Skip to main content
NCBI home page
Plant Diversity logoLink to Plant Diversity
. 2026 Mar 11;48(5):906–914. doi: 10.1016/j.pld.2026.02.009

Herbivore defense syndromes differ between native and invasive plant species

Yu-Mei Sun a, Feng He a, Yi-Wen Zhao a, Curtis C Daehler b, Ragan M Callaway c, Xiao Sun a,⁎
PMCID: PMC13615792  PMID: 42801111

Abstract

In their non-native ranges, invasive plant species may escape specialist herbivores and thus experience exclusive pressure from generalists. This is predicted to shape the defense strategies of successful non-native species, as proposed by the Shifting Defense Hypothesis. However, most research to date has quantified only limited sets of defense traits of non-native invasive plant species, and thus our understanding of their defense syndromes, which encompass suites of both physical and chemical traits, remains poorly understood. We evaluated a comprehensive array of leaf physicochemical traits, including quantitative structural defenses, qualitative chemical defenses, and nutritional properties, for 15 phylogenetically matched pairs of non-native species and native congeners co-occurring in abandoned farmland in China. We coupled this trait screening with assessments of herbivore damage in a common garden and laboratory larval growth assays using a generalist herbivore (Spodoptera littoralis; Lepidoptera). Non-native species experienced 50.8% less herbivory and reduced larval growth rate by 25.7%, relative to native congeners. This increase in resistance corresponded with a distinct defense syndrome: non-native species allocated more to broad spectrum, apparently qualitative, chemical defenses, exhibiting 43.1% higher phenolic concentrations, but less to quantitative structural defenses than natives. Hierarchical partitioning revealed that phenolics were the primary predictor of lower Spodoptera growth on non-native species, whereas leaf C content was the best predictor for lower Spodoptera growth on natives. We found differences in defense syndromes and strategies between non-native species and natives, and our findings were consistent with predictions of the Shifting Defense Hypothesis.

Keywords: Invasive species, Generalist herbivore, Plant defense, Qualitative defenses, Quantitative defenses, Trichomes

Highlights

  • •

    Invasive plants increase phenolic-based chemical defenses while reducing structural investments (e.g., lignin and trichomes).

  • •

    Coordinated shifts in defense traits reveal a functional trade-off between chemical and structural defenses.

  • •

    This provides the first syndrome-level evidence that integrated defense portfolios are reshaped during plant invasions.

1. Introduction

Upon arriving in a new geographic region, non-native invasive plant species may escape from specialist herbivores (Liu and Stiling, 2006) and exhibit rapid growth, particularly in disturbed environments characterized by fluctuating resources (Davis et al., 2000; Blumenthal, 2006). Under these circumstances, the success of non-native species may be jointly determined by defenses against generalist natural enemies and ability to grow rapidly. This combination of selection pressures could favor a particular defense syndrome, defined as an association of co-varying defensive traits (Agrawal and Fishbein, 2006). The defense syndromes of non-native species may be influenced by selection pressures that promote reduced investment in specialist defenses and increased allocation to growth and competitive ability; the Evolution of Increased Competitive Ability (EICA) (Blossey and Notzold, 1995; Agrawal et al., 2015). Simultaneously, the Shifting Defense Hypothesis (SDH) predicts increased investment in defenses against generalists after a biogeographical shift from native ranges containing both generalists and specialists to non-native ranges dominated by generalists (Joshi and Vrieling, 2005; Ridenour et al., 2008; Doorduin and Vrieling, 2011; Callaway et al., 2022; Sun et al., 2023a).

Specialist herbivore defenses are generally considered quantitative (dose dependent), although quantitative defenses may also affect generalists. Examples include leaf toughness, leaf thickness, and cellulose and lignin concentrations (Liao et al., 2013; Eichenberg et al., 2015; Lin et al., 2015; Müller, 2018). In contrast, specialists are often undeterred or are even attracted to the unique secondary biochemicals that are toxic to generalists even in low doses (Müller-Schärer et al., 2004; Zvereva and Kozlov, 2016) and commonly referred to as qualitative defenses. In non-native ranges, qualitative defense strategies against generalist herbivores are thought to be most important and are predominantly plant taxon-specific toxic biochemical compounds (Müller-Schärer et al., 2004; Sun et al., 2023a). For example, non-native plant species often produce higher concentrations of alkaloids and phenolics than native plants (Sun et al., 2020). There is some evidence for non-native species escaping generalists as well (Schaffner et al., 2011), suggesting that eco-evolutionary histories among plants and generalist consumers may affect invasions. These broad patterns are important, but binary approaches to defenses and specialist vs. generalist herbivores are heuristic, as neither defense syndromes nor herbivores fall neatly into these groups (Ali and Agrawal, 2012).

In addition to chemical and broad structural defenses, microscopic leaf structures like trichomes also play a crucial, though complex, defensive role. Trichomes serve as an essential defense against herbivores (Guan et al., 2024; Sun et al., 2024), but also do not fall neatly into qualitative or quantitative components of defense syndromes. Trichomes generally serve as physical barriers and quantitative deterrents to feeding, but some trichomes are glandular and may secrete qualitative defense chemicals. Some studies have reported lower trichome density on conspecifics sourced from non-native ranges (Blair and Wolfe, 2004; Valverde et al., 2015), while other studies reported the opposite (Fortuna et al., 2014; Liao et al., 2014; Kalske et al., 2022), or no difference between conspecifics sourced from native vs. non-native ranges (Alba et al., 2011; Sotes et al., 2015). Nevertheless, because trichome development requires substantial investment in physical structure, we expect successful non-native species to invest less in trichomes, on average. Funk and Throop (2010) compared trichome density on native and non-native species and found higher variation among natives, but we know of no study that has compared trichome-based defenses among large groups of native and non-native congeners. Also, most evaluations of trichomes have focused on counts and morphology, and rarely have integrated these metrics with herbivore performance, other structural defenses, leaf biochemistry, and leaf nutritional value.

