The tribe Inuleae is widely distributed within Asteraceae and exhibits considerable morphological variation, which complicates species classification. Pollen morphology provides relatively stable features for species delimitation, yet comprehensive palynological data for many species remain limited. In this study, the pollen morphology of 19 species from eight genera of Inuleae was investigated using light microscopy (LM) and scanning electron microscopy (SEM), with a focus on 10 quantitative traits. Multivariate analyses, including principal component analysis (PCA) and hierarchical clustering analysis (HCA), showed that both pollen size and exine ornamentation contribute to interspecific differentiation. Differences were especially notable between Blumea and Carpesium, with the latter showing larger pollen grains and more slender, scattered spines. This research also presents first-time palynological descriptions of Blumea, Carpesium, Inula, Laggera, Pentanema, and Pterocaulon. Overall, the findings indicate that pollen morphological traits are informative for species differentiation and lay a foundation for further palynological classification within Inuleae.

Blumea, Carpesium, HCA, palynology, PCA, SEM, taxonomy

Inuleae (Asteraceae: Asteroideae), sensu Anderberg and Eldenäs (2007), comprises approximately 690 species in 66 genera and represents a major lineage of the family. Most species are subshrubs, shrubs, or perennial herbs, mainly adapted to warm or temperate climates, and the tribe is primarily distributed in Eurasia, with additional occurrences in North Africa, Asia, and adjacent Arabia (Anderberg 1989, 1991, 2009; Anderberg and Eldenäs 2007; Pornpongrungrueng et al. 2007). The tribe exhibits considerable morphological heterogeneity, particularly in style types, bract morphology, and pollen wall structure (Merxmüller et al. 1977; Nylinder and Anderberg 2015). Although Merxmüller et al. (1977) provided a systematic overview based on palynology, cytology, and morphology, Inuleae does not form a distinct monophyletic group, highlighting the complexity of its classification (Bremer 1987; Anderberg 1989; Kim and Jansen 1995).

Pollen morphology has long contributed to plant taxonomy, supporting species and genus identification across diverse plant groups (Erdtman 1952; Larson and Skvarla 1962; Nowicke and Skvarla 1979; Ferguson 1985). In large and taxonomically complex families such as Asteraceae, it also provides insights into intergeneric relationships and evolutionary trends (Skvarla et al. 1977, 2005; Zhao et al. 2000; Pereira Coutinho and Dinis 2007, 2009; Wortley et al. 2007; Osman 2011). Pollen morphology is generally genetically conserved and shows limited variation across environmental conditions, which enhances its reliability as a taxonomic character (Zarrei and Zarre 2005; Blackmore 2007; Kadluczka et al. 2022). Key palynological traits, including size, shape, aperture type, exine ornamentation, and perforation dimensions, are widely used to infer genetic relationships and support taxonomic classification (Heidarian et al. 2021; Li et al. 2021). Although pollen morphology tends to be conserved within genera of Asteraceae, subtle trait differences can still assist in distinguishing closely related species (El-Ghazaly and Anderberg 1995; Magenta et al. 2010; Carrijo et al. 2013; Huang et al. 2017; Coutinho et al. 2020; Marques et al. 2021).

Recent studies on the palynology of the tribe Inuleae have made notable progress. Inuleae pollen typically exhibits a caveate, echinate–microperforate exine (Leins 1971; Anderberg 2009), though subtle interspecific variations hold phylogenetic significance. For example, Cyathocline has smaller grains with longer spines compared to Blumea (Peng et al. 2023). The three morphotypes of Pulicaria were identified based on spine and sculpturing features (Coutinho et al. 2011). Previous work has emphasized the importance of exine morphology, particularly spine shape and sculpturing patterns, in species classification and in providing consistent morphological evidence for phylogenetic analyses across several genera (Osman 2006; Pereira Coutinho and Dinis 2007; Zarin et al. 2010; Karlıoğlu Kılıç et al. 2021). However, despite these advances, palynological research on Inuleae remains relatively sparse (Perveen 1999; Pornpongrungrueng et al. 2007; Bahadur et al. 2022), and the pollen characteristics of many species are still unclear.

Accordingly, this study focuses on 19 species of Inuleae to analyze pollen morphology and its systematic implications. Detailed pollen morphological data for 11 species are presented here for the first time. Species sampling targeted lineages within Inuleae considered informative for interpreting pollen characters, enabling lineage-aware interpretation among the sampled taxa. This study forms the second part of an ongoing investigation into palynological and evolutionary patterns in Asteraceae, following our work on Astereae (Qu et al. 2025). The objectives are to (i) provide palynological data for Inuleae species using light microscopy (LM) and scanning electron microscopy (SEM); (ii) distinguish species with similar morphology based on pollen traits; and (iii) evaluate the role of palynology in the taxonomy of this group.

