Discussion

Biogeography and species diversity of Plusioglyphiulus in Cambodia

According to the latest comprehensive catalogue of Cambodian Diplopoda (Likhitrakarn et al. 2015), also incorporating the two new species described herein, the millipede fauna of Cambodia remains strikingly depauperate, currently standing at a mere 28 recorded species distributed across 17 genera, 13 families, and 8 orders (Likhitrakarn et al. 2020a, 2025a; Srisonchai et al. 2020, 2023). With the addition of Plusioglyphiulus parviserratus sp. nov. and P. battambangensis sp. nov., the total number of Plusioglyphiulus congeners known from the country is raised to six (Fig. 12), while the global diversity of the genus expands to 32 valid species. It is worth emphasizing that all existing records of this genus are strictly confined to the southern and western karst regions of Cambodia.

Such a surprisingly low nominal diversity is undoubtedly rooted in an insufficient sampling effort that has failed to grasp the true richness of the local fauna, representing instead a profound collecting bias and a lack of systematic field explorations rather than actual ecological poverty. When contrasted with neighboring Thailand, where more than 14 Plusioglyphiulus species have been documented from more than 300 sampled localities (Likhitrakarn et al. 2023), the Cambodian karst systems appear vastly understudied. Given that Thailand harbors many more than a thousand surveyed caves, the discovery of only a handful of cave-dwelling millipedes in Cambodia underscores the urgent need for intensive speleobiological investigations across its isolated limestone hills.

Karst isolation, micro-endemism, and transboundary faunal connections

The distribution patterns of the Cambodian Plusioglyphiulus heavily reinforce the concept of micro-endemism, a classic evolutionary trait observed in many cavernicolous and limestone-associated soil arthropods (Clements et al. 2006; Golovatch 2015). As illustrated in our updated distribution map (Fig. 12), the newly discovered taxa represent highly localized geographic isolates. The genus shows a profound propensity for parapatric or allopatric speciation driven by karst isolation; the severely fragmented nature of the limestone massifs in Battambang Province acts as an effective dispersal barrier, promoting rapid genetic divergence and morphogenetic specialization within individual cave systems or isolated mountain blocks. This geographic confinement and strict localization perfectly mirror the patterns observed in other highly specialized, karst-dwelling arthropod lineages across Indochina, where isolated limestone towers function as terrestrial islands that effectively halt gene flow (Clements et al. 2006; Srisonchai et al. 2018).

However, the syntopic or closely parapatric occurrence of related congeners within such localized environments warrants further eco-evolutionary consideration. Rather than assuming strict in-situ sympatric speciation, such distributional overlapping might heavily stem from complex, historical dispersal events from disparate microrefugia across shifting tropical corridors, culminating in subsequent secondary contact. While syntopy is a very rare phenomenon within the family Cambalopsidae, a notable parallel is documented in Borneo, where Plusioglyphiulus bedosae and P. pallidior stably coexist within the same cave system through marked body-size differentiation (Golovatch et al. 2009). A classic parallel in terms of montane habitat partitioning is observed in the high-montane forest fauna of northern Thailand, notably at Doi Inthanon and Doi Suthep, where up to ten sympatric species of the genus Tylopus Jeekel, 1968 successfully coexist, each maintaining a distinct phenological or microhabitat niche to avoid competitive exclusion (Likhitrakarn et al. 2010, 2014, 2016, 2025b). A similar ecological partition has also been well-documented in other Oriental paradoxosomatid lineages, such as Touranella Attems, 1937, where closely parapatric or sympatric congeners exhibit marked body-size differentiation and seasonal shifts in activity to mitigate intense ecological competition (Likhitrakarn et al. 2025c). We observed potential difference in body size between the two species, and niche differentiation remains a plausible explanation for their coexistence. However, this hypothesis requires ecological investigation with a larger data set. Further studies on microhabitat quantitative analysis are needed to better understand the mechanism underlying the coexistence of these two species.

In the case of the Plusioglyphiulus fauna treated herein, both new species demonstrate close morphological affinities with P. likhitrakarni, a species originally described from eastern Thailand (Golovatch et al. 2011). This morphological proximity strongly implies a historical transboundary faunal connection between the limestone karsts of eastern Thailand (e.g., Sa Kaeo and Chanthaburi provinces) and those of western Cambodia (Battambang Province). Such geographic patterns suggest that these regional karst systems might have shared historical ecological continuities prior to their more recent fragmentations.

Yet, despite their shared transboundary history and restricted distributions within the southern and western Cambodian karst systems, these micro-endemic species show profound, unequivocal morphological divergence. Most notably, they are easily distinguished from their syntopic or parapatric congeners, such as P. boutini Mauriès, 1970 and P. khmer Likhitrakarn, Golovatch, Thach, Chhuoy, Ngor, Srisonchai, Sutcharit & Panha, 2020, by the highly specialized architecture of the male gonopods and the specific carinotaxic ornamentation of the collum, providing explicit diagnostic features unique to each respective taxon.

