties. In recent years there has been a perceptible improvement in maize production in the state. However, this impact has been higher selective and confined largely to Kharif areas, where the spread of high yielding variety/hybrid is extensively grown due to the availability of a wide range of varieties and hybrids for different agro-climatic conditions.

Therefore, there is a need to compare all the available private and public hybrids and varieties to asses their potentiality and adaptability. This will help the extension workers and the farmers to select the suitable hybrids for commercial exploitation of hybrid vigour.

Keeping these things in view the present study on relative stability analysis of public and private bred hybrids of maize (Zea mays L.) comprising of 10 public and five private bred hybrids was planned and executed with the following objectives.

1. To estimate the genetic parameters; genotypic and phenotypic coefficients of variation, heritability and genetic advance

2. To study the relationship between different quantitative traits and yield

3. To work out path coefficients, so as to find out the relative contribution of different metric traits to the final grain yield

4. To evaluate the performance of maize hybrids across the locations for grain and fodder yield

5. To estimate stability of hybrids and identify the suitable hybrids for specific agro-climatic situations for increased production

2. REVIEW OF LITERATURE

Crop varieties show wide fluctuations in their yielding ability when grown over varied environments or agro-climatic zones. This causes difficulty in demonstrating the superiority of particular variety. Besides yield potential, yield stability over a range of environment is of major concern to the plant breeders and this has direct bearing on the spread of the variety, productivity and total production of the crop. Each genotype has a specific environment for its maximum performance and similarly in a specific environment, a specific genotype performs better.

The literature related to the objective of present study is reviewed below under the following headings.



2.3 Importance of G × E interactions, stability models and stability parameters

2.4 G × E interaction and stability analysis in maize

2.1 GENETIC VARIABILITY, HERITABILITY AND GENETIC ADVANCE

Possibility of achieving improvement in any crop plants depends heavily on the magnitude of genetic variability. Phenotypic variability expressed by a genotype or a group of genotypes in any species can be partitioned into genotypic and phenotypic components. The genotypic components being the heritable part of the total variability, its magnitude for yield and its component characters influences the selection strategies to be adopted by the breeders.

The characterwise chronological report of review of literature on phenotypic coefficient of variability, genotypic coefficient of variability, heritability and genetic advance per cent of mean is presented in the table 1.

2.2 CHARACTER ASSOCIATION AND PATH COEFFICIENT ANALYSIS

2.2.1 Character association

The inter relationship of quantitative characters with yield determine the efficiency of selection in breeding programmes. It merely indicates the intensity of association. Phenotypic correlation reflects the observed relationship, while genotypic correlation underline the true relationship among characters. Selection procedures could be varied depending on the relative contribution of each.

The following are the give review of literature on correlation in maize.

According to Appadurai and Nagarajan (1975), grain weight per ear and grain numbers per row had little effect on yield, while ear length and ear circumference had positive correlation with yield. Kim (1975) reported that 1000-grain weight was negatively correlated with days to silking and days to tasseling. Probecky (1976) reported that yield depends primarily on the number of grains per plant, which inturn depended mainly on the number of grains in the rows.

Sharma et al. (1982) reported that grain yield was positively correlated with grains per ear, 100-grain weight, plant height and ear height. El-Nagouly et al. (1983) concluded that phenotypic and genotypic correlation between yield and days to 50 per cent silking and ear height was positive and highly significant. Path analysis showed that yield was directly influenced by ear height and indirectly affected by days to 50 per cent silking via ear height.

Saha and Mukherjee (1985) observed that grain yield per plant was significantly correlated with ovules per ear, grains per ear and 100-grain weight. Malhotra and Khehra (1986) recorded positive correlation between grain yield and yield components like ear length,

Table 1. Character–wise chronological report on variability, heritability and genetic advance (as % of mean) in maize

Sl. No.

Character Experimental

material PCV (%)

GCV (%)

h2

(bs)

Genetic advance ( as % of mean)

References

36 genotypes ----- ---- 83.8 ----- Gyanendra et al. (1995)

45 F1s 1.39 0.55 15.62 0.45 Satyanaraya and Saikumar ( 1996)

24 hybrids 3.79 2.88 0.58 2.94 Pradeep and Satyanarana et al. (2001)

45 crosses 1034 7.78 72.28 2.97 Choudhary and Chaudhari (2002)

47 genotypes 5.88 5.84 98.6 15.31 Sumathi et al.(2005)

169 lines 6.83 6.37 86.9 12.91 Om prakash et al. (2006)

20 cultivars ----- ----- 0.27 ----- Salami et al. (2007)

1

Days to 50 % tasselling

20 genotypes 1.94 1.98 95.4 ------ Mohammad et al. (2008)

10 set of lines ----- ---- 40.32 ------- Reddy and Agarwal (1992)

45 F1s 1.61 1.12 48.65 1.62 Satyanaraya and Saikumar ( 1996)

38 lines 14.26 14.05 97.12 28.53 Mani and Bisht (1996)

66 genotypes 6.84 6.6 93.23 13.26 Jha and Ghosh (2001)

24 hybrids 3.79 2.8 0.57 2.9 Pradeep and Satyanarana et al.(2001)

20 genotypes 5.06 4.81 90.38 9.41 Satyanarayana et al. (2005)



20 cultivars ----- ----- 0.2 ------ Salami et al. (2007)

2

Days to 50 % silking

20 genotypes 2.34 2.43 92.6 ------- Mohammad et al. (2008)

10 set of lines ---- ---- 67.02 ----- Reddy and Agarwal (1992)

2 set of F3 proginies 8.1 10.55 58.93 12.8 Robin and Subramanian (1994)

36 genotypes 21.1 19.4 84.8 36.8 Gyanendra et al. (1995)


3 Plant height

66 lines 18.86 18.34 94.55 36.73 Jha and Ghosh (2001)

24 hybrids 10.1 8.08 0.84 12.27 Pradeep and Satyanarana et al.(2001) 45 crosses 11 10.21 86.08 13.26 Choudhary and Chaudhari (2002)




20 cultivars ------ ------ 0.34 ------- Salami et al. (2007) 20 hybrids 4.78 6.64 52 ----- Mohammad et al. (2008)

10 set of lines -- --- 71.28 --- Reddy and Agarwal (1992)



45 crosses 17.08 14.51 72.15 17.27 Choudhary and Chaudhari (2002)


169 lines 19.65 29 93.4 40.71 Om prakash et al. (2006)

20 cultivars --- ----- 0.35 ---- Salami et al. (2007)

4

Ear height

20 genotypes 7.25 11.51 39.6 --- Mohammad et al. (2008)

2 sets of F3 progenies 9.23 11.01 69.25 15.82 Robin and Subramanian (1994)


45 crosses 10.53 7.28 47.84 7.05 Choudhary and chaudhari (2002)


5

Cob length




6

Cob girth






7

Number of kernel rows per cob

169 lines 13.49 12.44 85 23.51 Om prakash et al. (2006)


36 genotypes --- 15.9 --- 29.3 Gyanendra et al. (1995)

8 Number of kernels per row


45 crosses 38.03 18.26 48.02 7 Choudhary and Chaudhari (2002) 47 genotypes 13.99 13.78 97.01 35.83 Sumathi et al.(2005)


41 S1 families --- --- 52 13.6 Alika (1994)


36 genotypes --- --- 82.1 27.6 Gyanendra et al. (1995)





9

100-grain weight



36 genotypes 24.6 15.9 --- --- Gyanendra et al. (1995)




10

Grain yield per plant

20 genotypes 10.85 11.92 82.8 ---- Mohammad et al. (2008)

66 lines 20.43 20.11 96.88 40.78 Jha and Ghosh (2001) 11

Grain yield per hectare

20 cultivars --- --- 0.24 ---- Salami et al. (2007)

ear circumference, number of rows per ear, 100-grain weight, shelling percentage, days to silking, ear height and plant height.

Tyagi et al. (1988) opined that grain yield was influenced more by ear weight, ear length, plant height, kernels per row and 100-grain weight. They also reported that 50 per cent pollen shedding and silking had a direct correlation with yield and so, early maturing genotypes had relatively low yield. Mahajan et al. (1990) concluded that grain yield was positively correlated with ear length, number of kernels per row and plant height.

Singh et al. (1991) noticed that grain yield per plant had significant positive correlations with plant height and ear weight in F1 and F2 generations under alkaline soil and with leaf area in both the generations under normal soil. Debnath and Khan (1991) revealed that days to silking, plant height, number of kernels per row and 1000-grain weight had strong positive contributions to grain yield.

Dash et al. (1992) reported that maturity traits showed a negative correlation with yield per plant. Path coefficient analysis revealed that ear diameter, plant height, ear length and 100-seed weight were the major factors contributing to yield.

Boraneog and Duara (1993) observed that plant height and ear height were found to be significant and positively correlated with yield.

Saha and Mukherjee (1993) reported significant positive correlations between grain yield per plant with 100-grain weight, ear length, ear circumference, number of grain rows per ear and number of grains per row. The ear circumference and number of grains per row had higher direct and indirect effects on grain yield.

Krishnan and Natarajan (1995) obtained highly significant positive association between grain yield and plant height, ear length, ear weight, number of kernels per row, dry matter production and harvest index. Path coefficient analysis revealed that selection on any trait in maize will influence the grain yield only through dry matter production.

Packiaraj (1995) observed direct positive correlation between grain yield and number of grains per row. Rahman et al. (1995) reported that grain yield was significantly and positively correlated with plant height, ear height, number of grains per ear and 1000-grain weight. Path analysis revealed that ear height, plant height and 1000-grain weight were the main contributors for grain yield.

The studies conducted by Sreekumar and Suma Bai (1995) revealed that plant population recorded high coefficient of variation, heritability and genetic advance for plant height indicating that selection based on these characters will result in improving fodder yield. Highest genotypic correlations were observed between plant population and dry fodder yield.

According to Satyanarayana (1996) grain yield was positively correlated with kernel rows per ear, ear length, ear circumference and 100-grain weight.

Guirong (1997) obtained direct correlation between yield and 1000-grain weight, rows per ear, ear circumference and kernels per row. According to Kumar and Kumar (1997) values of genotypes correlation were slightly higher than the corresponding phenotypic values. Significant positive correlation was recorded for plant height, days to 50 per cent silking, ear length and ear height with yield per plant.

Annapurna et al. (1998) showed that seed yield was positively and significantly correlated with plant height, ear circumference, number of seeds per row, number of seed rows per ear, number of seeds per ear and test weight.

Khakim et al. (1998) noticed that grain yield was positively correlated with plant and ear insertion height, leaf area, ear number, ear length, row number, grain number per row and per cob, grain weight per cob, ear weight and 1000-grain weight. Manivannan (1998) reported that ear circumference, number of kernels per row, 1000-grain weight, kernels per row and ear length had significant and positive correlation with grain yield. High and positive direct effects were observed for kernel rows and 1000-grain weight.

Alok Kumar et al. (1999) revealed that the number of grains per row, number of rows per ear, ear circumference, ear length, days to 50 per cent flowering and days to 50 per cent maturity had direct effect on grain yield.

Datu (1999) indicated that at the phenotypic and additive genetic level, growth period was strongly correlated with plant height and leaf number.

Firoza Khatun et al. (1999) reported that grain yield per plant was positively and significantly correlated with 1000-grain weight, number of kernels per ear, ear circumference and ear insertion height. Path analysis showed that 1000-grain weight, number of kernels per ear and number of ear per plant was more important components for determining grain yield.

Gautam et al. (1999) suggested that maximum correlation of grain yield was obtained with number of kernels per row followed by leaf area, plant height, tassel length and ear length. Path analysis revealed that the number of kernels per row, plant height, ear width, leaf area and 1000-grain weight had positive direct effect on grain yield.

Mani et al. (1999) reported that grain yield per plant indicated highly significant positive correlations with all the attributes and were highest with ear weight per plant. Path analysis also suggested that ear weight per plant followed by grains per row were the best direct contributors to grain yield per plant.

Mohammad et al. (1999) showed that correlation coefficients had positively and highly significant positive influence of plant height, number of leaves, leaf area per plant, stem girth, crude protein and ear height on forage production.

Nawar et al. (1999) observed that additive components were significant for row number. Highly significant positive correlation coefficients were detected among yield per plant. Components of ear and plant height, number of kernels per row imparted positive direct effect towards grain yield followed by plant height. Rather et al. (1999) observed that days to 50 per cent silking was positively correlated with ear height and grain yield.

Geetha and Jayaraman (2000) studied the effects of different quantitative traits on grain yield in 90 hybrids and suggested that the number of grains per row exerted maximum direct effect on grain yield. Kumar and Kumar (2000) suggested that selection based on plant height with greater ear weight, number of seeds rows per ear and number of seeds per ear was desirable for grain yield.

Netaji et al. (2000) reported that yield per plot was significantly and positively correlated with all the characters except days to 50 per cent tasseling, silking and dry husk. Maximum variability was observed for plant height, followed by ear height and test weight. Umakanth et al. (2000) observed that yield per plot was positively correlated with plant height, 1000-grain weight and kernels per ear, but days to 50 per cent silking had negative correlation with yield. Vaezi et al. (2000) noticed that grain yield was significantly and positively correlated to ear weight, ear circumference, kernel weight and number of kernels per row. Path analysis, for grain yield showed that kernel weight and kernel depth had the highest positive effect on grain yield.

Singh and Dash (2000) revealed high positive correlation of green fodder yield with plant height, leaf length and leaf stem ratio. Prodhan and Rai (2000) reported that significant positive correlation of popping expansion was found with popping percentage, tenderness and pericarp thickness, while popping expansion showed significant negative association with grain weight and non-significant negative association with grain yield. On the other hand, grain yield was strongly associated with grain weight.

Pradeep and Satyanarayana (2001) concluded that grain yield was positively associated with plant height, ear height, ear length, ear circumference, number of seed rows per ear and test weight. Swarnalatha and Shaik (2001) indicated that the plant height, days to 75 per cent silking and maturity, ear length, number of seeds per row and 100-grain weight positively influenced the yield directly and also indirectly through several yield components.

Umakanth and Khan (2001) observed that grain yield per plot showed significant and positive correlations with ear circumference, ear length, plant height and 100-seed weight. Path analysis revealed that plant height followed by number of seeds per row, 100-seed weight, ear length and ear circumference showed maximum positive direct genotypic effects as well as indirect contribution through other characters on grain yield.

Xie-Dasen et al. (2003) noticed significant positive correlation between the width and length of ear, width of axil and net weight of ear. Srivas and Singh (2004) observed that dry

fodder yield per plant was significantly and positively associated with green fodder yield and its contributing traits such as plant height, days to 50 per cent silking, number of leaves per plant, stem girth, leaf blade length, leaf width and sheath length. Shelake et al. (2005) noticed that grain yield was positively and highly correlated with number of grains per cob, biological yield per plant, harvest index, 100-grain weight, cob length, number of grain rows per cob and cob girth.

Ei-Shouny et al. (2005) showed that grain yield per plant correlated positively and significantly with ear diameter, ear length, number of kernels per row, 100-kernel weight, number of rows per ear, ear height, plant height and days to silking under normal planting date and with number of kernels per row, ear diameter, 100-kernel weight, ear length, number of rows per ear, ear height and days to silking under late planting date.

Sumathi et al. (2005) genotypic correlation studies indicated that ear weight, number of rows/ ear, number of kernels per row, and total number of kernels per ear were positively associated with grain yield. Oil per cent exhibited negatively non-significant correlation with grain yield, whereas it showed positive association with number of rows/ ear only. Path coefficient analysis revealed that number of kernels per row showed high direct effect on grain yield followed by 100 seed weight, number of rows per ear and total number of kernels per plant. Balbinot Junior et al. (2005) observed that the number of grains per row showed the highest total correlation with grain yield.

Wali et al. (2006) observed that yield was positively associated with plant height, ear length, ear circumference, number of kernels per row, fodder yield per plot and 100-grain weight, but was negatively correlated with number of days to 50 per cent silking at the phenotypic and genetic levels. The grain yield per plant was positively associated with plant height, ear length, ear circumference, number of kernels per row, fodder yield per plot and 100-grain weight at the phenotypic and genetic levels.

Harjinder Singh et al. (2006) reported significant positive correlations for grain yield with days to 75 per cent husk, plant height, ear height, and number of ears. Tan Heping et al. (2006) noticed that grain yield was significantly correlated with plant height, ear diameter, ear length, 100-kernel weight and grain production rate. Grain yield was most highly correlated with ear diameter, followed by 100-kernel weight, plant height, ear length and grain production rate. Li Yong Hong et al. (2006) showed significant positive correlation between kernels per row and kernel yield and between cob diameter and kernel yield.

Abirami et al. (2007) indicated that grain yield showed positive association with oil content and protein content. Path analysis showed that the weight of the cob contributed to the maximum direct effect to grain yield. It implied that selection for weight of the cob will be highly effective for the improvement of grain yield.

Sofi and Rather (2007) reported that the genotypic correlation coefficient revealed that ear diameter, 100-seed weight, ear length, number of kernel rows per ear and number of kernels per row showed the greatest correlation with grain yield. Path analysis indicated that 100-seed weight had greatest direct effect on grain yield, followed by number of kernels per row, number of kernel rows per ear, ear length and ear diameter.

Bhiote et al. (2007) showed positive and significant correlation between dry matter and crude protein yields with green forage yield and had positive direct influence on their correlation with green forage yield. Zhang Li et al. (2007) correlation analysis indicated that the test weight of kernel types was significantly and positively correlated with kernel weight and grain yield and negative correlation was obtained between test weight and gross fat content. Test weight was positively and significantly correlated with protein content and starch content.

Mohammad Akbar et al. (2008) noticed that plant height had highly significant genotypic and phenotypic association with cob height and days to 50 per cent tasseling with days to 50 per cent silking. All traits had significant genotypic association but not significant phenotypic association with grain yield.

2.2.2 Path coefficient analysis

Assuming yield is a contribution of several characters which are correlated among themselves and to the yield, path coefficient analysis was developed (Wright, 1921; Dewey

and Lu, 1959). Unlike the correlation coefficient which measures the extent of relationship, path coefficient measures the magnitude of direct and indirect contribution of a component character to a complex character and it has been defined as a standardized regression coefficient which splits the correlation coefficient into direct and indirect effects.

Jha et al. (1998) observed that plant height exhibited maximum direct positive effect on green fodder yield followed by ear leaf area and number of leaves per plant. Singh et al. (1999) indicated that the highest positive direct effect on yield was exhibited by kernel rows per ear, followed by plant height and ear diameter.

Vaezi et al. (2000) showed that 300-kernel weight and kernel depth had the highest positive effect on grain yield whereas ear diameter had a negative indirect effect on grain yield through some traits. Geetha and Jayaraman (2000) observed number of grains per row exerted a maximum direct effect on grain yield.

Swarnalatha and Shaik (2001) indicated that the plant height, days to 75 per cent silking and maturity, ear length, number of seed rows per ear, number of seeds per row and 100-seed weight positively influenced the yield directly and also indirectly through several yield components.

Guang et al. (2002) showed that importance of eight yield components to grain yield and suggested that more attention should be paid to cob length, cob diameter and kernel percentage.

Anees and Mohammad (2003) reported that vegetative phase had the highest positive direct contribution to grain yield per plant followed by growing degree days to tasseling and growing degree days to maturity. Growing degree days to the reproductive phase had the highest negative direct effect on grain yield. Venugopal et al. (2003) indicated that number of seeds per row followed by 100-seed weight, days to 50 per cent tasseling, ear girth and plant height contributed directly towards grain yield per plant. Number of seed rows per ear had a direct positive contribution towards grain yield, ear length, 100-seed weight and number of seeds per row had a indirect negative influence on grain yield.

Viola et al. (2003) revealed that early silking and harvesting of fresh cobs, greater plant height, cob length, cob weight, cob height and number of cobs per plant and lesser cob girth directly contributed to increased cob yield. Singh et al. (2003) observed that ear leaf area had the highest positive direct effect on green fodder yield per plant at genotypic and phenotypic levels followed by dry matter yield per plant, ear length and days to 50 per cent silking. Ear length had the maximum direct effect on grain yield followed by 500-kernel weight and ear leaf area. Number of leaves per plant, leaf:stem ratio and girth of basal internode had also highly positive direct effect on grain yield per plant.

Bao Heping et al. (2004) reported that maize yield was mainly influenced by ear length, followed by number of kernels per row, ear width, number of rows per ear, growth period and 1000-seed weight. Kernel percentage per ear and number of pointless ears had minimum effect on maize yield.

