The specimens were collected from Yanzhou Lijie Pier, Huizhou, Guangdong Province, China (22°43′12″N, 114°57′36″E), with a water temperature of 24°C, pH of 7.59, and salinity of 32‰ ( Supplementary Figure S1 ). Briefly, the flatworms attached to submerged rocks and oyster shells were rinsed in a water bucket and filtered using 200-mesh, 160-mesh, and 40-mesh plankton nets.

The flatworms were cultured in a 10-cm-diameter round buckle box with the addition of habitat sediment, wheat grains, and protozoa. The water was changed every 3 days, with the salinity maintained at 30‰ and the water temperature at 23°C ± 2°C. The behaviors of the flatworms, including movement, predation, mating, and spawn, were observed and recorded under an Olympus SZX16 stereomicroscope. The eggs were transferred to a clean culture dish for the continuous observation of embryonic development, hatching, and juvenile development under Olympus BX51 in conjunction with the Olympus DP72 camera system.

Three flatworms starved for 2 days were treated with liquid nitrogen for 15 s, and DNA was extracted using the E.Z.N.A. Mollusc DNA Kit (Omega, Norcross, GA, USA). PCR amplification was conducted using Vazyme 2×Taq Plus MasterMix II (Dye Plus). For the primers and procedures for amplifying 18S rDNA and 28S rDNA, refer to Wang et al. (2024). After quality examination using 1.2% agarose gel electrophoresis, the PCR products were sequenced in Tsingke Biological Technology (Beijing, China).

The 18S rDNA and 28S rDNA fragments of the 21 Plagiostomidae species [all molecular data currently available for Plagiostomidae species in National Center for Biotechnology Information (NCBI)] and two species of Rogneda Uljanin, 1870, as the outgroup were obtained from GenBank (https://www.ncbi.nlm.nih.gov/) ( Supplementary Table S1 ). Multiple sequence alignments were performed using MAFFT on the online server (https://mafft.cbrc.jp/alignment/server/) with the Q-INS-i algorithm and default parameters (Kuraku et al., 2013; Katoh et al., 2019), and sequence trimming was conducted using trimAl v1.2rev57 with the -automated1 parameter (Capella-Gutierrez et al., 2009). Base substitution saturation was tested using DAMBE v5.3.19 (Xia, 2013). The Concatenation Sequence program (Vaidya et al., 2011) in the PhyloSuite v1.2.3 (Xiang et al., 2023) was used to merge the 18S rDNA and 28S rDNA sequences of each species.

ModelFinder v2.2.0 (Kalyaanamoorthy et al., 2017) was used to select the best-fit substitution models for the 18S rDNA and 28S rDNA datasets using the Akaike information criterion (AIC) method (Lanfear et al., 2012): GTR+F+I+R3 (for 18S rDNA) and GTR+F+I+R4 (for 28S rDNA) for IQ-TREE; both datasets used GTR+F+I+G4 for MrBayes. IQ-TREE v1.6.2 (Nguyen et al., 2015) was used to construct the maximum likelihood (ML) phylogenetic tree using 1,000 ultrafast bootstrap replicates. MrBayes ver. 3.2.7a (Ronquist et al., 2012) was used to construct the Bayesian inference (BI) phylogenetic tree, employing the Markov chain Monte Carlo (MCMC) algorithm with four simultaneous chains, running for 2,000,000 generations, sampling every 1,000 generations, and discarding 25% of the burn-in samples. The average standard deviation of split frequencies was less than 1%. TRACER v1.7.1 (Rambaut et al., 2018) was used to check the.p files from each run, ensuring that all parameters had an effective sample size (ESS) greater than 200. The phylogenetic tree results were visualized using Figtree v1.4.3 (Rambaut, 2009).

To study the reproduction of the new species, 12 single-individual cultured groups, 10 three-individual cultured groups, and a population cultured group were established. Mature flatworms (larger than 600 μm in length) in the single-individual cultures were reared in six-well cell culture plates (well diameter 35 mm). The three individual cultured groups were conducted on a small scale in 90-mm glass evaporation dishes, while the population cultured group consisted of 200–300 mature flatworms in a 10-cm-diameter round buckle box. Habitat water and food were supplied daily. Flatworms at different cultured stages were fixed using Bouin’s fluid and then washed and soaked in 75% ethanol until the epidermis was transparent and the shadow of the copulatory apparatus was visible. The presence or absence of male copulatory apparatus was examined under an Olympus SZX16 stereomicroscope. Paraffin tissue sections were then prepared to observe the changes in the copulatory apparatus according to Yang et al. (2020). Additionally, 60 sexually mature flatworms freshly collected from the habitat were used as controls.

