Plant Breeding, 2025; 0:1–14 https://doi.org/10.1111/pbr.13291 1 of 14 Plant Breeding ORIGINAL ARTICLE OPEN ACCESS Screening for Fall Armyworm (Spodoptera frugiperda J. E. Smith) Resistance in Early-Maturing Tropical Maize Adapted to Sub-Saharan Africa Adamu Masari Abubakar1,2  | Idris Ishola Adejumobi1   | Muhyideen Oyekunle2  | Tegawende Odette Bonkoungou1  | Usman Muhammad2  | Onehireba Udah3  | Baffour Badu-Apraku1 1International Institute of Tropical Agriculture (IITA), Oyo State, Nigeria  |  2Department of Plant Science, Institute for Agricultural Research, Ahmadu Bello University, Zaria, Kaduna State, Nigeria  |  3Department of Crop Production and Protection, Faculty of Agriculture, Obafemi Awolowo University, Ile–Ife, Osun State, Nigeria Correspondence: Idris Ishola Adejumobi (i.adejumobi@cgiar.org) Received: 19 December 2024  |  Revised: 20 May 2025  |  Accepted: 25 May 2025 Funding: This work was supported by Bill and Melinda Gates Foundation, OPP1134248. Keywords: fall armyworm tolerance | ideotype | inbred lines | multiple-traits | selection index | source populations ABSTRACT Fall armyworm (FAW) (Spodoptera frugiperda J. E. Smith) has emerged as a serious pest since 2016 in Africa, affecting the food security and livelihoods of millions of smallholder farmers, especially those growing maize. Native genetic resistance to FAW is essential for reducing yield loss. The objectives of the study were to assess genetic variability, identify promising genotypes for FAW-resistance, and propose a new easy-to-use multi-traits selection index for ideotype selection under FAW infestation. Three hundred and ten genetic materials comprising inbred lines, hybrids, OPVs and landraces were evaluated under artificial fall armyworm infestation at Ile-Ife, in 2021 and 2022. All the trials were laid out in alpha lattice design with two replications. Data were collected on FAW foliar and ear damage and grain yield estimated. The Lme4 R package was used to perform an analysis of variance (ANOVA) using a mixed linear model with environment and replicate as random effect while genotype was kept as fixed effect. Three multi-traits selection strategies—desired index (DI), multiple traits selection index (MI) and multi-traits- genotype ideotype distance index (MGIDI) were used to identify FAW-resistant genotypes. Significant (p < 0.01) mean squares were detected among the genotypes (G), environments (E) and G × E interaction for grain yield and most measured traits. For each trial, FAW foliar damage scores at 6, 8 and 10 WAP varied. In the hybrid trial, the scores ranged from 3.8 to 6.6, 1.6 to 5.5 and 2.7 to 5.8, respectively. The score ranged from 4.2 to 7.2, 2.5 to 7.3 and 1.7 to 8.1, respectively among the inbred lines. In the landrace trial, the scores ranged from 3.5 to 6.5, 4.5 to 6.3 and 3.9 to 6.4, respectively. In the OPV trial, the scores at 6, 8 and 10 WAP were from 3.7 to 6.4, 2.4 to 5 and 3.8 to 5.5, respectively. At 15% selection intensity, the three selection indices identified 15, 14, 10 and 8 hybrids, inbreds, landraces and OPVs, respectively as ideotypes under FAW infestation. There was higher agreement and consistency between MGIDI and MI, often identifying more genotypes in common compared to DI. Significant negative correlations were detected between grain yield and FAW-induced foliar damage scores at 6, 8 and 10 WAP, and ear damage. Substantial genetic variability existed in the germplasms panel for FAW resistance improvement. The ideotypes identified in the diverse genetic backgrounds could facilitate the development of productive hybrids and serve as source populations for improved tolerance/resistance to FAW. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2025 The Author(s). Plant Breeding published by Wiley-VCH GmbH. https://doi.org/10.1111/pbr.13291 https://doi.org/10.1111/pbr.13291 mailto: https://orcid.org/0000-0002-6533-0612 mailto:i.adejumobi@cgiar.org http://creativecommons.org/licenses/by/4.0/ http://crossmark.crossref.org/dialog/?doi=10.1111%2Fpbr.13291&domain=pdf&date_stamp=2025-06-04 2 of 14 Plant Breeding, 2025 1   |   Introduction Maize is one of the world's most important food and feed crops and plays an important role in ensuring food security in sub- Saharan Africa (SSA). Despite the importance of maize, its yield potential is rarely realized due to several challenges, including drought, low soil fertility and parasitism by Striga hermonthica (Job et al. 2022). In addition to these recurrent constraints, fall armyworm (FAW; Spodoptera frugiperda (J.E. Smith)) invasion has become a threat to maize cultivation. During the past three decades, maize breeders at the International Institute of Tropical Agriculture (IITA) have utilized several breeding strategies to develop early (90–95 days) and extra-early (80–85 days) matur- ing populations, hybrids, and inbred lines with improved lev- els of tolerance/resistance to multiple stresses (Striga, drought and low N). The populations were developed to improve food security in SSA (Badu-Apraku and Fakorede 2017). The inva- sion, rapid spread, and large-scale damage of maize by FAW is likely to worsen food insecurity in Africa (Nagoshi et al. 2018; Prasanna et al. 2018). Fall armyworm emerged as a serious pest in 2016 in Africa, af- fecting the food security and livelihoods of millions of small- holder farmers, especially those growing maize (Nagoshi et al. 2018; Wightman 2018). The pest causes widespread dam- age across a wide range of host crops with, over 350 plant spe- cies including rice (Oryza sativa), wheat (Triticum aestivum) and sorghum (Sorghum bicolor) (Kumar et  al.  2022; Senay et  al.  2022). Several researchers have reported that maize is the most preferred host of FAW and the majority of culti- vars under production across SSA are susceptible to the pest (Baudron et al. 2019; Kasoma et al. 2020; Kumar et al. 2022). According to Day et  al.  (2017), the top 12 maize producing countries in Africa recorded an estimated 8.3 to 20.6 mil- lion tons of total grain production per year due to FAW dam- age, which accounts for 21% to 53% yield losses of the total production. The potential yield loss of maize due to FAW is projected to be between USD 2481 million and USD 6187 million, indicating a significant economic impact (Day et  al.  2017). In Nigeria, the average yield loss of 2.836 million tons of maize occurs each year, estimated at USD 877.6 million. Over the years, a wide range of technologies and management practices has been developed and are available for the control of FAW. This includes cultural practices, biological control, applications of bio-pesticides, synthetic pesticides and agro-ecological man- agement. However, these technologies may either not be ef- fective or environmentally not friendly. No single solution can provide sustainable management of a complex pest such as FAW on farmers' fields (Prasanna et al. 2022). Application of pesticides is the commonest method of controlling FAW in SSA. However, the method is unsustainable because most chemical pesticides are highly toxic to the environment and the FAW insect are capable of developing resistance to most of the available pesticides, thus rendering them inef- fective (Schlum et  al.  2021). Farmers often use these chem- icals without preventive measures (e.g., Personal Protective Equipment), thus harming humans (Tambo et  al.  2020), the environment (Togola et  al.  2018) and non-target organisms (Farrar et al. 2018). Sustainable control of FAW requires the implementation of in- tegrated pest management (IPM) strategies, in which host plant resistance is one of the key components (Prasanna et al. 2022). Host plant resistance could be achieved by exploring native re- sistance of maize germplasm or introduction of transgenic re- sistance traits into elite maize cultivars (Prasanna et al. 2022). However, different concerns have emerged about the safety, ac- cessibility, affordability and feasibility of growing transgenic Bt (Bacillus thuringiensis) maize that can exhibit FAW-inhibiting traits (Njuguna et