Arun and Singh (2004a) reported that days to 50 per cent silking and cob length had the maximum positive direct effect on grain yield. Whereas, days to 50 per cent tasseling and days to maturity had maximum negative effect on grain yield. Srivas and Singh (2004) observed that characters such as plant height, days to 50 per cent silking, stem girth, leaf length, leaf width and number of leaves per plant had positive direct effect on dry fodder yield at phenotypic levels.

Patel et al. (2005) reported that dry matter yield per plant, number of leaves per plant, days to 50 per cent silking and plant height had positive direct effects on green fodder yield. Shelake et al. (2005) revealed high magnitude of direct effects for all characters at the genotypic level. The number of days to 50 per cent tasseling, number of days to 50 per cent silking and harvest index showed higher genotypic direct effect. Biological yield per plant had the highest negative genotypic direct effect on grain yield.

Kumar et al. (2006) observed that days to 50 per cent tasseling, Anthesis Silking Interval (ASI), ear height and 100-seed weight had highest direct effect on grain yield. The days to 50 per cent silking exhibited negative direct effect on grain yield. Wang (2006) reported that kernel weight per ear mainly affected by ear length and ear diameter and the ear

length with bearing kernel played an important role on kernel weight per ear in high yielding combinations.

Jayakumar et al. (2007) noticed that grains per row recorded maximum positive direct effect on grain yield followed by ear length, ear girth, days to tasseling, total sugars and plant height. The maximum negative direct effect on grain yield was recorded by kernel rows followed by days to silking, crude protein, grain weight, days maturity, shelling percentage and leaves above upper most ear.

Abirami et al. (2007) showed that weight of the cob contributed to the maximum direct effect to grain yield. Sofi and Rather (2007) indicated that 100-seed weight had the greatest direct effect on grain yield followed by number of kernels per row, number of kernel rows per ear, ear length and ear diameter. Xie-Zhen Jiang et al. (2007) showed that kernels per plant was arranged for the tap position among the many agronomic traits that contributed to the yield enhancement of a single plant and was followed by kernels per row, 1000-kernel weight and leaf orientation value.

Mohammad et al. (2008) showed that all traits exerted positive direct effect on grain yield per plant except days to 50 per cent silking.

2.3 IMPORTANCE OF G × E INTERACTIONS, STABILITY MODELS AND STABILITY PARAMETERS

Various studies on genotype × environment interaction and its importance, stability models and stability parameters are presented in Table 2, 3a and 3b, respectively.

2.4 G × E INTERACTION AND STABILITY ANALYSIS IN MAIZE (Zea mays L.)

Ron Parra (1985) studied the performance of 23 varieties in 92 environments under rainfed and irrigated conditions at experimental stations and on farmer’s land and the results

showed variety × location and variety × location × year interactions were more important than

the variety × year. Stability analysis indicated that stability of varieties was influenced by maturity, source of germplasm. Correlations of the stability parameters revealed that the mean and regression coefficient were significantly correlated and repeatable over environments.

Jha et al. (1986) evaluated 36 experimental double cross hybrids and two commercial hybrids of maize for grain filling period, grain filling rate and grain yield at six locations. They found significant differences existed among genotypes for grain filling period and grain yield

for G × E interaction in case of all three traits.

Sain Dass et al. (1987) evaluated eight varieties of maize for stability and for

performance over six environments and observed that mean squares due to G × E interactions were significant for all characters except ear weight and days to silking.

Patel and Sanghi (1989) evaluated 64 genotypes of maize in four environments for

their stability. The result showed that Syn Do × (CM 202 × CM 111) and lowa 2 early H0 gave good grain oil yields under favourable environments, but under poor to average conditions,

lowa 2 early H0 × Vijay and Soan × Suwan 2 W were best for this trait. In general, H201-45 had good grain yield, oil content and earliness.

Mahajan et al. (1991) studied 28 hybrids obtained by crossing eight genetically diverse inbred lines of maize and tested in eight environments over two seasons. They reported that heterogeneity in the genetic material for stem diameter, internodal length and ear leaf area was contributed by non-linear component while for number of leaves, plant and ear height was contributed by both linear and non-linear component. Prasad and Singh (1991) analyzed 64 maize genotypes under six different environments for their adaptability

and they revealed G × E interaction for all characters.

Satyanarayana and Kumar (1995) reported significant variation for eight promising

genotypes of maize and G × E interaction for grain as well as fodder yields. Gharde and Deshmukh (1996) evaluated 14 genotypes of maize including eight hybrids and six composites for stability under eight artificially created environments. The result showed that G

Table 2. Importance of genotype × environment interactions in maize and other crops

Sl. No.

Crop Details of study Authors

Gave meaning of phenotype. 1. -

Importance of environmental process. Johannsen (1909)

2. - Coined the term ecotype. Turesson (1922).

3. Potato Existence of genotype × environment interaction was revealed. Fisher and Mackenzie (1923)

4. Corn Use of various components to sort out the effects of genotype, environment and their interactions to equate the observed means of ANOVA to the expected means on the random model.

Sprague and Federer (1951)

5. Lima bean Genetically was used as a quantitative measure of phenotypic stability to describe the varietal adaptability over a range of environments.

Allard (1961)

6. - Regression was used as a quantitative measure of phenotypic stability to describe the varietal adaptability over a range of environments.

Finlay and Wilkinson (1963)

7. - The genotypes environmental interactions are usually present irrespective of weather the material under test is pure line, hybrid, top crosses etc. and it reduces the progress from selection.

Comstock and Moll (1963)

8. - Implications of genotypes and environmental interactions in applied plant breeding.

Allard and Bradshaw (1964)

9. Maize Classification of environment into predictable and unpredictable components. Eberhart and Russell (1966)

10. Grasses Importance of genotypes × environmental interactions which posses serious problems in interpreting evolutionary trends and rationalization of policy and procedure in breeding for improved performance in economic crops.

Breese (1969)

11. Spring barley,

oats, wheat and bengalgram

Significance of both linear and non-linear portion of genotype × environment interaction was due to wide genetic variability.

Perkins and Jinks (1968a), Paroda and Hayes (1971), Paroda et al. (1973), Chaudary et al. (1978), Mehra and Ramanujam (1979) and Yadava and Kumar Prakash (1979)

Table 3a. Stability models

Sl. No.


1. Maize Use of linear regression approach, deviation from regression line was regarded as another component of varietal stability.

Eberhart and Russell (1966)

2. - Biometrical models were proposed to analyse stability genotype. Knight (1970), Freeman (1973), Hill (1975), Brain Westcott (1986) and Gautam et al. (1986)

3. - Phenotypic approach for partitioning G × E interaction component of variability into linear and non-linear portion – joint regression analysis joint regression analysis.

Hill (1975), Finlay and Wilkinson (1963), Eberhart and Russell (1966), Freeman and Perkins (1971) and Shukla (1972)

4. - Genotypic approach for stability analysis was reported. Hill (1975), Bucio Alanis and Hill (1966), Bucio Alanis et al. (1969), Perkins and Jinks (1968b), Breese (1969) and Perkins (1970)

5. - Basic stability model and various components approach model. Jain (1982), Yates and Cocharan (1938)

6. Rice Principal component analysis helps to provide a two dimensional representation of genotype environment pattern which allows the response of each environment to be directly identified.

Mahajan and Prasad (1986)

7. - Proposal of stability analysis of cultivars in long-term varietal traits. Qifa Zhang and Shu Geng (1986)

Table 3b. Stability parameters

Sl. No.


1. -

Phenotypic stability was defined as the ability of an individual to produce a certain narrow range of phenotypes in different environments stability factor for the i

th

genotype was defined as S.F. = XHE/XHL, where X – Mean, HE and HL are high and low environments. A unit value of S.F. indicates maximum phenotypic stability.

Lewis (1954)

2. -

The stability parameter of a genotype was found to be its phenotypic regression coefficient (bi). A genotype with unit bi value and higher mean value is said to be a stable variety for a range of environment. As the mean yield decreases, genotypes with high or low slopes are regarded as being adapted to favorable and unfavorable environments, respectively.

Finlay and Wilkinson (1963)

3. Maize

Deviation from the regression of ith varietal j

th environment was considered as

another important component of varietal stability. A stable variety being one with regression slope nearer to one, the least deviation from regression line and a higher mean yield.

Eberhart and Russell (1966)

4. Grasses, potato Application of this model in predicting the relative performance of a population over years and locations to find out differences in stability.

Breese (1969), Tai George (1971)

5. Rice Phenotypic index (Pi) was found to be an easier method of estimating the phenotypic stability Pi of a genotype being the average sum of deviation of the genotypic mean performances and the grand means.

Ram et al. (1970)

6. Potato

A method of phenotypic stability analysis using structural relationship analysis was

presented to partition the genotype × environmental effects of a variety into linear

(α) and non-linear (λ) components. A perfect stable variety is defined as the one

with α = 1 and λ = 1 average stable variety has α = 0 and λ = 1.

Tai george (1971)

7. -

The genotypes × environmental interaction sum of squares was partitioned into components for each genotype separately by considering stability variance of the i

th

genotype defined as the variance over environments of (gij – Biej + dij), where gij = the interaction between i

th genotype at j

th environment and eij is the mean of error

component over replications.

Shukla (1972)

8. Wheat A method to work out on adaptability index (Ai) as measure of stability was suggested. Magnitude of Ai indicates the degree of adaptability and its sign determines the nature of performance.

Choudary et al. (1972)

9. Wheat The relative ranking of genotypes in Eberhart and Russell and Perkins and Jinks models were found to be same.

Luthra and Singh (1974), Verma and Virk (1983)

10. Cotton Regression coefficient (bi) of yields was used on environmental indices as a measure of adoption and coefficient of determination (r

2) as a stability measure.

Bulbro and Ray (1976)

11. Peanut Parameters and regression analysis for comparison of stability statistics was identified as criteria for cultivar development in peanut.

Anderson et al. (1989)

× E interaction was significant for all the character under study. Gyanendra et al. (1996) analyzed ten genotypes of maize under six environments for their stability. The results

revealed that genotype × environment interaction was significant for all the traits except cob

girth, rows per cob, seeds per row and 100-seed weight, but the linear component of G × E interaction was significant for all of the traits except seeds per row.

Chen Xuejun (1997) analyzed 11 maize hybrids for yield stability in seven Chinese provinces. Zhongdan had the highest average yield and good stability.

Mani and Singh (1999) assessed the yield stability in 12 maize genotypes comprising of hybrids and components over three diverse environments. Composite ‘Navin’ followed by double top cross hybrid ‘EHF 1121’ were found to be most stable with greater yield than over all mean. Choukan (1999) studied nine medium maturity maize hybrids under ten locations for

three years and observed hybrids (B73 × K1264) × M017/11-1, (K1264/1 × L17/12-1) ×

M017/11-1, K1259/3 × B73 and (K2509 × B73 × M017 were classified as stable hybrids.

Gargi and Saikia (2000) studied the phenotypic stability of 21 baby corn genotypes over four environments for yield and yield attributing characters. The mean square due to

genotype × environment (G × E) interaction and both linear and non-linear components were significant for all the characters except stem girth. For all the characters except stem girth, both linear and non-linear components were significant. While, for the stem girth only non-

linear component contributed to G × E interaction variance. Gomes et al. (2000) studied 30 maize genotypes grown in 14 environments in Brazil. Significant differences in yield stability were observed among the genotypes studied.

Carvalho et al. (2000) evaluated 21 maize cultivars at 26 environments and observed

significant differences in yield due to environments, cultivars and cultivar × environment interactions. Choukan (2000) studied ten single cross hybrids of maize for their stability over 14 locations for two years. The result revealed that heterogeneity of regression coefficients was significant only for kernel depth and kernel rows, number per ear. The highest grain yield was produced by hybrid No. 6 and 10.

Burak and Broccoli (2001) evaluated fourteen pop corn hybrids in 5 locations for their

adaptability and results revealed significant differences in genotype × environment interaction, genotypes and environments. Ogunbodede et al. (2001) evaluated seven early maturity open pollinated and five yellow hybrid maize varieties over 22 locaitons. Stability analysis revealed significant location effects for grain yield in two sets of maize varieties tested.

Pixley and Bjarnason (2002) evaluated 18 single cross, 18 three-way and 18 double cross hybrids and eight open pollinated cultivars grown at 13 tropical locations on four

continents. Stability analysis showed genotype × environment interactions and sums of squares for deviations from linear regression for grain yield and protein concentration in grain and were largest (indicating least stability) for single cross hybrids, followed by three way, double cross and open pollinated cultivars.

Dodiya and Joshi (2003) studied 86 genotypes for genotype × environment (G × E)

interaction and stability parameters with respect to yield and maturity over three locations. G × E interaction was significant for days to 50 per cent maturity and grain yield.

Nirala and Jha (2003) conducted an experiment on 19 parents and 70 crosses to identify the high yielding and stable fodder maize genotypes. A highly significant mean squares due to genotypes and environments for al the traits indicated the presence of significant differences among the genotypes and environment.

Padam et al. (2003) evaluated 15 single cross maize hybrids and eight standard

controls for genotype × environment interactions. Vijay composite showed the highest grain yield and overall mean and regression value among the cultivars.

Reddy et al. (2004) evaluated ten recycled maize inbred lines in four environments for yield and some yield component traits. The stability analysis revealed that mean squares for

genotype × environment interaction effects were significant for all traits. Abera et al. (2004) evaluated ten maize genotypes across five locations and three years in Ethiopia. The stability analysis identified BH-660, Gibe-1 and E1 as the more stable genotypes, while Kulani, BH-140 and E5 were adapted to specific environments.

Arun and Singh (2004b) selected seven parents and their 21 single crosses and two standard checks of maize were used for estimating the stability parameters by raising crop at four different locations. The stability analysis exhibited highly significant variation for

genotypes, environments, G × E interactions and pooled deviation for most of the characters.

Reddy et al. (2004) carried out AMMI analysis to maize yield trials conducted with 45 hybrids over environments (years/location) to identify suitable and stable hybrids. Grain yield

data were subjected to AMMI analysis and revealed significant G × E interaction which could be attributed to different rankings of the genotypes across environments. Murakami et al. (2004) evaluated 31 maize hybrids for their stability and adaptability using multi-variate analysis.

Schmildt and Cruz (2005) compared two methods for evaluating the adaptability and stability of maize (Eberhart/Russel and Annichiarico) using 33 previous maize cultivars grown in eight environments. The Annichiarico method was determined as more effective than the Eberhart/Russel method.

Bhat and Singh (2005) analyzed 15 hybrids of six maize inbreds for stability under different moisture regimes and locations for different characters such as days to 50 per cent silking, anthesis – silking interval, plant height, grain number per row, test weight and grain

yield per plant. Ib 3 4 was stable for days to silking,

while Ib 2 × Ib < xub > 6 performed stability for plant height and Ib

 5 × Ib 3 sub <, Ib 3 × Ib 6 and Ib 3 × Ib 5 were stable for grain number per row. Ib

 2 × Ib 5 and In 3 × Ib 5 

were stable for 100-seed weight. Ib 5 × Ib 6 and Ib 

2 × Ib 3 were stable for grain yield per plant.

Abera et al. (2006) tested ten maize genotypes for grain yield across 15 environments. The combined analysis of variance for environment (E), genotypes (G) and GE interaction was highly significant. Kaundal and Sharma (2006) evaluated 21 elite genotype of maize in four environments to assess the genotype environment interactions for grain yield

and some other quantitative traits. The combined environment and genotype × environment variance were highly significant for all the traits.

Javed et al. (2006) analyzed stability six maize genotypes index six locations. Pooled analysis of variance for grain yield indicated significant differences for genotypes across the environment, environment across genotypes and their interactions. These significant interactions indicated uneven performance of the genotypes across the environment and year. R-2302 and R-2210 showed highly stable performance across the environments.

Abdulai et al. (2007) studied nine genotypes for four year at eight locations in Ghana. Stability analysis identified seven genotypes were stable, when b values alone were considered. When the b values and the deviations from regression (S

2d) were considered

(GH24 × 1368) × 5012 and (GH22 × 1368) × 5012 were the most stable, but when coefficient of determination was added to the b value and S

2d, GH132-28 was the most stable genotype.

Amit Dandeech and Joshi (2007) analyzed 74 genotypes of maize under four

environments. Variance due to genotype, environment, genotype × environment interactions were found significant for oil, starch and protein contents. Cardoso et al. (2007) assessed adaptability and yield stability of 15 maize cultivars and two triple maize hybrids under 41 environments and revealed the performance of the cultivars significantly varied among the environments.

3. MATERIAL AND METHODS

The details of materials used and methods followed in carrying out the present investigation are presented in this chapter.

3.1 SELECTION OF MATERIAL

The experimental material consisted of 15 hybrids, developed by both private and public institutions. The ten public bred hybrids viz., ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-4, ARBMH-5, ARBMH-6, ARBMH-7, ARBMH-8, ARBMH-9 and ARBMH-10 were developed by All India Co-ordinated Maize Improvement Project (AICMIP), ARS, Arabhavi. The five private bred maize hybrids from corporate sector were included they are 900 M, Bio-9681, Pinnacle, All rounder and K-235.

3.2 LOCATIONS

The present study was carried out by undertaking multi-location trials of above said 15 hybrids in three locations viz., Main Agricultural Research Station, Dharwad, Agricultural Research Station, Arabhavi and Agricultural Research Station, Bheemarayanagudi, University of Agricultural Sciences, Dharwad, during kharif 2007. The location details are given below.

E1: Irrigated environment, Main Agricultural Research Station, UAS, Dharwad located in Northern Transition Zone (Zone-8) of Karnataka.

E2: Irrigated environment, Agricultural Research Station, Arabhavi located in Northern Dry Zone (Zone-3) of Karnataka.

E3: Irrigated environment, Agricultural Research Station, Bheemarayanagudi located in North-eastern Dry Zone (Zone-2) of Karnataka.

The weather parameters like annual rainfall of 2007 and average rainfall are presented in Appendix I, II and III.

3.3 EXPERIMENTAL LAYOUT

The experiment under each environment was laid out in a Randomized Block Design (RBD) with three replications. The genotypes were allotted randomly to the 15 plots of each

replication. The plot size was 4.5 m × 4.0 m with inter and intra row spacing of 75 and 20 cm, respectively. The sowing dates at MARS, Dharwad, ARS, Arabhavi, and Bheemarayanagudi were 6

th, 3

rd and 18

th August, 2007, respectively. The recommended package of practices for

respective environment at each location was followed to raise good crop.

3.4 COLLECTION OF DATA

10 plants were tagged randomly for recording observations for each entry in each replication for all the quantitative characters except for days to 50 per cent tasseling and silking. Mean of 10 plants for each entry in each replication was worked out for each character at each location and used for statistical analysis.

3.5 OBSERVATIONS RECORDED

Observations on the following quantitative characters were recorded at appropriate stages of plant growth in all the three locations.

3.5.1 Days to 50 per cent tasseling

The number of days from sowing upto the day on which 50 per cent of the plants showed tassel emergence was recorded as days to 50 per cent tasseling.

3.5.2 Days to 50 per cent silking

The number of days from sowing upto the day on which 50 per cent of plants showed silk emergence was recorded as days to 50 per cent silking.

3.5.3 Plant height (cm)

Height of the plant from ground level upto the base of fully opened flag leaf was recorded in centimeters as plant height when plants were mature.

3.5.4 Ear height (cm)

Height from ground level upto the base of the upper most cob bearing internode was recorded as ear height in centimeters.

3.5.5 Cob length (cm)

Cob length was measured from bottom to the top of the cob and recorded in centimeters.

3.5.6 Cob girth (cm)

Cob girth was measured and recorded in centimeters as the thickness of the ear i.e., at the middle of the ear.

3.5.7 Number of kernel rows per cob

Number of kernel rows per cob was counted and recorded.

3.5.8 Number of kernels per row

Number of kernels per row was counted and average was recorded as number of kernels per row.

3.5.9 Hundred grain weight (g)

The weight of sun dried 100-grain samples drawn at randomly in each plot were recorded in grams at 15 per cent moisture content.

3.5.10 Grain yield per plant (g)

Grain yield per plant expressed in grams was recorded by weighing the grains obtained after shelling of cobs from individual plant.

3.5.11 Fodder yield (t/ha)

Fodder yield per plot was recorded (kg/plot) and converted and expressed in tonnes per hectare.

3.5.12 Grain yield (q/ha)

GY= (yield per plot (g) / Plot Size) *10000 m2

3.6 STATISTICAL ANALYSIS

The statistical analysis of the data on the individual character was carried out on the mean value of 10 randomly selected plants on each genotype from each of the three replications. The mean data was analyzed at computer centre, University of Agricultural Sciences, Dharwad. Different statistical methods employed for analysis are as follows.

