Field experiments were conducted during 2016/17 cropping at Kindo Koyisha (Altitude 1170 masl, annual rainfall 924 mm, 2100, significant crops cultivated in the study area include maize, sorghum, and sweet potato) and Humbo (Altitude 1800, annual rainfall 1295 with bimodal rainfall patterns, average temperature of 20°C).

Treatments used in this study were eight maize varieties (BH546, BH547, Gibe II, MH130, Melkasa IV, MH140, Melkasa II, and Melkasa 6Q) and three local cultivars (Local red, Local mixed, and Local white) of the total of eleven maize genotypes were evaluated at two moisture-prone areas in southern Ethiopia. The Treatments were laid out in a Randomized Complete Block Design (RCBD) with three replications. The plot size was 4 × 4 m with 1.5 m between replications and 1.0 m between plots. Planting was carried out as per the planting time of the respective area following the onset of rainfall. Maize was hand planted by placing two seeds per hill and thinned after emergence to maintain the proposed plant density per plot. Weed control was carried out by hand or hand hoeing, while diseases and insect damage were visually monitored during crop growing season. Phosphorus fertilizer in the form of DAP and N in the form of urea was applied as per recommendation for maize production. Moreover, other crop management practices carried out as desired.

Data recorded on yield components included ear length, ear diameter, number of seeds per row, kernels per ear, thousand seed weight, and prolificacy (ears per plant). Ear length and diameter were measured for five randomly selected plants from the base to the tip and approximately the middle of the ear at harvesting. The number of seeds per row was counted for five randomly selected plants. Seed number per ear was determined by multiplying the number of rows by the number of seeds per row. Thousand Seed Weight (TSW) was measured by counting a thousand seeds with a seed counter and weighing it with sensitive balance. Prolificacy is the property of producing more ears per plant and is estimated by dividing the number of ears by the number of plants per plot. Grain was manually harvested from the net plot and converted to kg/ha after adjusting the moisture content to 12.5%. Biomass yield was estimated as the sum of stover weighed and grain yield. Harvest Index (HI) is the ratio of grain yield to the total biomass yield, which was estimated by dividing grain yield by total biomass. Data were subjected to analysis of variance using the general linear model SAS version 9.1 (SAS Inst., 2003). Treatments means were compared using the least significant difference (LSD) at a 5% probability level.

The data for plant and ear heights as affected by location and varieties are depicted in Table 1.

Analysis o variance indicated that location had a significant effect on plant and ear height. Both parameters were higher at Kindo Koyisha as compared to Humbo. Similarly, maize varieties have significantly differed for plant height and ear height. The tallest plant height (235 cm) was recorded for variety BH547, followed by variety BH547 with a mean plant height of 229 cm. The shortest plant height (156 cm) was seen for variety Melkasa 6Q. In line with this, the tallest ear heights (117 cm) were observed for BH547, followed by a variety of Local mixed with a mean ear height of 58 cm.

On the other hand, location by variety interactions resulted in significant differences in on-ear heights (Table 1). Generally, all varieties exhibited taller ear heights at Kindo Koyisha as compared to Humbo. The tallest ear height (148 cm) was observed for variety Local white at Humbo, followed by variety BH547 at Kindo Koyisha with a mean ear height of 147 cm. The shortest ear height (34 cm) was seen for variety Melkasa 6Q at Humbo. In contrast, location by varieties interaction did not have a significant effect on plant heights.

The data for ear length and ear diameter as affected by location and varieties are presented in Table 2. Analysis of variance showed that the main effect of location and varieties had significant differences in length. The ear length of maize varieties was higher at Kindo Koyisha than that of Humbo. Averaged over locations, the longest cob length (15.13 cm) was obtained from variety BH546, followed by variety BH547 with a mean ear length of 13.74 cm. The shortest ear diameter (11.62 cm) was seen for variety Melkasa 6Q. In line with this, the interaction of location by varieties resulted in significant differences in ear length (Table 2). The most extended ear length (15.83 cm) was recorded for variety BH546 at Kindo Koyisha, followed by variety BH547 with a mean cob length of 14.45 cm at the exact location. The shortest ear length (9.69 cm) was seen for local white at Humbo.

On the other hand, the only main effect of varieties exhibited significant differences in ear length. The longest cob diameter (4.75 cm) was measured for variety BH546, followed by variety BH547 with a mean ear diameter of 4.34 cm. The shortest ear diameter (4.03 cm) was observed for local white. However, the main effect of location and its interactions with varieties did not significantly affect ear diameter.

The number of rows per ear seeds per row, seeds per ear, and TSW as affected by location and varieties are shown in Table 3.

Analysis of variance indicated that the main effect of location had a significant effect on the number of seeds per row and seeds per ear. Both parameters were higher at Kindo Koyisha as compared to that of Humbo. Similarly, varieties exhibited significant differences in rows per ear, seeds per row, seeds per ear, and TSW (Table 3). Variety BH547 produced the highest number per ear (15.1), followed by variety Melkasa 6Q with a mean number of rows per ear of 14.8. The least number of rows per cob (12.4) was seen for Local white. In line with this, the most significant number of seeds per row (34) and seeds per cob (469) were recorded for variety BH546, followed by variety BH547 with the mean number of seeds per row and seeds per ear of 30 and 445, respectively. Local white yielded the lowest number of seeds per row (26) and seeds per ear (295).

Moreover, location by variety interactions resulted in a significant number of seeds per row. The greatest number of seeds per row (36) was recorded that Kindo Koyisha for variety BH546, followed by the exact location for variety BH547 with the mean number of seeds per row of 34. The lowest number of seeds per row (22) was seen for Local white. In contrast, the main effect of location, variety, and interactions did not significantly affect TSW, seeds per row, and rows per ear (Table 3).