The effects of plant defense syndromes on herbivores interact with the nutritional value of different species (Strauss et al., 2002); for example, high nutritional value may reduce effectiveness of defenses (Huang et al., 2020; Lin et al., 2024). Thus, comprehensive evaluation of diverse defenses and nutritional quality of plants may shed light on the role of herbivory in invasions. Here, we use 15 pairs of non-native species and native congeners to compare herbivore damage, a suite of quantitative defenses including trichomes, an integrated set of secondary biochemicals relevant to qualitative defenses, leaf nutritional chemistry, and herbivore performance to explore whole leaf syndromes that correspond with invasion. We also explored the relative contributions of various leaf defense traits to insect resistance for non-native species. We tested the following hypotheses: compared to native species, non-native congeners will (1) experience less herbivore damage and reduce the larval weight gain of a target generalist due to greater relative investment in generalist defenses in non-native ranges, (2) produce greater concentrations of broad spectrum metabolites such as phenolics, alkaloids, and terpenoids in response to disproportionate generalist herbivory, and (3) produce weaker structural defenses, such as reduced leaf thickness, lower lignin content, and lower leaf trichome density. Integrating these components of defense syndromes should provide insight into plant–herbivore interactions and the defense strategies employed by non-native invasive plant species.

2. Materials and methods

2.1. Study system and plant growth

This study was conducted in the Henan Province, central China (34°49′N, 114°18′E). The mean annual temperature is 14°C and the mean annual precipitation is 670 mm (http://www.weather.com.cn/cityintro/101180801.shtml).

Drawing on prior field surveys, we identified the 15 most common non-native plant species, and native congeners, to serve in our experiment (Sun et al., 2023a). According to Ma and Li (2018), all non-native species in our study qualify as non-native invasive plant species. The 15 pairs of non-native plant species and their native congeners were in seven plant families: Asteraceae, Amaranthaceae, Convolvulaceae, Malvaceae, Phytolaccaceae, Solanaceae, Anacardiaceae (Table S1). In late April 2020, we sowed seeds into individual pots (19 cm diameter, 16 cm height) filled with local topsoil. Pots were placed in a greenhouse at 60% humidity, a 16/8 h light/dark cycle, and a 20–26°C day–night temperature cycle. Pots were watered once per two days to capacity. To minimize variation due to initial plant size, we selected uniform seedlings at the four-leaf stage seedlings for the following common garden survey and pot bioassays.

2.2. Herbivore damage in the common garden experiment

To assess herbivore damage on non-native plant species under natural conditions, we carried out a common garden experiment. We constructed 150 1 × 1 m plots divided into five blocks with plots 1 m apart and blocks 1.5 m apart. One plot was randomly assigned to one of the 30 plant species in each block. Two similar seedlings were planted, 50 cm apart, in each plot. Plants were exposed to herbivorous insects under natural conditions, allowing natural colonization of herbivores. We then irrigated the plots as needed using a sprinkler system, applying an equal volume of water to each plot, and weeded manually each week. Three months later, we evaluated leaf damage by identifying affected leaves exhibiting irregular holes, notches, or skeletonization.

For each individual plant, pictures were taken of one mature leaf selected randomly. The damaged leaf area was measured using ImageJ 1.54 g. There were 10 replicates for each species.

2.3. Herbivore growth

To compare the effects of non-native species and native congeners on the generalist herbivore, we measured the weight gain of larvae feeding on excised leaves (Hettenhausen et al., 2013; Sun et al., 2019). Lepidoptera species were the predominant generalist taxa associated with non-native plant species at our field sites (Sun et al., 2023a). Therefore, we selected Spodoptera littoralis (Lepidoptera: Noctuidae) as a model generalist herbivore. S. littoralis is widespread and is thought to be native to northern Africa and the greater Middle East (Shonouda and Osman, 2000; Ghoneim et al., 2020) and feeds on leaves from at least 40 plant families and now has a broad global geographical distribution (Saldamando and Marquez, 2012). We purchased S. littoralis larvae from Keyun Biological Control Co., LTD., Jiyuan, Henan Province, and fed them with soy-based artificial feed to minimize habituation to any of the plant species in our study.

We transplanted one seedling into one pot, with 75 pots for each of the 30 species. Seedlings that died were replaced within two days. All plants were cultivated in a naturally ventilated, rainproof, and semi-open shelter. Each week, we randomly relocated the pots. After plants had developed twenty leaves, herbivore bioassays were initiated. We randomly selected and harvested mature leaves and placed them in a Petri dish (9 cm diameter) on top of moist filter paper, with the adaxial surface of the leaf facing upward. A second instar caterpillar of Spodoptera littoralis was weighed (weight1) and placed in the Petri dish and the lid was sealed to prevent caterpillars from escaping. Leaves were replaced daily to maintain freshness. After four days, caterpillars were weighed again (weight2). Weight gain was calculated as caterpillar weight gain = weight2 – weight1. There were 20 replicates for each plant for a total of 600 plants used for the herbivory experiment (two nativities × 15 plant species × 20 replicates).

2.4. Leaf physicochemical traits

To correlate leaf traits with caterpillar growth, we measured a combination of leaf physicochemical traits of 15 pairs of non-native species and their native congeners.

2.4.1. Elements

When the plants had 20 leaves, we collected and dried two leaves from each of three sample plants per species at 45°C for 72 h. Dried plant leaves were pulverized to a powder. An elemental analyzer (Model 4010; Costech Analytical Valencia, CA, USA) was used to determine leaf total carbon (C) and total nitrogen (N) concentrations, and we calculated C/N ratios. In total, there were 180 plants used for the elemental measurements (C: two nativities × 15 plant species × three replicates; N: two nativities × 15 species × three replicates).

2.4.2. Primary metabolites

As above, we collected two leaves from each of three individuals per species, dried and pulverized them into powder. Soluble sugar was measured based on the anthrone colorimetric method (Wang, 2006), and soluble protein was measured with the Coomassie brilliant blue G-250 method (Deng et al., 2012). In total, there were 180 plants used for the leaf primary metabolites measurements (soluble sugar: two nativities × 15 plant species × three replicates; soluble protein: two nativities × 15 species × three replicates).