3.6.1 Mean

On the basis of individual plant observations, the population mean for each character was computed as follows.

X= 1/n (Σni=1Xi)

X=population mean

Xi= individual value

n= number of observations

3.6.2 Range

The minimum and maximum values on the basis of individual plant observations were used to indicate the range of given character.

3.6.3 Analysis of variance

The analysis of variance for different characters was carried out by using the mean data for each location separately in order to partition the variability due to different sources. The method given by Panse and Sukhatme (1964) was followed.

The structure of ANOVA is as follows

Source of variation d. f. MSS Expected value of MSS Cal F.

Replication

Genotypes

Error

(r-1)

(g-1)

(r-1) (g-1)

M1

M2

M3

--

σ2e + r σ

2g

σ2e

M2/M3

Total (rg-1)

3.6.4 Estimation of genetic parameters

Genetic parameters were estimated for different traits on maize genotypes.

3.6.4.1 Genotypic and phenotypic coefficient of variation

The genotypic and phenotypic coefficient of variation was computed according to Burton and Devane (1953) and expressed as percentage.

Genotypic coefficient of variation (GCV) = (σg/X) x100

Phenotypic coefficient of variation (PCV) = (σp/X) x100

Where,

σg = Genotypic standard deviation

σp = Phenotypic standard deviation

X =general mean of the character

PCV and GCV values were categorized as low, moderate and high values as indicated by Sivasubramanian and Menon (1973) as follows.

0-10 % : Low

10-20% : Moderate

>20 : High

3.6.4.2 Heritability (h2)

Heritability in broad sense was estimated as the ratio of genotypic variance to the phenotypic variance and expressed in percentage (Hanson et al., 1956).

Heritability (h2) = (Vg/Vp) x 100

Where,

Vg= Genotypic variance

Vp = Phenotypic variance

The heritability percentage was categorized as low, moderate and high as followed by Robinson et al. (1949), as follows.

0-30% : Low

30-60% : Moderate

>60% : High

3.6.4.3 Genetic advance

The extent of genetic advance to be expected by selecting five per cent of the superior progeny was calculated by using the following formula given by Robinson et al., (1949).

GA = i σp h2

Where,

i = efficacy of selection which is 2.06 at 5% selection intensity

σp = phenotypic standard deviation.

H2 = heritability in broad sense.

3.6.4.4 Genetic advance as per cent of mean

GA as per cent of mean = (GA/X) x 100

Where,

GA = genetic advance

X = general mean of character

The GA as per cent of mean was categorized as low, moderate and high as following by Johnson et al., (1955).

0-10 % : Low

10-20% : Moderate

20 and above : High

3.6.5 Association analysis

The correlation coefficients were calculated to determine the degree of association of characters with yield and also among the yield components themselves in each environment.

Phenotypic correlations were computed by using the formula given by Webber and Moorty (1952).

rp = Cov xyp / (Var xp x Var yp)1/2

Where,

rp = phenotypic correlation

Cov xyp =phenotypic covariance between the characters x and y.

Var xp and Var yp = phenotypic variance of the characters x and y

respectively.

rg = Cov xyg/ (Var xg x Var yg)1/2

where,

rg = genotypic correlation

Cov xyg =genotypic covariance between the characters x and y.

Var xg and Var yg = genotypic variance of the characters x and y respectively.

3.6.6 Path coefficient analysis

Path analysis is simple standardized partial regression coefficient, which splits the correlation coefficient into direct and indirect effects of the yield components on yield as suggested by Wright (1921) and elucidated by Dewey and Lu (1959).

Standard path coefficients are the standardized partial regression coefficients, which were obtained by solving the following set of ‘P’ simultaneous equations through use of ‘Doolittle technique’ as described by Goulden (1959).

3.6.7 Two way analysis of variance

The data obtained for twelve characters on 15 genotypes over 3 locations was subjected to two way analysis of variance using the method outlined by Sundararaj et al. (1972). This was done for each character in order to find out the variation due to genotypes and environments to reveal the existence of genotype x environment interaction, if any. Only after ascertaining the genotypic x environment interaction was significant in the two way analysis of variance, the data was further subjected to stability analysis.

The structure of pooled analysis of variance

Source of variation d.f. MSS Expected value of MSS Cal F.

Environments

Genotypes

Genotypes x environment

Pooled error

(e-1)

(g-1)

(g-1) (e-1)

M*

M1

M2

M3

-

σ2e + σ

2ge + eσ

2g

σ2e + σ

2ge

σ2e

-

-

-

-

* Degrees of freedom pooled over environments

3.6.9 Stability analysis

The stability model proposed by Eberhart and Russell (1966) was adopted to analyze the data over three environments. The model involves the estimation of three stability parameters like mean (X), regression coefficient (bi) and deviation from regression (S

2d),

which are defined by the following mathematical formula.

Yij = µi + βiIj + δij

Where,

Yij : Mean of the ith genotype at the j

th environment (i = 1, 2, 3, 4,

5………………….15, j = 1, 2, 3)

µi : The mean of ith genotype over all the environments

βi : The regression coefficient that measures the response of ith

genotype to varying environment

δij : The deviation from regression of the ith genotype of j

th environment and

Ij : The environmental index obtained by subtracting the regression of the ith

genotype the grand mean from the mean of all genotype at jth environment.

3.6.10 Stability parameters

The mean (µi), the regression coefficient (bi) and mean square deviation from linear regression line (S

2di) are the three stability parameters proposed by Eberhart and Russell

(1966) in their stability model. These parameters were computed by using the following formula.

µi (mean) = ΣjYij / n

bi (regression coefficient) =

S2di (deviation from the regression coefficient) = -

Where,

: Mean square for (estimate of) pooled error

n : Number of environments

Yij : Performance of ith genotype in j

th environment

Σjδ2ij : Sum of squares of deviations from the regression line

^

ΣjYijIj

ΣjI2

j

Σjδ2

ij

n-2

δ2e

r δ2

e

r

Ij : Environmental index (i.e., environmental mean – grand mean)

Ij = -

Where,


v : Number of genotypes with ΣjIj = 0

The total variation is partitioned into genotypes, environment, environment (linear),

genotype × environment (linear), pooled deviation and pooled error.

ANOVA for stability

Source d.f. M.S.S. F test

Genotype (V) (v-1) MS1 MS1/ MS3

Environment (G × E) v (n-1)

Environment (E) (linear) 1

Genotype × environment (linear) (v-1) MS2 MS2/MS3

Pooled deviations v (n-2) MS3

Pooled error n (r-1) (v-1) Me

Total (nv-1) MS1

Where,


v : Number of genotypes

r : Number of replications

F test

a. To test the significance of the differences among the genotypic means, the ‘F’ test followed was,

F = MS1 / MS3

Where,

MS1 : Mean sum of squares of varieties

MS3 : Mean sum of squares of pooled deviation

b. To test individual from linear regression, the formula is as follows,

F = Me

Where,


Σjδ2ij : Sum of squares of deviations from the regression line

Me : Pooled error

ΣjYij

v

ΣiYij

nv

Σjδ2

ij

n - 2

c. To test the hybrids/varieties not differ for their regression on the environmental index, the appropriate‘t’ test was,

t =

SE (b) =

Where,

X : Environmental index

N : Number of environments

A joint consideration of the three parameters that is,

1. The mean performance of the genotype over the environment (X)

2. The regression coefficient (bi) and

3. The deviation from linear regression (S2di) is used to define stability of a genotype

The estimate of deviations from regression (S2d) suggests the degree of reliance that

should be put to linear regression in interpretation of the data. If these values are significantly deviating from zero, the expected phenotype cannot be predicted satisfactorily. When, deviations (S

2d) are not significant the conclusion may be drawn by the joint consideration of

mean, yield and regression coefficient (bi) values (Finlay and Wilkinson, 1963 and Eberhart and Russell, 1966) as below.

Regression coefficient

Stability Mean yield

Remarks

bi = 1 Average High Well adopted to all environments

bi = 1 Average Low Poorly adopted to all environments

bi < 1 Below average High Specially adapted to favorable environments

bi > 1 Below average High Specially adapted to unfavorable environments

bi - 1

SE (b)

^

ΣY2 – (ΣY)

2/n – b2Σ (X – X)

2

(n – 2) Σ (X- X)2

½

4. EXPERIMENTAL RESULTS

The results of the present investigation entitled “Relative Stability analysis of public and private bred hybrids of maize (Zea mays L.)” are presented under following headings.

4.1 Mean performance of hybrids in different locations




4.1 MEAN PERFORMANCE OF HYBRIDS IN DIFFERENT LOCATIONS

The data collected on each of the morphological characters for 15 hybrids of maize were analyzed locationwise to find out performance of different hybrids across locations. Mean, range and environmental indices for different traits across different locations are presented in Table 4.

4.1.1 Days to 50 per cent tasseling

Out of three environments, maximum days to 50 per cent tasseling was recorded at Arabhavi (56.84 days) followed by Dharwad (55.44 days) and Bheemarayanagudi (51.62 days). The maximum range was observed at Arabhavi (55-58 days) and minimum at Bheemarayanagudi (50-53 days). Environmental indices ranged from -1.27 at Arabhavi to 2.55 at Dharwad.

4.1.2 Days to 50 per cent silking

Among three locations, maximum number of days to bear 50 per cent silking was observed at Arabhavi (59.69 days) followed by Dharwad (58.18 days) and Bheemarayanagudi (54.62 days). The range for this character was maximum at Dharwad (53 to 62 days) and minimum at Bheemarayanagudi (53 to 56 days). Environmental indices ranged from -2.87 at Bheemarayanagudi to 2.19 at Arabhavi.

4.1.3 Plant height (cm)

The average plant height was maximum at Bheemarayanagudi (200.13cm) and minimum at Dharwad (197.04).The maximum range for this character was observed at Dharwad (163 to 236 cm), while minimum was recorded at Arabhavi (178 to 235 cm). The environmental indices ranged from -1.87 to 1.21 at Dharwad and Bheemarayanagudi, respectively.

4.1.4 Ear height (cm)

Mean ear height was maximum at Dharwad (97.00 cm) while minimum was observed at Bheemarayanagudi (90.22cm). The range for this character was maximum at Bheemarayanagudi (72 to 115 cm) and minimum at Arabhavi (82 to 118cm). Environmental indices ranged from -4.36 at Bheemarayanagudi to 2.42 at Dharwad.

4.1.5 Cob length (cm)

Average length of the cob was maximum at Arabhavi (15.80cm), while minimum at Bheemarayanagudi (15.48cm). The maximum range was recorded at Arabhavi (11.35 to11.78) whereas, minimum range was observed at Dharwad (12.9 to 17.8). Environmental indices ranged from -0.14 at Bheemarayanagudi to 0.18 at Arabhavi.

Table 4. Mean, range and environmental index for traits in maize at different locations

Mean Range Environmental index Characters

DWR ARB BGD DWR ARB BGD DWR ARB BGD

Days to 50 per cent tasseling 55.44 56.84 51.62 50 -59 55-58 50-53 2.55 -1.27 -1.27

Days to 50 per cent silking 58.18 59.69 54.62 53-62 57-62 53-56 0.68 2.19 -2.87

Plant height (cm) 197.04 199.58 200.13 163-236 178-235 171-235 -1.87 0.66 1.21

Ear height (cm) 97.00 96.51 90.22 78-115 82-118 72-115 2.42 1.93 -4.36

Cob length (cm) 15.57 15.80 15.48 12.9-17.8 11.35-17.78 12.64-18.7 -0.05 0.18 -0.14

Cob girth (cm) 14.83 15.17 15.73 10.33-18.24 10.54-18.18 11.34-18.1 -0.41 -0.08 0.48

Number of kernel rows per cob 14.09 15.27 15.24 12.0-17.0 13-18 12.0-18.0 -0.78 0.40 0.38

Number of kernels per row 34.33 34.87 34.44 28-41 29-43 20-43 -0.21 0.32 -0.10

100 grain weight (g) 33.37 34.40 32.82 24-39.1 28.1-40.3 26-43.2 -0.16 0.87 -0.71

Grain yield per plant (g) 128.58 134.54 118.13 82.3-147.3 101.2-154.8 101.2-146.3 1.50 7.45 -8.95

Grain yield per hectare (q) 63.71 66.63 58.66 41.15-79.38 50.60-83.20 49.55-74.88 0.71 3.63 -4.34

Fodder yield per hectare (t) 8.91 9.32 8.23 6.17-11.91 7.52-12.48 6.6-11.82 0.09 0.50 -0.59

DWR-Dharwad ARB- Arabhavi BGD-Bheemayanagudi

4.1.6 Cob girth (cm)

Overall mean of hybrids for cob girth was maximum at Bheemarayanagudi (15.73), while minimum at Dharwad (14.83). Maximum range for this character was observed in Dharwad (10.33 to 18.24) and minimum range was at Bheemarayanagudi (11.34 to 18.1) and environmental indices ranged from -0.41 at Dharwad to 0.48 at Bheemarayanagudi.

4.1.7 Number of kernel rows per cob

Average number of kernel rows per cob was maximum at Arabhavi (15.27) and minimum was recorded at Dharwad (14.09). The maximum range was recorded at Bheemarayanagudi (12.0 to 18.0) and minimum was observed at Dharwad (12.0 to 17.0). Environmental indices ranged from -0.78 at Dharwad to 0.40 at Arabhavi.

4.1.8 Number of kernels per row

This character exhibited maximum mean at Arabhavi (34.87) followed by Bheemarayanagudi (34.44) and Dharwad (34.33). The maximum range value for this character was observed at Bheemarayanagudi (20 to 43), while, minimum range was recorded at Dharwad (28 to 41) and environmental indices ranged from -0.21 at Dharwad to 0.32 at Arabhavi.

4.1.9 100-grain weight (g)

The average 100-grain weight was maximum at Arabhavi (34.40g), while, minimum was recorded at Bheemarayanagudi (32.82 g). The wide range was observed at Bheemarayanagudi (26 to 43.2g) while, the range was minimum at Arabhavi (28.1 to 40.30g) and environmental indices ranged from -0.71 at Bheemarayanagudi to 0.83 at Arabhavi

4.1.10 Grain yield per plant (g)

Out of three locations, maximum average value for grain yield per plant was recorded at Arabhavi (134.54 g) whereas, minimum was at Bheemarayanagudi (118.13 g). The range of this character was maximum at Dharwad (82.3 to 147.3g), while minimum was recorded at Bheemarayanagudi (101.2 to 146.30g) and environmental indices ranged from -8.95 at Bheemarayanagudi to 7.45 at Arabhavi.

4.1.11 Grain yield per hectare (q)

Mean grain yield per hectare was maximum at Arabhavi (66.63 q). while, minimum was observed at Bheemarayanagudi (58.66q). Maximum range was recorded at Dharwad (41.15 to 79.38q) whereas, minimum range was observed at Bheemarayanagudi (49.55 to 74.88q). Environmental indices ranged from -4.34 at Bheemarayanagudi to 3.63 at Arabhavi.

4.1.12 Fodder yield per hectare (t)

Maximum mean fodder yield per hectare was observed at Arabhavi (9.32 t/ha) while minimum recorded at Bheemarayanagudi (8.23 t/ha). The range for this character was highest at Dharwad (6.17 to 11.91 t/ha). While, lowest range recorded at Arabhavi (7.52 to 12.48 t/ha) and environmental indices ranged from -0.59 at Bheemarayanagudi to 0.50 at Arabhavi.


The results pertaining to genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV), heritability and genetic advance as per cent of mean are presented in Table 5.

4.2.1 Genotypic (GCV) and phenotypic (PCV) coefficients of variation

Moderate estimates of PCV were recorded for traits like fodder yield per hectare (16.11%), number of kernel rows per cob (10.80%), number of kernels per row (12.46%), 100-grain weight (11.06%), ear height(10.26%) , cob girth (11.52%) and grain yield per plant (10.90%). whereas, low PCV of 2.51, 2.47, 7.95 and 8.49 per cent exhibited for the characters viz., days to 50 per cent tasseling, days to 50 per cent silking, plant height, and cob length, respectively.

Moderate estimates of GCV were recorded for traits like number of kernels per row (10.73), cob girth (10.25), fodder yield per hectare (15.86) and grain yield per plant (10.54) whereas, days to 50 per cent tasseling, days to 50 per cent silking, plant height, ear height, number of kernel rows per cob, 100-grain weight and cob length exhibited low GCV of 1.78, 1.72, 5.79, 8.96, 8.56, 9.51 and 6.83 per cent, respectively. .

4.2.2 Heritability

In the present study, the heritability estimates were invariably moderate to high for all the characters studied. Broad sense heritability estimates ranged from 43.47 per cent (days to 50 per cent tasseling) to 96.83 per cent (fodder yield per hectare).

Among 11 characters, only three characters showed moderate heritability i.e. days to 50 per cent tasseling (43.47%), days to 50 per cent silking (45.87%) and plant height (55.60%). All other showed high heritability i.e. ear height (75.97%), cob length (64.83%), cob girth (78.30%), number of kernel rows per cob (62.80%), number of kernels per row (74.30%), 100-grain weight (75.00%) and grain yield per plant (93.40%).

4.2.3 Genetic advance as per cent of mean (GAM)

The GAM values ranged from 2.55 per cent to 32.19 per cent (Table 5). The fodder yields per hectare (32.19%), grain yield per plant (20.99%) were recorded high magnitudes of GAM. Number of kernels per row (19.05%), cob girth (18.86%), 100-grain weight (16.87%), ear height (16.12%) and cob length (11.50%) were recorded moderate GAM. Whereas, plant height (8.93), days to 50 per cent tasseling (2.73%) and days 50 per cent silking (2.55%) showed low magnitude of GAM.


4.3.1 Character association

Results of phenotypic correlations among eleven quantitative characters are presented location wise in Table 6, 7 and 8 for Dharwad, Arabhavi and Bheemarayanagudi, respectively.

4.3.1.1 Dharwad

Association between yield and its component traits

The association of grain yield per plant was highly significant and positive with traits like cob length (0.555) and fodder yield per hectare (0.726) at phenotypic level.

Intercorrelation among yield components

At phenotypic level days to 50 per cent tasseling showed high significant positive association with days to 50 per cent silking (0.818) with grain yield per plant. While remaining characters showed non-significant association with grain yield per plant.

Days to 50 per cent silking and plant height exhibited non-significant negative correlation with grain yield per plant. The association between ear height and cob girth was significant (0.572) and positive at phenotypic level.

Table 5. Genetic variability, heritability and genetic advance parameters for grain yield and its components traits in maize

Traits PCV (%) GCV (%) h² (%) GAM (%)

Days to 50 per cent tasseling 2.51 1.78 43.47 2.73

Days to 50 per cent silking 2.47 1.72 45.87 2.55

Plant height (cm) 7.95 5.79 55.60 8.93

Ear height (cm) 10.26 8.96 75.97 16.12

Cob length (cm) 8.49 6.83 64.83 11.50

Cob girth (cm) 11.52 10.25 78.30 18.86

Number of kernel rows per cob 10.80 8.56 62.80 13.99

Number of kernels per row 12.46 10.73 74.30 19.05

100 grain weight (g) 11.06 9.51 75.00 16.87

Fodder yield per hectare (t) 16.11 15.86 96.83 32.19

Grain yield per plant (g) 10.90 10.54 93.40 20.99

PCV – Phenotypic coefficient of variation GCV – Genotypic coefficient of variation GAM – Genetic advance as mean h² – Heritability

Table 6. Phenotypic correlations among different quantitative traits in maize at Dharwad

Characters Days to 50

per cent tasseling

Days to 50 per cent

silking

Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)

Number of kernel rows per

cob

Number of

kernels per row

100-grain

weight (g)

Fodder yield per hectare

(t)

Grain yield per plant (g)

Days to 50 per cent tasseling 1.000 0.818** -0.122 0.141 -0.094 -0.094 0.195 0.354 -0.129 -0.022 -0.067

Days to 50 per cent silking 1.000 -0.078 0.202 -0.007 -0.007 0.171 0.386 -0.048 0.046 -0.112

Plant height (cm) 1.000 -0.027 -0.23 -0.052 0.200 0.008 0.37 0.076 -0.106

Ear height (cm) 1.000 0.018 0.572* 0.309 0.415 0.279 0.466 0.06

Cob length (cm) 1.000 -0.001 -0.265 0.317 -0.171 0.407 0.555*

Cob girth (cm) 1.000 0.212 0.461 0.276 0.538* 0.316

Number of kernel rows per cob 1.000 0.168 0.528* 0.274 0.02

Number of kernels per row 1.000 0.192 0.574* 0.379

100 grain weight (g) 1.000 0.397 -0.066

Fodder yield per hectare (t) 1.000 0.726**

Grain yield per plant (g) 1.000

* Significant at 5 % probability level ** Significant at 1% probability level

Cob length showed positive and significant association with grain yield per plant (0.555). Cob girth exhibited positive and significant correlation with fodder yield per hectare (0.538). Number of kernel rows per cob was found to be significant and positively associated with 100 grain weight (0.528). Number of kernels per row had significant and positive association with fodder yield per hectare (0.574).