b. brain, ca. common atrium, cvd. common vitelline duct, dp. distal penis, ds. distal sac, e. eye, eg. egg, fg-1. frontal gland-1 (orange-staining frontal gland), fg-2. frontal gland-2 (aniline blue-staining frontal gland), gg. gonopore gland, go. gonopore, in. intestine, m. mouth, o. ovary, p. pharynx, pg. prostate gland, pp. penial papilla, prp. proximal penis, ps. penile sheath, s. sperm, sg. shell gland, t. testis, v. vitellaria, va. vagina, vd. vitelline duct, vs. vesicula seminalis.

Order: Prolecithophora Karling, 1940;

Family: Plagiostomidae Graff, 1882;

Genus: Plagiostomum Schmidt, 1852;

The amplified 18S rDNA and 28S rDNA sequences of the two species range from 1,600 to 1,700 bp in length. After alignment and trimming, the lengths of each dataset were 1,753 bp (18S rDNA), 1,478 bp (28S rDNA), and 3,233 bp (concatenated dataset), with low base substitution saturation (ISS < ISS.c), making them suitable for phylogenetic tree construction. The Kimura two-parameter (K2P) distance between the two new species and Plagiostomum morgani Graff, 1911, is the shortest among Plagiostomum species, ranging from 5.3% to 6.5%. The ML and BI phylogenetic trees constructed from the three datasets had consistent topologies. Three individuals of P. nanhaiensis sp. nov. and P. plagae sp. nov. formed independent clades with solid support in both BI (posterior probability = 1.00) and ML (bootstrap value = 100) analyses, being sister to each other ( Figure 6 ). The clade of two new species clustered with Plagiostomum sp. SL-2020 and P. morgani Graff, 1911, close to the clade comprising P. vittatum Jenssen, 1878, Vorticeros auriculatum Müller, 1784, and Vorticeros ijima Tozawa, 1918. These results further supported the recognition of these two new species.

Our observation revealed that the two new species employed hypodermic impregnation as their fertilization model. Notably, self-fertilization was documented in both species during individual cultivation. In this instance, the tail curls upward, and the reproductive organ injects itself near the head and pharyngeal region. In contrast, triad and group cultivation facilitated cross-fertilization among these two species. Remarkably, we observed self-fertilization of P. nanhaiensis sp. nov. on two occasions when other conspecifics were present, suggesting that self-fertilization is not restricted to individual cultivation.

The mating behavior analysis in six pairs of P. nanhaiensis sp. nov. revealed that individuals engage in head-to-head contact before fertilization to assess each other. Following the initial interaction, both individuals rapidly curl up and separate. In scenarios involving multiple individuals, they may extend their copulatory organs; however, once one individual successfully inserts its copulatory organ subcutaneously into another, the remaining individuals retract their penises. During mating, the penis of the injecting individual protrudes from the gonopore ( Figure 7A ), and the blunt penial papilla firmly anchors to the recipient, preventing escape during movement ( Figure 7B ). Throughout the mating process, the lower midsection of the injecting individual remains noticeably contracted, facilitating direct sperm injection into the recipient. After fertilization, the injecting individual retracts its penis ( Figure 7C ). Occasionally, the recipient’s body contracts in response to the injection, resulting in the expulsion of sperm and the formation of a transparent sac-like protrusion while this phenomenon does not occur after every fertilization event ( Figure 7D ). In addition, we did not observe simultaneous bidirectional or multidirectional fertilization; there were instances where an individual that had just received sperm immediately became the injecting individual for another ( Figure 7D ).

In natural habitats, the hermaphroditic reproductive structures of the two new species of flatworms remain stable ( Figure 8A ). However, during cultivation, with the increase of culture time and the decrease of group density, both species exhibit degeneration of the male reproductive organ. The initial signs of degeneration include shrinkage of the vesicula seminalis, which ceases to store sperm, while no significant changes were observed in the prostate gland cells. Additionally, the structure of the testis degenerates, resulting in a decreased sperm count and eventually disappearance of the penis. The penile sheath and the common atrium also show signs of atrophy ( Figure 8B ). In the later stages of degeneration, the vesicula seminalis, prostate gland cells, and penile sheath disappear, and the common atrium shrinks into a crevice, ultimately transforming into solid tissue ( Figures 8C, D, G ). During this process, the intestine and female reproductive structures remain intact ( Figures 8B–F ).

