Standard Heterosis of Maize (Zea Mays L.) Genotypes for Grain Yield and Yield Related Traits in Mid-Altitude Agro-Ecology of Ethiopia

A line x tester analysis involving sixty four test-crosses generated by crossing 32 elite maize inbred lines with two testers and two standard checks was evaluated for yield and yield related traits in 6x11 alpha lattice design replicated twice during 2017 cropping season at Bako National Maize Research Center. The study was designed to estimate the amount of standard heterosis of the hybrids for grain yield and yield related traits. From this study, considerable standard heterosis for all traits over both commercial checks was manifested. For grain yield, the highest standard heterosis was recorded for L17xT1 (61.75 %) and (41.46%) over BH546 and BH547, respectively, indicating the presence of substantial heterotic potential that could be exploited in maize breeding program and possibility of developing desirable cross combinations through crossing of inbred lines with desirable traits of interest. Cross L3xT2 manifested negative standard heterosis over BH546 and BH547 for days to anthesis, silking and maturity that indicate earliness of the crosses in maturity as compared to both standard checks. Similarly, most of the crosses showed positive standard heterosis for biomass yield, harvest index, ear per plant, ear length, ear diameter, rows per ear, kernel per rows and thousand kernel weights over both commercial checks. In general, information from this study could be valuable for researchers who intend to develop high yielding maize varieties.

released by the same center. Table 1

Experimental Design and Field Management
The experiment was conducted in a 6 x11 alpha-lattice design with two replications. Each of the 11 incomplete blocks in replication was comprised of 6 plots having 5.1-meter row length with the spacing of 0.75 m between rows and 0.30 m between plants. All cultural practices such as land preparation, weeding, disease and insect pest control were done manually as per required.

Data Collected
Grain yield (GY): The total grain yield in kg per harvested plot and adjusted to 12.5% moisture level. Days to anthesis (DA): Number of days from planting to when 50% of the plant in a plot shed pollen. Days to silking (DS): Number of days from planting to when 50% of the plants in a plot produced 2-3 cm long silk. Days to maturity (DM): number of days from planting to when 50% of the kernels on the cob showed black layer on the tip the kernel where it attaches to the cob. Biomass yield (BY): Total aboveground biomass yield after oven drying to a constant weight in tons per hectare obtained from each plot at harvest. Harvest Index (HI): The ratio of dried grain weight per ha adjusted to 12.5% moisture content to the dried total aboveground biomass weight per ha. Ear length (EL): Average lengths of five randomly selected ears were measured in cm from the base to the tip of the ear. Ear diameter (ED): Average diameters of five randomly selected ears were measured in cm at mid-section along the length of the ear using a calliper. Number of ears per plant (EPP): Total number of harvested ears in each plot divided by the total number of stand count at harvest. Number of kernel rows per ear (KRPE): Total numbers of kernel rows of the ear were counted from five randomly taken ears and the average value was used as kernel rows per ear. Number of kernels per row (KPR): Number of kernels in each row from five randomly taken ears and the average values were recorded as kernels per row. Thousand kernel weight (TKW): 1000 randomly taken kernels were weighed from each plot using sensitive balance and was adjusted to 12.5 % moisture level.

Data Analysis Analysis of Variance (ANOVA)
The data collected for all yield and yield related traits were analyzed using PROC MIXED procedure in SAS (SAS, 2004).

Estimation of Standard Heterosis
Percent standard heterosis was calculated for traits showed statistically significant differences among genotypes as suggested by Falconer and Mackay (1996). This was computed as percentage increase or decrease of the cross performances over the best standard checks as follows. BH-546 and BH-547 were used as standard checks.
Where F1 ═ Mean value of the cross SV= Mean value of the standard variety SH = standard heterosis Test of significance for percent heterosis was made by using the t-test. The standard errors of the difference for heterosis and t-value were computed as follows: t (standard cross) = t (economic) = SH/SE SE(d)= (2ME/r) 1/2 Where, SE (d) is standard error of the difference Me = error mean square r = the number of replications and SH= standard heterosis The computed t value tested against the t-value at the degree of freedom of error.

RESULTS AND DISCUSSION Analysis of Variance (ANOVA)
Analysis of variance was conducted for all the traits and it showed that mean squares due to genotypes were significant (P<0.01 or P<0.05) for all traits ( Table 2). The significant mean squares due to genotypes for all the traits studied indicated the existence of variability among the genotypes, which could be exploited for the future improvement of the traits. The current finding is in line with the findings of (Shushay, 2014).