Imaging modalities in cambalopsid taxonomy

Modern millipede taxonomy increasingly relies on a combination of different imaging techniques to document and illustrate diagnostic somatic and gonopodal structures (Likhitrakarn et al. 2011, 2020b, 2022, 2024, 2025d, 2026a; Srisonchai et al. 2018; Huynh et al. 2023; Nguyen et al. 2023; Ng et al. 2025). In the present study, we evaluate and compare the utility, advantages, and limitations of four major modalities, including standard optical photography, line drawings, scanning electron microscopy (SEM), and micro-computed tomography (µCT) scanning, by utilizing the same cambalopsid species as a model.

Digital stereomicroscopic photography (Figs 3, 8) remains indispensable for capturing true-to-life coloration patterns and providing a comprehensive overview of the somatic habitus. However, even when deploying state-of-the-art optical systems, conventional light microscopy often fails to adequately resolve exceptionally minute, densely crowded, or overlapping appendages. Such limitations become particularly apparent when discerning the fine chaetotaxy on male leg-pairs 1–3, the intricate structures of the gnathochilarium, or the highly condensed, microscopic lamellae and processes characteristic of the anterior and posterior gonopods. Furthermore, subtle nuances of the cuticular sculpture, including the exact, detailed configuration of the metatergal ridges and crests, remain largely elusive under ordinary optical illumination.

Traditional line drawings, prepared with the aid of a camera lucida or executed from focus-stacked digital renderings, afford the ability to selectively emphasize and clarify specific diagnostic features. Nevertheless, the accuracy of such illustrations is heavily dependent upon the investigator’s personal interpretation and understanding of the structures under examination. This becomes particularly problematic when dealing with membranous or semi-translucent structures, which frequently remain visually elusive even under phase-contrast microscopy. For instance, the exact morphology of the bacilliform sensilla on antennomeres 5 and 6 (Figs 4C, 9C), the delicate penes on the male coxae 2 (Figs 4F, 9F), and especially the transparent flagellum process (f) (Figs 4K, 4L, 9K, 9L) on the posterior gonopods cannot be definitively resolved by means of standard light optics.

In sharp contrast, SEM provides unparalleled resolution and depth of field, bypassing any optical limitations associated with tissue translucency (Figs 5L, 5M, 10K–M). Because SEM utilizes an electron beam rather than light waves, the entire surface of the specimen is rendered with uniform, crisp focus, making it the gold standard for illustrating micro-characters. Its primary drawbacks, however, are that it provides no information regarding internal tissue density or variation in translucency within a single structure. Moreover, the prerequisite preparatory procedures (viz. critical point drying and gold sputter-coating) strictly demand structurally rigid specimens capable of withstanding high-vacuum conditions; otherwise, fragile or poorly calcified elements invariably risk distortion, fracturing, or artifactual displacement during the imaging process (e.g., Likhitrakarn et al. 2026b: fig. 13). The substantial maintenance costs and restricted availability of SEM facilities also present an obstacle to widespread, routine access.

Recently, micro-computed tomography (µCT scanning) has emerged as a powerful, increasingly accessible, and non-destructive alternative in myriapodology. In terms of external somatic architecture, the newly acquired 3DµCT data and their subsequent volume renderings yield highly satisfactory results (Figs 6, 11). The dataset provides a valuable non-destructive perspective, offering exceptional rotational capabilities and allowing for detailed observations of the cuticular patterns, rows of tubercles, and metatergal crests with remarkable clarity. Furthermore, virtual segmentation allows for the color-coded isolation of internal and closely appressed structures in situ, completely removing the necessity for physical dissection that might otherwise damage a unique specimen (Fig. 7A–D).

Nevertheless, regarding the visualization of intricate internal or highly localized structures, specifically the male leg-pairs 1–3 and the gonopods, the current resolution of non-invasive µCT data remains insufficient for critical taxonomic evaluation (Fig. 7E–M). The primary constraint inheres in the post-scanning segmentation and data processing, where manually delineating the boundaries and layers of densely packed structures is severely hampered by a lack of contrast and distinct separation. For instance, subtle micro-characters, such as the exact setation pattern on the mesal surface of leg 1 or the precise configuration of a vestigial second telopodite reduced to a mere rudimentary knob, cannot be successfully discriminated (Fig. 7E–H). Similarly, while the gonopod complex can be isolated as a distinct, color-coded volumetric mass (Fig. 7K–M), it fails to reveal the fine diagnostic features required for species-level discrimination and comparison, falling far short of the resolution captured via SEM (Fig. 5E–G). This limitation is undoubtedly exacerbated by the fact that the anterior and posterior gonopods remain naturally tightly appressed in situ (Figs 5E–G, 7A–D, 7K–M); without physical dissection prior to scanning, these overlapping elements mutually obscure one another, making the reconstruction of individual hidden surfaces from the virtual datasets exceedingly difficult.

