Correlation analysis for maize grain yield, other agronomic parameters and Striga affected traits under Striga infested/free environment

Evaluation of 15 F1 hybrids and six parents obtained from partial diallel crosses was carried out during 2011 rainy season at Institute of Agricultural Research (IAR) Samaru, Zaria Striga infested field and Striga free field (2 environments). Interrelationships of grain yield and its components under Striga infestation revealed that character ear height had positive and highly significant phenotypic correlation with grain yield (rph = 0.55)**. While the remaining six characters recorded no significant correlation with grain yield. Similarly, Striga related parameters with Striga affected traits indicated high and highly significant positive correlation between Striga count one with Striga count two (rg= 0.79, rph = 0.92 and re = 0.96)** and Striga rating (rg = 0.84, rph = 0.71 and re = 0.77)** and also had highly significant positive genotypic correlation with number of leaves per plant (rg = 0.79**) Striga count two exihibited high and highly significant positive phenotypic and environmental correlation with the traits Striga rating (rph = 0.78**and re = 0.71**) and number of leaves per plant (rph = 0.64** and re = 0.55**). While Striga rating recorded highly significant positive genotypic correlation (rg = 0.89**) coefficient rph = 0.78** and rph= 0.97** respectively. At Striga free environment grain yield was positively significantly correlated with 1000 grain weight (rph = 0.43* and re = 0.45*), highly significantly positively correlated at genotypic level (rg= 0.69)**, phenotypic level (rph = 0.52)* and environmental level (re = 0.43)* with number of leaves per plant. Furthermore, plant height with grain yield recorded highly significant positive correlation genotypically, phenotypically and environmentally with the following coefficients (rg = 0.86, rph= 0.67 and re = 0.58)**, and likewise ear height and grain yield revealed significantly positive correlation with the following coefficients rg = 0.45*, rph = 0.48* and re = 0.49*. Thus characters such as ear height, plant height, number of leaves per plant and 1000 grain weight should be taken into cognizance when selecting high yielding genotypes. Hybrids TZEEI 11 x TZEEI 3, TZEEI 11 x TZEEI 7 are outstanding in performance, especially grain yield and TZEEI 11 is good for hybrid formation.


INTRODUCTION
Maize (Zea mays L.) is one of the principal agricultural crops in many countries along with wheat and rice (FAO, 1982).In most of the developing countries of the world, maize constitutes the principal food for the majority of the people.Demand for maize globally and sub-Saharan Africa has been predicted to increase by 54 and 93% in 1995 to 2020 (Pingali and Pandey, 2001), while that of Nigeria is predicted to increase by about 84% by the year 2020 (Shaibu et al., 1997).This is because in Nigeria maize is a major staple food and makes an important contribution to the diet of the majority of the people constituting about 80 to 90% of the dietary profile of adults and over 80% of that of infants (Doswell et al., 1996).It is also a dependable cash for farmers in northern Nigeria (Ado et al., 1999).
Industrially wise maize is used as corn bread, corn chips, paper, insulator, card board pipe, chemicals, plastics, methanol, baby food, greencorn, starch, glucose and oil (Nuhu and Showemimo, 2008).The parasitic witch weed Striga hermonthica infestation is one of the most serious constraints to cereal production by small holder farmers in sub-Saharan Africa (Olaoye and Bello, 2008) infestation result in substantial yield losses usually more than 70% (Kim, 1994).Much of the damage occurs before the Striga emerges from the ground.The degree of damage depends on the susceptibility of the cultivar, the Striga species, the level of infestation and any additional stress imposed by the environment (Shinde and Kulkarni, 1982;Vasudeva Rao, 1982).Kim (1994) in a study conducted on genetics of maize tolerance to S. hermonthica reported that genetic control in the maize inbreds tested was inherited quantitatively and thus expected to be durable.It was concluded that, two different types of gene action are responsible for the inheritance of the traits studied.Additive gene action plays a major role in inheritance of Striga tolerance and non additive gene action plays a major role in Striga emergence (Kim, 1994).Correlation studies between yield and yield components are pre -requisite to plan a meaningful breeding programme (Muhammad et al., 2004).Many researchers studied association of characters for the selection of high yielding varieties.Khatun et al. (1999) studied that grain yield per plant was positively and significantly correlated with 1000 grain weight, number of kernels ear -1 and ear height.Orylan et al. (1999) studied that most important traits influencing grain yield were number of grains per row and number of grains per ear.Akanvou and Doku (1998) reported that, genetic correlation (r g ) for characters studied showed that yield was negatively correlated with Striga count one (r g = -0.22),Striga rating one (r g = -0.92)and with ear Striga rating (r g = 0.88).They further stated that, the negative association obtained were expected since Striga reduces yield through its adverse effects on the physiology of the infested plants.
In another investigation under Striga asiatica, Olakojo and Olaoye (2011) reported that, at phenotypic level, Striga count was negatively and significantly correlated with Striga rating (r p = -0.30)*,ear height (r p = -0.74)*andmaize grain yield (r p = 0.57)* indicating that the two traits are probably controlled by different genes.They further affirmed that, Striga syndrome rating was however positively count, was positively correlated with Striga genes and significantly correlated with ear height (r p = 0.31)*, days to silking (r p = 0.60)* and grain yield (r p = 0.33)*.Phenotypic correlation of ear height and grain yield was however negatively significantly correlated suggesting that, the two traits were not closely associated.Similarly, genotypic correlation coefficients shows that, Striga rating, plant height, days to silking, days to tasselling and kernel rows per cob had the following coefficients of 0.91*,0.30*,0.44** and 0.85** respectively.Striga rating on the other hand was positively correlated with plant height (r g = 0.36) and days to tasselling (r g = 0.50)** as compared with ear height which was negative and significantly correlated with Striga rating (r g = 0.72)**.The objectives of this study therefore, were (1) to determine the association of some maize agronomic traits to grain yield under S. hermonthica infestation and free environment.(2) to determine the association of Striga related parameters to Striga affected traits in maize plant.