Standard Heterosis
The estimates of standard heterosis over the standard checks (BH546 and BH547) were computed for grain yield and yield related traits that showed significant differences among genotypes. Among the 64 crosses, 12 crosses exhibited positive and significant or highly significant heterosis over BH546 for grain yield, whereas only 5 crosses exhibited positive and significant or highly significant heterosis over BH547 for the same trait ( (32.24%) and L10xT1(31.11%) showed positive and significant (P<0.05) standard heterosis over BH547. Positive and significant heterosis is advantageous as it indicates increased yield over the existing standard check. For grain yield, larger number of the crosses showed positive and significant standard heterosis over BH546 than BH547, indicating that BH547 was the highest yielding standard check. In agreement with the current finding, the expression of grain yield heterosis above the standard check in maize has been reported by several investigators (Berhanu, 2009;Chandana, 2013;Shushay,2014 andMesenbet et al., 2016).
For days to anthesis, standard heterosis over BH546 ranged from -6.47% for L3xT2, L16xT2 and L24xT2 to 10.79% for L32xT1. Regarding standard heterosis over BH547 for days to anthesis, 24 crosses showed negative and significant to highly significant heterosis while only 2 crosses (L32xT1and L32xT2) showed positive and significant to highly significant heterosis for this trait. Standard heterosis over BH547 for days to anthesis ranged from -8.45% to 8.45%.
For days to silking, 4 crosses (L1xT1, L3xT2, L15xT2 and L16xT2) showed negative and significant to highly significant standard heterosis over both standard checks (BH546 and BH547), except cross L32xT1(8.39 % and 9.15%) which showed positive and significant standard heterosis over BH546 and BH547 respectively. Standard heterosis for this trait ranged from -15.38% to 9.15%. Negative and significant standard heterosis is desirable direction as it indicates earlier anthesis and silking of the crosses than the standard check and the reverse is true for the crosses with positive and significant standard heterosis for both traits.
The estimates of standard heterosis for days to maturity over BH546 ranged from -6.47% (L3 xT2) to 10.79 % (L32 x T1). Out of 64 crosses studied for this trait, 18 of these crosses showed negative and positive significant to highly significant heterosis over BH546 (Table 3). Standard heterosis for days to maturity over BH547, ranged from -8.45% for L3xT2, L16xT2, and L24xT2 to 8.45% for L32xT1. 24 crosses exhibited negative and significant to highly significant heterosis while only 2 crosses (L32xT1 and L32xT2) showed positive and highly significant (P<0.01) and significant (P<0.05) standard heterosis for the trait respectively.
The extent of standard heterosis for biomass yield over BH546 varied from 109.27% (L32xT2) to -44.21% (L29 x T2). Cross L32x T2 also showed positive and highly significant standard heterosis for biomass yield over BH547 (Table 3). The estimates of standard heterosis for biomass yield over BH547 varied from 167.26% for L32xT2 to 37.08% for L15xT1. In conformity with the current results, both positive and negative standard heterosis for biomass yield has been reported by shushay (2014).
With respect to harvest index, 32 crosses exhibited positive and significant to highly significant standard heterosis over BH546 and it ranged from 115.43% for L10x T2 to 39.35% for L23x T1. This indicated that these crosses were the best as compared to standard check in order to increase harvest index. The standard heterosis computed for this trait over BH547 showed that only cross L10xT2 (54.51%) showed positive and highly significant standard heterosis while Cross L32xT2 (-36.00%) and L32xT1 (-33.83%) showed negative and highly significant to significant standard heterosis for this trait over BH547 respectively. The standard heterosis for harvest index over BH547 varied from 54.51% for L10xT2 to -36.00% for L32xT2 (Table 3). An agreement with the current finding Berhanu, (2009) in his study on heterosis and combining ability for yield, yield related parameters and stover quality traits for feed in maize adapted to the mid-altitude agro-ecology of Ethiopia and reported both positive and negative significant heterosis for harvest index.   * and ** = significant at 0.05 and 0.01 probability level, respectively df= degrees of freedom BY = biomass yield, DA = Days to Anthesis, DS = Days to Silking, DM = Days to Maturity, GY = Grain Yield, HI = Harvest Index. The extent of standard heterosis for ear length over BH546 varied from 9.90% for (L4xT2) to -24.75% for (L32xT2). For this trait, except cross L4xT2 (9.90%) and L9xT2 (8.42%) that showed positive and significant standard heterosis, most of the crosses showed negative and significant to highly significant standard heterosis over BH546. In contrast to this, most of the crosses showed positive and significant to highly significant standard heterosis over BH547 except, cross L6xT1(-13.26%), L32xT1(-12.71%) and L32xT2 (-16.02%) that showed negative and highly significant standard heterosis over BH547 for this trait. The extent of standard heterosis for ear length over BH547 varied from 22.65% for L4xT2 to -16.02% for (L32xT2). In line with this finding, Gudeta (2007), Melkamu (2013) and Kumar et al. (2014) reported significant positive standard heterosis for ear length.