al. 2021). Due to a lack of a biotechnology reg- ulatory framework, political obstacles and consumer preference, few transgenic maize varieties have been licensed for commer- cialization in SSA countries including Nigeria (ISAAA 2018). Therefore, there is a need for continuous screening and evalua- tion of diverse maize germplasm for FAW resistance across dif- ferent agro-ecological conditions with different FAW biotypes (Nuambote-Yobila et al. 2023). Early maturing maize varieties can contribute significantly to food security, as farmers in regions with short growing seasons or erratic rainfall patterns (Bonkoungou et al. 2024; Nuambote-Yobila et  al.  2023; Oyetunde et  al.  2020) prefer them. The early-maturing varieties are ready for harvest early in the season when other traditional crops such as sorghum and millet are not ready, and are thus used to fill the hunger gap in July in the savanna zone when all food reserves are depleted after the long dry period (Badu-Apraku et al. 2013). In addition, there is a high demand for the early varieties in the forest zone for peri-urban maize consumers. They provide farmers the opportunity to market the early crop as green maize at a premium price in addition to being compatible with cassava for intercropping. Therefore, there is an urgent need to screen and select early-maturing maize varieties that are adapted to SSA and are tolerant to FAW infestation to mini- mize yield losses, reduce the economic burden on smallholder farmers and enhance food security. Ideotype breeding, a holistic approach to crop improvement, aims to develop genotypes with a combination of traits that optimize performance across diverse environments. Ideotype breeding leverages modern breeding technologies, such as ge- nomic selection, marker-assisted selection and phenomics, to identify and select genotypes with the desired combination of traits more accurately and efficiently. This not only acceler- ates the breeding cycle but also improves the predictability of performance across diverse environments. Overall, it is a vital strategy for breeding programs aimed at developing resilient, adaptable, and high-value varieties for current and future agri- cultural challenges and sustainability. Several approaches have been employed for ideotype breeding. Of these approaches, breeders popularly employ four. These are multiple-trait se- lection index—MI (Badu-Apraku et al. 2016), desired selection index—DI (Ceron-Rojas and Crossa  2018), multi-trait index based on factor analysis and ideotype design—FAI-BLUP (Rocha, Machado, and Carneiro 2017) and multi-trait genotype- ideotype distance index—MGIDI (Olivoto and Nardino 2021). A major drawback of MGIDI, FAI-BLUP index and DI is the requirement for proficiency in R programming or bioinformat- ics workflows (bioflow), which may not be readily available in all breeding programs, particularly in resource-limited breed- ing programs. This reliance on technical expertise increases 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 3 of 14 the training demands and operational complexity of using these indices. Additionally, the computational intensity of MGIDI, FAI-BLUP and DI may require strong infrastructure and software systems, potentially limiting their accessibility to smaller or less technologically advanced breeding programs. These limitations highlight the trade-offs between the preci- sion offered by these indices and the resources required for their implementation. Furthermore, DI methods necessitate the evaluation of genotypes in more than two environments, which increases the cost and logistical challenges of field trials, including data collection, management and analysis. This re- quirement can be particularly challenging in regions with lim- ited access to diverse trial locations and financial constraints. In the present study, we proposed a modified MI incorporating FAW adaptive traits that is precise, straightforward, and simple for identifying promising genotypes under FAW infestation. Unlike MGIDI, FAI-BLUP and DI, this approach minimizes reliance on computational tools and complex workflows, mak- ing it more accessible to a broader range of breeding programs. The modified MI not only streamlines the selection process but also directly integrates key adaptive traits relevant to FAW re- sistance, ensuring relevance to current pest challenges. Its sim- plicity and reduced dependency on technical resources make it a more practical and efficient tool for breeding programs, es- pecially those prioritizing rapid deployment of FAW-resistant genotypes. Therefore, the objectives of the study were to assess genetic variability, select early-maturing maize genotype with promising levels of resistance to FAW infestation and compare the efficiency of proposed MI to MGIDI and DI. 2   |   Materials and Methods 2.1   |   Genetic Materials and Description of Study Location A total of 310 maize germplasm comprising 100 hybrids, 90 in- bred lines, 56 open-pollinated varieties (OPVs) from the Maize Improvement Programme of IITA and 64 landraces obtained from the Genetic Resource Center of IITA were used in the pres- ent study. The genetic materials were selected based on their out- standing performance in previous evaluations conducted under multiple stresses (i.e., Striga, drought and low N) and non-stress conditions. The genetic materials were screened under artificial FAW infestation at the Teaching and Research Farm (70° 28′ N, 50° 4′ E and 244 m asl) of the Obafemi Awolowo University (OAU), Ile–Ife, Osun State, Nigeria during the 2021 and 2022 rainy seasons. The colonies of FAW were cultured at the IITA Entomology Unit (7° 22′ 36.2496″ N, 3° 56′ 23.2296″ E and 230 m asl), Ibadan, Nigeria. 2.2   |   Field Layout and Experimental Design of the Trial All trials were conducted in 4 m single-row plots with two rep- lications per genotype. The intra- and inter-row spacing was 0.40 m and 0.75 m, respectively. The hybrids, inbreds, OPVs, and landraces were evaluated in 10 × 10, 9 × 10, 7 × 8 and 8 × 8 alpha lattice designs, respectively. Three seeds were planted per hill and seedlings were thinned to two plants per hill at two weeks after planting (WAP) to achieve a population density of 66,667 plants per hectare. A compound fertilizer NPK (15:15:15) was applied at the rate of 60 kg/ha at two WAP. Urea (46:0:0) was applied at 60 kg/ha at six WAP. Weeds were controlled through- out the experiment using a combination of Atrazine, Paraquat at 5 L/ha, and manual hand-weeding when required. 2.3   |   Establishment of the Colonies of FAW for Artificial Infestation in the Field In the culturing of the FAW at the IITA Entomology, atten- tion was given to the average temperature and relative humid- ity where the colony was raised following standard protocols (Sharanabasappa et  al.  2018). Also, their diets were prepared under hygienic conditions according to Kasoma et  al.  (2022). Following the methodology of Kasoma et  al.  (2022), larvae of mixed instars were collected from unsprayed maize fields using perforated plastic containers. In the insectary, the larvae were removed from the containers and transferred into rearing con- tainers containing chopped castor leaves. Diet replacements were carried out regularly to ensure a fresh supply of food for the growing larvae through the successive instars. Between each successive diet change, rearing containers were cleaned with a 5% hypochlorite solution to prevent microbial growth. To reduce pupal mortality during pupation, the temperature and humidity in the rearing room were adjusted to 26°C and 70 ± 5% relative humidity (RH) using an internal heating system and dehumid- ifier, respectively. In order to enhance mating and subsequent oviposition, both male and female FAW moths that emerged from the pupa were transferred into adult-rearing cages laden with paper towels. The moths were fed on a 5% sugar solution by soaking cotton wool balls in the sugar solution and placing these inside the cages on Petri dish covers. As soon as egg laying commenced, the fresh eggs were carefully collected off the sur- face of the cages using a clean spatula and placed in new rear- ing containers with tender maize leaves, for hatching. Neonates hatching from the eggs were sorted continually and reared until they were used for inoculation of seedlings. 