2.4.3. Secondary metabolites

As above, we collected two leaves from each of three individuals per species. First, freeze-dried samples were pulverized using a mixer mill at a frequency of 60 Hz for 1 min. Next, 10 mg of each sample was weighed and transferred into a 2 mL Eppendorf tube containing 1.0 mL of extraction solution (methanol/water in a 3:1 ratio, pre-chilled to −40°C and containing an internal standard). Subsequently, the samples were homogenized at 35 Hz for 4 min and then sonicated for 15 min in an ice-water bath after a 30-s vortex. After that, the samples were extracted overnight at 4°C on a shaker and then centrifuged at 12,000 rpm for 15 min, also at 4°C. The supernatant was meticulously filtered through a 0.22-micropore membrane. The obtained supernatants were diluted 10-fold with a methanol/water mixture (volume ratio 3:1, containing an internal standard), vortexed for 30 s, and transferred to 2 mL glass vials. Finally, 20 μL of each sample was mixed to form QC samples, which were stored at −80°C until analyzed by ultra-high-performance liquid chromatography-mass spectrometry (UHPLC–MS).

The UHPL separation was implemented utilizing an EXIONLC System (manufactured by Sciex). In this process, the mobile phase A consisted of 0.1% formic acid dissolved in water, while the mobile phase B was pure acetonitrile. The column temperature was maintained at 40°C, and the auto-sampler temperature was set to 4°C. A 2 μL injection volume was adopted. SCIEX Analyst Work Station Software (v.1.6.3) was utilized for the acquisition and processing of multiple reaction monitoring (MRM) data. Metabolomic determination and a series of preparations and organization protocols with the raw data preprocessing were performed by Biotree Biotech (Shanghai, China). Finally, we classified the measured substances into total phenolics, total alkaloids, and total terpenoids based on their groups (Dong et al., 2009).

2.4.4. Quantitative defenses

As above, we collected two leaves from each of three individuals per species. Leaf cellulose content was determined using the anthrone colorimetric method, and leaf lignin content was determined by sulfuric acid hydrolysis-titration method (Xiong et al., 2005; Wang, 2006). We randomly chose one mature leaf from each of 10 individuals per species for measurements of leaf thickness using a vernier caliper. We measured each leaf three times avoiding the veins and calculated the average value. We collected one leaf from each of 12 individuals per species to assess leaf toughness. We used a dynamometer Chatillon (AMETEK, San Diego, California) to measure leaf toughness avoiding the leaf veins. The force of the dynamometer required to puncture the leaves represented leaf toughness. In total, there were 840 plants used for measurements of leaf quantitative defenses (leaf cellulose: two nativities × 15 species × three replicates; leaf lignin: two nativities × 15 species × three replicates; leaf thickness: two nativities × 15 species × 10 replicates; leaf toughness: two nativities × 15 species × 12 replicates).

2.4.5. Leaf trichome density

Since trichomes can contribute to both quantitative and qualitative defense, we present their analysis in a separate section. For each species, we collected the third mature leaf from the top from each of three similar-sized plants. Leaf trichome densities were assessed on both the adaxial and abaxial surfaces of each leaf. Three 1-cm2 pieces were cut from each leaf on each side of the midrib at 2/3 of the distance from the attachment of the petiole to the leaf tip. The leaf midrib was avoided. The sample was fixed onto a microscope stage using double-sided adhesive tape, with both surfaces of the leaf exposed and then placed under a Miniscope TM 3000 (Hitachi High-Technologies, Tokyo, Japan). Under a low-vacuum mode, scanning electron micrographs were acquired using an accelerating voltage of 5 kV and a magnification range of 100–500×, with an image resolution of 1280 × 960 pixels; five microscopic fields were randomly selected for each piece. Finally, we calculated the average leaf trichome density for each individual.

2.5. Statistical analysis

2.5.1. Herbivore damage in the common garden experiment

To assess differences in damaged leaf area between non-native and native species, we applied a Wald Chi-square test to a zero-inflated beta generalized linear mixed model (GLMM), with origin (non-natives vs. natives) as a fixed effect in both the conditional and zero-inflation components. Species identity nested within native-non-native pairs and experimental block were included as random effects to account for non-independence with paired comparisons and heterogeneity among blocks. The zero-inflation component accommodated the excess of zeros (i.e., individuals that experienced no herbivory).

2.5.2. Herbivore growth in laboratory experiments

To test the differences in herbivore (Spodoptera littoralis caterpillars) growth when consuming non-native vs. native species, we used a Wald Chi-square test applied to a linear mixed model (LMM) that included origin as a fixed effect, species and native-non-native pairs as random effects. To explore the relationship between damaged leaf area and caterpillar growth on non-native species vs. native congeners, we used a linear regression model to examine the relationship between the damaged leaf area and the mass of caterpillars using mean values per species. Data were natural-log-transformed to improve homoscedasticity and normality of residuals.

2.5.3. Leaf physicochemical traits

To test differences in leaf physicochemical traits of non-native species and native congeners, we conducted LMMs to test leaf C, N, total phenolics, total terpenoids, leaf trichome densities, soluble sugars, cellulose and thickness with origin as a fixed effect, and species and native-non-native pair as random effects. Data were natural-log-transformed, or inv-transformed to improve homoscedasticity and normality of residuals. To evaluate differences in leaf C/N ratio, total alkaloids, lignin, toughness and soluble protein between 15 pairs of non-native species and native congeners, we fitted generalized linear mixed-effects models (GLMM) for each trait according to its distributional characteristics. Origin (non-native vs. native) was included as a fixed effect, and species identity and native-non-native pairs were included as random effects.

2.5.4. Relationships between leaf traits and herbivore growth

To further explore the relationships between leaf physicochemical traits and the growth of S. littoralis, we employed a generalized linear model (GLM) with a gaussian distribution to evaluate the effects of leaf physicochemical properties on the caterpillar growth of non-native or native species, using mean values per species. In regression analyses, leaf chemical and structural traits—including leaf total phenolics, total alkaloids, C, N, lignin content, and abaxial trichome density—were included as fixed-effect predictors. Model diagnostics were performed using variance inflation factors (VIF) to assess multicollinearity, and residual diagnostics were evaluated using the “simulateResiduals” function to ensure model assumptions were met. Furthermore, hierarchical partitioning analysis using the “glmm. hp” package was used to quantify the independent contribution of each predictor variable to variation in herbivore performance metrics. Data were natural-log-transformed to improve homoscedasticity and normality of residuals.

The DHARMa package was used for diagnostic checks of all models to evaluate residual behavior and detect any violations of model assumptions. We performed all analyses using the software R 4.5.0 (R Development Core Team 2017) with the “lme4” (Bates et al., 2014), “car” (Fox and Weisberg, 2011), “DHARMa” (Hartig, 2023), “glmmTMB” (Brooks et al., 2017), “glmm.hp” (Lai et al., 2022) and “MuMIn” (Bartoń, 2023) packages.