For ear diameter, standard heterosis over BH546 ranged from 4.21% for (L7xT2) to -12.63% for (L3xT1). For this trait, 4 crosses showed positive and significant to highly significant standard heterosis over BH546 whereas 5 crosses exhibited negative and significant to highly significant standard heterosis over BH546. In contrast, all crosses showed negative and highly significant standard heterosis over BH547 except, cross L7xT1, L23xT1 and L31xT1 that showed negative and non-significant standard heterosis for this trait and ranged from -25.63% for (L3xT1) to -7.71% for (L10xT1) and (L18xT1). For ear length and ear diameter, a cross with positive and significant standard heterosis was more advantageous as they increase the tendency of ear length and ear diameter than a standard check.
In case of number of ear per plant, standard heterosis over BH546 varied from 60.0% for (L3xT2) to 30.0% for nine (9) crosses including (L28xT1) and (L32xT1). For this trait, 18 crosses exhibited positive and significant to highly significant standard heterosis over BH546. Similarly, standard heterosis for this trait over BH547, nine (9) crosses showed positive and significant to highly significant standard heterosis and ranged from 45.45% for (L3xT2) to 27.27% for six (6) crosses including (L7xT1and L17xT2). This indicating that increased in number of ears per plant for these crosses as compared to the standard checks.
For number of rows per ear, only cross L19xT1 (20.00%) showed positive and significant standard heterosis over BH546 while cross L32xT2 (-20.00%) showed negative and significant standard heterosis for the same trait. Similarly, 43 crosses showed negative and significant to highly significant standard heterosis over BH547 for this trait and ranged from -29.41% for L32xT2 to -11.76% for L18xT2, L19xT2 and L20xT1 (Table 3). In line with this finding, Gudeta (2007) Berhanu (2009) and shushay (2014) reported positive and significant heterosis and negative and highly significant standard heterosis.
The estimates of standard heterosis for number of kernels per row, cross L19xT1 (63.53% and 73.32%) showed positive and highly significant standard heterosis over both (BH546 and BH547) standard checks, respectively while crosses L32xT1 (-25.88%) and L32xT2 (-35.29%) showed negative significant and highly significant standard heterosis over BH546 respectively. Negative and significant standard heterosis for number of kernels per row over BH547 was exhibited by cross L32xT2 (-31.42%).
For thousand kernel weight, nine (9) crosses exhibited positive and significant to highly significant standard heterosis over BH546 whereas 11 crosses exhibited positive and significant to highly significant standard heterosis over BH547. Cross L1xT2, L2xT2 and L31xT2 showed positive and highly significant (P<0.01) standard heterosis over both BH546 and BH547 whereas cross L3xT1 showed negative and significant (P<0.05) standard heterosis over both checks. The estimates of standard heterosis for this trait over both checks ranged from 44.81% for L2xT2 to -17.64% for L3xT1. Positive and significant standard heterosis for thousand kernel weight was a desirable character in maize while negative and significant standard heterosis was undesirable character for this trait. In line with the result of the current finding, Gudeta (2007) in his study on heterosis and combining abilities in QPM versions of early generation highland maize inbred lines, reported high levels of heterosis over both parents for 1000-kernels weight. Similar results were reported by Melkamu (2013), Kumar et al. (2014) and shushay (2014).

CONCLUSION
In this study, considerable standard heterosis for all traits over both commercial checks was manifested. The highest standard heterosis for grain yield observed from L17xT1 (61.75 %) and (41.46%) over BH546 and BH547, respectively. This indicated that, the presence of exploitable heterosis essential for this trait. Cross L3xT2 manifested negative standard heterosis over BH546 and BH547 for days to anthesis, days to silking and days to maturity that indicates earliness in maturity. Similarly, most of the crosses showed positive standard heterosis for biomass yield, harvest index, ear per plant, ear length, ear diameter, rows per ear, kernel per rows and thousand kernel weights over both commercial checks. The existence of genetic variation for grain yield and all yield related traits give further direction for maize breeders those who are interested in heterosis breeding. However, further evaluation at more locations and over years is advisable to confirm the promising results observed in present study.