Performances of selected tropical white maize single-cross hybrids for yield and yield attributing traits

In Ghana, the production of hybrid maize is at its infant stage, occupying only about 3% of the area devoted for maize production. Therefore, the yields of this crop are low, 1.7 t ha -1 due to the use of open pollinated varieties (OPVs), shortage of high yielding varieties, biotic and abiotic stresses. Thus, the present study was designed to identify superior crosses based on their agronomic performance. Thirtytwo hybrids and three checks were evaluated using a 5 x 7 alpha lattice design replicated twice at the University of Ghana, WACCI research farm during 2015/-16 offseason using drip irrigation. Analysis of variance revealed that genotype mean squares were highly significant (P < 0.001) for days to 50% anthesis and silking, plant and ear height, plant aspect, ear length, number of kernel rows ear -1 , number of kernels row -1 and yield. Similarly, mean squares of genotypes were significant (p ≤ 0.05) for ear rots, anthesis-silking interval, and husk cover. Based on the mean grain yield performance, six promising single crosses, L8 x T2, L1 x T2, L16 x T1, L16 x T2 L4 x T2, L9 x T1 having grain yield of 6377, 6011, 5848, 5222, 5150 and 5135 kg ha -1 , respectively were identified as possible candidates for release after establishing the stability of their performance in multi-locational trials and should be promoted for adoption and commercialization in the country.


INTRODUCTION
In Ghana, maize is the major crop in terms of area coverage and 2 nd staple food after rice where it is mainly, consumed among households (MoFA., 2011).However, low yields of 1.7 t ha -1 has been reported in Ghana as compared to 6.0 (China) and 9.9 t ha -1 (U.S.A) (FAOSTAT, 2013).This prominent difference in grain yields has been attributed partially to the use of open pollinated varieties (OPVs), shortage of high yielding varieties, biotic and abiotic stresses (Fening et al., 2011;Ragasa et al., 2013).MoFA (2011) reported that attainable grain yields of about 6 t ha -1 have been recorded in maize yield variety trials.This report therefore shows that the average maize yield of 1.7 t ha -1 currently found in Ghana, is about 70% less than that obtained in maize yield varietal trials by researchers.Efforts are being made to bridge the gap between the present low *Corresponding author.E-mail: yazachewgenet@gmail.com.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License grain yield and the attainable grain yield by promoting the use of superior hybrid maize varieties.At present, there is a growing demand for use of hybrid seeds and this has an effect in the driving of the emergence of seed companies in Ghana (Ragasa et al., 2014).Studies have shown that single cross hybrids have uniform production, are higher yielding and more stable in performance as well as in other plant characteristics.Divan et al. (2013) reported that single cross hybrid seed was higher yielding in many parts of the world, due to the expression of heterosis (hybrid vigor).
So far, in Ghana, the production of hybrid maize is at its infant stage, occupying only about 3% of the area devoted for maize production in Ghana (Ragasa et al., 2013).This may be due to lack of superior hybrids as compared to open pollinated varieties, which farmers can readily adopt.Ragasa et al. (2014) showed that there is a demand to develop superior white maize varieties because of the increasing population, urbanization, poultry and fish sectors in Ghana.
Therefore, the objective of this study was to identify superior single crosses based on their agronomic performances.

Description of experimental area
The experiment was conducted during 2015/16 offseason using drip irrigation at research field of West Africa Centre for Crop Improvement, University of Ghana.The University is located at 5.6508° N, latitude and 0.1869°W longitude and an altitude of 97 m above sea level (m.a.s.l).

Genetic materials used for the study
Eighteen white tropical maize inbred lines with diverse genetic backgrounds were selected from the pool of inbred lines at the West Africa Centre for Crop Improvement (WACCI).This comprises of ten lines from the International Institute for Tropical Agriculture (IITA), six from International Maize and Wheat improvement Center (CIMMYT) and the two testers 1368 from IITA and CML 444 from CIMMYT maize breeding programmes.The 16 inbred lines were crossed to the two testers using the line by tester mating method and it generated 32 (16 × 2) cross combinations.

Development of single cross hybrids
Emerging ears of each female parent were covered with shoot bags before silk emergence to prevent contamination with unwanted pollen.The tassels of the male parents (testers) were covered with tassel bag when one-fourth of the tassel had dehisced.Selfpollination from three to five plants per plot was done for all parental lines to bulk seeds of parental lines for future breeding.Pollination was done when a uniform growth of the silk was visible on the covered ears.The tassel bag containing freshly shed pollen was transferred over the silks after removing the shoot bags from the ears.Before putting tassel bags on the tassel, all details like date of pollination and labels were clearly written on the bags by waterproof pencil.In order to avoid contamination and to get enough fresh pollen, the tassel bags were put on tassel one day earlier than the intended pollination date.The tassel bags were held in position with the help of paper clips.

