This study was conducted at the Njala Agricultural Research Centre (NARC) experimental site, Njala in the Moyamba District of Southern Sierra Leone. Njala is about 47 km from the provincial headquarter town of Bo, Southern Province and about 255 km from Freetown, the capital city of Sierra Leone. Njala Campus experiences a tropical climate that is characterized by two main seasons: a rainy season and a dry season. The rainy season normally commences in May and ends in October while the dry season follows from November to April. The mean annual temperature ranges from a minimum of 28˚C to a maximum of 33˚C. The average annual rainfall is about 2500 mm, and the elevation is 128 m above sea level (altitude), on Latitude 8˚N and Longitude 12˚W.

The soil of the trial site belongs to the Njala soil series under secondary bush [39] . Njala soils are gravelly, well drained, highly weathered and acidic with a pH of 5.5 - 6.0 [40] . The clay content occurs from 20 cm depth in the soil profile. The Njala soil is classified as an Orthoxicpalehumult. However, soil test conducted by the Soil Science Department, Njala University revealed the pH to be 6.0 (acidic), low level of nitrogen (5.0 - 15 mg/l), medium level of phosphorus (25 - 50 mg) and low level of potassium (50 - 150 mg). The most dominant vegetation at Njala is grassland comprising mainly of Andropogon gayanus (Gamba grass) [41] .

A total of 3 genotypes comprising 2 improved white maize (2009 EVDT and

P = Phosphorus, N = Nitrogen, K = Potassium, DMR-ESR-Yellow (downy mildew resistance and early streak resistance), EVDT = early maturing variety drought tolerance, IWDC2 = intermediate white drought tolerance intermediate maturing.

IWDC2) and 1 local check variety (DMR-ESR-Yellow) were used. The improved white maize varieties were introductions from International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria. The experiment was a 3 × 7 factorial laid out in a split block design with three replications. The 21 plots comprised of three varieties × seven nitrogen rates (Table 1). Each plot measured 3 m × 5 m with an intra-plot spacing of 0.5 m and inter block spacing of 1.0 m. Each block/replication comprised 21 plots, making a total of 63 plots. Each plot had 40 plants and 840 plants per replication. In each trial year, standard agronomic practices including site selection, land preparation, planting, weeding and harvesting were done. The seeds were planted in August during the second cropping season of 2021 and 2022. Weeding was done before fertilizer application and this was followed by two subsequent weeding before the maturity of the crop.

WAP = weeks after planting; MAP = months after planting.

A total of 11 agro-morphological traits were collected. Data collection was based on protocols presented in the descriptor for maize variety performance evaluation trial [42] with slight modifications (Table 2).

The economic parameters were determined on grain yield. The parameters collected included partial factor productivity of nitrogen (PFP_N), increase in gross returns (GR) over control (PFP_N), agronomic efficiency of nitrogen (AE_N), grain yield value (GY_V), net return (NR), value cost ratio (VCR), marginal return (MR) (Table 3).

Data were subjected to analysis of variance (ANOVA) using the GENSTAT statistical program (GENSTAT, 15th release, Rothampstead, UK). The Least Significance Difference (LSD) was used to compare between treatment means using a significance level of α = 0.05. The residuals of data for the parameters were first checked for normality and homogeneity using the Shapiro-Wilk test and Bartlett’s test to ensure that data are normally distributed.

The variance component analysis was done using the procedure described by Patterson and Thompson [43] . Various genetic parameters such as broad sense (H²), genotypic coefficients of variation (GCV), phenotypic coefficient of variation (PCV), and expected genetic advance (GA) were determined. The GCV and PCV were determined following the formula described by Burton and Devane [44] . The GCV and PCV values were categorized using the technique proposed by Deshmukh et al. [45] as follows: values < 10% = low, values that are 10% - 20% = medium and values > 20% = high. The broad sense heritability was determined using a formula by Robinson et al. [46] . The expected genetic advance (GA) was estimated based on the equation given by Shukla et al. [47] .

The Partial Factor Productivity (the ratio of the grain yield to the applied rate of fertilizer) and Agronomic Efficiency (the ratio of the increase in grain yield over fertilizer-control plots to the applied rate of fertilizer) were determined according to the procedures described by Yadav [48] . Other economic analysis such as value cost ratio, net returns, marginal return, etc., were also estimated.