Our comparison of sex allocation in P. nanhaiensis sp. nov. between natural habitats and cultivation found that, among the 60 mature individuals of P. nanhaiensis sp. nov. collected from the wild, only two exhibited degeneration of the male reproductive organ, with 96.7% of individuals remaining hermaphroditic. In contrast, after 14 days of single, triad, and group cultivations, the percentages of hermaphroditic individuals increased to 40%, 55.6%, and 90%, respectively. Further observation revealed that the male reproductive organs of P. nanhaiensis sp. nov. degenerated completely after 10 to 21 days in the single cultivation (n = 10) and after 10 to 30 days in the triad cultivation (n = 21).

Currently, the species identification of Plagiostomidae mainly relies on morphology and histology, with phylogenetic analysis as a supplement. The two new species, P. nanhaiensis sp. nov. and P. plagae sp. nov., exhibit similarities in body size, life habits, juvenile characteristics, and egg development. However, they do not interbreed and display significant differences in pigmentation stripes and histological structures, indicating that they are two distinct species (Westram et al., 2022). Molecularly, the K2P distance among Plagiostomum species, based on 18S rDNA and 28S rDNA sequences, ranged from 2.2% to 22%. Although most distances exceed 10%, the genetic distance between three individuals of two species is approximately 0, and the genetic distance (5.4%) between P. nanhaiensis sp. nov. and P. plagae sp. nov. is similar to that between P. morgani Graff (5.3% and 6.5%, respectively), indicating clear genetic divergence. Meanwhile, phylogenetic analyses revealed that both P. nanhaiensis sp. nov. and P. plagae sp. nov. formed independent clades and were sister taxa to each other, exhibiting a certain evolutionary distance from other species. Based on the above evidence, we consider P. nanhaiensis sp. nov. and P. plagae sp. nov. to be two distinct new species.

To date, 85 species of Plagiostomum have been recorded, but little attention has been given to their mating behavior, egg-laying behavior, and individual development. A current study observed the mating behavior and individual developmental process of P. robusta, indicating that this species performs hypodermic impregnation (Wang et al., 2024). Therefore, we investigated the mating and egg-laying behaviors of P. nanhaiensis sp. nov. and P. plagae sp. nov. to further explore the biological characteristics and reproductive strategies of Plagiostomum species.

Hypodermic impregnation, being common in hermaphroditic animals, was considered as the response to sexual conflict allowing forced copulation, thereby increasing fertilization success, reducing sperm competition, and ensuring the survival of offspring (Ramm, 2017; Schärer and Janicke, 2009). The similar fertilization between P. nanhaiensis sp. nov., P. plagae sp. nov., and P. robusta indicated that most species in Plagiostomum may perform hypodermic impregnation (Wang et al., 2024). Among flatworms, hypodermic impregnation was first discovered in Macrostomum, and their mating behavior, reproductive structures, and other characteristics have been extensively studied (Brand et al., 2022; Schärer et al., 2004, 2011). The species of Macrostomum have sharp bony penile stylets and simple sperm structures, allowing them to quickly pierce the body surface of their mate and inject sperm (Brand et al., 2022; Schärer et al., 2004, 2011). Although P. nanhaiensis sp. nov. and P. plagae sp. nov. exhibit different sperm structures, both are similarly simple ( Figures 2 , 5 ). Additionally, the copulatory organs of P. nanhaiensis sp. nov., P. plagae sp. nov., and P. robusta are fleshy penises, lacking needle-like or stylet structures, being different from Macrostomum (Brand et al., 2022; Schärer et al., 2004, 2011; Wang et al., 2024). Nevertheless, species of Plagiostomum can still anchor their mating partner using the penial papilla, allowing the penis to inject their sperms. Hypodermic impregnation species often exhibit self-fertilization when the mating partner is unavailable, which is commonly regarded as a safeguard mechanism for reproduction (Giannakara and Ramm, 2017; Janicke et al., 2016; Noël et al., 2017). Recent studies on Macrostomum have revealed that the initiation of self-fertilization not only is caused by the female’s inability to receive sperm but also relates to the maintenance of male reproduction (Kaufmann and Schärer, 2020). During the cultivation of P. nanhaiensis sp. nov. and P. plagae sp. nov., we also observed their self-fertilization. Particularly, when multiple potential mates were present, P. nanhaiensis sp. nov. still exhibited self-fertilization behavior, further indicating that self-fertilization not merely is an emergency strategy triggered by specific environmental pressures (such as mate scarcity) but may be an inherent reproductive strategy.