In the present study, our post-processing of raw tomographic data benefited substantially from the advanced software suite AIVIA (v. 15) by Leica. While high-end commercial platforms like AIVIA yield vastly superior volumetric renderings and surface displays compared to standard open-source freeware, the inherent hardware and algorithmic thresholds required for resolving microscopic, naturally contiguous soft tissues or tightly appressed cuticular matrices remain a significant challenge.

Despite these spatial resolution constraints concerning the male genitalia, µCT scanning demonstrates immense potential for exploring structures deeply embedded within the body cavity, such as the female reproductive organ (vulva), which is situated internally within the third body segment. Given that male gonopods display hyper-diverse configurations routinely utilized to discriminate congeners, the degree of morphological variation in the corresponding female vulvae remains a fascinating, hitherto unresolved enigma in Plusioglyphiulus and related cambalopsids. Future refinements in µCT technology, scan geometries, and differential contrast-staining techniques will ultimately provide the keys to deciphering these internal structures.

To address the research objective regarding methodological efficiency, we present a qualitative comparison of the imaging methods applied in this study (Table 2). For standard identification, traditional light photography and line drawing continue to be the most economical but are limited by depth of field and subjective manual interpretation, respectively. SEM and µCT, in contrast, provide better non-destructive 3D visualization and high-resolution detail, but require much higher investments in equipment, specialized software and processing time. We conclude that a combined workflow that emphasizes routine photography for quick initial diagnosis, while leaving more advanced methods such as µCT and SEM for complex gonopod analysis, provides the best compromise between cost, time, and morphological resolution in cambalopsid taxonomy. As summarized in Table 2, this tiered approach allows researchers to select the imaging modality that best fits the specific taxonomic requirements of the specimen while effectively managing available resources and time constraints.

Imaging technique | Cost | Time consumption | Resolution / detail | Main limitation
Light photography | Low | Low | Low/Moderate | Shallow depth of field
Line drawings | Low | Very High | High | Subjective interpretation
SEM | Moderate/High | Moderate | Very High | Destructive preparation
µCT | High | High | High/Very High | Processing complexity

Figure 12.: Distribution of Plusioglyphiulus species in Cambodia. Asterisk: Plusioglyphiulus dubius (Attems, 1938); filled triangle: Plusioglyphiulus battambangensis sp. nov.; open square: Plusioglyphiulus parviserratus sp. nov.; filled square: Plusioglyphiulus khmer Likhitrakarn, Golovatch, Thach, Chhuoy, Ngor, Srisonchai, Sutcharit & Panha, 2020; open triangle: Plusioglyphiulus boutini Mauriès, 1970; open circle: Plusioglyphiulus biserratus Likhitrakarn, Golovatch, Thach, Chhuoy, Ngor, Srisonchai, Sutcharit & Panha, 2020.

Conservation implications for cavernicolous ecosystems

Limestone karsts and their subterranean networks in Southeast Asia represent critical hotspots of global biodiversity, supporting exceptionally high numbers of narrow-range endemics. However, these specialized ecosystems are extremely fragile and increasingly threatened by commercial limestone mining for cement manufacture, agricultural development, and unmanaged cave tourism (Clements et al. 2006). Such anthropogenic activities greatly destabilize underground microclimates and deplete essential organic resources, especially bat guano, a fundamental substrate with which many cambalopsid millipedes are intimately connected (Golovatch 2015). Recent ecological studies of the caves of Cambodia have highlighted the significant conservation conflict that occurs due to human visitation and disturbance of the reproduction of insectivorous bats, where even small changes to the microenvironment can lead to immediate localized extinctions (e.g., Lim et al. 2018).

Being microendemics strictly confined to isolated karst towers, Plusioglyphiulus parviserratus sp. nov. and Plusioglyphiulus battambangensis sp. nov. are highly vulnerable to habitat degradation. Any structural compromise to their type localities could result in extinction before their ecological roles are fully understood, which is a vulnerability shared by many other cave-adapted arthropod lineages across Indochina. Therefore, basic taxonomic inventories are an essential first step toward assessing the biological value of individual karsts, providing the empirical baseline needed to advocate for sustainable cave management and formal site protection in Cambodia.

Regarding regional Cambalopsidae, we have merely touched the tip of the diversity iceberg (Golovatch et al. 2007b). Cambalopsids are exceptionally diverse across Southeast Asian karsts, showing strong ecological coupling with bat guano. Future systematic explorations, particularly within Cambodia’s under-sampled limestone formations, will undoubtedly reveal many more undescribed diplopods.