Elaeidobius kamerunicus Faust. is the main pollinator of oil palm (Caudwell 2001; Prasetyo & Susanto 2016; Zulkefli et al. 2021) This pollinating insect originated from Africa and was introduced to Indonesia from Malaysia in 1982. Elaeidobius kamerunicus was officially released in Indonesia in 1983 (Lubis 1992). One of the reasons for the introduction of E. kamerunicus into Indonesia was because this insect has a larger body size, that it can carry more pollen compared to other species (Syed 1979). The presence of E. kamerunicus has a positive impact on oil palm plantations (Gintoron et al. 2023). Hutauruk et al. (1982) have reported that pollination by E. kamerunicus increases the oil palm fruit set from 44% to 75%. However, the success of pollination, as indicated by fruit set fluctuates under field conditions (Kurniawan 2010) and tends to decrease from year to year (Purba et al. 2010). The decrease in fruit set in Indonesia is also in line with the indication of a decline in the performance of E. kamerunicus (Rao & Law 1998; Prasetyo & Susanto 2012; Yousefi et al. 2020; Prasetyo et al. 2026). The performance of E. kamerunicus including the issue of body size is thought to be one of the factors causing decreased efficiency and effectiveness of pollination (Sagarra et al. 2001; Willmer & Finlayson 2014).

In more than 30 years since its introduction, the population of E. kamerunicus have become established and spread in various locations in Indonesia (Bakara et al. 2019) and Malaysia (Latip et al. 2019). Geographical differences among Africa, Malaysia, Sumatra, Java, Kalimantan, and Sulawesi are also thought to be factors that trigger intraspecific variations in body size (Bai et al. 2016). Therefore, we hypothesised that longterm geographic isolation has resulted in variation in body size among E. kamerunicus populations between native and introduced populations. To our knowledge, no quantitative comparative study has addressed intraspecific variation in body size of E. kamerunicus among Indonesia, Malaysia, and Cameroon, making this research necessary.

Measurement of variations in body size of insects was carried out using morphometric analysis (Daly 1985; Alvarenga et al. 2012). This morphometric analysis is reported to have been used successfully in understanding intra and interspecific body variations (Rattanawannee et al. 2007; Friedrich et al. 2014) from various groups of insects such as aphids (Foottit 1992), bees (Daly 1992), and beetles (Sanmartín & Martín-Piera 1999; Latip et al. 2019). Therefore, this study employs morphometric analysis to quantify and compare morphometric variation in E. kamerunicus populations from Indonesia, Malaysia, and Cameroon, thereby providing insights into the ecological and functional implications of morphological variation in this key pollinator species.

The study was conducted from October 2018 to May 2019 using six populations of E. kamerunicus from three countries: Cameroon (Littoral Region, Douala), Malaysia (Selangor, Dengkil), and Indonesia (Siantar, North Sumatra; Bogor, West Java; Tapin, South Kalimantan; and Morowali, Central Sulawesi) (Figure 1, Table 1). These locations represent both the native range (Cameroon) and introduced populations (Malaysia and Indonesia). Morphometric measurement was carried out at the Insect Taxonomy and Biosystematics Laboratory, Plant Protection Department, IPB University, Indonesia.

Adult E. kamerunicus from Indonesia populations were collected using purposive sampling from oil palm plantations aged 4–6 years. Weevils were captured directly from male inflorescences at anthesis using fine tweezers. A total of 200 individuals were collected per location (100 males and 100 females). Each specimen was temporarily placed in a small plastic container (2 cm × 2 cm) to prevent movement, transported to the laboratory, and preserved at −20 °C prior to measurement.

Specimens (100 males and 100 females) from Malaysia and Cameroon were obtained from collaborating researchers at national oil palm research institutes in each country. All sample were collected from commercial oil palm plantations and preserved using similar procedures before morphometric analysis.

One hundred males and one hundred females adult of E. kamerunicus from each region were observed using a stereo microscope and body parts were measured using Dino Capture 2.0 software. Each weevil measured 10 morphometric parameters, namely thorax length (TL), thorax width (TW), thorax thickness (TT), elytra length (EL), elytra width (EW), abdomen length (AL), abdomen width (AW), abdomen thickness (AT), body length (BL), and body width (BW) (Figure 2).

All morphometric data were tabulated using Microsoft Excel 2016. Differences in body size between males and females of E. kamerunicus were analyzed using principal component analysis (PCA) (Dillon & Goldstein 1984; McGarigal et al. 2013). PCA analysis was used to find the morphometric characters contributing in the separating between males and females of E. kamerunicus from the 10 measured body parts. In addition, T test was also carried out to distinguish the size of each body part between male and female E. kamerunicus.

Prior to statistical analysis, data were tested for normality and homogeneity of variance. Normality was assessed using the Shapiro–Wilk test (Shapiro & Wilk 1965), while homogeneity of variance was evaluated using Levene’s test. The difference in size of 10 body parts of males and females E. kamerunicus from 6 different locations were analyzed using one way analysis of variance (ANOVA), and means were separated by Duncan test at 95% confidence level using Genstat 18th Edition. Measurements data of body parts of E. kamerunicus from 6 locations were analyzed using a descriptive biplot analysis (Anuraga 2015) using Paleontological Statistics (PAST) version 3.18 (Hammer et al. 2001).