100 grain weight did not show any significant correlation with yield and yield components. Highly significant and positive association was exhibited between grain yield per plant and fodder yield per hectare (0.726).

4.3.1.2 Arabhavi

At phenotypic level, grain yield per plant exhibited highly significant positive association with fodder yield per hectare (0.684) while, other traits did not exhibit significant association with grain yield per plant.

Intercorrelation among yield and yield components

Days to 50 per cent tasseling exhibited highly significant positive correlation with day to 50 per cent silking (0.750) whereas, other characters showed positive non-significant association with grain yield per plant except cob girth, number of kernel rows per cob and number of kernels per row which exhibited negative non-significant association. Days to 50 per cent silking exhibited non-significant correlation with grain yield per plant.

Plant height had negatively significant correlation with cob length (-0.607) while, for remaining characters showed nonsignificant correlation. Ear height showed positive significant correlation with fodder grain yield per hectare (0.578).

The characters cob length, cob girth, number of kernel rows per cob, number of kernels per row and 100 grain weight showed nonsignificant correlation with grain yield per plant. However, fodder yield per hectare had highly significant correlation with grain yield per plant (0.684).

4.3.1.3 Bheemarayanagudi

Association between yield and its component traits

Grain yield per plant showed highly significant positive association with fodder yield per hectare (0.836). Whereas, remaining characters showed nonsignificant association with grain yield per plant.

Intercorrelation among yield components

Days to 50 per cent tasseling exhibited significant association with days to 50 per cent silking (0.910). Days to 50 per cent silking and plant height showed non-significant association with yield and yield components.

Ear height had significantly positive correlation with fodder yield per hectare (0.579). Cob length, cob girth, number of kernel rows per cob, number of kernels per row and 100 grain weight showed nonsignificant association with yield and yield components. Fodder yield per hectare had a highly positive significant correlation with grain yield per plant (0.836).

4.3.2 Path analysis

The phenotypic correlation among various character were subjected to path analysis for partitioning values into direct and indirect effect by considering the grain yield per plant as the dependant variable and other character as independent variable and are presented in the table 9, 10 and 11.

Table 7. Phenotypic correlations among different quantitative traits in maize at Arabhavi

Characters

Days to 50 per cent

tasseling

Days to 50 per cent

silking

Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)


cob

Number of

kernels per row

100-grain

weight (g)


(t)


Days to 50 per cent tasseling 1.000 0.750** -0.057 0.219 0.239 0.132 0.001 0.197 0.327 0.23 0.18

Days to 50 per cent silking 1.000 -0.289 0.273 0.478 0.163 -0.14 0.03 0.47 0.244 0.1

Plant height (cm) 1.000 -0.124 -0.607* -0.127 -0.238 0.189 -0.204 0.041 0.151

Ear height (cm) 1.000 0.203 -0.127 -0.362 0.049 0.352 0.578* 0.513

Cob length (cm) 1.000 0.187 0.101 -0.156 0.347 0.253 0.224

Cob girth (cm) 1.000 -0.039 0.286 0.179 0.139 -0.069

Number of kernel rows per cob 1.000 -0.012 -0.206 -0.452 -0.362

Number of kernels per row 1.000 0.111 0.328 -0.001

100 grain weight (g) 1.000 0.365 0.137




Table 8. Phenotypic correlations among different quantitative traits in maize at Bheemarayanagudi

Characters

Days to 50 per cent

tasseling

Days to 50 per cent

silking

Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)

Number of

kernel rows

per cob

Number of

kernels per row

100-grain

weight (g)


(t)


Days to 50 per cent tasseling 1.000 0.910** -0.342 -0.13 -0.298 -0.513 -0.092 -0.268 -0.04 -0.288 -0.073

Days to 50 per cent silking 1.000 -0.342 -0.13 -0.298 -0.513 -0.092 -0.268 -0.04 -0.288 -0.073

Plant height (cm) 1.000 0.223 0.153 0.223 0.137 0.386 0.027 0.463 0.312

Ear height (cm) 1.000 0.307 -0.259 -0.166 0.215 0.416 0.579* 0.442

Cob length (cm) 1.000 0.187 0.065 0.368 0.007 0.027 -0.145

Cob girth (cm) 1.000 0.142 0.285 0.058 0.242 0.246

Number of kernel rows per cob 1.000 0.042 -0.042 -0.084 -0.119

Number of kernels per row 1.000 0.082 0.403 0.339

100 grain weight (g) 1.000 0.482 0.355




Table 9. Direct (diagonal) and indirect effects of grain yield component traits on grain yield per plant at phenotypic level in maize at Dharwad

Characters

Days to 50 per cent

tasseling

Days to 50 per cent

silking

Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)


cob

Number of

kernels per row

100-grain

weight (g)


(t)


Days to 50 per cent tasseling 0.534 -0.533 -0.001 -0.045 -0.023 -0.017 0.016 -0.021 0.041 -0.019 -0.067

Days to 50 per cent silking 0.49 -0.581 -0.001 -0.064 -0.002 -0.001 0.014 -0.022 0.015 0.039 -0.112

Plant height (cm) -0.065 0.046 0.01 0.008 -0.057 -0.01 0.017 0.000 -0.118 0.064 -0.106

Ear height (cm) 0.075 -0.117 0.000 -0.316 0.004 0.105 0.026 -0.024 -0.089 0.395 0.06

Cob length (cm) -0.05 0.004 -0.002 -0.006 0.25 0.000 -0.022 -0.018 0.055 0.345 0.555*

Cob girth (cm) -0.05 0.004 -0.001 -0.18 0.000 0.184 0.018 -0.027 -0.088 0.456 0.316

Number of kernel rows per cob 0.104 -0.099 0.002 -0.098 -0.066 0.039 0.083 -0.01 -0.168 0.232 0.02

Number of kernels per row 0.189 -0.224 0.000 -0.131 0.079 0.085 0.014 -0.058 -0.061 0.486 0.379

100 grain weight (g) -0.069 0.028 0.004 -0.088 -0.043 0.051 0.044 -0.011 -0.319 0.336 -0.066

Fodder yield per hectare (t) -0.012 -0.027 0.001 -0.147 0.102 0.099 0.023 -0.033 -0.127 0.847 0.726**

RESIDUAL= 0.1782

4.3.2.1 Dharwad

The results of path coefficient analysis at phenotypic level are presented in table 9. Path analysis results at Dharwad revealed six traits viz. fodder yield per hectare (0.847), days to 50 per cent tasseling (0.534), cob length (0.25), cob girth (0.184) and number of kernel rows per cob (0.083) had positive direct effect on grain yield per plant whereas others had negative direct effect.

Fodder yield per hectare (0.726) had the highest direct effect on grain yield per plant followed by days to 50 per cent tasseling (0.534) and cob length (0.25). Plant height, ear height, cob length, cob girth, number of kernel rows per cob, number of kernels per row and 100 grain weight contributed to grain yield though their indirect influence via fodder yield per hectare (0.039, 0.064, 0.395, 0.345, 0.456, 0.232, 0.486 and 0.336, respectively). Whereas, days to 50 per cent silking contributed to grain yield per plant through their indirect influence via days to 50 per cent tasseling (0.49) and fodder yield per hectare (0.039). Other characters viz., fodder yield per hectare (0.847), days to 50 per cent tasseling (0.534), cob length (0.250) and cob girth (0.184) contributed mainly through their direct effects.

4.3.2.2 Arabhavi

The results of path coefficient analysis at phenotypic level are presented in table 10. Path analysis results at Arabhavi revealed five characters viz., fodder yield per hectare(0.602), cob length (0.290) days to 50 per cent tasseling (0.258), plant height (0.249) and ear height (0.185) had positive direct effect on grain yield per plant whereas, other characters had negative direct effects.

Fodder yield per hectare (0.602) had highest direct effect on grain yield per plant followed by cob length (0.290) and days to 50 per cent tasseling (0.258). days to 50 per cent silking, cob girth, number of kernels per row and 100 grain weight contributed to grain yield per plant through their indirect influence via fodder yield per hectare (0.147, 0.084, 0.198, and 0.22, respectively) whereas, number of kernel rows per cob contributed to grain yield per plant through their indirect influence via days to 50 per cent silking (0.041) and cob length (0.029). Other characters viz., fodder yield per hectare (0.602), days to 50 per cent tasseling (0.258) and cob length (0.290) contributed mainly through their direct effects.

4.3.2.3 Bheemarayanagudi

Path coefficient results of Bheemarayanagudi are presented in table 11.Among the direct effects, the highest direct effect on grain yield per plant was recorded for fodder yield per hectare (0.769) followed by days to 50 per cent silking (0.360), cob girth (0.284), ear height (0.213) and number of kernels per row (0.089).

Days to 50 per cent tasseling contributed indirectly to grain yield per plant through days to 50 per cent silking (0.36) and cob length (0.069). Days to 50 per cent silking had positive indirect effect on grain yield per plant through cob length (0.069). Plant height had positive indirect effect on grain yield per plant through fodder yield per hectare (0.356) and ear height (0.063). Ear height had positive indirect effect on grain yield per plant through fodder yield per hectare (0.445). Cob length contributed indirectly to the grain yield per plant through ear height (0.066) and cob girth (0.053). Cob girth and number of kernels per rows showed positive indirect effects on grain yield via fodder yield per hectare (0.186 and 0.310, respectively).while, number of kernel rows per cob contributed through indirect effect on grain yield per plant via 100-grain weight (0.005). 100-grain weight contributed indirectly to the grain yield per plant through fodder yield per hectare (0.370). Whereas, fodder yield per hectare had high positive effect on grain yield per hectare (0.769).

Table 10. Direct (diagonal) and indirect effects of grain yield component traits on grain yield per plant at phenotypic level in maize at Arabhavi

Characters

Days to 50 per cent

tasseling

Days to 50 per cent

silking

Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)


cob

Number of

kernels per row

100-grain

weight (g)


(t)


Days to 50 per cent tasseling 0.258 -0.22 -0.014 0.041 0.069 -0.007 0.000 -0.044 -0.04 0.139 0.18

Days to 50 per cent silking 0.193 -0.293 -0.072 0.051 0.139 -0.009 0.009 -0.007 -0.058 0.147 0.1

Plant height (cm) -0.015 0.085 0.249 -0.023 -0.176 0.007 0.015 -0.042 0.025 0.025 0.151

Ear height (cm) 0.056 -0.08 -0.031 0.185 0.059 0.007 0.023 -0.011 -0.044 0.348 0.513

Cob length (cm) 0.062 -0.14 -0.151 0.038 0.29 -0.01 -0.007 0.035 -0.043 0.152 0.224

Cob girth (cm) 0.034 -0.048 -0.032 -0.024 0.054 -0.054 0.003 -0.064 -0.022 0.084 -0.069

Number of kernel rows per cob 0.000 0.041 -0.059 -0.067 0.029 0.002 -0.065 0.003 0.026 -0.272 -0.362

Number of kernels per row 0.051 -0.009 0.047 0.009 -0.045 -0.016 0.001 -0.223 -0.014 0.198 -0.001

100 grain weight (g) 0.084 -0.138 -0.051 0.065 0.101 -0.01 0.013 -0.025 -0.124 0.22 0.137

Fodder yield per hectare (t) 0.059 -0.071 0.01 0.107 0.073 -0.008 0.029 -0.073 -0.045 0.602 0.684**

RESIDUAL= 0.2062

Table 11. Direct (diagonal) and indirect effects of grain yield component traits on grain yield per plant at phenotypic level in maize at Bheemarayanagudi

Characters

Days to 50 per cent

tasseling

Days to 50 per cent

silking

Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)


cob

Number of

kernels per row

100-grain

weight(g)


(t)


Days to 50 per cent tasseling -0.113 0.36 0.022 -0.028 0.069 -0.146 0.002 -0.024 0.005 -0.221 -0.073

Days to 50 per cent silking -0.113 0.36 0.022 -0.028 0.069 -0.146 0.002 -0.024 0.005 -0.221 -0.073

Plant height (cm) 0.039 -0.123 -0.064 0.048 -0.036 0.063 -0.003 0.034 -0.003 0.356 0.312

Ear height (cm) 0.015 -0.047 -0.014 0.213 -0.071 -0.073 0.003 0.019 -0.048 0.445 0.442

Cob length (cm) 0.034 -0.108 -0.01 0.066 -0.232 0.053 -0.001 0.033 -0.001 0.021 -0.145

Cob girth (cm) 0.058 -0.185 -0.014 -0.055 -0.043 0.284 -0.003 0.025 -0.007 0.186 0.246

Number of kernel rows per cob 0.01 -0.033 -0.009 -0.036 -0.015 0.04 -0.021 0.004 0.005 -0.064 -0.119

Number of kernels per row 0.03 -0.097 -0.025 0.046 -0.085 0.081 -0.001 0.089 -0.01 0.31 0.339

100 grain weight (g) 0.004 -0.014 -0.002 0.089 -0.002 0.017 0.001 0.007 -0.116 0.37 0.355

Fodder yield per hectare (t) 0.032 -0.104 -0.029 0.124 -0.006 0.069 0.002 0.036 -0.056 0.769 0.836**

RESIDUAL= 0.2062

4.4 STABILITY ANALYSIS

4.4.1 Analysis of variance

Location wise analysis of variance was carried out for 12 quantitative characters to partition the total variance (mean sum of squares) due to known and unknown causes were worked out using the method suggested by Lush (1949) and Chaudhary and Prasad (1967).

The analysis of variance for grain yield and its component characters under study indicated highly significant variation among the hybrids for all the characters across three locations (Table 12).

4.4.2 Pooled analysis of variance

Eberhart and Russell (1966) model of stability analysis was used for the assessment of environmental influence and hybrid x environmental interaction on hybrids for each character. When the hybrid x environment interaction were significant for the characters, then partitioning of total sum of squares due to hybrid x environment interactions into predictable and unpredictable source of variations was done using the procedure given by Eberhart and Russell (1966).

Pooled ANOVA for stability of different characters (Eberhart and Russell, 1966) are given in Table 13. Genotypic differences pooled over environments were significant for most of the characters except days to 50 per cent tasseling, cob girth and days to 50 per cent silking. Variance due to environments was significant for most of the characters except plant height, cob length, cob girth, number of kernels per row and 100 grain weight. Variance due to G x E was significant for most of the characters except plant height, ear height, number of kernel rows per cob and number of kernels per row. Environment linear component was significant for most of the traits except plant height, cob length, cob girth, number of kernels per row and 100 grain weight. whereas, G x E (linear) interaction was non-significant for most of the characters. As regard to pooled deviation (nonlinear portion of variance), which is unpredictable portion of G x E interaction was significant for all the characters except cob length under study.

4.4.3 Stability parameters

The three stability parameters viz, mean (X), regression coefficient bi and deviation from linear regression line (S

2di) were estimated for all the 12 traits and the results obtained

are present in the Table 14 to 19.

4.4.3.1 Days to 50 per cent tasseling

The data on location wise mean days to 50 per cent tasseling for 15 hybrids studied is presented in table 14. Out of 15 hybrids ARBMH-8 required maximum number of days to 50 per cent tasseling (56.00 days) while, ARBMH-7 exhibited minimum number of days to 50 per cent tasseling (53.44 days) and mean over three environments was 54.64 days.

The hybrids 900M and bio-9681 had regression value significantly deviating from unity. The deviations from regression values were significantly different from zero for the hybrids ARBMH-3, ARBMH-4, ARBMH-7, ARBMH-8, All-rounder and K-235. Hybrids ARBMH-1, ARBMH-2, ARBMH-5, ARBMH-6, ARBMH-9, ARBMH-10 and pinnacle had both non-significant regression coefficient and deviation from regression. The hybrids ARBMH-2, ARBMH-5, ARBMH-6, ARBMH-9 and ARBMH-10 were found to have higher mean value than population mean with bi and S

2di values not significantly different from 1 and 0, respectively.

4.4.3.2 Days to 50 per cent silking

Mean values and stability parameters for this trait are presented in Table 14. Out of 15 hybrids ARBMH-8 required maximum number of days to 50 per cent silking (58.89 days) while, All-rounder required minimum number of days to 50 percent silking (56.11 days) and mean over three environments was 57.50 days.

Table 12. Analysis of variance for grain yield and other quantitative characters at three locations in maize

Source

Replications (MSS) Genotypes (MSS) Error (MSS) Sl. No.

Characters

DWR ARB BGD DWR ARB BGD DWR ARB BGD

1 Days to 50 per cent tasselling 0.622 0.69 0.96 12.317** 1.33* 1.52* 1.194 0.64 0.69

2 Days to 50 per cent silking 0.29 0.56 0.96 9.18** 3.17** 1.52* 1.27 1.01 0.69

3 Plant height (cm) 13.62 0.29 32.47 437.66* 484.36** 649.61** 173.34 44.27 133.49

4 Ear height (cm) 7.47 5.42 23.09 209.10** 177.42** 332.46** 27.20 14.80 25.18

5 Cob length (cm) 0.04 0.14 0.20 2.71** 4.13** 5.57** 0.82 0.38 0.57

6 Cob girth (cm) 0.38 0.02 0.17 7.79** 11.08** 5.43** 0.54 0.43 0.86

7 Number of kernel rows per cob 0.02 0.27 1.16 5.17** 4.44** 8.17** 0.83 0.79 1.27

8 Number of kernels per row 2.60 0.20 0.09 33.29** 38.94** 69.51** 3.60 3.63 7.57

9 100-grain weight (g) 2.10 0.08 1.08 27.77** 27.59** 48.33** 2.96 1.74 6.49

10 Grain yield per plant (g) 0.09 2.58 7.36 619.76** 583.06** 456.38** 9.34 18.06 10.42

11 Grain yield per hectare (q/ha) 0.066 0.668 2.168 218.024** 196.694** 222.783** 2.247 4.550 2.553

12 Fodder yield (q/ha) 0.002 0.015 0.048 5.670** 5.280** 6.706** 0.044 0.090 0.049

DWR-Dharwad ARB- Arabhavi BGD- Bheemayanagudi * Significant at 5 % probability level ** Significant at 1% probability level

Table 13. Pooled analysis of variance for stability analysis (Eberhart and Russell, 1966) for twelve quantitative traits in maize over three locations

Source d.f. Days to

50 % tasselling


Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)

Number of

kernel rows

per cob

Number of

kernels per row

100 grain

weight (g)


Grain yield per

ha (q)

Fodder yield (t/ha)

Genotype 14 1.48 1.43 267.39* 171.14** 2.14* 3.24 3.50** 33.97** 22.39** 380.18** 170.86** 5.08**

Environment 2 109.60** 101.48** 40.37 214.31** 0.41 3.05 6.81** 1.19 9.69 1034.79** 243.79** 4.60**

Genotype X Environment(G X E)

28 1.79** 1.60* 128.24 34.26 0.99* 2.43** 1.21 6.64 6.09* 86.44** 20.81** 0.41**

Environment + (G X E) 30 8.98 8.25 122.38 46.26 0.96 2.47 1.59 6.28 6.33 149.66 35.68 0.69

Environment (linear) 1 219.21** 202.98** 81.34 428.64** 0.83 6.10 13.61** 2.38 19.38 2069.47** 487.49** 9.20**

Genotype X Environment (linear)

14 0.69 0.74 115.81 26.61 1.74** 2.43 1.54 5.06 5.02 150.73** 36.20** 0.69**

Pooled deviation 15 2.69** 2.29** 131.26** 39.11** 0.23 2.27** 0.83** 7.67** 6.68** 20.68** 5.07** 0.11**

Pooled error 90 0.84 0.99 117.03 22.40 0.59 0.61 0.97 4.93 3.73 12.61 3.12 0.06


Table 14. Stability parameters for days to 50 per cent tasseling and days to 50 per cent silking

Days to 50 per cent tasseling Days to 50 per cent silking Sl. No.