2.4   |   Artificial Infestation of the Maize in the Field Maize plants were artificially infested in the field using the sec- ond instar larvae of the FAW. The first infestation was done two WAP following thinning and fertilizer application. Four larvae were inoculated per plant stand. To prevent escape and ensure that the infestation is severe, six WAP, plants were re-infested at a rate of two juveniles per plant. Larvae were directly trans- ferred to the whorls of young maize plants using a paintbrush to initiate infestation and to ensure a controlled and consistent infestation of all plants. This procedure was a modification of Kasoma et al. (2022). 2.5   |   Data Collection Data were collected on 11 key traits, including grain yield (GY), FAW-induced foliar damage scores at 6, 8 and 10 WAP (FAW6, FAW8 and FAW10), ear damage at harvest (ED), days to 50% anthesis (DA), days to 50% silking (DS), plant height (PHT), ear 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4 of 14 Plant Breeding, 2025 height (EHT), anthesis-silking interval (ASI) and ears per plant (EPP) (Table 1). The plants were inspected on a plot basis for symptoms of FAW- induced foliar damage and severity, such as leaf-feeding damage, the presence of larvae and larval frass. The severity of leaf dam- age was rated using the scale of Davis and Williams (1992) and Prasanna et al. (2018). Damage was assessed on a 9-point scale, where 1 represented no visible damage and 9 represented severe defoliation (Table  2). Intermediate scores reflected increasing levels of leaf damage, including short holes, long lesions and overall leaf area loss. Ear damage at harvest (ED) was on a plot basis. The ears were inspected for signs of FAW damage, such as feeding marks, ear tip damage, the number of kernels damaged and the presence of larvae and their residues. The severity of ear damage was rated using a 9-point where 1 represented no visible damage to the ear and 9 represented almost 100% ear damage (Kamweru et al. 2022; Prasanna et al. 2018). 2.6   |   Data Curation and Analysis The phenotypic data were subjected to analysis of variance (ANOVA) using Lme4 (R Core Team 2023) as follows: where Yijk is the performance of the observed trait, μ is the pop- ulation mean, Gh is the effect of the hth genotypes, Si is the effect of the ith environment, ( Gh ∗Si ) is the genotypes × environment interaction effect, Rij is the effect of the jth replicate in the season ith, R ( Bk ) is the effect of the kth incomplete block within the jth replicate, and εhijk is the experimental error. In the model, genotypes were considered fixed, while other sources of variation were considered random. The error (�2e), ge- notypic (�2g) and phenotypic (�2p) variances were calculated from the expected mean squares (EMSs) of the ANOVA. Genotypic coefficients of variation, phenotypic coefficient of variation, and broad-sense heritability (H2) were estimated using the variabil- ity package in R based on the method of Kresovich (1990), where genotypes were considered random effects. Best linear unbiased estimation (BLUE) and best linear unbiased prediction (BLUP) were estimated using the Multi-environment Trials Analysis in R (META-R) 2.1 (Alvarado et al. 2020). The combined BLUEs of all the trials were used to compute Pearson's correlation co- efficients between grain yield and other traits using the package ‘PerformanceAnalytics’ in the R software (R Core Team 2023) (1)Yijk = � + Gh + Si + ( Gh ∗Si ) + Rij + R ( Bk ) + �hijk TABLE 1    |    Descriptions of the measured traits for genotypes evaluated under artificial FAW infestation at Ile-Ife, Nigeria during the 2021 and 2022 growing seasons. Trait Stage Unit Description Grain yield (GY) Harvest kg/ha Computed from the weight of the shelled grain adjusted to 80% shelling percentage and corrected for 15% moisture content Fall armyworm-induced foliar damage score at 6 WAP (FAW6) Vegetative Scale (1–9) Scored on per plot basis on a scale of 1 to 9, where 1 is no visible damage to leaves and 9 where plants showed damaged portions including short holes, shreds and completely damaged whorl Fall armyworm-induced foliar damage score at 8 WAP (FAW8) Flowering Scale (1–9) Scored on per plot basis on a scale of 1 to 9 Fall armyworm-induced foliar damage score at 10 WAP (FAW10) Post- flowering Scale (1–9) Scored on per plot basis on a scale of 1 to 9 Ear damage score (ED) Harvest Scale (1–9) Scored on per plot basis on a scale of 1 to 9, with 1 denoting ear that are tidy, uniform, big, and full, and 9 denoting ears with negative characteristics Days to 50% anthesis (DA) Flowering Days Day count to 50% pollen shed from planting Days to 50% silking (DS) Flowering Days Day count to 50% silk emergence from planting Plant height (PHT) Maturity cm Plant height (PHT) was measured as the distance in centimetres between the base of the plant and the first tassel branch. Ear height (EHT) Maturity cm Ear height (EHT) was measured as the distance in centimetres between the base of the plant and the top ear branch. Anthesis-silking interval (ASI) Flowering Days The time interval between onset of 50% anthesis and 50% silking Ears per plant (EPP) Post- flowering Numeric Calculated by dividing the number of harvested ears by the number of plants harvested per plot 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 5 of 14 and the adjusted means for each trial were visualized using the boxplot in R. To select ideotypes with the highest genetic merit across envi- ronments and multiple traits of interest, three models were em- ployed to build three independent selection indices for these populations. The first model was the Multi-traits Genotype- Ideotype Distance Index (MGIDI) (Olivoto and Nardino 2021), which is based on four key steps: (i) normalize traits to a 0–100 scale, (ii) reduce dimensionality and account for correlated traits based on factor analysis, (iii) plan an ideotype based on desired trait values and (iv) compute the distance between each geno- type and the planned ideotype. The MGIDI was done using the MGIDI function implemented in Metan R package as shown in Equation 2. where γij is the score of the ith genotype in the jth factor (i = 1, 2,…, g; j = 1, 2,…, f) and γj is the jth score of the ideotype. The genotypes with the lowest MGIDI values, i.e., genotypes closer to the ID, exhibited the desired values for all the traits studied. The second is the multiple-traits selection index (MI) (Badu- Apraku et al. 2016). This method involved assigning weights to grain yield under FAW infestation and calculating a composite score for each genotype using the standardized mean values of each trait across environments. The MI was estimated using Equation 3. where MI = multiple-trait selection index, GYFAW = grain yield under FAW infestation, EPP = number of ears per plant, ASI = anthesis–silking interval, ED = ear damage at harvest, FAW6, FAW8 and FAW10 = −Fall armyworm foliar damage scores at 6, 8 and 10 weeks after planting, respectively. The third is the Desired Selection Index (DI) implemented in bioflow (https://​biofl​ow.​ebspr​oject.​org/​). For the DI, a two- stage approach analysis was performed. The selection was done following data validation, single and multi-trial analysis. For the multi-environment trial analysis, each environment was replicated, making it a four environments trial. Genotypes were considered as fixed effects, while environment, blocks and replications were considered as random effects. Traits were standardized to a common scale, and economic weights were assigned to reflect their relative importance. For in- stance, a weight of 3 was assigned to grain yield, indicating its high priority, while weights of 1 and −1 were assigned to EPP and ASI, respectively, while FAW-induced foliar and ear damage scores were assigned −2, respectively. 