The estimated genetic parameter estimates of selected agronomic traits of maize showed that the phenotypic coefficient of variation (PCV) was higher than the genotypic coefficient of variation (GCV) values for all the traits studied (Table 4). Traits with high PCV were plant height, anthesis silking time, grain yield, number of kernel per cob, number of kernel rows, ear length and ear height; while those with high GCV were anthesis silking time and grain yield. The broad sense heritability estimates ranged between 9.65% (stem girth) and 58.84% (anthesis silking time). Traits with intermediate broad sense heritability (30% - 60%) were anthesis silking time and cob diameter, whereas the remaining traits had low broad sense heritability values > 30.0% (Table 4). High GA (>20.0%) was exhibited for anthesis silking time, whilst stem girth, number of kernel per row, ear diameter and ear height had low (<10.0%) GA (Table 4). Anthesis silking time combined intermediate broad sense heritability and high GA, whilst cob diameter combined intermediate broad sense heritability and intermediate GA, and the remaining traits combined low broad sense heritability and intermediate GA, except for stem girth, number of kernels per row, ear diameter and ear height which combined low broad sense heritability and low GA. This study revealed the presence of useful variation in the maize varieties that could be exploited through direct selection or population improvement scheme. The higher PCV values relative to the GCV indicate that the traits were sensitive to environmental modifications or effects. These results are consistent with the findings

of Sravani et al. [49] who found slightly higher PCV values for anthesis-silking advanced population of maize. The variations in some of the traits could be due to the different maize varieties and/or environmental factors such as different plant nutrients (nitrogen levels).

The knowledge of heritability guides the plant breeder about the selection procedure to be utilized under a given situation. Broad sense heritability is an estimate of the total contribution of the genetic variance to the total phenotypic variance of trait. It measures the relative amount of heritable portion of total variability and provides information on the extent to which a particular morphogenetic trait can be transmitted to successive generation. Traits with high broad sense heritability imply that these characters were less influenced by the environment. Thus, selection would be effective for the genetic improvement of these traits. These findings concur with the views of Kharel et al. [50] , Prakash et al. [51] and Supraja et al. [52] that characters that are less influenced by the environment are more heritable.

The Genetic advance is a more reliable index for understanding the effectiveness of selection in improving the traits because the derivation of this estimate involves heritability, phenotypic standard deviation and selection intensity. Thus, heritability coupled genetic advance provide clear picture about the effectiveness of selection for improving the plant characters [53] . Traits that exhibited high genetic advance could be considered as favorable attributes for genetic improvement through selection. These results are in conformity with those obtained by Bhadru et al. [54] , Lal et al. [55] and Wedwessen and Zeleke [56] for most of the yield attributing traits. The traits that had both high heritability and high genetic advance indicate that they were under the control of additive genes. Hence, direct selection would be an effective technique for further improvement of these traits. Traits with high heritability and low genetic advance are under the control of non-additive genes. These findings agree with the view that both heritability and genetic advance should be considered for efficient predictability of response to selection [7] .

The presence of genetic variation is a key prerequisite for genetic improvement in plant breeding and plays a pivotal role in germplasm usage in breeding programs. Thus, plant breeders and researchers benefit immensely from an understanding of sources of useful existing genetic variation in maize and ways of creating genetic variability, where it is limited, that can benefit farmers and other end users of the crop. Governments and other relevant research and development stakeholders support research and development efforts such as genetic improvement targeted at enhancing crop yields, quality and productivity parameters, as well as improving livelihoods, increasing income and reducing poverty.

The effects of variety and nitrogen rates on maize grain yields and economic parameter estimates are presented in Table 5. Generally, both the variety and nitrogen fertilization significantly affected (p < 0.05) maize grain yields, PFP_N,

GY = grain yield, PFP_N = partial factor productivity of nitrogen, GY_IOC = increase in gross returns (GR) over control, AE_N = agronomic efficiency of nitrogen, GY_V = grain yield value, NR = net return, VCR = value cost ratio, MR = marginal return, b1 = 30 kg∙N∙ha⁻¹, b2 = 60 kg∙N∙ha⁻¹, b3 = 90 kg∙N∙ha⁻¹, b4 = 120 kg∙N∙ha⁻¹, b5 = 150 kg∙N∙ha⁻¹, and b6 = 180 kg∙N∙ha⁻¹.

GR_IOC, AE_N, GY_V, NR, VCR and MR, whereas nitrogen rates significantly affected (p < 0.05) the GY_IOC. The mean maize grain yields and economic parameter estimates increased with increasing nitrogen rates, with plots amended with 30 N∙kg∙ha⁻¹ and 180 N∙kg∙ha⁻¹ exhibiting the lowest and highest maize grain yields of 1238 kg∙ha⁻¹ and 2098 kg∙ha⁻¹, respectively. Variety IWCD2 exhibited the highest grain yield of 2003 kg∙ha⁻¹ and economic parameter estimates, whereas the check variety DMR-ESR-Yellow exhibited the lowest values (Table 5).

The variety and nitrogen rates effects on economic parameter estimates (PFP_N, AE_N, NR, VCR and MR indicated that the mean values of these estimates were significantly affected by variety and nitrogen fertilization (Table 5). Maximum values of PFP_N (41.3 kg grain kg⁻¹∙N), and AE_N (29.4 kg grain kg⁻¹∙N), were obtained from plots of maize amended with 30 kg N∙ha⁻¹, whereas the maximum values of NR (SLL 30,000,000 ha⁻¹), VCR (2.83) and MR (SLL 1.83 SLL⁻¹ spent on N) were obtained in plots amended with 180 kg N∙ha^–1. The minimum mean values of PFP_N (13.8 kg grain kg⁻¹∙N), AE_N (14.9 kg∙grain∙kg⁻¹∙N) were found in plots amended with 120 kg∙N∙ha⁻¹ similar to plots amended with 150 kg∙N∙ha⁻¹ and 180 kg∙N∙ha⁻¹; whereas minimum values of NR (SLL 1,000,000 ha⁻¹), VCR (1.79) and MR (SLL 0.79 SLL⁻¹ spent on N) were found at 30 kg∙N∙ha⁻¹.