Variations in body size of male and female E. kamerunicus were analyzed using PCA. The first two principal components explained 94.741% of total variance. The PCA results clearly demonstrated a distinct separation between male and female individuals. As shown in Figure 3 based on 1,200 individuals measured (600 males and 600 females), the two groups were well differentiated.

Body length (BL) and elytra length (EL) showed highest loading scores both in PC1 and PC2 (Table 2), indicating that these two characters contributed most strongly to the differentiation between male and female among the ten morphometric characters examined.

The T-test analysis further revealed highly significant difference (p < 0.001) in all measured body parts between male and female E. kamerunicus (Table 3). Males generally exhibited larger body size than females across all locations.

Morphometric measurements of ten body characters revealed clear variations in female E. kamerunicus from Indonesia, Malaysia, and Cameroon, with all variables meeting ANOVA assumptions. Female from Cameroon consistently exhibited the largest body size across most measured characters. For example, BL in Cameroon populations reached 2.97–3.11 mm, compared to 2.74–2.90 mm in Indonesian populations and 2.65–2.80 mm in Malaysian populations. A similar pattern was observed for elytra and thorax dimensions, where Cameroon populations showed higher mean values across all characters.

In contrast, female from Malaysia consistenly exhibited the smallest body size for most characters. For instance, EW and TW were generally lower (0.59–0.63 mm and 0.89–0.95 mm, respectively) compared to Indonesia (0.62–0.67 mm and 0.94–1.00 mm respectively) and Cameroon (0.66–0.71 mm and 1.01–1.06 mm, respectively). However, an exception wasobserved in AL, where Malaysian population showed slighty higher values relative to Indonesian population (Figure 4A).

Biplot analysis explained 97.466% and 91.438% of total variance in females and males, respectively (Figure 5).

BL, TL, and EL showed the highest contribution to variance, while AW, AT, and TT contributed relatively less in both sexes.

This study revealed clear geographic variation in body size among E. kamerunicus populationfrom Indonesia, Malaysia, and Cameroon. Based on clustering patterns, three distinct groupswere identified for both males and females (Figure 5). Group 1 consisted of individuals with a smaller body size, represented by populations from Selangor, Malaysia. Group 2 included individuals with intermediate body size, represented by populations from Siantar, Dramaga, Tapin, Morowali, Indonesia. Group 3 comprised individuals with larger body size, represented by populations from Douala, Cameroon.

This study demonstrates clear variations in body size of E. kamerunicus across populations from Indonesia,

Malaysia, and Cameroon. Individuals from Cameroon consistently exhibited the largest body size compared to those from Indonesia and Malaysia both males and females. Similarly, Nasir et al. (2020) reported that the E. kamerunicus population from Cameroon was larger and morphometrically distinct from Malaysian populations, with PCA revealing a clear separation between the Cameroon and Malaysian populations. Together, these results suggest a morphological divergence in body size between native and introduced populations. However, this inference should be interpreted cautiously, as temporal data and controlled comparisons are required to confirm directional changes over time.

Variation in body size may have important functional implications for pollination. Body size is positively correlated with pollen load, as larger individuals possess greater body surface area for pollen attachment ( (Nabilah et al., 2011): (Tambunan et al., 2020)). Multiple studies on E. kamerunicus have demonstrated that larger individuals carry significantly higher pollen loads, indicating that body size is a key determinant of pollination efficiency in oil palm (Agenginardi 2011; Lesmana et al. 2014; Fikra et al. 2015; Tambunan et al. 2020b). This relationship is primarily driven by increased body surface area and the presence of setae, particularly on the elytra and thorax, which enhance pollen adherence and transport. Consequently, larger E. kamerunicus individuals may contribute more effectively to pollination.

Similar patterns have been observed in other pollinator taxa, such as bumblebees, where larger individuals carry and transfer more pollen than smaller ones (Willmer & Finlayson 2014; Neto & Schlindwein 2025). Since fruit set is influenced by pollen load, with higher pollen deposition generally increasing fruit set (Nerson 2009), a reduction in body size may potentially affect pollination effectiveness. Nevertheless, direct causal relationships between body size variation and field-level pollination outcomes require further empirical validation.

From an ecological and evolutionary perspective, variation in insect body size is shaped by both genetic factors and environmental conditions (Atkinson 1996; Chown & Gaston 2010). A previous genetic study based on the COI gene (Tambunan et al. 2020a) did not identify specific genetic mechanisms regulating body size in E. kamerunicus. Therefore, the observed size variation in Indonesia and Malaysia may be influenced by environmental factors such as climate, host plant conditions, or resource availability, rather than genetic divergence alone. Continuous adaptation to new environments can lead to shifts in morphological traits over time (Jannah 2014), although further integrative studies combining morphometric, genetic, and ecological data are needed to confirm these mechanisms.

Intraspecific body size variations of E. kamerunicus formed three distinct groups, with Cameroon population exhibiting the largest body size, Malaysia the smallest, and Indonesia intermediate values in both males and females. These results, supported by key morphometric characters such as body length (BL) and elytra length (EL), indicate morphological divergence between native and introduced populations. Notably, the difference in body length (BL) between native and introduced populations reached 0.22 mm (male) and 0.32 mm (female). This variation may have important implications for pollination efficiency, highlighting the need for further studies linking body size with ecological performance.