MATERIALS AND METHODS
The study was conducted during 2011 rainy season (May to October) at the Institute for Agricultural Research (IAR) Farm Samaru, Zaria (Northern guinea savanna 11° 11'N 7° 38'E, 686 m above sea level).The genetic materials used comprised six Striga resistant inbred lines that are extra-early maturing developed by International Institute of Tropical Agriculture (IITA) Ibadan.Paired parents mating according to Stuber (1980) was used to intermate the six inbred lines.The six inbred lines were crossed in a partial diallel mating.15F 1 hybrids were obtained, processed and stored prior to field evaluation.The field study was carried out at Striga sick plot of Institute for Agricultural Research Samaru, Zaria and Striga free field at the same Institute.Entries which included the hybrids and the parents were planted on 9th July 2011 under Striga infested field and 10th July, 2011 under Striga free field.Planting was done in a row plot of 5 x 1.5 m each and planted at inter row spacing of 75 cm and within row spacing of 50 cm to enhance a plant population of 53, 333 plant per hectare.The experiment was laid out using 5 x 5 incomplete lattice design pattern with three replications.Striga inoculum was prepared by thoroughly mixing Striga seeds with sieved sand at a ratio of 1:39 by weight which can be translated as approximately 2000 germinable seeds per 2.5 g of sand to seeds mixture (Kim, 1994).Uniform artificial inoculation was mixing Striga seeds with sieved sand at a ratio of 1:39 by weight which can be translated as approximately 2000 germinable seeds per 2.5 g of sand to seeds mixture (Kim, 1994).Uniform artificial inoculation was accomplished using one full coca cola bottle cap of Striga inoculum per planting hill of 3cm deep and 5cm wide.In the Striga infested trial, three seeds were sown per hole and later thinned to two seedlings per stand, at two weeks after planting (WAP).The same trial was repeated at the Striga free field.NPK fertilizer was applied at the rate of 60 kgN ha -1 , 30 kg P205 ha -1 and 30 kg K20 ha -1 in split-doses, half of the N and all of the P205 and K20 were applied at two WAP, while the remaining half of N was applied six WAP.In the Striga inoculated plot, Striga shoot counts at eight and 10 WAP and Striga syndrome rating at 10 WAP were recorded using scale 1-9 described by Kim (1994).For the rating, 1 = normal plant growth, no visible symptom; 9 = complete scorching of leaves causing premature death or collapse of the host plant and no ear formation.At both the Striga inoculated and the non-inoculated fields, agronomic parameters recorded were: Days to 50% tasseling (number of days from planting till the time 50% have tasseled), days to 50% silking (number of days from planting to when 50% of the population have silked), plant height measured from soil level to the base of the tassel, Ear height from the soil level to the node bearing the top most ear, number of leaves per plant, 1000 grain weight (weight of 1000 grains randomly taken from each plot after shelling) and grain yield per plot transformed to kilograms per hectare.
Genotypic, phenotypic and environmental correlations were estimated using the formula described by Singh and Chaudhary (1985):