3. Results

3.1. Herbivore damage, herbivore growth, and leaf traits

In the common garden, herbivore damage on non-native species was 50.8% less than on native congeners (χ2 = 9.12, p = 0.003, Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Differences in the damaged leaf area of leaves on non-native species and native congeners in the common garden. Boxplots represent interquartile ranges, the solid lines in the box represent the median, and error bars show the min–max value calculated as the 1.5× quartile range. Dots represent raw data points; black circles represent species means. ∗∗p < 0.01.

In experiments, the mass of Spodoptera littoralis caterpillars feeding on excised leaves of non-native species was 25.7% lower than that of those feeding on native congeners (χ2 = 5.23, p = 0.022, Fig. 2a). The mass of caterpillars grown on excised leaves from non-native species was correlated with the degree of herbivore damage on the same species in the common garden (R2 = 0.25, p = 0.049), whereas there was no significant correlation for native congeners (R2 = 0.05, p = 0.430) (Fig. 2b).

Fig. 2.

Fig. 2

Open in a new tab

Differences in the growth of Spodoptera littoralis larvae on leaves of non-native species and native congeners (a). Boxplots represent interquartile ranges, the solid lines in the box represent the median, and error bars show the min–max value calculated as the 1.5× quartile range. Dots represent raw data points; black circles indicate species means. Relationship between the percentage of damaged leaf area in the common garden experiment and the S. littoralis larval weight gain on leaves of non-native species and native congeners (b). Each point represents individual non-native species (blue points) and their native congeners (red points). Regression lines are shown only for significant relationships, with shaded areas representing 95% confidence intervals. The blue line indicates significant linear regression relationships across non-native plant species. ∗p < 0.05.

For elements, C in leaves of non-native species was 5.5% higher (χ2 = 9.00, p = 0.003) and N was 18.4% lower in non-native species than in native congeners (χ2 = 4.48, p = 0.034). Thus, the C/N ratio was 29.7% higher in the leaves of non-natives compared to their native congeners (χ2 = 5.60, p = 0.018) (Fig. 3a).

Fig. 3.

Fig. 3

Open in a new tab

Differences in leaf physicochemical traits of non-native species and native congeners. Differences in total carbon, total nitrogen, carbon to nitrogen ratio (C/N), soluble protein, soluble sugar, total alkaloids, total phenolics, and total terpenoids on the leaves between non-native species and native congeners (a). Differences in leaf cellulose, lignin, thickness, toughness, trichome intensity on both surfaces, adaxial surface, and abaxial surface between non-native species and native congeners (b). The blue dots in solid represent the means of the leaf traits of non-native species, and the red dots in solid represent the means of the leaf traits of native species. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

For primary metabolites, there was no difference in soluble protein (χ2 = 0.42, p = 0.515) or soluble sugars (χ2 = 1.26, p = 0.262) between non-natives and natives (Fig. 3a).

For secondary biochemistry, total phenolic concentrations were 43.1% higher (χ2 = 8.53, p = 0.003) and total alkaloid concentrations were 8.6% lower (χ2 = 4.04, p = 0.044) in leaves of non-native species than native congeners. Total terpenoid concentrations did not differ between natives and non-natives (χ2 = 1.70, p = 0.193) (Fig. 3a).

For quantitative structural defenses, lignin content in the leaves of non-native species was 14.9% lower than that of natives (χ2 = 4.87, p = 0.027). Non-native species showed no significant differences from natives in leaf toughness (χ2 = 1.38, p = 0.240), cellulose content (χ2 = 0.84, p = 0.360), or thickness (χ2 = 0.70, p = 0.401) (Fig. 3b).

There were no differences in total trichomes (χ2 = 3.32, p = 0.068) or trichome densities (χ2 = 1.35, p = 0.246) on the adaxial leaf surfaces. However, trichome densities on the abaxial leaf surfaces of non-native species were 43.4% lower than those of native congeners (χ2 = 3.98, p = 0.046) (Fig. 3b).

3.2. Relationships between leaf traits and herbivore growth

Multiple linear regression and the associated variance partitioning analysis indicated that the combined effects of total phenolics, total alkaloids, C, N, lignin, and abaxial trichomes of leaves accounted for 33.1% of the variation in Spodoptera littoralis caterpillar mass among non-native plant species. Furthermore, the regression model identified total phenolics as a significant negative predictor of caterpillar mass for non-native species (parameter estimates = −0.26, 95% CI = [−0.5136, −0.0070], p = 0.044) (Fig. 4a).

Fig. 4.

Fig. 4

Open in a new tab

Effects of total phenolics, total alkaloids, total carbon, total nitrogen, lignin and abaxial trichomes of leaves on Spodopteralittoralis larval weight gain when grown on leaves of non-native species (a) vs. on leaves of native congeners (b). Colors in the bars show the relative importance of each predictor, expressed as the percentage of explained variance, and symbols to the right of the bars show the average parameter estimates (standardized regression coefficients) of the model’s predictors and their 95% confidence intervals. R2 of the averaged model and the p value of each predictor are presented as: ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

For native congeners, leaf traits explained 58.2% of the variation in caterpillar mass, with leaf C emerging as a significant negative predictor of caterpillar mass for the native species (parameter estimates = −0.24, 95% CI = [−0.3872, −0.0848], p = 0.002) (Fig. 4b).

4. Discussion

Our findings are consistent with the idea that non-native invasive plant species have defense syndromes that are different than native congeners. By integrating generalist herbivore performance with a comprehensive suite of leaf traits, we show a functional trade-off: the enhancement of phenolic-based chemical defense associated with a reduction of investment in structural components including lignin and trichomes (Fig. 5). These results suggest a shift away from defenses targeting specialists to those targeting generalists and corresponding with rapid growth. However, since the species in our study were from highly disturbed habitats, the full spectrum of potential non-native species life history strategies may not be well represented. Thus, non-native species from less disturbed systems might not exhibit similar defense syndromes. That said, our results are consistent with the SDH (Doorduin and Vrieling, 2011), although we cannot exclude the possibility that the success of non-native species reflects pre-adaptation or plasticity in the expression of a defensive traits. Non-native species defense syndromes enhanced protection from generalists through greater broad-spectrum chemical defenses, while promoting faster growth by reducing investment in more costly (in the short term) physical defenses that may better deter specialists.