Genotypes

Mean bi S²di Mean bi S²di

1 ARBMH-1 54.56 1.25 0.02 56.89 1.08 -0.11

2 ARBMH-2 54.78 1.23 -0.20 57.78 1.28 -0.13

3 ARBMH-3 53.89 0.70 6.98* 57.56 0.97 5.28*

4 ARBMH-4 54.33 1.12 1.65* 56.67 0.94 1.34*

5 ARBMH-5 54.67 0.71 0.66 57.67 0.72 0.79

6 ARBMH-6 55.56 1.25 0.02 57.89 1.08 -0.11

7 ARBMH-7 53.44 0.72 11.36* 57.11 0.99 9.18*

8 ARBMH-8 56.00 0.85 4.09* 58.89 0.88 1.86*

9 ARBMH-9 54.67 1.03 0.84 57.22 0.96 0.01

10 ARBMH-10 55.56 1.01 0.35 57.89 0.83 0.14

11 900M 54.89 1.20* -0.22 57.89 1.24* -0.32

12 Bio-9681 54.56 1.28* -0.27 58.22 1.54 -0.09

13 Pinnacle 54.56 0.93 -0.07 56.89 0.75 -0.21

14 All-rounder 53.56 0.74 9.64* 56.11 0.65 10.00*

15 K-235 54.56 0.98 1.28* 57.78 1.08 1.78*

Population mean 54.637 57.496

* - Significant at 5% probability level ** - Significant at 1% probability level bi – Regression coefficient S²di – Deviation from regression

The hybrid 900M exhibited significant regression coefficient values higher than unity. The deviations from regression values were significantly different from zero for hybrids ARBMH-3, ARBMH-4, ARBMH-7, ARBMH-8, All-rounder and K-235. The hybrids ARBMH-1, ARBMH-2, ARBMH-5, ARBMH-6, ARBMH-9, ARBMH-10, Bio-9681 and pinnacle had both non-significant regression coefficient and deviation from regression. The hybrids ARBMH-2, ARBMH-5, ARBMH-6, ARBMH-10 and Bio-9681 were found to have higher mean value than population mean with bi and S


4.4.3.3 Plant height

The stability parameters bi and s2di for plant height for the hybrids across locations

along with mean values are presented in table 15. The hybrid ARBMH-1 with mean plant height of 216.33cm was the tallest and ARBMH-8 with a mean plant height of 185.78cm was the shortest among 15 hybrids and average plant height over the environments was 198.92cm.

None of the hybrids had the significant regression values either lower or higher than unity while, the hybrids ARBMH-2, ARBMH-7, ARBMH-10 and Pinnacle had significant deviation from regression. The hybrids ARBMH-1, ARBMH-6, ARBMH-9 and 900M were found to have higher mean value than population mean with bi and S

2di values not

significantly different from 1 and 0, respectively.

4.4.3.4 Ear height

Mean across locations along with corresponding bi and S2di value for each hybrid in

respect of ear height are presented in table 15. The mean ear height over three environments was 94.58 cm. The maximum (111.78 cm) and minimum (85.56 cm) ear height were recorded in the hybrids ARBMH-2 and All-rounder, respectively.

None of the hybrids had the significant regression values either lower or higher than unity. While the hybrids ARBMH-3, ARBMH-4, ARBMH-6, ARBMH-10, pinnacle and K-235 had significantly deviation from regression. The hybrids ARBMH-2, ARBMH-5, ARBMH-8, 900M and bio-9681 were found to have higher mean value than population mean with bi and S


4.4.3.5 Cob length

Mean across the locations and stability parameters for cob length are presented in table 16. Among 15 hybrids tested over three environments ARBMH-4 and Pinnacle exhibited maximum (16.87 cm) and minimum (13.99 cm) cob length, respectively. The mean ear height over three environments was 15.62cm.

All the hybrids showed nonsignificant regression coefficient (bi) value and deviation from regression. The hybrids ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-4, ARBMH-5, ARBMH-7, and ARBMH-9, 900M and Bio-9681 were found to have higher mean value than population mean with bi and S

2di values not significantly different from 1 and 0, respectively

(Table 16 and Fig. 1).

4.4.3.6 Cob girth

Mean cob girth across the locations and stability parameters for each hybrid are presented in table 16. The maximum (16.67 cm) and minimum (13.01 cm) cob girth were shown by 900M and ARBMH-8, respectively. The mean ear height over three environments was 15.24 cm.

None of the hybrids differed significantly for regression coefficient values. While, hybrids ARBMH-1, ARBMH-3, ARBMH-6, ARBMH-7, ARBMH-8, ARBMH-9, Pinnacle and K-235 showed significant deviation from regression. The hybrids ARBMH-2, ARBMH-4, 900M, Bio-9681 and All-rounder were found to have higher mean value than population mean with bi and S

2di values not significantly different from 1 and 0, respectively (Table 16 and Fig. 2).

Table 15. Stability parameters for plant height and ear height

Plant height Ear height Sl. No.

Genotypes


1 ARBMH-1 216.33 0.36 -32.81 93.56 1.63 -0.24

2 ARBMH-2 204.78 4.41 380.62* 111.78 -0.21 -6.70

3 ARBMH-3 188.56 -6.65 -14.19 87.44 1.31 45.89*

4 ARBMH-4 189.11 10.43 89.69 90.22 2.35 158.56*

5 ARBMH-5 196.67 -0.02 -32.12 106.33 0.6 -6.14

6 ARBMH-6 202.00 -0.99 -37.40 93.67 0.68 46.12*

7 ARBMH-7 188.22 -2.17 321.33* 93.89 0.71 -1.84

8 ARBMH-8 185.78 2.45 35.51 97.67 0.26 7.50

9 ARBMH-9 205.22 2.54 72.97 89.11 0.84 -2.71

10 ARBMH-10 194.11 2.81 205.06* 88.00 0.33 113.05*

11 900M 206.44 3.72 86.30 102.78 -0.02 -1.85

12 Bio-9681 191.89 -0.84 -23.65 98.67 0.05 6.70

13 Pinnacle 213.56 7.78 166.88* 91.11 3.26 54.10*

14 All-rounder 203.44 -5.17 -23.42 85.56 1.78 9.39

15 K-235 197.67 -3.70 188.95 88.89 1.45 52.84*


* - Significant at 5% probability level ** - Significant at 1% probability level bi – Regression coefficient S2di – Deviation from regression

Table 16. Stability parameters for cob length and cob girth

Cob length Cob girth Sl. No.

Genotypes


1 ARBMH-1 15.92 -1.59 -0.12 14.38 3.28 1.11*

2 ARBMH-2 16.86 1.73 -0.10 16.33 -0.26 0.01

3 ARBMH-3 16.30 3.43 0.44 16.09 1.39 1.64*

4 ARBMH-4 16.87 -5.22 -0.16 15.76 3.63 -0.10

5 ARBMH-5 16.08 -4.70 0.48 14.95 -2.53 0.07

6 ARBMH-6 15.04 4.59 -0.03 16.00 0.03 3.21*

7 ARBMH-7 15.79 4.04 0.06 14.94 1.75 2.50*

8 ARBMH-8 15.17 6.15 -0.08 13.01 -4.07 9.74*

9 ARBMH-9 16.18 -5.63 -0.13 15.65 0.94 5.39*

10 ARBMH-10 15.15 4.30 0.45 13.36 5.50 0.23

11 900M 15.77 0.49 -0.18 16.67 -1.23 0.11

12 Bio-9681 16.03 2.78 -0.16 15.98 0.25 0.65

13 Pinnacle 13.99 -11.27 0.16 15.21 2.67 1.74*

14 All-rounder 14.52 7.04 -0.06 15.39 2.05 -0.15

15 K-235 14.63 8.84 -0.02 14.91 1.58 4.93*



13

49

5

1

11

2

123

7610

814

15

-15

-10

-5

0

5

10

1 3 5 7 9 11 13 15 17 19

Reg

ress

ion

co

eff

icie

nt

(bi)

Fig. 1. Relationship between the regression coefficient (bi) and cob length (cm)

Cob length (cm)

Red coloured figures indicate stable hybrids

Fig. 1. Relationship between the regression coefficient (bi) and cob length (cm)

4.4.3.7 Number of kernel rows per cob

The data on location wise mean number of kernel rows per cob for 15 hybrids studied along with bi and S

2di values are present in Table 17. The hybrids ARBMH-6 and ARBMH-5

showed maximum (16.67) and minimum (13.11) number of kernel rows per cob. While, population mean for number of kernel rows per cob was 14.87.

The regression coefficients were found to be nonsignificant for all the hybrids. The hybrids ARBMH-8, ARBMH-9 and Pinnacle showed significantly deviate from regression (S

2di≠0). The hybrids ARBMH-2, ARBMH-4, ARBMH-6, All-rounder and K-235 were found to

have higher mean value than population mean with bi and S2di values not significantly

different from 1 and 0, respectively (Table 17 and Fig.3).

4.4.3.8 Number of kernels per row

The stability parameters for number of kernels per cob are presented in table 17. Among the 15 hybrids tested over three environments 900M and All-rounder exhibited maximum (40.78) and minimum (30.67) number of kernels per cob, respectively. The mean ear height over three environments was 34.55 cm.

None of the hybrids had the significant regression coefficient value either lower or higher than unity. While, hybrids ARBMH-1, ARBMH-2, ARBMH-4, ARBMH-5, ARBMH-9, Bio-9681, Pinnacle and K-235 had values significantly different from zero (S

2di≠0). The

hybrids ARBMH-6 and 900M were found to have higher mean value than population mean with bi and S

2di values not significantly different from 1 and 0, respectively (Table 17 and Fig.

4).

4.4.3.9 100-grain weight (g)

Stability parameters in respect of 100-grain weight are presented in table 18. Maximum (38.66 g) and minimum (30.21 g) 100-grain weight was shown by hybrids 900M and Pinnacle, respectively. While, population mean for 100-grain weight was 33.53 g.

Among 15 hybrids, none of the hybrids showed significant difference for regression coefficient. While, hybrids ARBMH-1, ARBMH-3, ARBMH-5 and All-rounder had values significantly different from zero (S

2di≠0). The hybrids ARBMH-2, ARBMH-6, 900M

and bio-9681 were found to have higher mean value than population mean with bi and S2di

values not significantly different from 1 and 0, respectively (Table 18 and Fig. 5).

4.4.3.10 Grain yield per plant (g)

Stability parameters for grain yield per plant are presented in Table 18. The hybrids ARBMH-2 showed the highest (146.99 g) mean value, while, K-235 showed the lowest (97.49 g) mean value for grain yield per plant. The average grain yield per plant over three environments was 127.10 g.

None of the hybrids had the significant regression coefficient values either lower or higher than unity. While hybrids ARBMH-4, ARBMH-9 and K-235 had values significantly different from zero (S

2di≠0). The hybrids ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-6,

ARBMH-7 and 900M were found to have higher mean value than population mean with bi and S


4.4.3.11 Grain yield per hectare (q)

Stability parameters for grain yield per hectare are presented in Table 19. The hybrids ARBMH-2 showed the highest (79.21 q/ha) mean value, while, K-235 showed the lowest (48.74 q/ha) mean value for grain yield per hectare. The average grain yield per hectare over three environments was 63.00 q/ha.

Hybrids ARBMH-7, 900M, ARBMH-5, ARBMH-8 and Bio-9681 showed significant regression coefficient less than and higher than unity, respectively. The hybrids ARBMH-4,

10

41

13

14

39

12

157

2

11

5

8

6

-6

-4

-2

0

2

4

6

1 3 5 7 9 11 13 15 17 19

Reg

ress

ion

co

eff

icie

nt

(bi)

Fig. 2. Relationship between the regression coefficient (bi) and cob girth (cm)

Cob girth (cm)


Fig. 2. Relationship between the regression coefficient (bi) and cob girth (cm)

Table 17. Stability parameters for number of kernel rows per cob and number of kernels per row

Number of kernel rows per cob Number of kernels per row Sl. No.

Genotypes


1 ARBMH-1 13.44 -1.13 0.60 34.33 -6.41 13.41*

2 ARBMH-2 16.22 0.27 0.58 38.67 0.19 1.24*

3 ARBMH-3 13.44 1.43 -0.32 31.89 0.45 -1.60

4 ARBMH-4 15.67 2.56 -0.06 37.67 4.07 1.73*

5 ARBMH-5 13.11 0.15 -0.27 32.89 -10.40 20.41*

6 ARBMH-6 16.67 0.42 -0.26 38.00 -0.89 -1.55

7 ARBMH-7 14.22 2.01 -0.13 31.78 2.26 -0.60

8 ARBMH-8 15.00 0.88 1.64* 31.67 -1.36 -1.27

9 ARBMH-9 15.00 3.39 3.44* 33.56 -4.97 3.40*

10 ARBMH-10 14.44 1.88 1.00 32.33 4.49 -1.28

11 900M 14.00 0.00 -0.32 40.78 2.26 -0.60

12 Bio-9681 14.78 -0.72 -0.27 37.11 -0.59 11.71*

13 Pinnacle 15.33 0.83 1.72* 37.00 11.27 11.79*

14 All-rounder 15.89 0.30 0.56 30.67 5.80 -0.75

15 K-235 15.78 2.72 -0.28 29.89 8.83 34.32*



145

8

214

1

12

6

13

3

107

415

9

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

1 3 5 7 9 11 13 15 17 19

Reg

ressio

n c

oeff

icie

nt

(bi)

Fig. 3. Relationship between the regression coefficient (bi) and number of kernel rows per cob

Number of kernel rows per cob


Fig. 3. Relationship between the regression coefficient (bi) and number of kernel rows per cob

13

15

14

10 4

11

212

7

3

8

9

1

5

6

-15

-10

-5

0

5

10

15

1 6 11 16 21 26 31 36 41 46

Re

gre

ss

ion

co

eff

icie

nt

(bi)

Number of kernels per row

Fig. 4. Relationship between the regression coefficient (bi) and number of kernels per row


Fig. 4. Relationship between the regression coefficient (bi) and number of kernels per row

Table 18. Stability parameters for 100- grain weight and grain yield per plant

100- grain weight Grain yield per plant Sl. No.

Genotypes


1 ARBMH-1 31.04 1.73 15.57* 137.97 2.12 11.98

2 ARBMH-2 37.39 0.17 -1.24 146.99 0.66 0.88

3 ARBMH-3 35.74 -1.25 41.79* 136.29 1.72 4.81

4 ARBMH-4 30.58 0.03 -1.13 129.12 2.66 23.16*

5 ARBMH-5 33.53 1.13 17.55* 125.21 2.45 3.51

6 ARBMH-6 33.94 1.86 0.81 131.5 0.98 -0.25

7 ARBMH-7 33.37 3.68 -0.29 129.23 0.94 -2.08

8 ARBMH-8 31.03 3.11 -0.86 126.08 1.42 -4.09

9 ARBMH-9 30.37 1.15 -0.72 123.64 1.73 28.18*

10 ARBMH-10 31.79 0.03 1.3 126.32 0.21 -2.2

11 900M 38.66 0.08 -1.18 136.17 -0.65 -4.05

12 Bio-9681 36.58 0.27 -0.8 123.72 0.93 -4.12

13 Pinnacle 30.21 2.43 -0.89 116.87 0.73 8.13

14 All-rounder 35.44 4.12 6.8* 119.63 -0.67 9.34

15 K-235 33.25 -3.55 4.83 97.49 -0.26 173.94*



15

3

2 1112

4 10

7 5

61

13

8

7

14

-4

-3

-2

-1

0

1

2

3

4

5

1 6 11 16 21 26 31 36 41

Re

gre

ss

ion

co

eff

icie

nt

(bi)

100-grain weight (g)

Fig. 5. Relationship between the regression coefficient (bi) and 100-grain weight (g)


Fig. 5. Relationship between the regression coefficient (bi) 100-grain weight (g)

ARBMH-9 and K-235 exhibited significant deviation from regression (S2di≠0). The hybrids

ARBMH-1, ARBMH-2, ARBMH-3 and ARBMH-6 were found to have higher mean value than population mean with bi and S

2di values not significantly different from 1 and 0, respectively

(Table 19 and Fig. 6).

4.4.3.12 Fodder yield per hectare (t)

Stability parameters for fodder yield per hectare are presented in table 19. The hybrids ARBMH-2 and K-235 showed the highest (11.88 t/ha) and lowest (7.3 t/ha) mean value for fodder yield per hectare, respectively. Mean fodder yield per hectare over three environments was 8.82 t/ha.

Among 15 hybrids, ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-4, ARBMH-5, ARBMH-6, ARBMH-7, ARBMH-9, Pinnacle and All-rounder exhibited significant deviation from regression (S2di≠0). The hybrids 900M and Bio-9681 were found to have higher mean value than population mean with bi and S

2di values not significantly different from 1 and 0,

respectively (Table 19 and Fig. 7).

Table 19. Stability parameters for grain yield per hectare and fodder yield per hectare

Grain yield per hectare Fodder yield per hectare Sl. No.

Genotypes


1 ARBMH-1 68.98 2.19 3.16 9.8 2.26* 0.07

2 ARBMH-2 79.21 0.74 0.4 11.88 0.81* 0.01

3 ARBMH-3 66.63 1.74 1.2 8.66 1.64* 0.02

4 ARBMH-4 61.69 2.62 5.44* 8.33 2.57* 0.11*

5 ARBMH-5 61.91 2.5* 0.96 8.67 2.54* 0.03

6 ARBMH-6 70.1 0.95 -0.19 8.2 0.92* -0.01

7 ARBMH-7 61.75 0.93* -0.58 8.27 0.9* -0.01

8 ARBMH-8 58.14 1.35* -1.03 7.9 1.34 -0.02

9 ARBMH-9 57.7 1.66 6.16* 7.91 1.66* 0.12*

10 ARBMH-10 59.65 0.21 -0.6 7.76 0.2 -0.01

11 900M 74.13 -0.73* -0.98 11.05 -0.8 -0.02

12 Bio-9681 68.73 1.07* -1.01 10.1 1.14 -0.02

13 Pinnacle 56.48 0.73 1.81 8.25 0.78* 0.04

14 All-rounder 59.15 -0.68 2.23 8.22 -0.69* 0.04

15 K-235 48.74 -0.27 43.56* 7.31 -0.28 0.99*



1114

15

10

13

7

2

612

8

93

1

54

-1

-0.5

0

0.5

1

1.5

2

2.5

3

1 11 21 31 41 51 61 71 81 91

Grain yield (q/ha)

Reg

res

sio

n c

oeff

icie

nt

(bi)

Fig. 6. Relationship between the regression coefficient (bi) and grain yield (q/ha)


Fig. 6. Relationship between the regression coefficient (bi) and grain yield (q/ha)

6

1114

15

10

713

8

9 3

2

12

1

54

-1

-0.5

0

0.5

1

1.5

2

2.5

3

1 3 5 7 9 11 13

Re

gre

ssio

n c

oe

ffic

ien

t (b

i)

Fodder yield (t/ha)

Fig. 7. Relationship between the regression coefficient (bi) and fodder yield (t/ha)


Fig. 7. Relationship between the regression coefficient (bi) and fodder yield (t/ha)

5. DISCUSSION

The knowledge about the extent of fluctuations of yield and yield attributes over environments is very important to identify genotypes which are widely adapted. Grain yield is quantitatively inherited character and there is considerable interaction between genotypes and environments. Some of the crop varieties are widely adapted, whereas others do not. Multilocation testing of genotypes provides an opportunity to the plant breeders to study the adaptability of genotypes to a particular environment and the stability of the genotype over different environments. The information on genotype x environment interaction is of major importance to the plant breeder in developing an improved stable variety.

Since, yield is a complex trait influenced by different yield contributing traits, it is necessary to measure the nature of association among these traits and their direct and indirect contributions to grain yield.

In the present investigation, an attempt has been made to assess the variability and its parameters in maize hybrids. The results obtained from the present investigation are discussed under the following headings.

5.1 Mean performance of genotypes in different locations




5.1 MEAN PERFORMANCE OF GENOTYPES IN DIFFERENT LOCATIONS

Among the environments, Bheemarayanagudi location had taken minimum days to 50 per cent tasseling followed by Arabhavi and Dharwad, while for days to 50 per cent silking, Bheemarayanagudi location taken minimum days followed by Dharwad and Arabhavi.

For plant height, Bheemarayanagudi location was more favourable followed by Arabhavi and Dharwad. In respect of ear height, genotypic expression was better at Dharwad as compared to other location as indicated by higher location mean.