3   |   Results 3.1   |   Analysis of Variance of the Measured Traits The combined analysis of variance (ANOVA) of the 100 hybrids evaluated under artificial FAW infestation revealed significant (p < 0.01) mean square for grain yield, FAW-induced foliar damage scores at 6, 8 and 10 WAP, ear damage and all other measured traits (Table 3). Environment and hybrid by environment mean square were significant (p < 0.01) for all the traits recorded. Similar results were observed all the trials, except environment mean squares for DS and ASI in inbred trial FAW10 and ED in OPV trial and geno- type × environment mean squares for FAW6 and PHT in landrace trial, which were not significant (Table 3). 3.2   |   Mean Performance, Genotypic and Phenotypic Coefficients of Variation, and Broad-Sense Heritability Estimate for the Measured Traits Grain yield varied from 55.9 to 5902.5 kg/ha with the mean of 3480.6 kg/ha for the hybrids and ranged from 221.0 to 3288.9 kg/ha with the mean of 1738.8 kg/ha for the inbred lines (Table 4). The grain yield of the landraces ranged from 414.8 to 3153.4 kg/ha with the mean of 1289.6 kg/ha, while the yield of the OPV ranged from 862.0 to 4685.0 kg/ha with the mean of 2606.5 kg/ha (Table 4). The fall armyworm scores in the various weeks of assessments, (FAW6, FAW8 and FAW10) ranged from 3.8 to 6.6 with the mean of 5.1, 1.6 to 5.5 with the mean of 3.4, and 2.7 to 5.8 with the mean of 4.1, respec- tively among the hybrids. For the inbred trial, FAW6, FAW8 and FAW10 varied from 4.2 to 7.2 with the mean of 5.8, 2.5 to 7.3 with the mean of 4.1 and 1.7 to 8.1 with the mean of 4.6, respectively. Among the landraces, the foliar damage scores (FAW6, FAW8 and FAW10) were 3.5 to 6.5 with the mean of 3.5, 4.5 to 6.3 with the mean of 4.5, and 3.9 to 6.4 with the mean of 3.9, respectively. Furthermore, among the OPVs, the FAW damage scores at 6, 8 and 10 WAP varied from 3.7 to 6.4 with the mean of 5.1, 2.4 to 5.0 with the mean of 3.8, and (2)MGIDIi = ∑f j=1 [( yij−yj ) 2 ]0.5 (3) MI= ( 2×GYFAW ) +EPP−ASI−ED −FAW6−FAW8−FAW10 TABLE 2    |    A scale developed for rating leaf damage of maize genotypes artificially infested with fall armyworm larvae for host plant resistance and susceptibility. Symptoms Score No visible damage 1 < 25% of plants in the plot showed shot holes damage on leaves only 2 ≥ 25% but < 50% showed shot hole damage on leaves only 3 ≥ 50% but less than 75% showed short hole damage on leaves only 4 ≥ 25% but < 50% showed short hole and shred damage on leaves 5 ≥ 50% but less than 75% showed short holes and shreds damage on leaves 6 ≥ 50% but less than 75% showed short holes, shreds and traces of whorl damage on leaves 7 ≥ 75% window-pane damaged portions, short holes, shreds and moderately damaged whorl 8 ≥ 90% of plants showed damaged portions including short holes, shreds and completely damaged whorl 9 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://bioflow.ebsproject.org/ 6 of 14 Plant Breeding, 2025 T A B L E 3      |     M ea n sq ua re fo r g ra in y ie ld , F AW -in du ce d fo lia r a nd e ar da m ag e sc or es , a nd o th er a gr on om ic tr ai ts o f m ai ze h yb ri ds , i nb re d lin es , l an dr ac es a nd o pe n- po lli na te d va ri et ie s ev al ua te d un de r ar tif ic ia l F AW in fe st at io n at Il e- If e, N ig er ia d ur in g th e 20 21 a nd 2 02 2 gr ow in g se as on s. So ur ce D F G Y FA W 6 FA W 8 FA W 10 E D D A D S PH T E H T A SI E PP H yb ri d ge no ty pe s EN V 1 71 80 83 01 5* * 34 .2 ** 56 .9 ** 6. 6* * 42 3. 2* * 10 56 8. 5* * 12 99 2. 9* * 16 27 18 .6 ** 88 86 2. 4* * 54 .6 ** 1. 3* * R EP (E N V ) 2 12 87 03 90 ** 10 .8 ** 1. 4 n s 3. 8* 5. 5* * 57 .3 ** 58 .1 ** 40 92 .0 ** 95 8. 4* * 0. 1 n s 0. 2* * BL O C K (R EP *E N V ) 36 1, 81 8, 98 8 n s 2. 3* * 1. 9* 1. 0* 1. 3* * 6. 9* * 7.7 ** 44 3. 5* * 12 0. 6* * 0. 5* 0. 04  n s G EN 98 42 85 26 0* * 0. 96 * 2. 0* * 1. 6* * 2. 4* * 8. 6* * 11 .4 ** 76 3. 0* * 25 9. 5* * 0. 6* * 0. 03  n s G EN :E N V 94 23 05 14 3* * 1. 67 ** 1. 9* * 0. 8* 0. 9* * 5. 2* * 5. 5* * 34 9. 7* * 13 9. 6* * 0. 6* * 0. 04 * R es id ua ls 15 6 12 33 73 0. 00 0. 71 0. 73 0. 52 0. 54 2. 45 2. 86 20 3. 52 54 .6 9 0. 36 0. 03 In br ed g en ot yp es EN V 1 43 75 19 57 .8 ** 3. 8* 18 .1 ** 62 .1 ** 20 3. 2* * 27 .0 ** 4. 1 n s 26 45 0. 7* * 11 85 9. 1* * 0. 3 n s 0. 5* R EP (E N V ) 2 27 54 02 8. 3* * 1. 3 n s 0. 1 n s 0. 4 n s 2. 9 n s 1. 2 n s 6. 4 n s 13 4. ns 31 3. 7 n s 0. 5 n s 0. 2 n s BL O C K (R EP *E N V ) 36 61 25 91 .1  n s 1. 2 n s 1. 2 n s 1. 2 n s 1. 2 n s 3. 5* 4. 4* 17 9. 6 n s 10 2. 6 n s 0. 5 n s 0. 1 n s G EN 89 23 53 20 9. 9* * 1. 6* * 2. 2* * 4. 5* * 3. 8* * 7. 4* * 7. 9* * 45 4. 7* * 15 4. 5* * 0. 6* * 0. 2* * G EN :E N V 79 13 38 02 8. 5* * 1. 7* * 1. 9* * 2. 4* * 2. 7* * 7. 1* * 7. 5* * 58 1. 4* * 14 0. 3* * 0. 7* * 0. 2* * R es id ua ls 13 2 52 04 23 .3 0 0. 96 1. 08 1. 13 1. 27 2. 37 2. 35 15 9. 38 73 .5 8 0. 35 0. 10 La nd ra ce g en ot yp es EN V 1 68 49 12 9. 2* * 4. 1* 34 .3 ** 30 .7 ** 9. 1* * 46 88 .8 ** 40 88 .2 ** 18 72 .9 * 28 20 .0 ** 25 .4 ** 16 .4 ** R EP (E N V ) 2 10 32 65 .8  n s 5. 8* * 1. 8* 5. 6* * 1. 8* 2. 0 n s 2. 1 n s 96 9. 6 n s 55 0. 7 n s 0. 1 n s 0. 6* * BL O C K (R EP *E N V ) 28 45 55 78 .8  n s 1. 9* * 0. 9* 1. 0* * 0. 6 n s 6. 5* * 6. 6* * 78 7. 3* 33 5. 8 n s 0. 7 n s 0. 1 n s G EN 63 17 33 37 1. 1* * 2. 2* * 0. 7* 1. 0* * 1. 1* * 31 .0 ** 43 .0 ** 13 34 .1 ** 83 0. 8* * 2. 9* * 0. 2* * G EN :E N V 62 14 10 15 9. 1* * 1. 0 n s 1. 0* * 0. 8* 0. 8* 8. 5* * 12 .4 ** 59 0. 1 n s 44 1. 0* * 2. 3* * 0. 2* * R es id ua ls 97 46 29 44 .8 0 0. 92 0. 44 0. 35 0. 46 1. 89 2. 32 45 7. 40 24 5. 95 0. 84 0. 08 O PV g en ot yp es EN V 1 17 61 35 16 2. 1* * 9. 0* * 96 .5 ** 0. 01  n s 0. 02  n s 39 78 .3 ** 48 47 .2 ** 37 31 1. 9* * 17 83 9. 2* * 23 .1 ** 1. 6* * R EP (E N V ) 2 14 71 74 98 .1 ** 7. 5* * 0. 7 n s 1. 4* 5. 2* * 1. 6 n s 0. 7 n s 99 .8  n s 85 .1  n s 0. 6 n s 0. 1* BL O C K (R EP *E N V ) 28 10 72 73 2. 9* 1. 4* * 0. 6 n s 0. 6 n s 0. 6 n s 3. 2 n s 3. 9* 21 9. 1* 92 .6  n s 0. 7* 0. 02  n s (C on tin ue s) 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 7 of 14 3.8 to 5.5 with the mean of 4.7, respectively. The ear damage scores at harvest ranged from 4.1 to 8.0 with the mean of 5.9 for the hybrids, 2.2 to 7.2 with the mean of 4.0 for the inbred lines, 4.6 to 7.5 with the mean of 6.0 for the landraces, and 4.0 to 6.7 with the mean of 5.1 for the OPVs (Table 4). Generally, the landraces exhibited the highest foliar damage scores at FAW6, FAW8 and FAW10, respectively, followed by the inbred lines, the OPVs, while the hybrids had the lowest foliar dam- age scores (Figure 1). The heritability estimates for grain yield ranged from 21% in the landrace trial to 60% in the OPV trial. The FAW damage score at 6 WAP had the heritability estimates varying from 10% in the OPV trial to 43% in the landrace trial. Similarly, heritability estimates for FAW damage scores at 8 WAP var- ied from 14% in the OPV trial to 45% in the landrace trial, while FAW damage scores at 10 WAP had heritability esti- mate ranging from 34% in the OPV trial to 50% in the inbred trial (Table  4). High (≥ 20%) genotypic coefficients of varia- tion (GCV) were obtained for grain yield and anthesis silking interval. The GCV ranged from moderate (10%–20%) to low (≤ 10) for the FAW-induced foliar damage scores at 6, 8 and 10 WAP and ear damage scores at harvest. Generally, the pheno- typic coefficients of variation (PCV) follow a similar trend as GCV (Table 4). 