Fig. 5.

Fig. 5

Open in a new tab

A conceptual overview of leaf physicochemical traits in defense against herbivores between non-native plant species and native congeners based on difference analysis and hierarchical partitioning analysis. Blue dots with + indicate leaf traits that are higher for the non-native species and blue dots with – indicate leaf traits are higher for the native species. The blue irregular polygon represents the contribution of the leaf traits of non-native species to experimental herbivore growth, and the red irregular polygon indicates the contribution of leaf traits of native species to herbivore growth.

Non-native species had lower values for traits generally associated with quantitative defense primarily against specialists – leaf lignin and trichome density, and one trait often associated with qualitative defense against generalists – phenolic concentrations – was higher for non-native species. However, some phenolics appear to defend against specialists and the roles of phenolics are not always clear (Müller-Schärer et al., 2004; Massad et al., 2011). The lower weight gain of the generalist herbivore on non-native species observed in our experiments is consistent with the observed differences in defense traits between native and non-native species. This pattern observed in non-native species may reflect the combined effects of constitutive and induced defense mechanisms.

Not all defensive traits differed among non-native and native species in ways that were consistent with the SDH. Alkaloids, a key group of qualitative defense chemicals (Wei et al., 2015; Fridley et al., 2025), were actually higher in natives, although the percent change was less than 10%, so it is not clear whether this difference is ecologically meaningful. Terpenoids, another group of qualitative defenses (Eilers and Heger, 2019), did not differ among native and non-native species. These findings might be explained by the multifunctionality of alkaloids and terpenoids. Besides deterring herbivores, they certain alkaloids can function in nitrogen storage and turnover (Bhambhani et al., 2021), and both chemicals may also be allelopathic, affecting plant–plant and plant–soil interactions (Wink et al., 2023; Luo et al., 2024; Bo et al., 2025). In new biogeographical ranges —where competitors and soil biota may be novel —selection may favor metabolic investment in these compounds for functions beyond or alongside herbivore defense (Ridenour et al., 2008). This functional flexibility may decouple their expression from predictions based solely on herbivore pressure, helping explain the observed patterns. Leaf toughness and thickness, two traits associated with quantitative defenses (Hanley et al., 2007), also did not differ among native and non-native species. Even though not all measured defenses fit predictions, when they were considered as defense complexes or syndromes (Agrawal and Fishbein, 2006; Agrawal, 2007), the overall native-non-native differences aligned with the SDH.

C and N contents in plant tissues influence insect growth and development (Kurokawa et al., 2010). Our results showed non-native species had higher C concentrations, lower N concentrations, and higher C/N ratios than native congeners, which independently of other factors should make non-native plant species less nutritious (Cronin et al., 2015; Sun et al., 2020). High C/N ratios in plant tissues are thought to reduce the fitness of herbivores (Awmack and Leather, 2002). Interestingly, our results indicated that leaf C content was the key predictor of herbivory in natives. Mechanistically, the lower C content in native species might reflect reduced need for structural and/or defensive carbon compounds, such as lignin and terpenes which have higher energetic costs (Coley et al., 1985; Osunkoya et al., 2008), leading to tissues that are more palatable and digestible for generalist herbivores (Awmack and Leather, 2002). This aligns with the observed pattern of higher herbivore consumption on natives. Notably, our results contrast with some previous reports of lower C/N ratios in non-native species (Kurokawa et al., 2010) or in invasive genotypes (Cronin et al., 2015), but these studies do not involve paired native and non-native congeners, which might account for the differing pattern. Our results for leaf C and N content should be interpreted in the context of the entire array of defense traits we measured. Non-native species that escape specialist herbivores may produce qualitative carbon-based chemical defenses to resist generalists in new habitats, as observed in our study. Thus, a high C/N ratio may be an indication of a strong defense syndrome tuned to generalist herbivores.

Plant trichomes are an important line of defense against herbivorous insects (Andama et al., 2020; Kaur and Kariyat, 2023). However, their roles are complex as they both produce chemicals and provide physical protection. Therefore, different selection pressures work on their number, form, and function (Goldberg et al., 2020; Johnson and Baucom, 2023). In the absence of competition and herbivores, Sun et al. (2023b) found that the invasive plant Alternanthera philoxeroides had more leaf trichomes than native plants. In our case, we found a decreased leaf trichome density on the abaxial leaf surface of the non-native species, indicating structural defense is likely decreased in non-native species. These results differ from those reported by Ridenour et al. (2008) but are consistent with the EICA hypothesis.

An important limitation of our results for defense differences between non-native and native species is that they are not biogeographical (Hierro et al., 2005); we did not compare conspecifics from the native and non-native ranges in common conditions. Thus, we did not measure changes in conspecifics in non-native ranges. Instead, we sought evidence of defense syndromes for non-native invasive plants, i.e., different patterns of defenses between non-native species and native congeners. While the observed cross-species differences align with selection on defense syndromes, they are also consistent with pre-adaptation or plasticity. Therefore, our results reveal systematic differences in defense profiles between non-native and native species, but do not provide direct evidence for post-introduction evolutionary changes. These patterns may alternatively reflect inherent phylogenetic differences, pre-adaptation, or plastic trait expression under common environmental conditions. A second limitation is that we focused on a single generalist herbivore species, Spodoptera littoralis. Although we found statistical relationships between success of the generalist and specific defense traits in our non-native and native plant species, we cannot be sure that identical relationships would be found for other generalist herbivores. We acknowledge that such relationships may be herbivore-specific. Therefore, future studies should incorporate native generalists to evaluate the broader applicability and generality of our findings. Regarding trichome quantification, we did not classify the very wide range of trichome types in our native-to-non-native comparisons. Different trichome types can have very different effects on herbivores. Thus, in future multi-species studies, categorizing different trichome types might provide insightful relationships between trichomes and herbivore damage. Last, an important methodological consideration concerns the initial size of the study plants. Although initial seedling sizes in our experiment were estimated to be very small and relatively uniform across seedlings, we acknowledge that variability in initial biomass could contribute to differences in subsequent traits. Including initial plant size as a covariate in statistical models would enhance the robustness of future analyses.