RESULTS
Mean performance of parents and crosses are shown in Table 1 under Striga infested environment inbred parent TZEEI 4 (57 days) x TZEEI 4 (52 days) had the least number of days to 50% tasselling respectively.For days 50% silking inbred parent TZEEI 6 and TZEEI 4 (59 days each) and hybrid TZEEI 6 x TZEEI 3 (59 days) are the least for the trait.TZEEI 8 (9.89) plant and TZEEI 11 x TZEEI 7 (10.11)hybrid had the highest mean value for number of leaves per plant.For the trait plant height TZEEI 6 (86.78 cm) is the shortest parent however for crosses TZEEI 11 x TZEEI 7 (137.11)Under Striga free environment (Table 2) TZEEI 7 and TZEEI 6 (52 days each) together with the hybrid TZEEI 11 x TZEEI 3 (49 days) recorded the least number of days to 50% tasselling respectively for days to 50% silking inbred parent TZEEI 7 (57 days) and cross TZEEI 11 x TZEEI 4 (55 days) had the least mean value respectively.TZEEI 8 (9) parent had the highest number of leaves per plant, while hybrid TZEEI 6 x TZEEI 3 (9) recorded the highest mean value for the trait.TZEEI 6 (69.89 cm) and cross TZEEI 8 x TZEEI 6 (27.22 cm) had the lowest mean respectively.TZEEI 6 (18.11 cm) had the lowest mean value respectively for ear placement.For grain yield TZEEI 8 (737.77kg/ha) is the highest inbred parent for the trait and hybrid TZEEI 11 x TZEEI 3 (1111.10kg/ha) is the highest.For 1000 grin weight parent TZEEI 8 (114.53 g) is the highest for the trait, while cross TZEEI 7 x TZEEI 3 (115.7 g) had the highest mean value for the trait.
The genotypic (r g ), phenotypic (r ph ) and environmental (r e ) correlations measured in Striga infested environment and Striga free environment are presented in Tables 3, 4 and 5.All the correlation coefficient determined in this study were either positive or negative correlation coefficients.Under Striga infestation (Table 3) 1000 grain weight had significant positive genotypic correlation with days to 50% tasselling (r = 0.44).The trait also indicated highly significant positive phenotypic association with plant height (r = 0.66) and ear height (r = 0.55).Highly significant positive environmental correlation between the trait and days to 50% tasseling were observed (r = 0.88).Days to 50% tasseling exhibited highly significant negative environmental correlation with plant height (r = -0.78).Number of leaves per plant showed highly significant positive genotypic (r = 0.98) phenotypic (r = 0.75) and environmental (r = 0.72) association with ear height.Plant height showed highly significant positive phenotypic (r = 0.75) and environmental correlation (r = 0.98) with ear height.Ear height showed highly significant positive phenotypic association with grain yield (r = 0.55) (Table 3).For Striga parameters and Striga affected traitsin maize (Table 3).Striga count one showed highly significant positive genotypic correlation with Striga count two (r = 0.76), Striga rating (r = 0.84) and number of leaves per plant (r = 79).The trait also indicated highly significant positive phenotypic association with Striga count two (r = 0.92) and Striga rating (r = 0.71) Striga count one showed highly significant positive environmental correlation with Striga count two (r = 0.96) and Striga rating (0.77).Striga count two exhibited highly significant positive association with ear height (r =  0.48).The trait also showed highly significant positive phenotypic correlation with Striga rating and number of leaves per plant.The trait had highly significant positive environmental correlation with Striga rating and number of leaves per plant.Striga rating showed highly significant positive genotypic correlation with plant height (r = 0.89).The trait also had highly positive genotypic correlation with ear height (r = 0.67) and   significant positive genotypic correlation with number of leaves number per plant (r = 0.47).The trait also showed significant positive phenotypic association with ear height (r=0.44) and grain yield (r=0.43) and significant positive significant positive phenotypic correlation with ear height (r=0.78) and grain yield (r = 0.97).For Striga free environment (Table 5) 1000 grain weight indicated highly significant environmental correlation between the trait and grain yield (r = 0.45) was also observed.Days to 50% silking showed significant positive phenotypic correlation with plant height (r = 0.44).Number of leaves per plant exhibited highly significant positive genotypic association with ear height (r = 0.96) and grain yield (r = 0.69).The trait also had highly significant positive phenotypic correlation with plant height (r = 0.88) and ear height (r = 0.71) and highly significant positive genotypic correlation(r = 0.69) and significant phenotypic correlation (r=0.52) with grain yield and highly significant positive environmental correlation with plant height (r = 0.62) and ear height (r=0.63) and significant positive environmental association with grain yield (r =0.43).Plant height showed highly significant positive genotypic correlation with grain yield (r = 0.86) and highly significant positive phenotypic association with ear height (r = 0.71) and grain yield (r = 0.67).The trait also showed highly significant positive environmental correlation with ear height (r = 0.67) and grain yield (r = 0.58).Ear height showed significant positive genotypic correlation with grain yield (r = 0.45) and significant positive phenotypic correlation between the trait and grain yield (r =0.48).The trait also had significant positive environmental correlation with grain yield (r = 0.49).