3.3   |   Selection of Ideotypes Using Multi-Trait Genotype-Ideotype Distance Index (MGIDI), Multi-Trait Selection Index (MI) and Desire Selection Index (DI) The ideotypes identified from each trial across environments com- bined high grain yield, increased ears per plant, decreased days for anthesis-silking interval, low FAW-induced foliar damage scores at 6, 8 and 10 WAP, as well as ear damage scores. In the hybrid trial, MI ranged from −9.7 (HYB91) to 8.8 (HYB21) and DI varied from −9.2 in hybrid HYB87 to 9.2 in hybrid HYB73. In the inbred trial, MI ranged from −14.3 (INB7) to 9.4 (INB29) and DI varied from −8.1 (INB11) to 8.4 (INB83). For landrace trial, MI ranged from −8.7 (LR34) to 13.6 (LR52) and DI ranged from −7.1 (LR35) to 12.7 (LR64). For the OPV trial, MI ranged from −9.3 (OPV31) to 7.4 (OPV30) while DI ranged from −7.1 (LR35) to 12.7 (LR64). Based on 15% selection intensity implemented in the three multi-traits selection models (MI, MGIDI and DI), 15 hybrids, 14 inbred lines, 10 landraces and nine open-pollinated varieties were selected s (Figure 2 and Supplementary Tables S1–S4). Of the 15 hybrids selected (Supplementary Tables  S5), 10 hybrids were common to all three selection indices, indicating 67% con- cordance among the three models (Figure 3a). For the inbreds trials, 14 inbred genotypes were selected with 43% (six inbreds) level of concordance among the models (Table S6). In addition, MGIDI and MI showed 79% level of concordance. Similarly, MGIDI and DI, as well as MI and DI, showed 64% concordance in inbred selection (Figure  3b). In the landrace trials, 10 gen- otypes were selected corresponding to 15% selection intensity implemented in the model (Table S7). Of the 10 landraces, the three selection models MGIDI, MI and DI showed 80% level of concordance (Figure 3c). In the OPV trial, nine genotypes were selected corresponding to the selection intensity implemented So ur ce D F G Y FA W 6 FA W 8 FA W 10 E D D A D S PH T E H T A SI E PP G EN 55 36 73 40 6. 5* * 2. 0* * 1. 8* * 0. 9* * 1. 7* * 16 .3 ** 16 .4 ** 64 5. 0* * 33 9. 3* * 1. 1* * 0. 2* * G EN :E N V 55 15 71 54 8* * 1. 5* * 1. 6* * 0. 9* * 0. 8* 13 .7 ** 16 .0 ** 42 2. 4* * 20 4. 8* * 0. 7* * 0. 2* * R es id ua ls 82 60 53 59 .8 0 0. 41 0. 45 0. 38 0. 54 1. 54 2. 15 11 4. 45 11 5. 56 0. 40 0. 02 A bb re vi at io ns : A SI : a nt he si s s ilk in g in te rv al ; D A : d ay s t o 50 % a nt he si s; D F: d eg re e of fr ee do m ; D S: d ay s t o 50 % si lk in g; E D : e ar /c ob d am ag e sc or e at h ar ve st ; E H T: e ar h ei gh t; E N V: e nv ir on m en t; EP P: e ar s p er p la nt ; F AW 6: F AW - in du ce d fo lia r d am ag e sc or es a t 6 W A P; F AW 8: F AW fo lia d am ag e ra tin g at 8 W A P; F AW 10 : F AW -in du ce d fo lia r d am ag e sc or e at 1 0 W A P; G E N : g en ot yp e; G Y: g ra in y ie ld ; n s: no t s ig ni fic an t; PH T: p la nt h ei gh t; R ep : r ep lic at io n. *S ig ni fic an t a t p  <  0. 05 . ** H ig hl y si gn if ic an t a t p  <  0. 01 . T A B L E 3      |     (C on tin ue d) 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 8 of 14 Plant Breeding, 2025 TABLE 4    |    Mean, genotypic coefficients, phenotypic coefficients and broad-sense heritability of measured traits of the hybrid, inbred, landrace and open-pollinated varieties evaluated under artificial FAW infestation at Ile-Ife, Nigeria during the 2021 and 2022 growing seasons. Statistics GY FAW6 FAW8 FAW10 ED DA DS PHT EHT ASI EPP Hybrid genotypes Max 5902.5 6.55 5.5 5.8 8.0 62.0 63.0 177.6 91.1 2.0 1.2 Ave 3480.6 5.1 3.4 4.1 5.9 58.5 59.4 149.2 69.5 1.0 0.9 Min 55.9 3.77 1.6 2.7 4.1 56.0 56.0 116.3 51.8 0.0 0.6 H2 0.50 0.18 0.35 0.50 0.62 0.14 0.35 0.60 0.45 0.21 0.0 GCV (%) 21.3 7.8 13.9 9.3 10.0 6.7 8.2 7.0 7.5 33.9 0.0 PCV (%) 62.0 16.3 30.1 11.8 27.8 12.8 14.0 21.2 33.1 68.9 0.0 LSD 2141.5 1.8 2.0 1.3 1.4 3.0 3.1 25.1 16.8 1.1 0.3 CV 31.8 20.8 25.5 15.9 12.4 2.7 2.8 9.5 10.7 57.3 18.5 Inbred genotypes Max 3288.9 7.2 7.3 8.1 7.2 61.0 60.9 196.7 79.5 1.8 1.5 Ave 1738.8 5.8 4.1 4.6 4.0 57.5 57.5 137.5 59.5 0.5 0.8 Min 221.0 4.2 2.5 1.8 2.2 53.6 53.2 108.6 43.3 0.0 0.1 H2 0.39 0.13 0.25 0.41 0.12 0.21 0.54 0.65 0.48 0.02 0.01 GCV (%) 42.6 7.0 13.9 9.5 14.8 1.7 1.4 7.6 8.8 23.1 12.9 PCV (%) 64.0 14.1 26.2 12.1 40.7 13.1 14.5 23.0 38.7 49.2 18.3 LSD 1709.0 1.9 2.0 2.3 2.4 3.9 3.9 35.9 17.7 1.2 0.6 CV 41.0 20.5 24.7 23.1 27.7 2.7 2.7 9.2 14.7 123.3 40.3 Landrace genotypes Max 3153.4 6.5 6.3 6.4 7.5 70.4 75.2 229.6 126.5 5.0 1.5 Ave 1289.6 5.0 5.5 5.6 6.0 64.3 66.4 189.6 100.2 2.1 0.9 Min 414.8 3.5 4.5 3.9 4.6 57.5 58.8 152.7 66.4 0.0 0.5 H2 0.21 0.43 0.45 0.41 0.29 0.77 0.77 0.65 0.55 0.22 0.07 GCV (%) 23.1 13.1 8.8 9.3 4.4 3.8 4.3 7.4 10.2 18.8 11.6 PCV (%) 59.9 16.1 14.6 14.0 8.5 10.5 10.0 7.4 13.3 48.6 49.0 LSD 1682.9 1.4 1.3 1.1 1.2 4.0 4.6 31.4 27.6 2.1 0.6 CV 51.7 24.0 12.0 10.5 11.2 2.2 2.3 11.0 15.2 41.6 31.2 OPV genotypes Max 4685.0 6.4 5.0 5.5 6.7 63.4 64.7 186.7 102.6 2.2 1.7 Ave 2606.5 5.1 3.8 4.7 5.2 57.7 58.7 155.1 74.0 1.1 0.8 Min 862.0 3.7 2.4 3.8 4.0 50.6 51.7 121.7 51.7 0.0 0.3 H2 0.58 0.10 0.13 0.34 0.53 0.20 0.36 0.35 0.40 0.32 0.08 GCV (%) 28.1 15.5 20.4 13.7 9.4 7.1 7.3 7.9 7.7 27.5 12.0 PCV (%) 59.7 18.5 29.8 18.9 12.0 13.2 14.1 16.0 20.7 60.2 37.9 LSD 1798.8 1.7 1.7 1.3 1.3 5.2 5.6 28.2 20.1 1.2 0.5 CV 30.3 16.1 17.3 14.1 14.7 2.2 2.5 6.9 14.0 59.0 15.1 Abbreviations: ASI: anthesis silking interval; Ave: average; CV: coefficient of variation; DA: days to anthesis; DS: days to silking; ED: ear damage score at harvest; EHT: ear height; EPP: ears per plant; FAW6: FAW-induced foliar damage scores at 6 WAP; FAW8: FAW-induced foliar damage scores at 8 WAP; FAW10: FAW-induced foliar damage score at 10 WAP; GCV: genotypic coefficient of variation; GY: grain yield; H2: heritability; LSD: least significant difference; Max: maximum; Min: minimum; PCV: phenotypic coefficient of variation; PHT: plant height. 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 9 of 14 in other trials (Table S8). Of the selected genotypes, we observed 56% concordance among the three models. Additionally, MI and DI showed 78% consonance, while 67% selection agreement ex- isted between MGIDI and MI (Figure 3d). 3.4   |   Genetic Correlations Between Grain Yield, Agronomic and Fall Armyworm Damage Traits of Maize Genotypes Significant negative correlation was observed between GY and FAW8 (r = −0.95; p < 0.001), FAW10 (r = −0.95; p < 0.001), ED (r = −0.70; p < 0.05), DA (r = −0.66; p < 0.05) and DS (r = −0.63; p < 0.05). However, significant positive correlation (r = 0.66; p < 0.05) was detected between GY and EPP. FAW-induced fo- liar damage score at 8 WAP had significant positive correlation with FAW10 (r = 0.90; p < 0.001), DA (r = 0.75; p < 0.01) and DS (r = 0.72; p < 0.05), while significant negative relationship was shown between FAW8, FAW10 and EPP (r = −0.66; p < 0.05, r = −0.72; p < 0.05). The FAW-induced foliar damage score at 10 WAP had significant positive correlation with ED (r = 0.79; p < 0.01), DA (r = 0.63; p < 0.05), DS (r = 0.61; p < 0.05) and a significant negative correlation with EPP (r = −0.72; p < 0.05) (Figure 4). 4   |   Discussion To initiate the development of FAW resistant maize populations, screening of diverse germplasm is needed to identify multiple sources of FAW resistance (Kamweru et al. 2022). Therefore, it is essential to combine resistance from various germplasm sources to improve the stability and durability of host plant resistance to FAW in maize. This is because FAW has the potential to become a multi-generational pest of significant economic importance in SSA, where the climate is favourable and host plants are abun- dant (Prasanna et al. 2018). In the present study, t diverse early- maturing populations were evaluated to identify promising hybrids and OPVs that could be promoted for further testing and commercialization to mitigate the urgent need of native FAW resistance. The inbred lines and landraces identified could be useful sources of resistance for the development of FAW resis- tant populations. The observed variability among the maize ger- mplasm provides basis for effective selection, enhancing genetic improvement and ultimately ensuring food security and eco- nomic benefits to farmers. The combined analysis of variance (ANOVA) of the hybrids, inbred lines, OPVs and landraces under artificial FAW infestation provided comprehensive insights into the variability and interaction of different factors affecting grain yield and other traits under FAW infestation. Significant mean FIGURE 1    |    Box plot showing distribution Fall armyworm (FAW) damage scores at 6, 8 and 10 WAP and ear damage at harvest of the maize hy- brids, inbred lines, landraces and OPVs evaluated under artificial FAW infestation during the 2021 and 2022 in Nigeria. Abbreviations: ED: ear dam- age score at harvest; FAW6: FAW-induced foliar damage score at 6 WAP; FAW8: FAW-induced foliar damage score at 8 WAP; FAW10: FAW-induced foliar damage score at 10 WAP. 