This study documented substantial inter-individual variation in pollen morphology across 19 Inuleae species, particularly in grain size and exine ornamentation. All species exhibited spherical pollen grains with tricolporate apertures, and the exine surfaces were echinate. Variations in spine morphology provided measurable differences between species, consistent with previous studies on Asteraceae (Skvarla et al. 1977; Osman 2006; Pereira Coutinho and Dinis 2007; Coutinho et al. 2011; Peng et al. 2023). These traits contribute meaningfully to species delimitation and lay a morphological foundation for future taxonomic refinements within the tribe (Bahadur et al. 2022; Huang et al. 2023; Lin et al. 2023; Qu et al. 2025).

A comparative review was conducted to evaluate the consistency of morphological traits across genera. Among the Blumea species—B. balsamifera, B. megacephala, B. lacera, and B. fistulosa—we observed tricolporate pollen with echinate ornamentation, consistent with the descriptions in Peng et al. (2023). Our measurements of P and E were generally larger than those reported in their study, although the P/E ratios were similar. Since their study lacked illustrations, these results should be further validated. The pollen morphology of Laggera crispata, for which data are reported for the first time, showed subtle differences when compared to L. alata, as reported by Meo and Khan (2009). Minor variations in P and E were noted, likely reflecting genetic diversity, environmental adaptation, or methodological differences in pollen processing, as previously suggested (Mo et al. 1997; Zhang and Qian 2011; Lu et al. 2022). The pollen traits of Inula japonica conform to the morphological range established by Karlıoğlu Kılıç et al. (2021) for the genus Inula. Similarly, Pulicaria dysenterica shows patterns consistent with observations by Coutinho et al. (2011), supporting trait stability. The morphology of Karelinia caspia, the only species in its monotypic genus, corresponds to data reported by Lu et al. (2018). These consistencies support the reliability of our measurements. In the genus Carpesium, in addition to C. cernuum, whose pollen morphology was studied by Zarin et al. (2010), we provide the first descriptions for C. szechuanense, C. triste, C. cordatum, and C. longifolium, contributing to the understanding of pollen morphology within the genus. Additionally, this study establishes the first pollen morphological dataset for the genus Pentanema and extends previous work on Pterocaulon (Corrêa et al. 2008) by offering quantitative data on P. redolens, transitioning from qualitative to quantitative documentation.

The genera Blumea DC. and Carpesium L. are major members of the subtribe Inulinae within Inuleae. Blumea is the largest genus in the tribe (Anderberg 1991), predominantly consisting of herbs or shrubs, with the highest diversity in the tropics of the Old World (Chen and Anderberg 2011a; Pornpongrungrueng et al. 2016). Many species have notable economic and ecological value, with over half possessing medicinal or ethnobotanical uses (Jia and Li 2005; Zhang et al. 2019). However, taxonomic confusion persists due to limited morphological studies (Gagnepain 1920; Pornpongrungrueng et al. 2007, 2016). Carpesium, with approximately 20 species (Zhang et al. 2015), is distributed in Asia and Europe, particularly in mountainous regions of Southwest China, where several species are endemic (Chen and Anderberg 2011b). The genus is also valued for its medicinal applications in traditional herbal medicine (Kim et al. 1997; Moon 2007; Chung et al. 2008; Moon and Zee 2011), but its palynology remains understudied (Zarin et al. 2010; Peng et al. 2023). Comparative analysis reveals distinct differences in pollen morphology between the two genera. The maximum polar length (P) in Blumea reaches 32.90 µm and equatorial width (E) 29.79 µm, while in Carpesium, P is 35.06 µm and E is 32.42 µm. Pollen length (L) in Blumea ranges from 27.07 to 30.35 µm, whereas in Carpesium it ranges from 30.16 to 33.49 µm. These measurements indicate that Blumea pollen grains are generally smaller. Differences in exine ornamentation are useful for species identification, though further study of additional species is necessary for confirmation. The spine base diameter (D) in Blumea ranges from 3.00 to 3.69 µm, with spine height (H) from 3.58 to 4.09 µm, yielding a D/H ratio of 0.75 to 0.95. In Carpesium, while D (2.96 to 3.75 µm) is similar, H is higher (3.74 to 5.17 µm), resulting in a lower D/H ratio (0.68 to 0.80) and more slender spines. Furthermore, spine spacing (Ss) in Blumea is narrower (4.79 to 6.01 µm) compared to Carpesium (5.08 to 6.34 µm), reflecting a denser ornamentation pattern. These pollen traits provide morphological evidence that may aid in distinguishing the two genera, although further study is needed to confirm their diagnostic value in generic delimitation within Inuleae.