The data for biomass, grain yield, and HI as affected by location and variety are depicted in Table 4. The location did not have a significant effect on the biomass yield of maize varieties. However, varieties exhibited significant differences in biomass yield. Biomass yield for maize varieties ranged from 7083 to 14792 kg/ha, with the highest biomass yield recorded (14792 kg/ha) for variety BH546, followed by variety BH547 biomass yield of 14688 kg/ha. The lowest biomass yield (7083 kg/ha) was obtained from variety Melkasa 6Q. In line with this, location by varieties interactions resulted in significant differences in biomass yield. The most excellent biomass yield (17188 kg/ha) was recorded at Kindo Koyisha for variety BH547, followed by variety BH546 at the exact location with a mean biomass yield of 15938 kg/ha. The lowest biomass yield (6979 kg/ha) was seen for variety Melkasa 6Q at Kindo Koyisha.

Grain yield has significantly differed in response to the location where higher grain yield was obtained from Kindo Koyisha than Humbo (Table 3). Similarly, maize varieties exhibited significant differences in grain yield. The highest grain yield (5208 kg/ha) was recorded at Kindo Koyisha for variety BH546, followed by MH140 with a mean grain yield of 5000 kg/ha at the exact location. The lowest grain yield (2396 kg/ha) was achieved from Local red at Humbo. In general, maize varieties tested for moisture responded differently to respective environments. At Kindo Koyisha, varieties BH546, MH140, and BH547 showed good performance in a such moisture-stress-prone environments. On the other hand, MH140 and MH130 relatively exhibited superiority over others at Humbo.

Maize varieties exhibited differently for agronomic traits measured in response to the location regarding their genetic variability (Table 1, 2, 3, and 4). Generally, almost all maize varieties showed superior performance at Kindo Koyisha compared to Humbo for agronomic traits. The grain yield differences recorded were 730 kg/ha between Kindo Koyisha and Humbo. Thus, relatively the performances of varieties were poor at Humbo, which probably suggests that Kindo Koyisha was a relatively better environment with plant growth conditions. Moreover, this illustrated that subjecting plants to favourable growing conditions increased the ability of varieties to capture resources, which was reflected as evident in their increased agronomic performance. The significant effects of environments indicated that the genotypes performed differently across locations. Thus, the mean yield of genotypes differed from location to location.

Similarly, maize varieties, averaged over locations, showed significant differences in plant height, ear height, rows per cob, seeds per row, seeds per cob, ear length, and ear diameter (Table 1, 2 3). Relatively higher plant height (≥ 200 cm) was recorded for varieties BH546, BH547, and MH140, whereas ear heights (≥ 100 cm) were recorded for varieties BH547, Local mixed, and Local white. Variety BH546 gave the most extended cob length, while BH547 produced the highest cob diameter. Variety BH547 gave the highest number of rows per cob, while variety BH547 produced the most significant number of seeds per row and seeds per cob. Maize varieties averaged over locations tended to express a wide range of their genetic variability for grain yield. Grain yield variations ranged from 2396 to 4063 kg/ha. Variety BH 546 out yielded which MH140 followed. Local red was the least concerning grain yield performance. The significant difference among the genotypes showed variations in their response (yield potential) to different locations.

Location by variety interactions resulted in significant differences on-ear height, cob length, seeds per row, biomass, and grain yield (Table 1, 2 & 3). For parameters as mentioned above, varieties had relatively superiority at Kindo Koyisha as compared to Humbo. In general, the performance of varieties was poor at Humbo, with the grain yield variability ranged from 2396 to 3021 kg/ha. At Humbo varieties, MH140, MH130, BH546, and Melkasa IV gave relatively higher grain yield with HI (Physiological efficiency and ability to convert total dry matter into economic yield) values 0.30, 0.40, 0.28, and 0.36, respectively. This variability might be attributed to varietal differences in maize genotypes in response to the prevailing environmental conditions. Hence, the Humbo location could be considered a stressful environment with profound limitations in the potential performance of maize varieties. At Kindo Koyisha, maize varieties expressed relatively better performance concerning grain yield. Grain yield variability ranged from 2604 to 5208 kg/ha from lowest to the highest. At this location, varieties with superior performance with sounding grain yield were BH546, MH140, BH547, and MH130. This probably indicates that genotypes describe the complete set of genes inherited by an individual important for expressing a trait under consideration in a particular environment.

In general, maize varieties at Kindo Koyisha performed best to their potential as compared to Humbo. Maize varieties BH546, MH130, and MH140 showed relative stability across the location with superiority of grain yield. The research results indicated that genotype by environment (G × E) interactions is a differential genotypic expression across environments which affects the genotypes rankings within each environment and hence relevant for identifying mega environments and targeting genotypes. Moreover, G × E’s significance indicates fluctuation of genotypes performance across environments or testing sites with inconsistent performance. Similar results were recorded by Akcura et al. (2005), Acura and Kaya (2008), Asfaw (2008), Dagne (2008), Solomon et al. (2008), Abdurhaman (2009), and Muluken (2009). The relationship between selected agronomic traits with the grain is depicted in Table 5. The correlation coefficient (r) values of selected agronomic traits with grain yield ranged from−0.05 to 0.82. Plant and ear height were positively significantly (P ≤ 0.05) correlated, suggesting that the traits are closely associated with grain yield.

Similarly, the number of seeds per row, seeds per cob, ear length, ear diameter, biomass, and TSW were positively associated with yield. In contrast, the number of rows per cob with grain yield correlation was not significant. The correlation of almost all agronomic traits with grain yield was relatively strong, indicating that their contribution towards grain yield was considerable.