Sex allocation is a critical reproductive strategy in hermaphroditic organisms, as it determines how resources are distributed between male and female functions to optimize reproductive success (Charnov, 1979, 1984; Fischer, 1981). However, accurately measuring these allocations in hermaphrodites is complex due to the need to quantify both male and female organs, considering both static and dynamic changes over time (Schärer, 2009). Currently, flatworms of the order Macrostomida are primarily assessed based on the relative proportions of sexual organs and glands to measure sex allocation (Brand et al., 2022; Janicke et al., 2013; Vellnow et al., 2018). Interestingly, during the cultivation of the two new species, we observed the complete degeneration of male reproductive structures, providing a clear indicator of female-only individuals. Complete feminization in hermaphrodites is rare and may represent an extreme sex allocation strategy.

Our cultivation revealed that the proportion of hermaphroditic individuals increased with the expansion of the group. Additionally, our observation of the flatworms collected from the wild in different seasons, from March to June and from October to November, showed that their hermaphroditic reproductive structures remained stable in the natural habitat. However, the male structures degenerated after the small-scale laboratory cultivation, indicating that seasonal changes and temperature fluctuations are not decisive factors for their sex allocation. Furthermore, in individuals with degenerated male reproductive structures, the intestine and other organ structures remained intact, ruling out the hypothesis that starvation was the cause of autophagy and organ disappearance. In summary, we suggested that sex allocation in flatworms is plastic, with mating group size being the most significant influencing factor corresponding to the traditional sex allocation theories (Charnov, 1984; Schärer, 2009). This phenomenon suggests that in natural environments, with the increase of mating group size, individuals tend to allocate more resources to male reproductive functions to overcome sexual selection pressures and to ensure reproduction during multiple male competition, thereby increasing the success probability of fertilization (Brand et al., 2022; Charnov, 1984; Janicke et al., 2013; Schärer, 2009). In contrast, in smaller or single-cultured conditions, the competing sperm are highly related. This leads to diminished returns on investment in sperm production, as producing additional sperm primarily results in competition among one’s own sperm—a phenomenon known as local sperm competition (LSC) (Brand et al., 2022; Janicke et al., 2013). The LSC reduces the effectiveness of additional sperm investment, leading to a shift in resource allocation toward female reproductive functions and thus decreasing the level of sperm competition (Brand et al., 2022; Janicke et al., 2013). This strategy allows the individuals to adapt to the environment by increasing the quantity and quality of offspring when the mating opportunities are limited (Brand et al., 2022; Charnov, 1984; Janicke et al., 2013; Schärer, 2009). Compared to closed laboratory cultures, the group size of flatworms in natural habitats is larger, so the proportion of hermaphroditic individuals among mature flatworms collected from the wild is extremely high. This indicates that the evolution of sex allocation in flatworms is closely related to their social and environmental conditions and the sex allocation strategies exhibit adaptive changes under different ecological contexts.

Similar to those flatworms of the family Dugesiidae, the eggs of these two new species connect the egg stalk and base through mucus secreted by the reproductive glands (Solà et al., 2022). Their egg base adheres to the walls of the culture dish, which securely anchors the egg and reduces dispersion. In addition, the eggs undergo color change and hardening after being laid, which better protects the embryo inside the eggshell and improves the survival rate of the embryo. In addition to increasing embryo survival through egg fixation and eggshell hardening, flatworms also continuously collect and transport surrounding debris during and after egg-laying to cover the embryos, reducing the likelihood of predation by predators. The two new species in this study exhibit considerable similarity in diet, egg development, and juvenile morphology, indicating a trend of convergent evolution in the same habitat, with significant niche overlap and intense interspecies competition. However, during laboratory cultivation, slight differences in habitat preferences were observed between the two species: P. nanhaiensis sp. nov. often resides in the sediment, while P. plagae sp. nov. prefers to inhabit the side walls near the water surface. Multiple field collections have shown that P. nanhaiensis sp. nov. is more abundant than P. plagae sp. nov., indicating that P. nanhaiensis sp. nov. is the competitive dominant species between the two.