Correlation
result under Striga infested environment indicated that 1000 grain weight was positively and significantly correlated with days to 50% tasseling genotypically (rg = 0.44)* and environmentally (re =0.88)**.1000 grain weight also have positive and significant association with plant height (r ph = 0.66)** and ear height (r ph = 0.55)** at phenotypic level.Similar result was reported by Ibitome (2010) where he indicated that 1000 grain weight increases with increase in plant and ear height which indicates more leaves, photosynthesis and heavier grains.Days to 50% tasselling have negative but significant environmental correlation with plant height (r e = 0.78)**.This indicates that the two traits are probably controlled by different genes.Days to 50% silking had the same effect as days to 50% tasselling on plant height, ear height and grain yield which showed that the longer the days to 50% tasseling and days to 50% silking, the taller the plant and ear heights the more the number of leaves and the more the assimilates for grain yield (Ibitome, 2010;Devi et al., 2001).Number of leaves per plant have positive and significant correlation with ear height at genotypic, phenotypic and environmental level with the following coefficients 0.98**, 0.75** and 0.72** respectively.The physiological implication of this association is the more the number of leaves the greater the ability of the plant to photosynthesize and consequently the production of assimilates for grain yield.Plant height had highly positive significant association with ear height at phenotypic (r p = 0.75)** and environmental level (r e = 0.98)** and this signifies that increase in plant height may increase grain yield (Hallauer and Miranda, 1988).Ear height also had positive significant phenotypic correlation with grain yield (r p = 0.55)** indicating that increase in ear height contributed to increase in grain yield as reported by Ibitome (2010) and Rahman et al. (1995).Correlation on Striga parameters was carried out taking into cognizance of the affected portion of the plant that were use to assess the damage caused by S. hermonthica as reported by Berner et al. (1997) the plant parts are leaves, stalk and cobs and it is on this note that correlation studies were carried out on number of leaves per plant, plant height, ear height and grain yield.
The trait Striga count one had positive significant correlation with Striga count two at genotypic (r g = 0.79)** phenotypic (r p = 0.92)** and environmental level (r e = 0.96)** and it also had positive association with Striga rating with following coefficients r g = 0.84**, r p = 0.71** and r e = 0.77**.The positive and significant correlation between Striga count one, count two and Striga rating concur with the findings of Gethi and Smith (2004) where they suggested that one could use one of these traits to select for the others.Similarly, Striga count one had significant and positive genotypic correlation with the number of leaves per plant (r g = 0.79)**.Striga count two had positive significant correlation with Striga rating (r p = 0.78** and r e = 0.71**), number of leaves per plant (r p = 0.6** and r e = 0.55**) plant height (r g = 0.83)**.This finding is almost in agreement with that of Olakojo and Olaoye (2011) where they reported significant positive correlation of Striga count on Striga rating and plant height under Striga lutea infestation.Furthermore, there is significant but negative correlation between Striga count two and ear height at genotypic level (r g = -0.48)*this indicate that according to Akanvou and Doku (1998) negative association were expected since Striga reduces yield through its adverse effects on the physiology of the infested plants.For Striga rating there are significant positive association with plant height at genotypic level (r g = 0.89)**, ear height and grain yield at phenotypic level with coefficient 0.78** and 0.97** respectively.This result concur with that of Olakojo and Olaoye (2011) in which they affirmed that Striga syndrome rating being positively counted was however positively correlated with Striga genes and significantly correlated with ear height and grain yield and positively significantly correlated with plant height.