In conclusion, our multi-trait, syndrome-level analysis demonstrates that the successful defense of disturbance-adapted non-native invasive plant species against herbivores does not appear to be merely a matter of strengthening any single defense, but instead is characterized by a complex resource allocation syndrome. The observed native–non–native contrasts are characterized by a functional rebalancing in which investment appears to shift from structural components toward phenolic-based chemical defenses. This pattern advances the SDH by illustrating a quantifiable rebalancing among an integrated syndrome. Thus, our findings underscore the importance of suites of defense traits among disturbance-adapted non-native species.

CRediT authorship contribution statement

Yumei Sun: Writing – original draft, Visualization, Investigation, Formal analysis, Data curation. Feng He: Data curation. Yiwen Zhao: Investigation. Curtis C. Daehler: Conceptual development, writing – review & editing. Ragan M. Callaway: Conceptual development, writing – review & editing. Xiao Sun: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

Data availability statement

All data supporting the findings of this study are publicly accessible in the figshare (https://doi.org/10.6084/m9.figshare.31085551).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We are grateful to Xiang Zhao for his invaluable contributions to the methodology and theoretical development of this work. We thank Jianqing Ding for his invaluable guidance and insightful suggestions regarding the manuscript. We also thank the Undergraduate-Graduate Research Group for conducting the chemical analyses. This work was supported by the National Key Research and Development Program (2023YFE0122100), National Natural Science Foundation of China (32371752), and Outstanding Young Scientists Fund of Henan Province (232300421027). The authors utilized DeepSeek for grammar checking and some vocabulary in this manuscript.

Footnotes

Peer review under the responsibility of Editorial Office of Plant Diversity.

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pld.2026.02.009.

Contributor Information

Yu-Mei Sun, Email: sunyumei@henu.edu.cn.

Feng He, Email: hefeng0331@henu.edu.cn.

Yi-Wen Zhao, Email: 17737491986@163.com.

Curtis C. Daehler, Email: daehler@hawaii.edu.

Ragan M. Callaway, Email: ray.callaway@mso.umt.edu.

Xiao Sun, Email: sunxiao@henu.edu.cn.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Multimedia component 1
mmc1.docx (16.8KB, docx)