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 10 of 14 Plant Breeding, 2025 squares (p < 0.01) observed for grain yield, FAW-induced foliar damage scores at 6, 8 and 10 weeks after planting (WAP), FAW- induced eardamage, and other traits in the different trial types indicated that there are potential differences among the hybrids, inbred lines, OPVs and Landraces. This suggests the presence of potential genetic variation in the germplasm panel studied that could be explored for FAW resistance breeding in tropical germ- plasm. Kamweru et al. (2022) and Kasoma et al. (2020) reported similar observation. Significant environment and hybrid by environment mean squares (p < 0.01) for all traits implied that the environment has a substantial impact on the expression of these traits, and the performance of genotypes is influenced by specific envi- ronmental conditions. Earlier research has shown that the analysis of genotypes by environmental interactions could facilitate the identification of genotypes whose stability is re- lated to the linear effect of an environmental index (Kamweru et  al.  2022). Even though measured traits from the inbred evaluation were influenced by genetic and environmental factors, days to silking and anthesis silking interval are less affected by environmental factors. Kamweru et al. (2023) re- ported a similar observation in 160 medium maturing tropi- cal maize hybrids evaluated under artificial FAW infestation. The non-significant environment effect for FAW-induced fo- liar damage at 10 WAP and ear damage in the OPV genotypes suggests that damage scores at later stages are less sensitive to environmental differences. The wide range in grain yields for the different maize germplasm reflects significant genetic variability among the genotypes. The hybrids had the high- est grain yield, which is consistent with their known vigour and productivity (Badu-Apraku and Fakorede 2017). The in- bred lines had the lowest yields, reflecting the less vigour of the germplasm due to several cycles of selection and selfing to attain homozygosity (Badu-Apraku and Fakorede  2017). Landraces and OPVs showed intermediate performance. The mean grain yield of the four populations revealed that hybrid maize is superior and should be given higher consideration FIGURE 2    |    Selected maize hybrids, inbred lines, landraces and open-pollinated varieties evaluated under artificial FAW infestation at Ile-Ife in 2021 and 2022 using the multi-trait genotype-ideotype distance index analysis. The selected genotypes are shown as green dots, while the unselected genotypes are the red dots. The green circle represents the cut-off point based on the selection pressure. 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 11 of 14 for commercialization and cultivation by farmers to enhance food productivity. The main reason for this is that hybrids are typically more genetically resilient than inbred lines and can better tolerate FAW attacks. Previous studies report that gen- otypes with high heterosis outperform their inbred line coun- terparts that often exhibit inbreeding depression and poor growth (Fu et al. 2015; Matova et al. 2022). The FAW-induced foliar and ear damage score show variability in FAW resistance across different genetic backgrounds. The Inbred lines and landraces generally exhibited higher damage scores when compared to the hybrids and OPVs. All the four groups of the genotypes (hybrids, inbred lines, landraces and OPVs) were most vulnerable to FAW foliar damage at early stages of growth, but, by the mid-to late-whorl stages, they were able to recover from the severe damage revealing the pos- sibility for late-whorl stage resistance. This finding is partially consistent with Matova et al. (2022) who reported that hybrids and OPVs were more vulnerable to FAW-induced foliar dam- age at early growth stages, but they grew out of it at the mid to late stages, while inbred lines showed increasing damage as they grew to maturity. Both hybrids and inbred lines showed varying response to FAW-induced foliar and ear damage. The FAW-induced foliar damage was lower in hybrids than in inbred lines suggesting that hybrid vigour could be a factor in the performance of the hybrid compared to inbred lines. However, the inbred lines showed lower ear damage scores compared to the hybrid genotypes. This was unexpected as it can be inferred that the resistance to FAW-induced foliar and ear damages are under different genetic control. Perhaps cob damage resistance is primarily controlled by additive gene action, where the inbreds, through selection, have accumu- lated favourable alleles. This observation highlights the need for evaluating hybrids under FAW infestation to select hybrids with lower ear damage for improved yield and quality traits. Generally, the landraces exhibited increased rating scores to FAW-induced foliar and ear damage compared to the hy- brids and inbreds. The contrasting performance between the landrace and the breeding populations suggests that breeders could find promising genotypes that are resistant to FAW from the existing improved varieties in their breeding populations compared to a new collection from wild types. Heritability estimates for grain yield and FAW damage scores ranged from low to moderate, indicating a direct selection for FAW resistance germplasm could be ineffective. The herita- bility estimate of FAW-induced foliar damage scores increases with advancing developmental stage from 6 (vegetative) to 10 WAP (flowering). This implies that genetic resistance to FAW improvement through direct selection may be more reliable in the later stages of growth of maize plants than in the early grow- ing period of maize genotypes. High GCV for grain yield and anthesis-silking interval observed in this study suggests sub- stantial genetic variability, offering potential for selection and improvement. Moderate to low GCV for FAW damage scores and ear damage indicates less genetic variability, which could make improvement through selection more challenging (Prasanna et  al.  2022). PCV values were higher than GCV for all mea- sured traits, indicating that environmental factors contribute significantly to the total phenotypic variation. This reinforces FIGURE 3    |    (a–d) Venn diagram of the common (a) hybrids, (b) inbred lines, (c) landraces and (d) open-pollinated varieties evaluated under arti- ficial FAW infestation at Ile-Ife in 2022 and 2023. Abbreviations: DI: Desire Selection Index; MGIDI: Multi-Trait Genotype-Ideotype Distance Index; MI: Multi-trait Selection index. 