Experimental site

The study was conducted at the Plant Protection Laboratory of the Assosa Agricultural Research Center (AsARC), located in Benishangul Gumuz, Ethiopia (latitude: 10°03′ N, longitude: 34°59′ E, altitude: 1553 m.a.s.l.). The site experiences an average annual rainfall of 1177 mm and a mean temperature of 26.79°C.

Insect colony maintenance

Third-generation adults of C. chinensis used for infestation trials were sourced from a laboratory colony maintained at the AsARC. The colony originated from bruchids recovered from infested soybean grain stores in Assosa and was sustained under controlled conditions using Gizo soybean seeds. Insects were reared in 2-L plastic containers covered with muslin cloth to allow air circulation while preventing escape. To ensure continuous breeding and avoid age bias, adults were periodically transferred to fresh seeds. Morphological identification of C. chinensis L. was conducted based on diagnostic criteria established by Farrell et al. (Reference Farrell, Brier and Colins2015), confirming the species’ identity prior to experimentation.

Soybean genotypes and experimental design

This study evaluated 50 soybean genotypes, including 48 genotypes from the Jimma Agricultural Research Center and 2 genotypes (Belesa-95 and Gishama) obtained from the AsARC (table 1). To ensure the exclusion of pre-existing insect contamination, all seed samples were oven-dried at 30°C for 24 h following the method described by Amusa et al. (Reference Amusa, Ogunkanmi, Adetunbi, Akinyosoye, Bolarinwa and Ogundipe2014). The dried seeds were then conditioned on laboratory shelves for 7 days at ambient temperature to equilibrate moisture levels before infestation (Kananji, Reference Kananji2007).

Table 1.

List of 50 soybean genotypes used in this study

From each genotype, 50 seeds were randomly selected and weighed to establish initial seed mass. The 50 seeds were placed in plastic Petri dishes (9 cm diameter) and infested using the no-choice test protocol outlined by Somta et al. (Reference Somta, Somta, Tomooka, Ooi, Vaughan and Srinives2008). Each dish received one female and two to three male adult C. chinensis (1–2 days old), with sexes identified based on external morphological characteristics (Farrell et al., Reference Farrell, Brier and Colins2015). The experiment followed a randomised complete block design with two replications per genotype. Infesting adults were removed after 10 days to prevent overlapping generations and ensure consistent pest pressure (Kananji, Reference Kananji2007).

Data collection and resistance indices

On the 11th day following infestation, the number of eggs deposited on the seeds was recorded, following the procedure outlined by Kananji (Reference Kananji2007). Emerged adult insects were subsequently counted and removed daily until no emergence was observed for 5 consecutive days, as described by Lephale et al. (Reference Lephale, Addo-Bediako and Ayodele2012). Following the final emergence assessment, the remaining seeds in each Petri dish were weighed to determine post-infestation weight:

1. a. Grain weight loss percentage (GWL%): This represents the economic loss due to insect damage and was calculated using the formula proposed by Amusa et al. (Reference Amusa, Ogunkanmi, Adetunbi, Akinyosoye, Bolarinwa and Ogundipe2014):

${\text{GWL}}\% = \frac{{{\text{IGW}} - {\text{FGW}}}}{{{\text{IGW}}}}{\text{ }} \times {\text{ }}100$, where IGW is the initial grain weight (g) and FGW is the final grain weight (g).

1. b. Growth index (GI): This reflects the suitability of each genotype for the development of C. chinensis, based on the formula by Régnière et al. (Reference Régnière, Powell, Bentz and Nealis2012):

\begin{equation*}{\text{GI = }}\frac{{\% {\text{Adult emergence}}}}{{{\text{Median development period}}}}\end{equation*}

The median development period (MDP) was defined as the number of days elapsed from the fifth day after oviposition to the emergence of the first adult, following Kananji (Reference Kananji2007).

1. c. Dobie’s susceptibility index (DSI): This quantifies the susceptibility of seed samples to bruchid damage, incorporating both the number of emerged adults and the MDP, as defined by Dobie (Reference Dobie1974):

\begin{equation*}{\text{DSI}} = {\text{ }}\frac{{{\text{Lo}}{{\text{g}}_{\text{e}}}\left( {{\text{Total adults emerged}}} \right){\text{ }} \times {\text{ }}100}}{{{\text{MDP}}}}\end{equation*}

A DSI score of 0 was assigned if no adult insects emerged (Derera et al., Reference Derera, Giga and Pixley2001). Based on DSI values, soybean genotypes were classified into susceptibility categories as follows (Kananji, Reference Kananji2007; Radha and Susheela, Reference Radha and Susheela2014): resistant when DSI = 0–1, moderately resistant (DSI = 2–3), susceptible (DSI = 4–5), and highly susceptible (DSI = 6–9).

Data analysis

Statistical analyses were conducted to evaluate genotypic variation in resistance traits using a randomised complete block design model across 2 years as follows:

\begin{equation*}{y_{ijk}} = \mu + {Y_i} + {B_{j\left( i \right)}} + {G_k} + {\left( {GY} \right)_{ik}} + {\varepsilon _{ijk}},\end{equation*}

where ${y_{ijk}}$ is the observed resistance trait, $\mu {\text{ }}$is the overall mean, ${Y_i}$ is the fixed effect of year _i, ${B_{j\left( i \right)}}{\text{ }}$is the effect of replication _j nested within year _i, ${G_k}$ is the fixed effect of genotype _k, ${\left( {GY} \right)_{ik}}$ is the genotype × year interaction, and ε_ijk is the random error. Analysis of variance was performed in R Statistical software (R Core Team, 2023) to determine the significance of differences among soybean genotypes for each resistance parameter. Treatment means were separated using Tukey’s Honestly Significant Difference (HSD) at a 5% probability level. Based on the mean values of DSI, genotypes were classified into distinct resistance categories. To further investigate the relationships among measured traits, such as grain weight loss, adult emergence, MDP, and GI, regression and Pearson’s correlation analyses were conducted, following the approach of Amusa et al. (Reference Amusa, Ogunkanmi, Bolarinwa and Ojobo2013).