For better expression of cob length, the location Arabhavi was most favourable followed by Dharwad and Bheemarayanagudi, while for cob girth expression, Bheemarayanagudi was favourable followed Arabhavi and Dharwad.

In respect of number of kernel rows per cob and number of kernels per row, genotype expression was better at Arabhavi as compared to other two locations as indicated by highest mean of 15.27 to 34.87 respectively.

The mean 100-grain weight at Arabhavi was highest than Dharwad and Bheemarayanagudi locations. Environment had greater influence on grain yield per plant as indicated by the range of environmental index. Among the three locations, Arabhavi was found to be more favourable for higher grain yield per plant followed by Dharwad and Bheemarayanagudi.

Among the three environments, Arabhavi was found to be most favourable with better expression for grain yield per hectare and fodder yield per hectare followed by Dharwad and Bheemarayanagudi.

Among three environments, for better expression of nine out of characters viz., days to 50 per cent tasseling, days to 50 per cent silking, cob length, number of kernel rows per cob, number of kernels per row,100- grain weight, grain yield per plant, grain yield per hectare and fodder yield per hectare, second environment Arabhavi was more favourable. For two characters viz., plant height and cob girth, third environment Bheemarayanagudi was more favourable and for the one character viz., ear height, first environment Dharwad was more favourable.


Fifteen genotypes were evaluated at three locations during kharif 2007, to know the amount of variability for yield and yield contributing characters. The analysis of variance indicated highly significant differences among genotypes for all the characters except days to 50 per cent tasseling, days to 50 per cent silking and cob girth at all locations. Among the characters studied viz., days to 50 per cent tasseling, days 50 per cent silking, plant height, ear height, cob length, cob girth, number of kernel rows per cob, number of kernels per row, 100-grain weight, grain yield per plant, grain yield per hectare and fodder yield per hectare. The results clearly indicated presence of higher amount of variability for yield and yield components among the genotypes studied. Therefore, there is lot of scope for selection for majority of the traits in the progenies. Absolute variability of different characters does not reveal which of the particular characters are showing the highest variability. This could be assessed through standardizing the genotypic and phenotypic variance by obtaining coefficients of variability. Thus, components of variation, such as genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) were computed.

Further, it is essential to know about the selection by separating out the environmental influence from the total variability. This indicates the accuracy with which a genotype can be identified by its phenotypic performance. The estimates of heritability alone fail to indicate the response to selection (Johnson et al., 1955). Therefore, the heritability estimates appears to be more meaningful when accompanied by estimates of genetic advance. The genetic advance as per cent mean (GAM) was also estimated.

A wide range of variation for grain yield per plant was noticed in the present investigation. A narrow difference between GCV and PCV indicate that they are less influenced by environment. The character also showed moderate GCV and PCV values with high heritability and GAM, indicating that it is controlled by additive gene action and less influenced by environment. In addition, this trait can be improved through selection. But, yield is a complex character and is function of several component characters and their interaction with environment. Direct selection based on yield alone will not be very effective in crop improvement programmes. Grafius (1959) pointed out that structure of yield probed through its components rather than directly would be more efficient. Study of yield components and their inter relationship along with yield and their direct and indirect contribution towards yield is of immense importance. Falconer (1981) obtained that this helps to base selection procedure to strike a balance when two opposite desirable characters affecting the principal characters are being selected. Also, it helps to improve different characters simultaneously.

The other characters differed in the extent of their variation as measured by the coefficients of variation. The magnitude of genotypic coefficient of variation was low as compared to that of phenotypic coefficients of variation in all the traits studied.

Moderate values of genotypic coefficient of variation and phenotypic coefficient of variation were noticed for cob girth, number of kernels per row, grain yield per plant and fodder yield per hectare. Similar results of moderate GCV and PCV were noticed by Robin and Subramanian (1994), Gyanendra et al. (1995), Mani et al. (1996), Pradeepkumar and Satyanarayana (2001), Sumathi et al. (2005), Chaudhary and Choudhary (2002) and Om Prakash et al. (2006). Whereas for number of kernel rows per cob, 100 grain weight and ear height the GCV was low and PCV was moderate. For days to 50 per cent tasseling, days to 50 per cent silking, cob length and plant height the GCV and PCV was low. Similar results of low GCV and PCV were reported by Satyanarayana et al. (2005) for days to 50 per cent silking and plant height, Satyanarayana and Saikumar (1996) and Gargi and Sharma (2001) for plant height, Pradeepkumar and Satyanarayana (2001) for days to 50 per cent tasseling, days to 50 per cent silking and plant height and Chaudhary and Choudhary (2002) for cob length.

Heritability values for all the characters were found to be high except days to 50 per cent tasseling, days to 50 per cent silking and plant height and the values ranged from 43.47 to 96.83 per cent. Fodder yield per hectare recorded the highest heritability value (96.83%) and days to 50 per cent tasseling recorded the lowest heritability value (43.47%). The results suggest that the yield components in maize are less influenced by environmental conditions.

High heritability values for these traits indicate that the variation observed was mainly under genetic control and less influenced by environment.

High genetic advance as per cent mean was also noticed for fodder yield per hectare and grain yield per plant. Mani et al. (1999), Jha and Ghosh (2001), Chaudhary and Choudhary (2002), Sumathi et al. (2005), Kumar et al. (2006) and Om Prakash et al. (2006) reported similar results of high heritability coupled with high genetic advance.


5.3.1 Association analysis

Correlation among different quantitative characters

The phenotype of a plant is the result of interaction of a large number of factors. Hence, the final yield is some total of effects of several components factors. Therefore, it is important to know the extent and nature of interrelationship between grain yield and its contributing characters and also among themselves. The correlation coefficient helps the breeder in determining the direction and number of characters to be considered in improving the grain yield. In the present study, phenotypic correlations were computed of various characters at all the three environments.

Days to 50 per cent tasseling and days to 50 per cent silking were negatively correlated with grain yield per plant except in Arabhavi location but for cob girth, number of kernel rows per cob and number of kernels per row association varied with the three locations (Netaji et al., 2000, Pradeepkumar and Satyanarayana, 2001, Venugopal et al., 2003).

Fodder yield per hectare showed significant positive association with grain yield per plant in all the three locations (Gautham et al., 1999, Venugopal et al., 2003 and Wali et al.,

2006). Ear height exhibited positive but non-significant correlation with grain yield per plant in three locations (Venugopal et al., 2003). Plant height showed significantly negative correlation with cob length except in Bheemarayanagudi (Singh et al., 1991, Venugopal et al., 2003 and Wali et al., 2006). 100 grain weight showed non-significant association with all the characters in all three locations. These results are similar with the results reported by Venugopal et al. (2003). Number of kernel rows per cob showed non-significant association with all the characters in all three locations except with 100 grain weight in Dharwad location. These results are similar with Venugopal et al. (2003). Number of kernels per row showed non-significant association with all the characters in all three locations except with fodder yield per hectare in Dharwad location.

5.3.2 Path analysis

Correlation between different characters is important in planning selection programme. However, the correlation coefficient only devotes the total association existing between a pair of characters which themselves are the result of the interaction between various features of the plant. Physiology and linkage also cause association between characters that are not easily separable from physiological relationship. Path analysis is one useful technique by which Dewey and Lu (1959) tried to find out the relative contribution of component characters directly to grain yield per plant and indirectly through various other features of the plant. Since, then, the technique is being extensively used in understanding the genetic architecture of the crop plant. In the present investigation, among the characters studied, fodder yield per hectare exerted maximum direct effect on grain yield per plant in all the locations. This indicates that other features are held constant on increase in fodder yield per hectare will reflect in an increased grain yield. 100-grain weight and number of kernel per row also showed positive directs effect on yield. Similar results were obtained by Venugopal et al. (2003).

However, the fodder yield per hectare showed significant association with grain yield but and exhibited positive direct effect in all the three locations. Cob girth exhibited positive direct effect on grain yield per plant in all the locations except in Arabhavi location. Days to 50 per cent tasseling, days to 50 per cent silk and number of kernel rows per cob recorded non-significant association with grain yield per plant in all the location, but number of kernels per row showed positive direct effect in Bheemarayanagudi but negative direct effect in both

Arabhavi and Dharwad. (Gyanendra et al., 1995, Swarnalatha et al., 2001 and Venugopal et

al., 2003).

Number of kernel rows per cob also had negative direct effect on grain yield in Bheemarayanagudi and Arabhavi but positive direct effect in Dharwad. Ear height had direct positive effect on grain yield except in, Dharwad. Cob length showed positive direct effect on grain yield per plant in Arabhavi and Dharwad, where in Bheemarayanagudi had direct negative effect on grain yield per plant (Harijinder et al., 2006).

5.4 STABILITY ANALYSIS

The ability of a genotype to produce a narrow range of phenotype in different environments can be called as stable (Lewis, 1954). The genotypes will be stable in the absence of the environmental influence as well as genotype X environment interaction.

Eberhart and Russell (1966) model of stability analysis was used for the assessment of environmental influence and genotype x environmental interaction on genotypes for each character. When the genotype x environment interaction were significant for the characters, then partitioning of total sum of squares due to genotype x environment interactions into predictable and unpredictable source of variations was done using the procedure given by Eberhart and Russell (1966).

The main objective of the present investigation was to identify the stable genotype(s) over the three locations for different quantitative characters. The stability analysis was carried out by employing linear regression model given by Eberhart and Russell (1966).

5.4.1 Pooled analysis of variance

The pooled analysis of variance (Table 13) revealed that mean sum of squares (MSS) due to genotype and environment were highly significant indicating the presence of variability among the genotypes and environments for most of the characters studied indicating the variability among the environments.

The genotype x environment interaction was significant for all the traits except plant height, ear height, number of kernel rows per cob and number of kernels per row indicating differential response between the genotypes and environments. The MSS due to environment + (genotype x environment) was significant for all the characters except plant height, ear height, number of kernel rows per cob, number of kernels per row, cob girth and 100 grain weight. MSS due to environment (liner) was significant for major character like days to 50 per cent tasseling, days to 50 per cent silking, ear height, number of kernel rows per cob, grain yield per plant, grain yield per hectare and fodder yield per hectare indicating that environment effects are additive. The linear component of G x E interaction was also significant for four-character viz., cob length, grain yield per plant, grain yield per hectare and fodder yield per hectare indicating significant rate of linear response of the genotypes to environmental changes. The pooled deviation was also significant for all the characters except cob length indicating that non-linear component of G x E interaction was predominant. Similar result were reported by Sain et al. (1987), Arun and Singh (2004b), Kaundal and Sharma (2006).

5.4.2 Stability analysis of individual characters

Stability analysis was carried out by employing the linear regression model proposed by Eberhart and Russell (1966). Although there are number of models available to characterize the genotypes for their G x E interactions, this model is widely used for its simplicity and reliability. Many workers have employed this model in maize like Jha et al. (1986), Satyanarayana and Kumar (1995), Mani et al. (1999), Arun and Singh (2004b) and Kaundal and Sharma (2006).

An ideal genotype is defined as the one possessing high mean performance, with regression coefficient around unity (bi=1) and deviation from regression (S

2di) close to zero.

The linear regression is regarded as the measure of linear response of a particular genotype to the changing environment. If the regression coefficient (bi) is greater than unity, the genotype is said to be highly sensitive to environmental fluctuations but adapted to high yielding environments. If regression coefficient (bi) is equal to unity, it indicates the average sensitivity to environmental fluctuations and adaptable to all environments. If the regression

coefficient (bi) is less than unity, it indicates less sensitivity to environmental changes and if this is accomplished by a high mean value, then the genotype is said to be better adapted for poor conditions.

In the present study stability parameters such as mean (X), regression coefficient (bi) and deviation from regression (S

2di), as suggested by Eberhart and Russell (1966) were

considered to explain and discuss the stability of different genotypes for various characters under consideration.

5.4.2.1 Days to 50 per cent tasseling

The stability parameters for days to 50 per cent tasseling showed the genotypes ARBMH-1, ARBMH-2, ARBMH-5, ARBMH-6, ARBMH-9, ARBMH-10 and Pinnacle as stable across the environments. Among these ARBMH-1 was the earliest to 50 per cent tasseling while, ARBMH-10 was late to 50 per cent tasseling. The genotypes which require minimum number of days to 50 per cent tasseling are more desirable. So, ARBMH-1 was the stable and ideal genotype for days to 50 per cent tasseling. The hybrid 900M and Bio-9681 had the regression value more than unity, indicating its suitability to favourable environments. These results are in line with the reports of Gargi and Saikia (2001), Arun and Singh (2004b) and Kaundal and Sharma (2006).

5.4.2.2 Days to 50 per cent silking

The stability parameters for days to 50 per cent silking showed the genotypes ARBMH-1, ARBMH-2, ARBMH-6, ARBMH-9, ARBMH-10 and bio-9681 as stable across the environments. Among these ARBMH-1 was the earliest to 50 per cent tasseling while, line Bio-9681 was late to 50 per cent silking. The genotypes which require minimum number of days to 50 per cent silking are more desirable. So, ARBMH-1 was the stable and ideal genotype for days to 50 per cent silking. The hybrid 900M had the regression value more than unity, indicating its suitability to favourable environments. These findings are in accordance with Arun and Singh (2004b) and Kaundal and Sharma (2006).

5.4.2.3 Plant height

Considering stability parameters of 15 genotypes tested over environments, none of the genotypes had stable performance. While, the genotypes ARBMH-2, ARBMH-10 and Pinnacle had higher mean performance with nonsignificant regression values and significant deviation from regression indicating their suitability for all environments with unpredictable performance. while, the hybrids ARBMH-1 and 900M had high mean value with non significant regression coefficient and deviation from regression indicating adaptability across environments with predictable performance. The present findings are in agreement with Sain et al. (1987), Mahajan et al. (1991), Gargi and Saikia (2001), Arun and Singh (2004b) and Kaundal and Sharma (2006).

5.4.2.4 Ear height

Regression coefficient and deviation from regression were nonsignificant for hybrids ARBMH-1, ARBMH-2, ARBMH-5, ARBMH-7, ARBMH-8, 900M,bio-9681 and All-rounder suggesting that these are responsive genotypes for environments and suitable for all environments. Among these the hybrids ARBMH-2, ARBMH-5, ARBMH-8, 900M and bio-9681 were ideal as they had higher mean than population mean. However, the genotypes ARBMH-3, ARBMH-4, ARBMH-6, ARBMH-10 Pinnacle and K-235 had significant deviation from regression indicating suitability of these genotypes over the environments with unpredictable performance. Similar findings were also reported by Sain Dass et al. (1987), Mahajan et al. (1991) and Arun and Singh (2004b).

5.4.2.5 Cob length

For cob length, all the genotypes were responsive to environments and suitable for all the environments as both the stability statistics parameters were nonsignificant indicating these could be widely adaptable for all the environments. Among these ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-4, ARBMH-5, ARBMH-7, ARBMH-9, 900M and Bio-9681 were ideal as they had higher mean values than population mean.

5.4.2.6 Cob girth

The hybrids ARBMH-2, ARBMH-4, ARBMH-5, ARBMH-10, 900M, bio-9681 and All-rounder were considered as responsive to environments as they had nonsignificant bi and S2di value. These hybrids had better adoptability to widely differing conditions with predictable performance, when grown in different environments. While, the genotypes ARBMH-2, ARBMH-4, 900M, Bio-9681 and All-rounder had higher mean value, nonsignificant regression values and nonsignificant deviation from linear regression, a desirable combination of stability. The hybrids ARBMH-1, ARBMH-3, ARBMH-6, ARBMH-7, ARBMH-8, ARBMH-9, pinnacle and K-235 had significant deviation from regression. These hybrids are classified as average sensitive genotypes indicating their suitability to all the environments with unpredictable performance. Significant observations were also made by Arun and Singh (2004b) and Kaundal and Sharma (2006) confirming the present investigation.

5.4.2.7 Number of kernel rows per cob

The regression coefficient and deviation from regression were nonsignificant for most of the hybrids except ARBMH-8, ARBMH-9 and Pinnacle suggesting that these are responsive for environment and suitable for all environments. Among these eight genotypes viz., ARBMH-2, ARBMH-4, ARBMH-6, ARBMH-8, ARBMH-9, Pinnacle, All-rounder and K-235 were ideal as they had higher mean than the population mean. However, the hybrids ARBMH-8, ARBMH-9 and pinnacle had significant deviation from regression indicating suitability of these genotypes over the environments with unpredictable performance. Arun and Singh (2004b) and Kaundal Sharma (2006) also made similar observations confirming the present results.

5.4.2.8 Number of kernels per row

Stability parameters for this character revealed that two genotypes viz., ARBMH-6 and 900M had higher mean performance, nonsignificant regression coefficient and deviation from regression, which revealed their adaptation to all the environments. While the genotypes ARBMH-1, ARBMH-2, ARBMH-4, ARBMH-5, ARBMH-9, Bio-9681, Pinnacle and K-235 had significant deviation from regression indicating their average sensitivity to environments with unpredictable performance. Similar findings were also made by Arun and Singh (2004b) and Kaundal Sharma (2006).

5.4.2.9 100-grain weight

The genotypes ARBMH-2, ARBMH-6, 900M and Bio-9681 showed stable performance with higher mean 100-grain weight across environments which revealed their adaptation to all the environments. However, the hybrids viz., ARBMH-1, ARBMH-3, ARBMH-5 and All-rounder had significant deviation from regression indicating their average sensitivity to environments with unpredictable performance. Arun and Singh (2004b) also reported same results in their studies, which endorsed the present investigation.

5.4.2.10 Grain yield per plant

Stability parameters for these character revealed that six hybrids viz., ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-6, ARBMH-7 and 900M had higher mean performance, nonsignificant regression coefficient and deviation from regression, which revealed their adaptation to all the environments. While the hybrids ARBMH-4, ARBMH-9 and K-235 had significant deviation from regression indicating their average sensitivity to environments with unpredictable performance. These findings are in line with Jha et al. (1986), Sain Dass et al. (1987), Arun and Singh (2004b) and Kaundal and Sharma (2006), which enlightened the present outcome on grain yield per plant.

5.4.2.11 Grain yield per hectare

Considering stability parameters of 15 genotypes tested over environments, the hybrids ARBMH-7 and 900M which had regression value less than unity whereas, the hybrids ARBMH-5, ARBMH-8 and Bio-9681 had regression value more than unity are suitable for poor and favourable environments respectively. Four hybrids ARBMH-1, ARBMH- 2, ARBMH-3and ARBMH-6 had ideal stability parameters i.e., higher mean value, nonsignificant regression coefficient and deviation from regression, indicating their wider adaptability and

stable performance across locations. Similar results are also reported by Satyanarayan and Kumar (1995) which encouraged present results.

5.4.2.12 Fodder yield per hectare

For fodder yield per hectare, only two hybrids 900M and bio-9681 exhibited higher mean performance along with nonsignificant regression coefficient and deviation from regression indicating their wider adaptability to all the environments. The hybrids ARBMH-8 and ARBMH-10 also exhibited nonsignificant bi and S2di values but had lower mean value compare to overall mean of fodder yield per hectare. The genotypes ARBMH-2, ARBMH-6, ARBMH-7, pinnacle, All-rounder and ARBMH-1, ARBMH-3, ARBMH-4, ARBMH-5 ARBMH-9 had significant bi value lower and higher than unity indicating their suitability to poor and favourable environments, respectively. The findings of Satyanarayan and Kumar (1995) and Gargi and Saikia (2001) recorded the present investigation.

5.4.3 Stability of genotypes for different characters and frequency of stable genotypes for different traits

The hybrids exhibiting stability for each character was given a score 1 and total score obtained by each hybrid across 12 characters are calculated. Similarly for each trait total number of stable hybrids was calculated (Table 21) of the 15 hybrids the hybrids ARBMH-2 and ARBMH-6 were found to be stable for as many as eight characters followed by 900 M (7 characters) and ARBMH-1 and Bio 9681 (6 characters).

Among the 15 genotypes as many as seven hybrids exhibited stability for days to 50 per cent tasseling and days to 50 per cent silking, while four hybrids showed stability for plant height, 100-grain weight and grain yield per hectare. For ear height and grain yield per plant six hybrids were found stable while, nine hybrids exhibited stability for cob length. Two hybrids showed stability for cob girth, number of kernels per row and fodder yield per hectare, while three hybrids showed stability for number of kernel rows per cob.

FUTURE LINE OF WORK

1. The stable hybrids identified for different traits to be analyzed for stability over seasons, years and locations

2. The stable hybrids may be screened against pest and diseases

3. The top performing hybrids may also be studied for response to integrated nutrient management

Table 20. Stable genotypes identified for each characters over locations

Sl. No.