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 12 of 14 Plant Breeding, 2025 the importance of multi-environment trials to accurately assess and select desirable genotypes. The observed variability in grain yield and FAW damage scores across different trial genotypes emphasizes the need for utilizing genetic variation available for breeding programs. High heritability estimates observed for grain yield and FAW resistance traits indicate that these traits could be effectively improved through selection. However, the higher PCV compared to GCV highlights the influence of en- vironmental factors on trait expression, necessitating multi- environment testing to identify stable and high-performing genotypes. The results of the ideotype selection based on the MGIDI, MI, and DI highlight the comparative performance and compat- ibility of the three multi-trait selection indices in identify- ing high-performing genotypes under FAW infestation. The MGIDI also integrates multiple traits but focuses on the dis- tance between the observed genotype performance and an ideotype (a hypothetical ideal genotype). The MI and DI com- posite measures incorporating multiple traits weighted for high grain yield, increasing ears per plant, decreasing ASI, ear damage, and FAW-induced foliar damage scores at 6, 8 and 10 WAP. In the present study, the degree of overlap between indi- ces reflects their consistency in identifying genotypes. There was relatively higher agreement among the indices. However, MGIDI seems closely aligned with MI, often identifying more genotypes in common compared to DI. The three indices also showed appreciable level of concordance, suggesting that genotypes for which they agreed upon actually possess some useful genes for FAW resistance improvement. This implies that the three indices are both reliable and effective for iden- tification superior genotypes. The alignment of results from both indices indicates that breeders can confidently use ei- ther index for selection purposes, potentially simplifying the breeding process and decision-making. This study demon- strates the effectiveness of MGIDI, MI, and DI in selecting high-performing genotypes under FAW infestation. However, MI is straight forward, and simple for identifying promising genotypes under FAW infestation. Unlike MGIDI and DI, the MI approach minimizes reliance on computational tools and complex workflows, making it more accessible to any breed- ing programs. The significant negative correlation observed between grain yield and FAW-induced foliar damage scores indicated that high grain yield is associated with lower FAW-induced foliar dam- age at 8 and 10 WAP, lower ear damage at harvest, and shorter days to anthesis and silking as well as high ear numbers. This suggests that plants that are likely resistant to FAW mature faster and tend to have higher yields. The correlation among FAW damage scores and other traits indicates that higher FAW damage at 10 WAP is associated with higher FAW damage at 8 WAP and longer times to anthesis and silking. This suggests that plants more susceptible to early FAW damage are likely to experience continued damage and have delayed flowering. A significant negative correlation between FAW damage at 8 WAP FIGURE 4    |    Correlation coefficients between grain yield, agronomic and FAW damage traits of maize hybrids, inbred lines, landraces and OPVs evaluated under artificial FAW infestation at Ile-Ife during the 2021 and 2022 in Nigeria. Abbreviations: ASI: anthesis silking interval; DA: days to anthesis; DS: days to silking; ED: ear damage score at harvest; EHT: ear height; EPP: ears per plant; FAW6: FAW damage score at 6 WAP; FAW8: FAW damage score at 8 WAP; FAW10: FAW damage score at 10 WAP; GY: grain yield; PHT: plant height. 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 13 of 14 and ears per plant indicates that higher FAW damage early on is associated with fewer ears per plant. 5   |   Conclusion Significant genetic variability was revealed among the hy- brids, inbreds, OPVs and landrace for all measured traits. Environment and hybrid by environment interactions were also significant. At 15% selection intensity, MGIDI, MI and DI in- dices each identified 15 hybrids, 14 inbred lines, 10 landraces, and nine open-pollinated varieties. Of the selected genotypes, MI and MGIDI had 93% hybrid, 71% inbred, 90% landrace and 67% OPV concordance, whereas MI and DI had 93% hybrid, 57% inbred, 90% landrace and 67% OPV concordance. Significant negative correlations were observed between grain yield and FAW-induced foliar damage scores at 8 and 10 WAP, and ear damage at harvest. In comparison to the more computationally demanding models (MGIDI and DI), the MI demonstrated simi- lar accuracy in identifying promising ideotypes for further test- ing and commercialization. All the genotypes were vulnerable to FAW damage at early whorl-stage but exhibited late whorl- stage resistance as development progressed, suggesting that di- rect selection for FAW resistance should be targeted towards the late vegetative stage. Conflicts of Interest The authors declare no conflicts of interest. Data Availability Statement The data that support the findings of this study are openly available in IITA-CKAN repositories at https://​doi.​org/​10.​25502/​​znc7-​8z14/​d. References Alvarado, G., F. M. Rodríguez, A. Pacheco, et  al. 2020. “META-R: A Software to Analyze Data From Multi-Environment Plant Breeding Trials.” Crop Journal 8, no. 5: 745–756. https://​doi.​org/​10.​1016/j.​cj.​2020.​ 03.​010. Badu-Apraku, B., and Fakorede, M. A. B. 2017. Advances in Genetic Enhancement of Early and Extra-Early Maize for Sub-Saharan Africa. https://​doi.​org/​10.​1007/​978-​3-​319-​64852​-​1. Badu-Apraku, B., M. A. B. Fakorede, A. O. Talabi, et  al. 2016. “Gene Action and Heterotic Groups of Early White Quality Protein Maize Inbreds Under Multiple Stress Environments.” Crop Science 56, no. 1: 183–199. https://​doi.​org/​10.​2135/​crops​ci2015.​05.​0276. Badu-Apraku, B., M. Oyekunle, A. Menkir, et al. 2013. “Comparative Performance of Early-Maturing Maize Cultivars Developed in Three Eras Under Drought Stress and Well-Watered Environments in West Africa.” Crop Science 53, no. 4: 1298–1311. https://​doi.​org/​10.​2135/​crops​ ci2012.​11.​0640. Baudron, F., M. A. Zaman-Allah, I. Chaipa, N. Chari, and P. Chinwada. 2019. “Understanding the Factors Influencing Fall Armyworm (Spodoptera Frugiperda J.E. Smith) Damage in African Smallholder Maize Fields and Quantifying Its Impact on Yield. A Case Study in Eastern Zimbabwe.” Crop Protection 120: 141–150. https://​doi.​org/​10.​ 1016/j.​cropro.​2019.​01.​028. Bonkoungou, T. O., B. Badu-Apraku, V. O. Adetimirin, K. R. Nanema, and I. I. Adejumobi. 2024. “Genetic Analysis of Grain Yield and Related Traits of Extra-Early Orange Maize Inbred Lines and Their Hybrids Under Drought and Rain-Fed Conditions.” Frontiers in Plant Science 15: 1463924. https://​doi.​org/​10.​3389/​fpls.​2024.​1463924 39678008. Ceron-Rojas, J. J., and J. Crossa. 2018. Linear Selection Indices in Modern Plant Breeding, 256. Springer Nature. Davis, F. M., and W. P. Williams. 1992. Visual Rating Scales for Screening Whorl-Stage Corn for Resistance to Fall Armyworm. Mississippi Agricultural and Forestry Experiment Station, Technical Bulletin 186. Mississippi State University. Day, R., P. Abrahams, M. Bateman, et  al. 2017. “Fall Armyworm: Impacts and Implications for Africa.” Outlooks on Pest Management 28, no. 5: 196–201. https://​doi.​org/​10.​1564/​v28_​oct_​02. Farrar, J. J., P. C. Ellsworth, R. Sisco, et  al. 2018. “Assessing Compatibility of a Pesticide in an IPM Program.” Journal of Integrated Pest Management 9, no. 1: 3. https://​doi.​org/​10.​1093/​jipm/​pmx032. Fu, D., M. Xiao, A. Hayward, et al. 2015. “What Is Crop Heterosis: New Insights Into an Old Topic.” Journal of Applied Genetics 56: 1–13. https://​ doi.​org/​10.​1007/​s1335​3-​014-​0231-​z. ISAAA. 2018. Pocket K No. 16: Biotech Crop Highlights in 2017. Inter national Service for the Acquisition of Agri-biotech Applications (ISAAA). Job, A., I. Iseghohi, A. Abe, et al. 2022. “Genetic Analysis of Agronomic and Fall Armyworm-Resistance Traits in Maize Hybrids With Varying Levels of Resistance to Stem Borer.” Agronomy 12, no. 1–18: 3042. https://​doi.​org/​10.​3390/​agron​omy12​123042. Kamweru, I., B. Y. Anani, Y. Beyene, et  al. 2022. “Genomic Analysis of Resistance to Fall Armyworm (Spodoptera frugiperda) in CIMMYT Maize Lines.” Genes 13, no. 2: 251. https://​doi.​org/​10.​3390/​genes​ 13020251. Kamweru, I., Y. Beyene, A. Y. Bruce, et al. 2023. “Genetic Analyses of Tropical Maize Lines Under Artificial Infestation of Fall Armyworm and Foliar Diseases Under Optimum Conditions.” Frontiers in Plant Science 14: 1086757. https://​doi.​org/​10.​3389/​fpls.​2023.​1086757. Kasoma, C., H. Shimelis, M. Laing, A. I. T. Shayanowako, and I. Mathew. 2020. “Screening of Inbred Lines of Tropical Maize for Resistance to Fall Armyworm, and for Yield and Yield-Related Traits.” Crop Protection 136: 105218. https://​doi.​org/​10.​1016/j.​cropro.​2020.​105218. Kasoma, C., H. Shimelis, M. D. Laing, and B. Mekonnen. 