Under Striga free environment 1000 grain weight had positive significant genotypic correlation with number of leaves per plant (r g = 0.47)* and positive significant genotypic and phenotypic correlation with ear height (r g = 0.67** and r p = 0.44*).It is similarly had an association with grain yield at environmental level (r e = 0.45*).The data agree with the findings of Rafique et al. (1999) and Orylan et al. (1999).Days to 50% silking had positive significant association with plant height at phenotypic level (r p 0.44)* number of leaves per plant had positive significant phenotypic and environmental correlation with plant height (r p = 0.88** and r e = 0.62**) and similarly, also had positive significant association with ear height (r g = 0.96)**, (r p = 0.71)** and (r e = 0.63)** and grain yield (r g = 0.69)**(r p = 0.52)* and (r e = 0.43)* genotypically phenotypically and environmentally.For plant height there is significant positive phenotypic and environmental correlation with ear height (r p = 0.71** and r e = 0.67**).It also had significant positive association genotypically (r g = 0.86)**, phenotypically (r p = 0.67)** and environmentally (re = 0.58)** with grain yield.Furthermore, the trait ear height recorded significantly positive correlation to grain yield at genotypic, phenotypic and environmental level with following coefficients 0.45*, 0.48* and 0.49* respectively.The physiological implication of days to 50% silking to plant height, ear height and grain yield was reported by Mohan et al. (2002) as days to 50% tasselling increases with days to 50% silking and had an inverse association with plant height, ear height and grain yield.Ibitome (2010) and Devi et al. (2001) indicated that days to 50% silking had the same effect as days to 50% tasselling on plant height, ear height and grain yield which showed that, the longer the days to 50% tasselling and 50% silking the taller the plant and ear heights, the more the number of leaves and the more the assimilate for grain yield.Also Troyer and Larkins (1985) observed that plant height was strongly associated with the flowering dates, both morphologically and ontogenically, because internode formation stops at floral initiation which means that earlier flowering maize is usually shorter.

Conclusion
From the result of this study it is concluded that effective selection for superior genotypes is possible taking into cognizance of days to 50% tasselling, days to 50% silking, number of leaves per plant, ear height, plant height and 1000 grain weight.
Y = environmental covariance between traits X and Y, σ 2 e X = Environmental variance of traits X, σ 2 e Y = Environmental variance of traits Y.
is the tallest plant while the smallest hybrid was TZEEI 8 x TZEEI 4 (69.89Halidu et al. 11 cm).TZEEI 4 (28.22)parent had the lowest ear placement.For grain yield TZEEI 8 (1413 kg/ha) inbred parent had the highest mean value while hybrid TZEEI 11 x TZEEI 7 (1093.33kg/ha) had the highest mean value.1000 grain weight TZEEI 8 (130.96g) recorded the highest mean value, TZEEI 11 x TZEEI 3 (128.17g) hybrid recorded the highest mean value for the trait.

Table 1 .
Mean performance of parents, four checks and fifteen crosses evaluated under Striga infestation in Samaru (2011).

Table 2 .
Mean performance of six parents, four checks and fifteen crosses evaluated under Striga free environment in Samaru (2011).

Table 5 .
Correlation among the seven traits of Maize grown under Striga free environment in Samaru (2011).