Data analysis and procedures
Analysis of variance for all agronomic parameters studied was calculated using the PROC GLM procedure and test for significant differences among the genotypes was performed using SAS software (SAS, 2002).

RESULTS AND DISCUSSION
The mean performances of genotypes included in the study were computed for traits, which showed significant mean squares difference.Days to 50% anthesis ranged from 52 (L8 × T1 and L8 × T2) to 59 days (L3 × T1) with the overall mean of 56 days (Table 1).Testcrosses, L5 × T1, L5 × T2, L6 × T1, L6 × T2, L9 × T1, L11 × T1, L11 × T2 and L15 × T1 demonstrated significant fewer days than the hybrid between the two testers check (1368 x CML 444).L8 × T1 and L8 × T2 showed significant fewer days to 50% anthesis than the checks Obatanpa and Mamaba.On the other hand, 13 crosses revealed statistically the same days to 50% anthesis with checks Obatanpa and Mamaba.Days to 50% silking showed a similar pattern to days to anthesis and varied from 53 to 60 for L8 × T1 and L3 × T1, respectively.L5 × T1, L5 × T2, L8 × T1, L8 × T2, L9 × T1, L11 × T2 and L15 × T1 had significant fewer days to 50% silking than the three checks.Five testcrosses had the same days to 50% silking as the hybrid check 1368 × CML 444.Mean performance for anthesis-silking interval manifested that most of crosses had shorter anthesis-silking intervals than checks Obatanpa and Mamaba which indicate that there was better synchronization of anthesis and silking for crosses.Similar findings to this result were reported by pervious investigators (Abrha et al., 2013;Hosana et al., 2015) Mean performance of maize streak disease scored demonstrated that there were crosses, which had less disease scored than cheeks.It indicated that some testcrosses had better gene resistance to maize streak virus.From the results, except for L15 × T2, all testcrosses showed significant better performances than the commercial hybrid check (Mamaba).Similarly, L8 × T2, L7 × T1 and L11 × T2 performed better than the check Obatanpa.This finding is in agreement with previous work by Gichuru et al., (2011) who reported the genetic variation of germplasm in their reaction to maize streak virus disease.
Plant height among the studied testcrosses ranged from 124 (L10 × T1) to 207 cm (Obatanpa), respectively with over all mean of 154.9 cm.All genotypes were significantly shorter than Obatanpa with L10 × T1 (123 cm) and L6 × T2 (135 cm) being the shortest crosses among the 32 crosses.On the other hand, L16 × T1 (157 G.mean = mean, CV = coefficient of variation, SE (d) = standard error difference, R 2 = the model explain the variability of the response data around its mean, AD= days to 50% anthesis, SD = days to 50% silking, ASI = anthesis -silking interval, MSD = maize streak virus disease, PH = plant height, EH = ear height, PLASP = plant aspect, HC% = husk cover in percentrot % = ear rot in percent, EL = ear length, RE = number of kernel rows ear -1 , NKR = number of kernels row -1 and Yld = grain yield kg ha -1 , Check1 = 1368 x CML 444, Check 2 = Obatanpa, Check3 = Mamaba cm), L4 × T2 (159 cm), L12 × T2 (159 cm) and L8 × T2 (161 cm) were significantly taller than the hybrid between the two testers (144 cm).The remainder of the testcrosses' plant height, which ranged from 163 to 170 cm, was not significantly different from the check Obatanpa.However, L1 × T1 was significantly taller than Mamaba (Table 1).Hybrids with shorter plant height indicate less stem and root lodging.Ear height varied from 52 (L3 × T1) to 99 cm (Obatanpa).L3 × T1 (59 cm), L6 × T2 (59 cm), L10 × T1 (53 cm), L14 × T1 (57 cm) recorded significantly shorter ear heights from the check Mamaba and none of the crosses showed significantly different ear height than check 1368 × CML 444.Similarly, all genotypes had shorter ear heights than the check Obatanpa.The overall mean of ear height was 72.65 cm, whereas check 1368 × CML 444, Obatanpa and Mamaba had 62, 99 and 70 cm, respectively.Testcrosses that have short plant height and ear length indicate less stem and root lodging.Mean performance of plant aspect of genotypes included under study, four crosses, L8 x T2, L2 x T2 , L11 x T2 , L12 x T2 (2) exhibited good and significant plant aspect.It indicates that these testcrosses had better uniformity of plant height and ear height, less disease occurrence and good grain filling.Mean performance of genotypes for husk cover percentage ranged from zero to 50%, indicating that there were genetic variations among genotypes included under study for this trait.Most of crosses tested revealed good and significant husk cover percentages.Mean performance of genotypes for ear rots manifested considerable significant differences and varied from 0 to 57%.The testcrosses L13 x T2 (57%) and L1 x T1 (26%) had significantly high number of ears with ear rot, while the remaining testcrosses were not damaged significantly by ear rots.Thus, selecting these crosses with good husk cover may reduce yield losses caused by ear rots and weevils before harvesting.The current investigation is in agreement with the pervious researchers report (Hosana et al., 2015) Mean performance of ear length varied from 10 (L11 x T2) to 19 cm (L12 x T1 and L12 x T2) with overall mean of 15.5 cm.Of 32 genotypes tested, eight testcrosses showed significant higher ear length than the overall mean.Thus, testcross L1 x T2, L12 x T1, L12 x T2, L4 x T2 and L13 x T2 had the longest ear length.Among 32 genotypes included in the study, L1 x T2, L12 x T1 L12 x T2, L4 x T2, L13 x T2, L2 xT2, L5 x T2 and L15 x T2 showed significantly longer ear length than the three checks included in the study.Testcrosses, L4 x T1, L10 x T2, L11 x T2, 15 x T1 showed significantly shorter ear length than the checks.The means of checks for ear length were 15, 15 and 16 cm for 1368 x CML 444, Obatanpa and Mamaba, respectively.
Genotype mean performance of the number of kernel rows ear -1 was significant among tested materials in the study.Mean performance of this trait varied from 11 to 18, kernel rows ear -1 with the grand mean of 14 kernel rows.Of the 35 genotypes tested, the maximum number of kernel rows ear -1 was recorded from a cross L16 x T2 with 18 kernel rows ear -1 , while L6 x T1 and L12 x T1 had the least number of kernel rows ear -1 .