Characters Stable hybrids

1 Days to 50 per cent tasseling ARBMH-1, ARBMH-2, ARBMH-5, ARBMH-6, ARBMH-9, ARBMH-10 and pinnacle

2 Days to 50 per cent silking ARBMH-1, ARBMH-2, ARBMH-6, ARBMH-9, ARBMH-10, Bio-9681

3 Plant height (cm) ARBMH-1, ARBMH-6, ARBMH-9,900M

4 Ear height (cm) ARBMH-2, ARBMH-5. ARBMH-8, 900M, Bio-9681

5 Cob length (cm) ARBMH-1, ARBMH-2,900M,Bio-9681

6 Cob girth (cm) ARBMH-2,900M,All-rounder

7 Number of kernel rows per cob ARBMH-4, ARBMH-6,Bio-9681

8 Number of kernels per row ARBMH-6,900M

9 100 grain weight (g) ARBMH-2, ARBMH-6,900M,Bio9681

10 Grain yield per plant (g) ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-6, ARBMH-7, 900M

11 Grain yield per hectare(q) ARBMH-1, ARBMH-2, ARBMH-3, ARBMH-6

12 Fodder yield per hectare (t) 900M,Bio9681

Table 21. Stability of hybrids for different characters

Hybrids Days to

50 % tasselling


Plant height (cm)

Ear height (cm)

Cob length (cm)

Cob girth (cm)

Number of

kernel rows

per cob

Number of

kernels per row

100 grain

weight (g)


Grain yield

per ha (q)

Fodder yield (t/ha)

Total score

ARBMH-1 1 1 1 - 1 - - - - 1 1 - 6

ARBMH-2 1 1 - 1 1 1 - - 1 1 1 - 8

ARBMH-3 - - - - 1 - - - - 1 1 - 3

ARBMH-4 - - - 1 - - 1 - - - - - 2

ARBMH-5 1 1 - 1 1 - - - - - - - 4

ARBMH-6 1 1 1 - - - 1 1 1 1 1 - 8

ARBMH-7 - - - - 1 - - - - 1 - - 2

ARBMH-8 - - - 1 - - - - - - - - 1

ARBMH-9 1 1 1 - 1 1 - - - - - - 5

ARBMH-10 1 1 - - - - - - - - - - 2

900M 1 - 1 1 1 - - 1 - 1 - 1 7

Bio-9681 - 1 - 1 1 - 1 - 1 - - 1 6

Pinnacle - - - - - - - - 1 - - - 1

All-rounder - - - - 1 - - - - - - - 1

K-235 - - - - - - - - - - - - 0

6. SUMMARY AND CONCLUSIONS

The present investigation was taken up to elucidate the information on the amount of variability present in the component traits of yield and the amount to which it is heritable, nature of association and their direct and indirect effect of their component character on grain yield per plant, genotype x environment interaction in order to assess the relative stability of maize hybrids for productivity traits.

The experiment was conducted with the public and private bred maize hybrids in a randomized block design with three replication in three different locations viz., Dharwad, Arabhavi and Bheemarayanagudi. The observations were recorded for 12 quantitative traits viz., days to 50 per cent tasseling, days to 50 per cent silking, plant height, ear height, cob length, cob girth, number of kernel rows per ear, number of kernels per row, 100-grain weight, grain yield per plant, grain yield per hectare and fodder yield per hectare. The results obtained in the percent study were summarized below.

Mean performance of hybrids in three different locations indicated that Arabhavi was the most favourable environment for the better expression of most of the characters except plant height, ear height and cob girth.

Genetic variability analysis revealed that, grain yield per plant and fodder yield per hectare had higher heritability and expected GAM. Ear height, cob girth, number of kernel row per cob, number of kernels per row, 100-grain weight, fodder yield per hectare and grain yield per plant had moderate variability. While, days to 50 per cent tasseling, days to 50 per cent silking, plant height and cob length had lower variability. All the character under study had high variability except days to 50 per cent tasseling and days to 50 per cent silking. Grain yield per plant and fodder yield per hectare had high GAM. Other characters viz., ear height, cob length, cob girth, number of kernel rows per cob, number of kernels per row and 100 grain weight had moderate GAM.

The association analysis revealed that grain yield per plant exhibited significant positive association with fodder yield per hectare at all the locations and path analysis indicated that fodder yield per hectare exerted high direct effect at all the three locations. Therefore irrespective of locations, this trait can be considered as principal yield determining component and it is suggested to use this as selection criteria for grain yield improvement in maize.

Analysis of variance revealed significant differences among the hybrids for all the quantitative traits at all the locations suggesting the presence of desired variability among the genotypes tested. The variability of the selected experimental locations was evidenced by the significant variances due to environment.

The pooled analysis of variance revealed significant differences among the genotypes and environments for most of the characters among the hybrids except plant height, ear height, number of kernel rows per cob and number of kernel per row indicating genotypes and environments tested were diverse in nature whereas, genotype x environment interaction was also significant for all the traits suggesting that genotypes interacted significantly with the environments. While, genotype x environment (linear) interaction was nonsignificant for all the characters indicating genotypes responded nonlinearly to the changing environment. The non linear component (pooled deviation) was highly significant for all the traits except cob length indicating that this part of variation in terms of performance of the hybrids is unpredictable.

The estimates of stability parameters viz., mean (X), regression coefficient (bi) and deviation from regression (s

2di) were analyzed by using stability model of Eberhart and

Russel (1966). The stability analysis for grain yield revealed that the hybrids viz., ARBMH- 1, ARBMH- 2, ARBMH- 3 and ARBMH- 6 had high mean, unit regression coefficient (bi) and associated with non significant deviation from regression (s

2di) indicating that these hybrids

were well adapted to all environments. Similarly the hybrids 900M and Bio-9681 were found to be most suitable for fodder yield, whereas, the hybrids ARBMH- 3, ARBMH- 5 and ARBMH- 9 were found to be specifically adapted to unfavorable environments for fodder yield.

As indicated by significant linear and nonlinear genotype x environment interactions, the observed performance of the hybrids in the present study cannot be predicted and recommended. Therefore, it is suggested to study the performance of the hybrids in still more diverse environments to confirm the stability of the hybrids.

REFERENCES

Abdulai, M. S., Sallah, P. Y. K. and Safo-Karnataka, O., 2007, Int. J. Agric. Biol., 9(1): 41-45.

Abera, W., Labuschagne, M. T. and Maartnes, H., 2006, Evaluation of maize genotypes using parametric and non-parametric stability estimates. Cereal Res. Commun., 34(2/3): 925-931.

Abera, W., Rensburg, J. B. J-Van., Labuschagne, M. T. and Maartnes, H., 2004, Genotype – environment interactions and yield stability of new maize in Ethiopia. South African J. Pl. Soil, 21(4): 251-254.

Abirami, S., Vanniarajan, C., Arumugachamy, S. and Uma, D., 2007, Correlation and path analysis for morphological and biochemical traits in maize genotypes. Plant Archives, 7(1): 109-113.

Alika, J.E., 1994, Genetic variability among S1families for ogi yield in maize (Zea mays L.). Indian J. Genet., 54(1):27-31.

Allard, R. W. and Bradshaw, A. D., 1964, Implication of genotype × environmental interaction in applied plant breeding. Crop Sci., 4: 503-505.

Allard, R. W., 1961, Relationship between genetic diversity and consistency of performance in different environments. Crop Sci., 1: 127-129.

Alok Kumar, Gangashetti and Arju Dahiya, H. G., 1999, Analysis of direct and indirect effects of quantitative traits in diallel crosses of maize. Ann. Biol., 15: 173-176.

Amit Dadheech and Joshi, V. N., 2007, Stability analysis for quality traits in maize (Zea mays L.). Agric. Sci. Dig., 27(1): 26-29.

Anderson, W. F., Mozingo, R. W. and Wyne, J. C., 1989, Comparison of stability statistics as criteria for cultivars development in peanut. Peanut Sci., 16: 21-25.

Anees-Ahmad and Muhammad Saleem, 2003, Path coefficient analysis in Zea mays L. Int. J. Agric. Biol., 5(3): 245-248.

Annapurna, D., Khan, K. H. A. and Shaik Mohammad, 1998, Genotypic, phenotype correlations and path coefficient analysis between seed yield and other associated characters in tall genotypes of maize. Crop Res., 16: 205-209.

Anonymous, 2007, Agriculture, Centre for Monitoring Indian Economy, p.70.

Appadurai, R. and Nagarajan, R., 1975, Hybrid vigour in pop corn. J. Madras Agric. Univ., 62: 122-126.

Arun Kumar and Singh, N. N., 2004a, Identification of yield contributors through path analysis in maize. Ann. Agric. Res., 25(3): 448-450.

Arun kumar and Singh, N. N., 2004b, Stability studies of inbred and single crosses in maize. Ann. Agric. Res., 25: 142-148.

Balbinot Junior, A. A., Backes, R. L., Alves, A. G., Ogliari, J. B. and Fonreca, J. A. da, 2005, Contribution of yield components on grain yield in maize open pollinated varieties. Revista Brasileria-de-Agrociencia, 11(2): 161-166.

Bao-Heping, Zhang Jun, Zhao-Lizhi, Yu-Yan Shen and Dou-Maohai, 2004, Path analysis of ear characters in spring inbred maize lines. J. Jilin Agric. Univ., 26(1): 16-18.

Bhat, I. S. and Singh, R. O., 2005, Stability analysis of maize hybrids under different moisture regimes. New Botanist, 32(1/4): 79-85.

Bhiote, K. D., Pardeshi, S. R. and Mhaske, B. M., 2007, Correlation and path analysis studies in forage maize. Agric. Sci. Dig., 27(2): 146-147.

Bilbro, S. A. and Ray, L. L., 1976, Environmental stability and adoption of seven cotton cultivars. Crop Sci., 16: 821-824.

Boraneog, S. and Duara, P. K., 1993, Phenotypic performance and correlation in maize. Indian J. Agric. Sci., 6: 29-33.

Brian Westcott, 1986, Some methods of analyzing genotype environmental component of variability II heterozygotes. Heredity, 21: 339-405.

Breese, E. L., 1969, The measurement and significance of genotype environmental interactions in grasses. Heredity, 24: 181-200.

Bucio Alanis, L. and Hill, J., 1966, Environmental and genotype –environmental components of variability II heterozygotes. Heredity, 21: 339-405.

Bucio Alanis, L., Perkins, Jean, M. and Kinks, J. L., 1969, Environmental components of variability II heterozygotes. Heredity, 24: 399-405.

Burak, R. and Broccoli, A. M., 2001, Stability analysis for yield and expansion volume in pop corn hybrids. Maize Genetics Co-operation News lett., 75: 37-38.

Burton, G. W. and Devane, E.M., 1953, Estimating heritability in fall fescue (Festuca circunclinaceae) from replicated clonal-material. Agron. J., 45: 478-481.

Cardoso, M. J., Carvalho, H. W. L-de, Gomes-e-Gama, E. E. and Souza-E-M-de, 2007, Stability of the grain yields of Zea mays L. cultivars in the Brazilian Middle north region. Revista Ciencia Agronomica, 38(1): 78-83.

Carvallo, H. W. L. de, Silva Leal-M-de-L-da, Santos-M-X-dos, Cardoso, M. J., Monterio, A. A. T. and Tabosa, J. N., 2000, Adaptability and stability of corn cultivars in the Brazillian Northeast. Pesquisa Agropecuaria Brasileira, 35(6): 1115-1123.

Castanon, G., Tosquy, H., Arano, R. and Raygoza, B., 1999, Cluster analysis in tropical maize hybrids and synthetics in Veracruz, Mexico. Agronomia Mesoamericana, 9(2): 77-81.

Chaudhary, B. S., Paroda, R. S. and Singh, V. P., 1978, Stability and genetic architecture of harvest index in wheat. Z. Plfanzenzuchty, 81: 312-318.

Chen FaBo, Yang KeCheng, Rong TingZhao and Pan GuangTang, 2007, Analysis of genetic diversity of maize hybrids in the regional tests of Sichuan and South West China. Acta Agronomica Sinica, 33(6): 991-998.

Chen-Xuejun, 1997, Yield stability analysis of new maize hybrids selected in Northeast and North China. J. Jilim Agric. Univ., 19(2): 24-27.

Choudary, A.K. and Chaudhary, L.B., 2002, Genetic studies in some crosses of maize (Zea mays L.). J. Res. (BAU), 14(1):87-90.

Choudary, R. C., Nanda, J. S. and Nagendra and Gupta, T. B. R., 1972, Adaptability index in rice. Oryza, 9: 13-21.

Choukan, R., 1999, Stability analysis of yield of grain maize hybrids using different stability indices. Seed and Plant, pp. 170-183.

Choukan, R., 2000, Stability of grain yield and yield components of maize hybrids. Seed and Plant, 16(3): 269-284.

Comstock, R. L. and Moll, R. H., 1963, Genotype × environment interactions, statistical genetics and plant breeding (Eds. Hanson, W. D. and Robinson H. F.). Nation. Acad. Sci., National Research Council Publication, 982: 164-196.

Dash, B., Singh, S. V. and Shashi, J. P., 1992, Character association and path analysis in S < ovid:sub.|</ovid: sub lines of maize (Zea mays L.). Orissa J. Agric. Res., 5(1/2): 10-16.

Datu, H., 1998, Phenotypic and genotypic correlations between vegetative period and grain yield and other agronomic trait in early maize. Analete Institului Ceretari pentra Cerela Si Plante Technice Fundulea, 65: 101-109.

Datu, M., 1999, Phenotypic and genotypic correlations between growing season, plant height and leaf number in maize. Corcetari Agronomice in Moldova, 31: 69-78.

Debnath, S. G. and Khan, M. F., 1991, Genotypic variation, covariance and path analysis in maize. Pakistan J. Sci. and Indust. Res., 34: 391-394.

Dewey, D. and Lu, K. H., 1959, A correlation and path coefficient analysis of components of crested wheat grass seed production. Agron. J., 51:515-518.

Dodiya, N. S. and Joshi, V. N., 2003, Genotype × environment interaction and stability analysis for yield and maturity in maize (Zea mays L.). Crop Res., Hisar, 26(1): 110-113.

Downsell, C.r., Paliwali, R.C. and Contrell, R.P., 1996, Maize in the third world. West View Press, pp.1-37.

Eberhart, S. A. and Russell, R. A., 1966, Stability parameters for comparing varieties. Crop Sci., 6: 36-40.

Ei-Nagouly, O. O., Abul-Fadol, M. A., Ismail, A. A. and Khamis, M. N., 1983, Genotypic and phenotypic correlations and path analysis in maize and their implications in selection. Agronomy Abstract, Madison Square, Wisconsin, USA, American Soc. Agron., 15: 62-63.

Ei-Shouny, K. A., Ei-Bagowly, O. H., Ibrahim, K. I. M. and Ai-Ahmad, S. A., 2005, Correlation and path analysis in four yellow maize crosses under two planting dates. Arab Univ. J. Agric. Sci., 13(2): 327-339.

Falconer, D. S., 1981, Introduction to Quantitative Genetics. 2nd

Edition, Oliver and Boyd, Edinburg, London, p. 316.

Finlay, K. W. and Wilkinson, G. N., 1963, The analysis of adaptation in a plant breeding programme. Australian J. Agric. Res., 14: 742-754.

Firoz Khatum, Sajeda begum, Abdul Motin, Suraja Yasmin and Islam, M. R., 1999, Correlation coefficient and path analysis of some maize hybrids. Bangladesh J. Bot., 28: 9-15.

Fisher, R. A. and Mackenzie, W. A., 1923, Studies in crop variation II. The manorial response of different potato varieties. J. Agric. Sci., Cambridge, 13: 311-320.

Freeman, G. H. and Perkins, J. M., 1971, Environmental and genotype environmental components of variability VIII relation between genotype grown in different environments and measure of these environments. Heredity, 27: 15-23.

Freeman, G. H., 1973, Statistical methods for the analysis of genotype environment interaction. Heredity, 27: 15-23.

Gargi Sharma and Saikia, R. B., 2001, Genetic variability and stability of fodder yield in baby corn (Zea mays L.). The Andhra J., 48 (3&4):165-168.

Gargi Sharma and Saikia, R. B., 2000, Stability analysis for yield attributing characters in baby corn (Zea mays L.). Indian J. Hill Farm., 13(1/2): 30-34.

Gautam, A. S., Mittal, R. K. and Bhandari, J. G., 1999, Correlations and path coefficient analysis in maize (Zea mays L.). Ann. Agric. Biotechnol. Res., 4: 169-171.

Gautam, P. L., Garg, D. K., Pal, S., Malik, S. K. and Manglik, V. P., 1986, Genotype × environment interactions and crop improvement A review in genetics and crop improvements (Eds. P. K. Gupta and J. R. Bahl), Rastogi and Company, Subhas Bazar, Meerut, India.

Geetha, K. and Jayaraman, N., 2000, Path analysis in maize (Zea mays L.). Agric. Sci. Dig., 20: 60-61.

Gharde, G. N. and Deshmukh, P., 1996, Stability analysis in maize for grain yield and its component traits. PKV Res. J., 20(2): 176-178.

Gomes-e-Gama, E. E., Parentoni, S. N., Pachecco, C. A. P., Olivera, A. C. de, Oliveira Guimaraes, P. E. de and Santos, M. X., dos, 2000, Pesquisa Agropecuaria Brasileira, 35(6): 1143-1149.

Goulden, C. H.,1959, Methods of Statistical Analysis, Asia Publishing House, Calcutta.

Grafius, J, F., 1959, Heterosis in barley. Agron. J., 5: 551-554.

Guang Cheng, Xue-Yan and Gou-Sheng Xue, 2002, Path analysis of elite yield components of maize. J. Maize Sci., 10(3): 33-35.

Guirong, W., Shui, Z. De., Decai, C. and Yi, Y. G., 1997, Genetic analyses of various yield components of maize hybrids. J. Shandong Agric. Unvi., 28: 33-36.

Gyanendra Singh, Major Singh, Bhutta, D. T. and Awasthi, R. P., 1996, Stability analysis in maize under mid hills of Sikkim. J. Hill Res., 9(1): 65-68.

Gyanendra Singh, Major Singh and Dhiman, K.R., 1995, Genetic analysis of maize (Zea mays L.) in Sikkim. Indian J. Agric. Sci., 65(4):293-294.

Hanson, G. H., Robinson, H. F. and Comstock, R. E., 1956, Biometrical studies of yield in segregating populations of Korean hespedeza. Agron. J., 48:267-282.

Harjinder Singh, Chawla J.S. and Grewal, M.S., 2006, Correlation and path coefficient analysis on some elite genotypes. Crop. Improv., 33(1):31-33.

Hill, J., 1975, Genotype × environment interaction a challenge for plant breeding. J. Agric. Sci., Cambridge, 85: 477-493.

Jain, I. P., 1982, Genotype environment interaction. In Statistical Techniques in Quantitative Genetics, Tata McGraw Hill Publishing Company, Limited, New Delhi.

Javed, H. I., Masood, M. A., Chughtai, S. R., Malik, H. N., Hussain, M. and Saleem, A., 2006, Performance of maize genotypes on the basis of stability analysis in Pakistan. Asian J. Pl. Sci., 5(2): 207-210.

Jayakumar, J., Sunderam, T., Ranguramarajan, A. and Kannan,S., 2007, Studies on path analysis in maize (Zea mays L.) for grain yield and other yield attributes. Plant Archienes, 7(1):279-282.

Jha, P.B. and Ghosh, J., 2001, Variability and component analysis for fodder or dual purpose maize (Zea mays L.).J. Res., 13(1):65-67.

Jha, P B., Khehra, A. S. and Akhter, S. A., 1986, Stability analysis for grain development and yield in maize. Crop Res., 13: 16-19.

Jha, P.B., Ghosh, J. and Nirala, R. B. P., 1998, Genetic variability and character association in fodder maize. J. Res. BAU, 10(2): 139-143.

Johnson, H. W., Robinson, H. F. and Comstock, R. E., 1955, Genotypic and phenotypic correlations and their implications in selection of soybean. Agron. J., 47: 477-483.

Johnnsen, H. W., 1909, Elementa dor Exakten Erblickkoitalohru Int. Vol. 151, Jene Gustav Fisher.

Kaundal, R. and Sharma, B. K., 2006, Genotype × environment interaction and stability analysis for yield and other quantitative traits in maize (Zea mays L.) under rainfed and high rainfall valley areas of the submontane. Research on Crops., 7(1): 171-180.