References

  1. Agrawal A.A. Macroevolution of plant defense strategies. Trends Ecol. Evol. 2007;22:103–109. doi: 10.1016/j.tree.2006.10.012. [DOI] [PubMed] [Google Scholar]
  2. Agrawal A.A., Fishbein M. Plant defense syndromes. Ecology. 2006;87:S132–S149. doi: 10.1890/0012-9658(2006)87[132:pds]2.0.co;2. [DOI] [PubMed] [Google Scholar]
  3. Agrawal A.A., Hastings A.P., Bradburd G.S., et al. Evolution of plant growth and defense in a continental introduction. Am. Nat. 2015;186:E1–E15. doi: 10.1086/681622. [DOI] [PubMed] [Google Scholar]
  4. Alba C., Bowers M.D., Blumenthal D., et al. Evolution of growth but not structural or chemical defense in Verbascum thapsus (common mullein) following introduction to North America. Biol. Invasions. 2011;13:2379–2389. [Google Scholar]
  5. Ali J.G., Agrawal A.A. Specialist versus generalist insect herbivores and plant defense. Trends Plant Sci. 2012;17:293–302. doi: 10.1016/j.tplants.2012.02.006. [DOI] [PubMed] [Google Scholar]
  6. Andama J.B., Mujiono K., Hojo Y., et al. Nonglandular silicified trichomes are essential for rice defense against chewing herbivores. Plant Cell Environ. 2020;43:2019–2032. doi: 10.1111/pce.13775. [DOI] [PubMed] [Google Scholar]
  7. Awmack C.S., Leather S.R. Host plant quality and fecundity in herbivorous insects. Annu. Rev. Entomol. 2002;47:817–844. doi: 10.1146/annurev.ento.47.091201.145300. [DOI] [PubMed] [Google Scholar]
  8. Bartoń K. MuMIn: multi-model inference. R package version 1.47.5. 2023 [Google Scholar]
  9. Bates D., Mächler M., Bolker B., et al. Fitting linear mixed-effects models using lme4. Stat. Comput. arXiv. 2014;1406:133–199. [Google Scholar]
  10. Bhambhani S., Kondhare K.R., Giri A.P. Diversity in chemical structures and biological properties of plant alkaloids. Molecules. 2021;26:3374. doi: 10.3390/molecules26113374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Blair A.C., Wolfe L.M. The evolution of an invasive plant: an experimental study with Silene latifolia. Ecology. 2004;85:3035–3042. [Google Scholar]
  12. Blossey B., Notzold R. Evolution of increased competitive ability in invasive nonindigenous plants: a hypothesis. J. Ecol. 1995;83:887–889. [Google Scholar]
  13. Blumenthal D.M. Interactions between resource availability and enemy release in plant invasion. Ecol. Lett. 2006;9:887–895. doi: 10.1111/j.1461-0248.2006.00934.x. [DOI] [PubMed] [Google Scholar]
  14. Bo Y., Liao Y., Pawlett M., et al. Solidago canadensis modifies microbial community and soil physicochemical properties through litter leachates and root exudates. J. Plant Ecol. 2025;18:rtaf017. [Google Scholar]
  15. Brooks M.E., Kristensen K., Van Benthem K.J., et al. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R J. 2017;9:378–400. [Google Scholar]
  16. Callaway R.M., Lucero J.E., Hierro J.L., et al. The EICA is dead? Long live the EICA. Ecol. Lett. 2022;25:2289–2302. doi: 10.1111/ele.14088. [DOI] [PubMed] [Google Scholar]
  17. Coley P.D., Bryant J.P., Chapin III F.S. Resource availability and plant antiherbivore defense. Science. 1985;230:895–899. doi: 10.1126/science.230.4728.895. [DOI] [PubMed] [Google Scholar]
  18. Cronin J.T., Bhattarai G.P., Allen W.J., et al. Biogeography of a plant invasion: plant–herbivore interactions. Ecology. 2015;96:1115–1127. doi: 10.1890/14-1091.1. [DOI] [PubMed] [Google Scholar]
  19. Davis M.A., Grime J.P., Thompson K. Fluctuating resources in plant communities: a general theory of invasibility. J. Ecol. 2000;88:528–534. [Google Scholar]
  20. Deng L., Pan X., Sheng J., et al. Optimization of conditions for determining trace - amount soluble protein content in apple tissue by Coomassie Brilliant Blue method. Food Sci. 2012;33:185–189. [Google Scholar]
  21. Dong J.E., Zhang K.J., Liang Z. Northwest A&F University Press; 2009. Plant Secondary Metabolism and Its Regulation. [Google Scholar]
  22. Doorduin L.J., Vrieling K. A review of the phytochemical support for the shifting defence hypothesis. Phytochem. Rev. 2011;10:99–106. doi: 10.1007/s11101-010-9195-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Eichenberg D., Purschke O., Ristok C., et al. Trade-offs between physical and chemical carbon-based leaf defence: of intraspecific variation and trait evolution. J. Ecol. 2015;103:1667–1679. [Google Scholar]
  24. Eilers E.J., Heger T. Past competition affects offspring foliar terpenoid concentrations, seed traits, and fitness in the invasive forb Erodium cicutarium (Geraniaceae) Front. Ecol. Evol. 2019;7:392. [Google Scholar]
  25. Fortuna T.M., Eckert S., Harvey J.A., et al. Variation in plant defences among populations of a range-expanding plant: consequences for trophic interactions. New Phytol. 2014;204:989–999. doi: 10.1111/nph.12983. [DOI] [PubMed] [Google Scholar]
  26. Fox J., Weisberg S. An R Companion to Applied Regression. second ed. SAGE Publications; Thousand Oaks, CA, USA: 2011. [Google Scholar]
  27. Fridley J., Griffin-Nolan R.J., Bensaddek L., et al. Joining the locals: plant invaders shift leaf defenses to match native neighbors. Ecology. 2025;106 doi: 10.1002/ecy.70129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Funk J.L., Throop H.L. Enemy release and plant invasion: patterns of defensive traits and leaf damage in Hawaii. Oecologia. 2010;162:815–823. doi: 10.1007/s00442-009-1497-4. [DOI] [PubMed] [Google Scholar]
  29. Ghoneim K., Hamadah K., Waheeb H. Bioefficacy of Farnesol, a common sesquiterpene, on the survival, growth, development, and morphogenesis of Spodoptera littoralis (Lepidoptera: Noctuidae). Egypt. Acad. J. Biolog. Sci. (F. Toxicol. Pest Contr.) 2020;12:71–99. [Google Scholar]
  30. Goldberg J.K., Lively C.M., Sternlieb S.R., et al. Herbivore-mediated negative frequency-dependent selection underlies a trichome dimorphism in nature. Evol. Lett. 2020;4:83–90. doi: 10.1002/evl3.157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Guan Y., Jiang L., Wang Y., et al. CmMYC2–CmMYBML1 module orchestrates the resistance to herbivory by synchronously regulating the trichome development and constitutive terpene biosynthesis in Chrysanthemum. New Phytol. 2024;244:914–933. doi: 10.1111/nph.20081. [DOI] [PubMed] [Google Scholar]
  32. Hanley M.E., Lamont B.B., Fairbanks M.M., et al. Plant structural traits and their role in anti-herbivore defence. Perspect. Plant Ecol. 2007;8:157–178. [Google Scholar]
  33. Hartig F. DHARMa: residual diagnostics for hierarchical (multi-level/mixed) regression models. R package version 0.4.7. 2023 [Google Scholar]
  34. Hettenhausen C., Baldwin I.T., Wu J. Nicotiana attenuata MPK 4 suppresses a novel jasmonic acid (JA) signaling-independent defense pathway against the specialist insect Manduca sexta, but is not required for the resistance to the generalist Spodoptera littoralis. New Phytol. 2013;199:787–799. doi: 10.1111/nph.12312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Hierro J.L., Maron J.L., Callaway R.M. A biogeographical approach to plant invasions: the importance of studying exotics in their introduced and native range. J. Ecol. 2005;93:5–15. [Google Scholar]
  36. Huang K., Kong D.-L., Lu X.-R., et al. Lesser leaf herbivore damage and structural defense and greater nutrient concentrations for invasive alien plants: evidence from 47 pairs of invasive and non-invasive plants. Sci. Total Environ. 2020;723 doi: 10.1016/j.scitotenv.2020.137829. [DOI] [PubMed] [Google Scholar]