2022. “Fall Armyworm Infestation and Development: Screening Tropical Maize Genotypes for Resistance in Zambia.” Insects 13, no. 11: 1020. https://​ doi.​org/​10.​3390/​insec​ts131​11020​. Kresovich, S. 1990. “Quantitative genetics in maize breeding.” Field Crops Research 23, no. 1: 105218. https://​doi.​org/​10.​1016/​0378-​4290(90)​ 90102​-​h. Kumar, R. M., B. G. Gadratagi, V. Paramesh, et al. 2022. “Sustainable Management of Invasive Fall Armyworm, Spodoptera frugiperda.” Agronomy 12, no. 9: 2150. https://​doi.​org/​10.​3390/​agron​omy12​092150. Matova, P. M., C. N. Kamutando, D. Kutywayo, C. Magorokosho, and M. Labuschagne. 2022. “Fall Armyworm Tolerance of Maize Parental Lines, Experimental Hybrids, and Commercial Cultivars in Southern Africa.” Agronomy 12, no. 6: 1463. https://​doi.​org/​10.​3390/​agron​omy12​ 061463. Nagoshi, R. N., G. Goergen, K. A. Tounou, K. Agboka, D. Koffi, and R. L. Meagher. 2018. “Analysis of Strain Distribution, Migratory Potential, and Invasion History of Fall Armyworm Populations in Northern sub- Saharan Africa.” Scientific Reports 8, no. 1: 3710. https://​doi.​org/​10.​ 1038/​s4159​8-​018-​21954​-​1. Njuguna, E., P. Nethononda, K. Maredia, et  al. 2021. “Experiences and Perspectives on Spodoptera frugiperda (Lepidoptera: Noctuidae) Management in sub-Saharan Africa.” Journal of Integrated Pest Management 12, no. 1: 7. https://​doi.​org/​10.​1093/​jipm/​pmab002. Nuambote-Yobila, O., A. Y. Bruce, G. O. Okuku, et al. 2023. “Assessment of Resistance Mechanisms to Fall Armyworm, Spodoptera frugiperda in 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.25502/znc7-8z14/d https://doi.org/10.1016/j.cj.2020.03.010 https://doi.org/10.1016/j.cj.2020.03.010 https://doi.org/10.1007/978-3-319-64852-1 https://doi.org/10.2135/cropsci2015.05.0276 https://doi.org/10.2135/cropsci2012.11.0640 https://doi.org/10.2135/cropsci2012.11.0640 https://doi.org/10.1016/j.cropro.2019.01.028 https://doi.org/10.1016/j.cropro.2019.01.028 https://doi.org/10.3389/fpls.2024.1463924 https://doi.org/10.1564/v28_oct_02 https://doi.org/10.1093/jipm/pmx032 https://doi.org/10.1007/s13353-014-0231-z https://doi.org/10.1007/s13353-014-0231-z https://doi.org/10.3390/agronomy12123042 https://doi.org/10.3390/genes13020251 https://doi.org/10.3390/genes13020251 https://doi.org/10.3389/fpls.2023.1086757 https://doi.org/10.1016/j.cropro.2020.105218 https://doi.org/10.3390/insects13111020 https://doi.org/10.3390/insects13111020 https://doi.org/10.1016/0378-4290(90)90102-h https://doi.org/10.1016/0378-4290(90)90102-h https://doi.org/10.3390/agronomy12092150 https://doi.org/10.3390/agronomy12061463 https://doi.org/10.3390/agronomy12061463 https://doi.org/10.1038/s41598-018-21954-1 https://doi.org/10.1038/s41598-018-21954-1 https://doi.org/10.1093/jipm/pmab002 14 of 14 Plant Breeding, 2025 Tropical Maize Inbred Lines.” Agronomy 13, no. 1: 203. https://​doi.​org/​ 10.​3390/​agron​omy13​010203. Olivoto, T., and M. Nardino. 2021. “MGIDI: Toward an Effective Multivariate Selection in Biological Experiments.” Bioinformatics 37, no. 10: 1383–1389. https://​doi.​org/​10.​1093/​bioin​forma​tics/​btaa981 33226063. Oyetunde, O. A., B. Badu-Apraku, O. J. Ariyo, and C. O. Alake. 2020. “Efficiencies of Heterotic Grouping Methods for Classifying Early Maturing Maize Inbred Lines.” Agronomy 10, no. 8: 1198. https://​doi.​ org/​10.​3390/​agron​omy10​081198. Prasanna, B. M., A. Bruce, Y. Beyene, et al. 2022. “Host Plant Resistance for Fall Armyworm Management in Maize: Relevance, Status and Prospects in Africa and Asia.” In Theoretical and Applied Genetics, vol. 135, Issue 11, 3897–3916. Springer Science and Business Media Deutschland GmbH. https://​doi.​org/​10.​1007/​s0012​2-​022-​04073​-​4. Prasanna, B. M., J. E. Huesing, R. Eddy, and V. M. Preschke. 2018. “Fall Armyworm in Africa: A Guide for Integrated Pest Management.” In West Africa Regional Training of Trainers and Awareness Generation Workshop on Fall Armyworm Management, IITA, Cotonou, Bénin: Vol. First Edit, 1st ed. CIMMYT. R Core Team. 2023. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Rocha, J., J. Machado, and P. Carneiro. 2017. “Multi-Trait Index Based on Factor Analysis and Ideotype-Design: Proposal and Application on Elephantgrass Breeding for Bioenergy.” GCB Bioenergy 10, no. 1: 52–60. https://​doi.​org/​10.​1111/​gcbb.​12443​. Schlum, K., K. Lamour, P. Tandy, et  al. 2021. “Genetic Screening to Identify Candidate Resistance Alleles to cry1f Corn in Fall Armyworm Using Targeted Sequencing.” Insects 12, no. 7: 618. https://​doi.​org/​10.​ 3390/​insec​ts120​70618​ 34357278. Senay, S. D., P. G. Pardey, Y. Chai, L. Doughty, and R. Day. 2022. “Fall Armyworm From a Maize Multi-Peril Pest Risk Perspective.” Frontiers in Insect Science 2: 971396. https://​doi.​org/​10.​3389/​finsc.​2022.​971396. Sharanabasappa, S. D., C. M. Kalleshwaraswamy, M. S. Maruthi, and H. B. Pavithra. 2018. “Biology of Invasive Fall Army Worm Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctuidae) on Maize.” Indian Journal Of Entomology 80, no. 3: 540–543. https://​doi.​org/​10.​5958/​0974-​ 8172.​2018.​00238.​9. Tambo, J. A., M. K. Kansiime, I. Mugambi, et al. 2020. “Understanding Smallholders' Responses to Fall Armyworm (Spodoptera Frugiperda) Invasion: Evidence From Five African Countries.” Science of the Total Environment 740: 140015. https://​doi.​org/​10.​1016/j.​scito​tenv.​2020.​ 140015. Togola, A., S. Meseka, A. Menkir, et al. 2018. “Measurement of Pesticide Residues From Chemical Control of the Invasive Spodoptera Frugiperda (Lepidoptera: Noctuidae) in a Maize Experimental Field in Mokwa, Nigeria.” International Journal of Environmental Research and Public Health 15, no. 5: 849. https://​doi.​org/​10.​3390/​ijerp​h1505​0849. Wightman, J. A. 2018. “Can Lessons Learned 30 Years ago Contribute to Reducing the Impact of the Fall Army Worm Spodoptera Frugiperda in Africa and India?” Outlook on Agriculture 47, no. 4: 259–269. https://​ doi.​org/​10.​1177/​00307​27018​814849. Supporting Information Additional supporting information can be found online in the Supporting Information section. 14390523, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/pbr.13291 by N igeria H inari N PL , W iley O nline L ibrary on [07/08/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.3390/agronomy13010203 https://doi.org/10.3390/agronomy13010203 https://doi.org/10.1093/bioinformatics/btaa981 https://doi.org/10.3390/agronomy10081198 https://doi.org/10.3390/agronomy10081198 https://doi.org/10.1007/s00122-022-04073-4 https://doi.org/10.1111/gcbb.12443 https://doi.org/10.3390/insects12070618 https://doi.org/10.3390/insects12070618 https://doi.org/10.3389/finsc.2022.971396 https://doi.org/10.5958/0974-8172.2018.00238.9 https://doi.org/10.5958/0974-8172.2018.00238.9 https://doi.org/10.1016/j.scitotenv.2020.140015 https://doi.org/10.1016/j.scitotenv.2020.140015 https://doi.org/10.3390/ijerph15050849 https://doi.org/10.1177/0030727018814849 https://doi.org/10.1177/0030727018814849 Screening for Fall Armyworm (Spodoptera frugiperda J. E. Smith) Resistance in Early-Maturing Tropical Maize Adapted to Sub-Saharan Africa ABSTRACT 1   |   Introduction 2   |   Materials and Methods 2.1   |   Genetic Materials and Description of Study Location 2.2   |   Field Layout and Experimental Design of the Trial 2.3   |   Establishment of the Colonies of FAW for Artificial Infestation in the Field 2.4   |   Artificial Infestation of the Maize in the Field 2.5   |   Data Collection 2.6   |   Data Curation and Analysis 3   |   Results 3.1   |   Analysis of Variance of the Measured Traits 3.2   |   Mean Performance, Genotypic and Phenotypic Coefficients of Variation, and Broad-Sense Heritability Estimate for the Measured Traits 3.3   |   Selection of Ideotypes Using Multi-Trait Genotype-Ideotype Distance Index (MGIDI), Multi-Trait Selection Index (MI) and Desire Selection Index (DI) 3.4   |   Genetic Correlations Between Grain Yield, Agronomic and Fall Armyworm Damage Traits of Maize Genotypes 4   |   Discussion 5   |   Conclusion Conflicts of Interest Data Availability Statement References