Khakim, A., Stoyanova, S. And Tsankova, G., 1998, Establishing the correlation between yield and some morphological reproductive and biochemical characters in maize. Rantenie dini Nauki, 35: 419-422.

Kim, S. K., 1975, Genotype × environment interactions of several agronomic traits of maize. Korean J. Breed., 7: 163-169.

Knight, R., 1970, The measurement and interpretation of genotype environment interactions. Euphytica, 19: 225-235.

Krishnan, V. and Natarajan, N., 1995, Correlation and component analysis in maize. Madras Agric. J., 82: 391-393.

Kumar, A. and Kumar, D., 1997, Correlation studies in maize (Zea mays L.). Ann. Soil, Ludiayana, 13: 271-273.

Kumar, N. M. V. and Kumar, S., 2000, Studies on characters association and path coefficient for grain yield and oil content in maize (Zea mays L.). Ann. Agric. Res. 21: 73-78.

Kumar, S., Shashi, J. P., Singh, J. and Singh, S. P., 2006, Correlation and path analysis in early generation inbreds of maize (Zea mays L.). Crop Improv., 33(2): 156-160.

Lewis, D., 1954, Genotype × environment interaction – A relationship between heterosis, phenotypic stability and variability. Heredity, 8: 333-356.

Li-Yong Hong, Li Yue Hua, Zheng Ming, Kong Liang Liang and Chen Guomin, 2006, Analysis on correlation between yield and main agronomic traits of waxy maize hybrids. J. Maize Sci., 14(4): 56-60.

Luthra, O. P. and Singh, R. K., 1974, A comparison of different stability models in wheat. Theor. Appl. Genet., 45: 143-147.

Mahajan, K. and Prasad, A. S. R., 1986, Principal component analysis of genotype environment interaction in rice. Crop Improv., 13: 58-62.

Mahajan, Y., Khehra, A. S. and Pal, S. S., 1991, Stability analysis in maize over diverse seasons. Crop Res., 4: 267-284.

Maharajan, V., Khehra, A., Dhillon, B. S. and Sharma, V. K., 1990, Inter relationship of yield and other traits in maize in monsoon and winter season. Crop Improv., 17: 128-132.

Malhotra, V. V. and Khehra, A. S., 1986, Genotypic variation and co-variation in indigenous germplasm of maize. Indian J. Agric. Sci., 56: 811-816.

Mani, V. P. and Bisht, G.S., 1996, Genetic variability in local maize (Zea mays L.) germplasm of uttar Pradesh hills. J. Hill Res, 9(1): 131-134.

Mani, V. P. and Singh, N. K., 1999, Stability analysis of yield in maize. Indian J. Agric. Sci., 69: 34-35.

Mani, V. P., Singh, N. K., Bisht, G. S. And Sinha, M. K., 1999, Variability and path coefficient study in indigenous maize (Zea mays L.) germplasm. Environment and Ecology, 17: 653-658.

Manivannan, N., 1998, Character association and component analysis in maize. Madras Agric. J., 85: 293-294.

Mehra, R. B. and Ramanujam, S., 1979, Adaptation in bengalgram. Indian J. Genet., 30: 492-500.

Mohammad Akbar, M., Shabbir Shakoor, Amer Hussain and Mohammad Sarwar, 2008, Evaluation of maize three way crosses through genetic variability, broad sense heritability, character association and path analysis. J. Agric. Res., 46(1): 39-45.

Mohammad Basheruddin, M., Balakrishna Reddy and Phaik Mohammad, 1999, Correlation coefficient and path analysis of component character as influenced by the environments in forage maize. Crop Res., 17: 85-89.

Murakami, D. M., Cardoso, A. A., Cruz, C. D. and Bizao, N., 2004, Some considerations about two methodologies for stability and adaptability analysis. Ciencia Rural, 34(1): 71-78.

Nawar, A. A., Fahmi, A. I. and Salma, S. A., 1999, Genetic analysis of yield components and callus growth characters in maize (Zea mays L.). J. Genet. Pl. Breed., 53: 119-127.

Netaji, S. V. S. R. K., Satyanarayana, E. and Suneetha, V., 2000, Heterosis studies for yield and yield component characters in maize (Zea mays L.). The Andhra Agric. J., 47: 39-42.

Nirala, R. B. P. and Jha, P. B., 2003, Phenotypic stability for fodder traits in maize (Zea mays L.). Range Management and Agroforestry, 24(1): 9-13.

Ogunbodede, B. A., Ajibade, S. R. and Olakojo, S. A., 2001, Grain yield stability of new maize varieties in Nigeria. Africa Crop Sci. J., 9(4): 685-692.

Om Prakash, Shanthi, P., Satyanarayana, E. and Saikumar, R., 2006, Studies on genetic variability exploitation for quality traits and agronomic characters on quality protein maize (QPM) germplasm (Zea mays L.). Ann. Agric. Res., 27(2):147-153.

Packiaraj, D., 1995, Genetic studies of yield and its components in maize (Zea mays L.). Ph. D. Thesis, Tamil Nadu Agric. Univ., Coimbatore, India.

Padam Singh, Dass, S., Dwivedi, V. K. and Kumar, Y., 2003, Genotype × environment interactions for grain yield in maize (Zea mays L.). Nation. J. Pl. Improv., 5(2): 135-136.

Panse, V.G. and Sukhatme, D. V., 1964, Statistical Methods for Agricultural Workers. ICAR, Pub .New Delhi, p. 115.

Paroda, R. S. and Hayes, J. D., 1971, An investigation of genotype environment interactions for rate of ear emergence in spring barley. Heredity, 26: 157-175.

Paroda, R. S., Karwasra, R. R. and Yadav, H. R., 1973, Prediction of phenotypic performance through genotypes environment interaction studies in fodder cowpea (Vigna sinensis Brak). Paper presented at the Second Congress of SABRAO.

Patel, D. A., Patel, J. S., Bhatt, M. M. and Bhatt, H. M., 2005, Correlation and path analysis in forage maize (Zea mays L.). Res. on Crops, 6(3): 502-504.

Patel, D. R. and Sanghi, A. K., 1989, Stability analysis for oil and other characters in maize. Gujarat Agric. Univ. Res. J., 15(1): 30-35.

Perkins, J. A., 1970, Environmental and genotype-environmental components of variability IV Diallel sets crosses. Heredity, 25: 29-40.

Perkins, J. M. and Jinks, J. D., 1968a, Environment and genotype environmental components of variability III multiple, line and crosses. Heredity, 23: 339-356.

Perkins, J. M. and Jinks, J. L., 1968b, Environmental and genotype environmental components of variability IV. Non-linear interactions for multiple inbred lines. Heredity, 23: 525-535.

Pixley, K. V. and Bjarnason, M. S., 2002, Stability of grain yield, endosperm modification and protein quality of hybrid and open pollinated quality protein maize (QPM) cultivars. Crop Sci., 42(6): 1882-1890.

Pradeep Kumar, P. and Satyanarayana, E., 2001, Variability and correlation studies of full season inbred lines of maize. J. Res., Acharya N. G. Ranga Agric. Unvi., Hyderabad, India, 29: 71-75.

Prasad, S. K. and Singh, N. K., 1991, Phenotypic stability of some Indian and exotic genotypes of maize. Crop Improv., 18: 41-45.

Probecky, O., 1976, Yield and its components in newly bred high yielding maize grain hybrids. Rostlinna Vjroba, 22: 1021-1027.

Prodhan, H. S. and Rai, R., 2000, Character association in pop corn. Indian Agriculturist, 44(1/2): 101-103.

Qifa Zhang and Shu Geng, 1986, A method of estimating varietal stability for data of longterm trails. Theor. Appl. Genet., 71: 810-814.

Rahman, M. M., Islam, M. R., Sultan, M. K. and Mitra, B., 1995, Correlation and path coefficient in maize (Zea mays L.). Sci. and Indust. Res., 30: 87-92.

Ram, J., Jain, O. P. and Murthy, B. R., 1970, Stability performance of some varieties and hybrid derivatives of rice under high yielding varieties programme. Ann. Cog. Ind. Soc. Agric. Stat., Patna.

Rather, A. G., Sheikh, F. A. and Wani, S. A., 1999, Variability and correlation studies in an improved population of maize (Zea mays L.) under rainfed conditions. Adv. in Pl. Sci., 12: 539-542.

Reddy, K.H.E. and Agarwal, B.D., 1992, Estimation of genetic variation in an improved population of maize (Zea mays L.). Madras Agric. J., 79(12):714-719.

Reddy, D. M., Ahuja, V. P. and Mukherjee, B. K., 2004, AAMI analysis for grain yield stability of maize hybrids. Ann. Agric. Res., 25: 218-222.

Robin, S. and Subramanian, M., 1994, Genetic variability study in biparental progenies in maize (Zea mays L.). Crop Res., 7(1):79-83.

Robinson, H. F., Comstock, R. E. and Harvey, P. H., 1949, Estimates of heritability and degree of dominance in corn. Agron. J., 41: 353-359.

Ron Parra, J., 1985, Variety-environment interactions for maize in Bajio, Mexico, Dissertation Abstracts Int. B-Sci. and Engg. 45(7): 1974B.

Saha, B. C. and Mukherjee, B. K., 1985, Analysis of heterosis for number of grains in maize (Zea mays L.). Indian J. Pl. Breed., 45: 240-246.

Saha, B. C. and Mukherjee, B. K., 1993, Grain yield of maize in relation to grain farming potential and other traits. J. Res., Birsa Agric. Univ., 5: 27-31.

Sain Dass, Mohinder Singh, Ashok, K., Sarial and Aneja, D. R., 1987, Stability analysis in maize. Crop Res., 14: 185-187.

Salami, A.E., Adegoke, S.A.O. and Adegbite, O.A., 2007, Genetic variability among Maize Cultivars Grown in Ekiti-State, Nigeria. Middle-East J. Sci. Res., 2(1): 9-13.

Satyanarayana, E., 1996, Combining ability for quantitative traits of early single cross maize hybrids under rainfed conditions. Mysore J. Agric. Sci., 30: 204-208.

Satyanarana, E. and Kumar, R.S., 1995, Phenotypic stability studies for grain and straw yields in maize composites. J. Maharashtra Agric. Univ., 20: 225-226.

Satyanarayana, E., Shanthi,P. and Saikumar, R., 2005, Genetic variability in baby corn (Zea mays L.) genotypes. J. Res. AGRAU., 33(1):83-86.

Satyanarayana, E. and Saikumar, R., 1996, Genetic variability of yield and maturity components in maize hybrids. Curr. Res., 25(1-12):10-11.

Schmildt, E. R. and Cruz, C. D., 2005, Adaptability and stability of maize using Eberhart/Russeel and Annichiarico methods. Revista Ceres, 52(299): 45-58.

Sharma, S. R., Khera, A. S., Dhillon, B. S. And Malhotra, V. V., 1982, Evaluation of S1 lines of maize crossed in a diallelic system. Crop Improv., 9: 42-47.

Shelake, D. V., Bhave, S. G., Bendale, V. W., Madav, R. R. and Pethe, U. B., 2005, Genetic factors influencing grain yield in maize. J. Ecobiol., 17(6): 521-528.

Shukla, G. K., 1972, Some statistical aspects of partitioning genotype environment components of variability. Heredity, 29: 254-273.

Singh, J. M. and Dash, B., 2000, Analysis of genetic variability and character association in maize (Zea mays L.). Environ. Ecol., 18(2): 503-505.

Singh, P. K., Jha, P. B. and Kumar, P., 2003, Path coefficient for green fodder yield and grain yield in maize (Zea mays L.). J. Appl. Biol., 13(1/2): 29-32.

Singh, P. K., Prasad, M. K. and Chaudary, L. B., 1999, Association analysis in winter maize. J. Appl. Biol., 9(2): 133-136.

Singh, S. N., Singh, K. N. and Singh, H. G., 1991, Genetic variability and inter relationship in maize. Narendra Deva J. Agric. Res., 6: 233-237.

Sivasubramanian, S. and Menon, M., 1973,Heterosis and inbreeding depression in rice. Madras Agric. J., 60:1139.

Sofi, P. A. and Rather, A. G., 2007, Studies on genetic variability, correlation and path analysis in maize (Zea mays L.). Maize Genet. Co-op. News lett., 81: 26-27.

Sprague, G. E. and Federer, W. T., 1951, A comparison of variance components in corn yield

trials; II error, year × variety, location × variety components. Agron. J., 43: 535-541.

Sree Kumar, S. G. and Suma Bai, D. I., 1995, Genetic analysis of yield and its components in fodder maize. J. Trop. Agric., 33: 8-10.

Srivas, S. K. and Singh, U. P., 2004, Genetic variability, character association and path analysis of yield and its component traits in forage maize (Zea mays L.). Range Manag. Agroforestry, 25(2): 149-153.

Sumathi, P., Nirmalakumari, A. and Mohanraj, K., 2005, Genetic variability and traits interrelationship studies in industrially utilized oil rich CYMMIT lines of maize (zea mays l). Madras Agric. J. 92 (10-12): 612 – 617.

Sunderaraj, N., Nagaraj,V. S., Venkataramu, M. N. and Jagannath, M.K., 1972, Design and Analysis of Field Experiments. University of Agricultural Sciences, Bangalore, p. 424.

Swarnalatha Devi, I. And Shaik Mohammed, 2001, Character association and path coefficient analysis of grain yield and components in double crosses of maize. Crop Res., 21: 255-359.

Tai George, 1971, Genotype stability analysis and its application to potato regional traits. Crop Sci., 11: 184-190.

Tan Heping, Wang Guiyue, Hu-Xiamnu and Xu-Qiaoxian, 2006, Multiple regression and path analysis of effective factors affecting maize yield. Acta Agriculturae Zhejiangensis, 18(4): 238-240.

Turesson, C., 1922, The species and the variety as ecological unit. Heredity, 3: 110-113.

Tyagi, A. P., Pokhariyal, G. P. and Odongo, O. M., 1988, Correlation and path coefficient analysis for yield components and maturity traits in maize (Zea mays L.). Maydica, 33: 109-119.

Umakanth, A. V. and Khan, H. A., 2001, Correlation path analysis of grain yield and yield components in maize (Zea mays L.). J. Res., Acharya N. G. Raga Agric. Univ., Hyderabad, 29: 87-0.

Umakanth, A. V., Satyanarayana, E. and Kumar, M. V. N., 2000, Correlation and heritability studies in Ashwini maize composite. Ann. Agric. Res., 21: 328-330.

Vaezi, S., Abd-Mishani, C., Yazdi-Samadi, B. and Ghannadha, M. R., 2000, Correlation and path analysis of grain yield and its components in maize. Iranian J. Agric. Sci., 31(1): 71-83.

Venugopal, M., Ansari, N. A. and Rajanikanth, T., 2003, Correlation and path analysis in maize (Zea mays L.). Crop Res., Hisar, 25(3): 525-529.

Verma, M. M. and Virk, D. S., 1983, Regression analysis of varietal adaptation. Crop Improv., 10: 67-78

Viola, G., Ganesh, M., Reddy, S. S. and Kumar, C. V. S., 2003, Studies on correlation and path coefficient analysis of elite baby corn (Zea mays L.) lines. Prog. Agric., 3(1/2): 22-24.

Wali, M. C., Salimath, P. M., Prashanth, M. and Harlapur, S. I., 2006, Studies on character association as influenced by yield, starch and oil in maize (Zea mays L.). Karnataka J. Agric. Sci., 19(4): 932-935.

Wang Dachum, 2006, Hereditary correlation and path analysis of main traits in maize at different yield levels. J. Maize Sci., 14(2): 40-41.

Webber, C.R. and Moorthy, B.R., 1952, Heritable and non-heritable relationship and variability of oil content and agronomic characteristics in the F2 generation of soybean crosses. J. Agron. 44:202-209.

Wright, S., 1921, Correlation and causation. J. Agric. Res., 20: 202-209.

Xie-Dasen, He-Xiao Ming, Peng-Qing Wu and He-Xin Zhou, 2003,Correlation and path analysis of agronomic characters in glutinous maize. Acta Agriculture Uiversitatis-Jiangxiensis, 25(4): 498-500.

Xie-Zhen Jiang, Li-Ming Shun, Li-XinHai, Zhang-Shi Huang and Zhang-Baoshi, 2007, Relativity between yields and agronomic traits of major maize inbred lines of north China. J. Shenyang Agric. Univ., 38(3): 265-268.

Yadava, T. P. and Kumar Prakash, 1979, Phenotypic stability of yield components and oil content in bunch group of groundnut. Indian J. Agric. Sci., 49: 318-321.

Yates, F. and Cocharan, W. G., 1938, The analysis of groups of experiments. J. Agric. Sci., Cambridge, 28: 556-580.

Yin ZhiTong, Xue Lin, Deng DeXiang, Bian YunLong, Chen GuoQing and Lu HuHua, 2004, Cluster analysis of plant traits in maize inbreds. Southwest China J. Agric. Sci., 17(5): 563-566.

Zhang-Li, Dong Shu Ting, Liu Cun Hui, Wang Kongjum, Zhang Jiwang and Liu-Peng, 2007, Correlation analysis on maize test weight, yield and quality. Scientia Agricultura Sinica, 40(2): 405-411.

APPENDIX

APPENDIX I

Monthly meteorological data during crop growth period (2007-08) at Main Agricultural Research Station, University of Agricultural Sciences, Dharwad

Temperature (°C)

Months Rainfall (mm)

Mean maximum Mean minimum

Relative humidity (%)

April - 07 86.4 36.7 21.4 55.0

May 65.0 34.6 21.3 61.0

June 220.1 29.7 21.3 80.0

July 211.2 27.0 21.1 85.0

August 176.0 27.1 20.5 85.0

September 18.08 27.2 20.3 83.0

October 74.8 29.7 19.4 68.0

November 54.0 29.5 15.1 53.0

December 0.0 29.0 14.6 65.0

APPENDIX II

Monthly meteorological data during crop growth period (2007-08) at ARS, Arabhavi (Gokak)

Temperature (°C)




June – 07 106.82 36.78 23.80 78.6

July 18.1 31.32 23.03 82.2

August 39.8 32.85 22.24 87.2

September 62.7 32.92 22.27 85.9

October 34.7 32.45 19.35 81.4

November 4.6 32.72 14.52 79.4

December 0.0 31.69 14.83 75.9

APPENDIX III

Monthly meteorological data during crop growth period (2007-08) at ARS, Bheemarayanagudi

Temperature (°C)




June – 07 80.50 33.58 25.80 60.70

July 83.00 32.03 23.35 65.95

August 97.50 31.35 22.32 70.11

September 152.50 30.03 21.71 74.54

October 13.50 31.58 20.29 63.29

November 18.50 29.55 15.80 56.11

December 0.00 30.32 16.30 55.83

RELATIVE STABILITY ANALYSIS OF PUBLIC AND PRIVATE BRED HYBRIDS OF MAIZE (Zea mays L.)

NAGABHUSHAN 2008 Dr. M. C. WALI

MAJOR ADVISOR

ABSTRACT

An experiment was carried out involving 15 hybrids of maize including five private bred maize hybrids during kharif 2007 at three locations to know the magnitude of variability present and to assess the stability of hybrids for productivity traits. The hybrids were also assessed for the nature of inter-relationships, path effects of different characters towards grain yield.

Mean performance of hybrids indicated that Arabhavi was most favourable environment for better expression of most of the characters. Analysis of variance revealed significant differences among most of the characters. Genetic variability analysis revealed grain yield per plant and fodder yield per hectare had moderate variability, higher heritability and expected GAM.

The association analysis revealed that grain yield per plant exhibited significant positive association with fodder yield per hectare at all the locations and path analysis indicated that fodder yield per hectare exerted maximum direct effect at all the locations.

The pooled analysis of variance revealed significant differences among the genotypes and environments for most of the characters indicating the genotypes and environments tested are diverse in nature. G x E interaction was significant for most of the characters suggesting genotypes interacted significantly with the environments. G x E (linear) interaction was significant for cob length, grain yield per plant, grain yield per hectare and fodder yield per hectare indicating significant rate of linear response of genotypes to environmental changes. The non-linear component was highly significant for all the traits indicating variance in terms of performance of genotypes is unpredictable.

On the basis of stability parameters, ARBMH-2, ARBMH-6 and 900M were promising genotypes for maximum number of characters. While, ARBMH-2 was the best yielder for grain and fodder across the environments.

sa in maize

Documents

characterwise chronological report

randomized block design

red coloured figures

agricultural research station

environmental indices ranged

monthly meteorological data

variability ii heterozygotes

negative direct effect