PCA highlighted pollen size and exine ornamentation as key contributors to morphological differentiation among species (Wang et al. 2009; Zhang et al. 2017; Maciejewska-Rutkowska et al. 2021; Jardine et al. 2022). PC1 (67.39% variance) mainly reflects size-related traits, while PC2 (13.67%) captures variation in exine ornamentation. Species on the positive PC1 axis, such as C. longifolium and C. cernuum, exhibit larger pollen grains. In contrast, species on the negative axis, such as Pulicaria dysenterica and Pentanema cernuum, show smaller pollen. PC2 further separates species by exine ornamentation, especially D/H and T/L ratios. For instance, B. megacephala and B. balsamifera have high D/H and T/L ratios, which may enhance pollen rigidity under arid or stressful conditions (Wodehouse 1935; Zhang and Zhou 2016). Conversely, C. szechuanense and C. triste occupy negative PC2 positions, with distinct ornamentation patterns possibly reflecting environmental adaptations (Tanaka et al. 2004; Ejsmond et al. 2011; Ďurišová et al. 2023).

According to Fu et al. (2016), the Inuleae tribe is divided into two subtribes, Inulinae and Plucheinae. The 19 species in this study belong to these two subtribes (Table 1). To explore the role of pollen traits in species differentiation, we performed hierarchical cluster analysis. The HCA results divided the 19 species into two main groups (Group A and Group B), with Group A further split into A1 and A2 subgroups. This division primarily aligns with a pollen size gradient, as strongly indicated by PC1 (Fig. 8), but also reveals intriguing patterns when compared to phylogenetic classifications. Group A1 includes species from Blumea, Laggera, and Inula. Nylinder and Anderberg (2015) noted that the boundaries between Blumea and Laggera are unclear, and some species show high morphological similarity. Our palynological data provide quantitative support for this observation, as these genera share intermediate values in key traits like Ss and D/H ratio (Table 2), which may explain their clustering in the HCA. Group A2 includes all Carpesium species, along with B. fistulosa, B. formosana, and Karelinia caspia (Plucheinae). The clustering of all five Carpesium species suggests strong palynological consistency within the genus. The placement of K. caspia here is incongruent with its subtribal classification but is robustly supported by shared traits like large pollen size and wide spine spacing (Table 2), highlighting the complex role of pollen evolution in taxonomy. Group B consists of four species. The two Pentanema species are clustered together, while the Pterocaulon and Pulicaria species are in separate branches. These species have smaller pollen sizes, with P values ranging from 22.64 to 23.85 µm. PCA analysis supports this, with Group B species primarily located in the negative region of the PC1 axis (Fig. 8). The tight clustering of the two Pentanema species, in particular, underscores the utility of pollen morphology for delimiting closely related species. These results suggest that pollen size is a key feature for species differentiation (Lu et al. 2009; Chung et al. 2010; Usma et al. 2022). Taken together, the HCA results are largely consistent with the groupings suggested by PCA, confirming the role of pollen traits in species differentiation. However, they do not fully align with molecular phylogenetic groups (Lin et al. 2023), and the incongruence in plant systematics may need integrative taxonomic approaches (Keating et al. 2023). PCA identified pollen size and exine ornamentation as the most influential traits, while HCA revealed species groupings based on overall pollen morphological similarity. Together, they confirm the significant value of pollen morphological data for the classification within Inuleae.

Quantitative pollen morphological data from 19 Inuleae species provide valuable insights into the systematic palynology of Asteraceae. In this study, one specimen per species was analyzed, a common practice in palynological research (Bahadur et al. 2022; Lu et al. 2022; Peng et al. 2023). While this approach facilitates interspecific comparisons, future studies could expand sampling to better assess the range and consistency of pollen traits, including ploidy counting and the evaluation of defective or sterile pollen. To capture potential nonlinear variation, dimensionality reduction techniques such as UMAP may complement PCA (McInnes et al. 2018; Becht et al. 2019; Erb et al. 2024). Higher-resolution imaging methods, such as transmission electron microscopy (TEM) combined with SEM, could resolve fine-scale exine structures (Jia et al. 2017; Gabarayeva et al. 2024). Combining morphological, genomic, and transcriptomic data may reveal the genetic basis of pollen variation and its evolutionary significance (Keating et al. 2023; Qu et al. 2025). These approaches could refine species delimitation and clarify pollen evolution within a broader phylogenetic context.

This study documented variation in pollen morphology among 19 Inuleae species, focusing on grain size, shape, and exine ornamentation. These traits proved useful in distinguishing species and contributed to taxonomic classification. While HCA and PCA refined species clustering based on pollen traits, the taxonomic relationships among species cannot be fully clarified by pollen morphology alone. These results provide a foundation for future research on the role of pollen morphology in species differentiation within Inuleae.

The authors thank Dr. Xiuping Xu (Institute of Botany, Chinese Academy of Sciences) for expert guidance in SEM and the anonymous reviewers for their invaluable comments, which greatly improved the quality of this manuscript.