Mean of number of kernels row
-1 varied from 23 (L11 x T2) to 40 (L4 x T2 and L12 x T1) with over all mean of 33 kernels row -1 .Among the genotypes tested, L4 x T2 and L12 x T1, L1 x T1 , L5 x T2, L8 x T2, L15 x T2 with 40, 40, 38, 38, 38 and 38 kernels row -1 , respectively, had higher and significant number of kernels row -1 as compared to the three checks used in the experiment.Similarly, L5 x T1 and L12 x T2 with 37 kernels row -1 Ejigu et al. 163 each showed significant and higher number of kernels than check 1368 x CML 444 and Obatanpa.On the other hand, L11 x T2, L10 x T2, L9 x T2, L3 x T1, L4 x T1 with 23, 24, 26, 27 and 27 kernels per row, respectively showed significant and lower number of kernels as compared to the three checks.Thus, it is possible that crosses with higher ear length, higher number of kernel rows ear -1 and higher number of kernels row -1 can increase grain yield of maize.
The mean grain yield of genotypes tested in this study ranged from 2872 (L14 x T1) to 6377kg ha -1 (L8 x T2) with overall mean of 4668 kg ha -1 .Among 32 crosses evaluated, 17 testcrosses manifested significant higher grain yield than the best commercial hybrid check Mamaba.In general, L8 x T2, L1 x T2, L16 x T2, L16 x T1, L4 x T2 and L9 x T1 revealed significantly higher grain yield as compared to all the three checks included in the study and are identified as the potential hybrids for production after further testing to confirm stability.The means of three checks were not significantly different.These indicate that there were genetic variations among genotypes for this trait in agreement with several authors report (Hosana et al., 2015;Miranda et al., 2012;Vah, 2013).

CONCLUSION AND RECOMMENDATIONS
This study was conducted to identify superior singlecross hybrids (SCH) developed from line x tester mating design.The experiment had two phases.In the first phase of the experiment, thirty-two (16 x 2) cross combinations were generated through line x tester mating design.In the second phase, the 32 F 1 crosses including the hybrid between the two testers, one popular open pollinated variety and a standard hybrid checks were evaluated for their agronomical performance using a 5 x 7 alpha lattice design at WACCI research field in 2015.
The results obtained in the present investigation were encouraging and tremendous increase in grain yield was obtained in most of the hybrids.Six promising testcrosses (L8 x T2, L1 x T2, L16 x T1, L16 x T2, L4 x T2 and L9 x T2) which had higher yield as compared to the checks were identified based on their mean performance which can improve the production and productivity of maize yield in the country.
Therefore, promising testcrosses (L8 x T2, L1 x T2, L16 x T1, L16 x T2, L4 x T2, and L9 x T1) identified in this study should be used in maize research programme as possible candidates for release after confirming the stability of their performance in multi-locations and one more season.

Table 1 .
Mean performance of single crosses for yield and yield-contributed traits of selected tropical white maize.