  37. Johnson N., Baucom R. 2023. Building a Barrier: Investigating the Defensive Role of Induced Trichome Production in Herbicide Exposed Plants. [Google Scholar]
  38. Joshi J., Vrieling K. The enemy release and EICA hypothesis revisited: incorporating the fundamental difference between specialist and generalist herbivores. Ecol. Lett. 2005;8:704–714. [Google Scholar]
  39. Kalske A., Luntamo N., Salminen J.-P., et al. Introduced populations of the garden lupine are adapted to local generalist snails but have lost alkaloid diversity. Biol. Invasions. 2022;24:51–65. [Google Scholar]
  40. Kaur I., Kariyat R. Trichomes mediate plant–herbivore interactions in two Cucurbitaceae species through pre-and post-ingestive ways. J. Pest. Sci. 2023;96:1077–1089. doi: 10.1007/s10340-023-01611-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Kurokawa H., Peltzer D.A., Wardle D.A. Plant traits, leaf palatability and litter decomposability for co-occurring woody species differing in invasion status and nitrogen fixation ability. Funct. Ecol. 2010;24:513–523. [Google Scholar]
  42. Lai J., Zou Y., Zhang S., et al. glmm.hp: an R package for computing individual effect of predictors in generalized linear mixed models. J. Plant Ecol. 2022;15:1302–1307. [Google Scholar]
  43. Liao Z.-Y., Zhang R., Barclay G.F., et al. Differences in competitive ability between plants from nonnative and native populations of a tropical invader relates to adaptive responses in abiotic and biotic environments. PLoS One. 2013;8 doi: 10.1371/journal.pone.0071767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Liao Z.-Y., Zheng Y.-L., Lei Y.-B., et al. Evolutionary increases in defense during a biological invasion. Oecologia. 2014;174:1205–1214. doi: 10.1007/s00442-013-2852-z. [DOI] [PubMed] [Google Scholar]
  45. Lin T., Doorduin L., Temme A., et al. Enemies lost: parallel evolution in structural defense and tolerance to herbivory of invasive Jacobaea vulgaris. Biol. Invasions. 2015;17:2339–2355. [Google Scholar]
  46. Lin T., Wang Z., Ke M., et al. Simulated nitrogen deposition increased herbivore resistance and competitive ability of an invasive plant species over a native co-occurring congeneric species. Plant Soil. 2024;505:795–810. [Google Scholar]
  47. Liu H., Stiling P. Testing the enemy release hypothesis: a review and meta-analysis. Biol. Invasions. 2006;8:1535–1545. [Google Scholar]
  48. Luo S.-H., Hua J., Liu Y., et al. The chemical ecology of plant natural products. Prog. Chem. Org. Nat. Prod. 2024;124:57–183. doi: 10.1007/978-3-031-59567-7_2. [DOI] [PubMed] [Google Scholar]
  49. Ma J., Li H. Higher Education Press; Beijing: 2018. The Checklist of the Alien Invasive Plants in China. [Google Scholar]
  50. Massad T.J., Fincher R.M., Smilanich A.M., et al. A quantitative evaluation of major plant defense hypotheses, nature versus nurture, and chemistry versus ants. Arthropod-Plant Inte. 2011;5:125–139. [Google Scholar]
  51. Müller-Schärer H., Schaffner U., Steinger T. Evolution in invasive plants: implications for biological control. Trends Ecol. Evol. 2004;19:417–422. doi: 10.1016/j.tree.2004.05.010. [DOI] [PubMed] [Google Scholar]
  52. Müller C. 2018. Evolution of Increased Competitive Ability and Shifting Defence Hypotheses, Invasion Biology: Hypotheses and Evidence; pp. 103–123. [Google Scholar]
  53. Osunkoya O.O., Daud S.D., Wimmer F.L. Longevity, lignin content and construction cost of the assimilatory organs of Nepenthes species. Ann. Bot. 2008;102:845–853. doi: 10.1093/aob/mcn162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Ridenour W.M., Vivanco J.M., Feng Y., et al. No evidence for trade-offs: Centaurea plants from America are better competitors and defenders. Ecol. Monogr. 2008;78:369–386. [Google Scholar]
  55. Saldamando C.I., Marquez E.J. Approach to Spodoptera (Lepidoptera: Noctuidae) phylogeny based on the sequence of the cytocrhome oxydase I (COI) mitochondrial gene. Rev. Biol. Trop. 2012;60:1237–1248. [PubMed] [Google Scholar]
  56. Schaffner U., Ridenour W.M., Wolf V.C., et al. Plant invasions, generalist herbivores, and novel defense weapons. Ecology. 2011;92:829–835. doi: 10.1890/10-1230.1. [DOI] [PubMed] [Google Scholar]
  57. Shonouda M., Osman S. New botanical derivatives, used in medicinal reparations, showing bioactive action on insect pests. Ⅰ-Toxicological effect on the development of Spodoptera littoralis boisd. J. Egypt. Ger. Soc. Zool. 2000;31:227–234. [Google Scholar]
  58. Sotes G.J., Cavieres L.A., Montesinos D., et al. Inter-regional variation on leaf surface defenses in native and non-native Centaurea solstitialis plants. Biochem. Syst. Ecol. 2015;62:208–218. [Google Scholar]
  59. Strauss S.Y., Rudgers J.A., Lau J.A., et al. Direct and ecological costs of resistance to herbivory. Trends Ecol. Evol. 2002;17:278–285. [Google Scholar]
  60. Sun C., Wei J., Gu X., et al. Different multicellular trichome types coordinate herbivore mechanosensing and defense in tomato. Plant Cell. 2024;36:4952–4969. doi: 10.1093/plcell/koae269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Sun K.K., Yu W.S., Jiang J.J., et al. Mismatches between the resources for adult herbivores and their offspring suggest invasive Spartina alterniflora is an ecological trap. J. Ecol. 2020;108:719–732. [Google Scholar]
  62. Sun X., Siemann E., Liu Z., et al. Root-feeding larvae increase their performance by inducing leaf volatiles that attract above-ground conspecific adults. J. Ecol. 2019;107:2713–2723. [Google Scholar]
  63. Sun X., Sun Y., Cao X., et al. Trade-offs in non-native plant herbivore defences enhance performance. Ecol. Lett. 2023;26:1584–1596. doi: 10.1111/ele.14283. [DOI] [PubMed] [Google Scholar]
  64. Sun Y., Pan Y., Liu M., et al. The specialist-to-generalist ratio affects growth and defense strategy of invasive plant Alternanthera philoxeroides. Biodivers. Sci. 2023;31 [Google Scholar]
  65. Valverde P.L., Arroyo J., Núñez-Farfán J., et al. Natural selection on plant resistance to herbivores in the native and introduced range. AoB Plant. 2015;7:1–13. doi: 10.1093/aobpla/plv090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Wang X. Higher Education Press; Beijing: 2006. Experimental Principles and Techniques of Plant Physiology and Biochemistry. [Google Scholar]
  67. Wei X., Vrieling K., Mulder P.P., et al. Testing the generalist-specialist dilemma: the role of pyrrolizidine alkaloids in resistance to invertebrate herbivores in Jacobaea species. J. Chem. Ecol. 2015;41:159–167. doi: 10.1007/s10886-015-0551-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Wink M., Latz-Brüning B., Schmeller T. Biochemical effects of allelopathic alkaloids. Prin. Prac. Plant Ecol. 2023:411–422. CRC Press. [Google Scholar]
  69. Xiong S., Zuo X., Zhu Y. Determination of cellulose, hemicellulose and lignin in rice hull. Cereal Feed Ind. 2005;8:40–41. [Google Scholar]
  70. Zvereva E.L., Kozlov M.V. The costs and effectiveness of chemical defenses in herbivorous insects: a meta-analysis. Ecol. Monogr. 2016;86:107–124. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1
mmc1.docx (16.8KB, docx)

Data Availability Statement

All data supporting the findings of this study are publicly accessible in the figshare (https://doi.org/10.6084/m9.figshare.31085551).

Articles from Plant Diversity are provided here courtesy of Kunming Institute of Botany, Chinese Academy of Sciences

RESOURCES