Findings on the grain yield and economic parameter estimates indicate that efficient utilization of fertilization for maize production is important for increasing maize grain yields, and maximizing economic returns. The higher grain yields in the plots amended with higher nitrogen rates relative to those with lower rates could be attributed partly to the higher nutrient availability that supported both vegetative growth and reproductive organ development of the crop. These findings are consistent with the view that improvement in leaf area index (LAI), crop growth rate (CGR) and light interception with higher N rate of maize produce remarkable increase in yield and yield components of maize as well as maximum benefits in terms of PFP_N, AE_N, NR, VCR and MR [57] .

The nitrogen rates application in plots of three varieties of maize produced sharp variations with IWDC2 exhibiting the highest grain yield and economic parameter estimates whereas the check variety DMR-ESR-Yellow had the lowest values. These variations are possibly attributed to the inherent genetic variability of the varieties and difference in the amount of nitrogen and N availability, which necessitate site specific recommendation for improved and profitable nutrient management. These findings concur with Ahmad and Mahdi [58] who opined the relevance of site-specific recommendations for crops due to differential response to various nutrient inputs across growing environments. Sanchez [59] also suggested that a good understanding of concepts of ideal soil fertility level and response to nutrient management provides practical guidelines for improving nutrient management under the increasing climate variability conditions of smallholder farmers. Generally, some of the ways of improving soil fertility that could be exploited include incorporation of cover crops that add organic matter to the soil, crop rotation, tillage reduction, soil examination, intercropping legumes with cereals, utilization of inorganic and organic (crop residues, compost, farm yard manure, green manure, etc.) fertilizers. The compost, manure and other organic materials stimulate soil microbe activity, improve soil structure and supply necessary nutrients.

Findings of this study demonstrate that the application of a unit urea was economical, since the value of the increase in the grain yields due to the quantity of fertilizer added is greater than the cost of fertilizer used. Findings agree with the view that if a unit of urea applied does not increase the yield enough to pay for its cost, its application is uneconomical and unprofitable even after a constant increase in the yield [60] . Results also support the view that the efficient use of N for maize production is important for increasing grain yield, maximizing economic returns, and minimizing NO₃ leaching to ground water [61] [62] . Treatments with a VCR of 2 represents 100% return on the money invested on N-fertilizer. For smallholder farmers with low technology and limited capital, a fertilizer rate exhibiting a VCR greater than 2 is recommended [63] .

The interactive effects of variety and nitrogen rates on maize grain yields and economic parameter estimates are presented in Table 6. Interactive effects of variety into N rates of application (a × b) in Table 6 showed that when N was applied at the low rate of 30 kg∙N∙ha⁻¹, each PFP_N, AE_N, NR, VCR and MR showed positive relationship with increase in amounts of N application, except variety DMR-ESR-Yellow which exhibited negative NR (−500,000) and MR (−0.03). At the recommended (120 kg∙N∙ha⁻¹) and highest rates (180 kg∙N∙ha⁻¹) of N, maize exhibited better performance in terms of NR, VCR and MR for all the three varieties studied. Varieties 2009 EVDT, IWCD2 and DMR-ESR-Yellow exhibited the highest net return at 150 kg∙N∙ha⁻¹, 180 kg∙N∙ha⁻¹ and 180 kg∙N∙ha⁻¹, respectively. Applications at 180 kg∙N∙ha⁻¹ and recommended nitrogen fertilization (120 kg∙N∙ha⁻¹) in varieties IWCD2 and DMR-ESR-Yellow had similar net returns. These findings are in concurrence with Shah et al. [57] , who opined that improvement in leaf area index (LAI), crop growth rate (CGR) and light interception with higher N rate of maize resulted in remarkable increase in yield and yield components of maize consequently producing maximum benefits in terms of the various economic parameters studied. These findings are consistent with those drawn by Mariga et al. [61] and Gehl et al. [62] , who suggested that efficient use of N for maize production is important for increasing grain yield, maximizing economic returns, and minimizing NO₃ leaching to ground water.

GY = grain yield, PFPN = partial factor productivity of nitrogen, GY_IOC = increase in gross returns (GR) over control, AE_N = agronomic efficiency of nitrogen, GY_V = grain yield value, NR = net return, VCR = value cost ratio, MR = marginal return, a0 = 2009 EVDT; a1 = IWCD2; a2 = DMR-ESR-Yellow (check variety); b1 = 30 kg∙N∙ha⁻¹, b2 = 60 kg∙N∙ha^–1, b3 = 90 kg∙N∙ha^–1, b4 = 120 kg∙N∙ha^–1, b5 = 150 kg∙N∙ha^–1, and b6 = 180 kg∙N∙ha^–1.