All voucher specimens are deposited in the collection of the Insect Biosystematics Laboratory, Research Institute for Agriculture and Life Science, Seoul National University, Korea (SNU). Digital images of external characteristics were taken with a DMC 5400 digital camera attached to a Leica Z16 APO motorized microscope. Genital structures were dissected and observed under a Leica DM 4000B microscope, and images were taken using a digital camera combined with the microscope (Lumenera Infinity 3). All measurements (mean and range) are provided in millimeters. Terminology for male and female genital structures follows Oh and Lee (2018).

Genomic DNA was extracted from the abdominal segments IV–VI and legs of specimens preserved in 95% to 99% alcohol using a DNeasy Blood and Tissue Kit (QIAGEN, Inc.) following the manufacturer’s protocols. Examined specimens were labeled and stored in alcohol depending on their condition. After DNA extraction, a sequencing library was prepared using the TruSeq Nano DNA Kit, based on the TruSeq Nano DNA Sample Preparation Guide, Part #15041110 Rev. D (Illumina, Inc.). NovaSeq X Plus was used as the sequencing platform. Read length was 151 bp (30 million reads per lane) according to the Macrogen Company (South Korea) sequencing facility. Sequences were aligned using Geneious Prime 2025.1.3 software (Kearse et al. 2012). A total of six reference sequences were used for alignment (Table 1), and the resulting data were annotated using MITOS2 (Bernt et al. 2013) and visualized as a circular map using the webserver Proksee (https://proksee.ca/, accessed on 26 July 2024) (Grant et al. 2023).

A COI-based dataset, comprising 10 sequences from the NCBI database and four from the present study, was compiled. To evaluate inter- and intraspecific genetic distances, 563 bp of the mitochondrial protein-coding COI (cytochrome oxidase subunit I) gene was used (Folmer et al. 1994), with LCO1490 (5′-GGTCAACAAATCATAAAGATATTGG-3′) and HCO2198 (5′-TAAACTTCAGGGTGACCAAAAAATCA-3′) primers. During alignment, an L-INS-i strategy was used, without any manipulation except trimming the ends of reads. A neighbor-joining (NJ) tree (Saitou and Nei 1987) was generated, and sequence divergences were evaluated using the Kimura 2-parameter model (K2P) (Kimura 1980) in MEGA 7.0 (Kumar et al. 2016). Detailed information for the sequences used is provided in Table 2.

Heterotoma planicornis is morphologically similar to H. merioptera (Scopoli, 1763), leading to considerable taxonomic confusion reviewed by Kment and Bryja (2006, 2012). Current diagnostic characters primarily rely on differences in genital structures and the length ratio of antennal segments II and III. Heterotoma planicornis tends to have a shorter second segment and a longer third segment compared to H. merioptera (Wagner 1968, 1974; Goula 1990). Additionally, Kment and Bryja (2006) proposed the ratio of vertex width to compound eye width as an alternative diagnostic character, but they noted that it overlaps between the two species.

The lengths of antennal segments III:II in the examined Korean specimens (0.289–0.303) were consistent with the value reported for H. planicornis by Kment and Bryja (2006). The presence of two apically sinuate, unfused endosomal sclerites and parameres in the examined specimens was consistent with the illustrations of H. planicornis provided by Tamanini (1962) and Goula (1990). However, a medially bifurcated sclerite, as in fig. 2 of Kment and Bryja (2012), was not observed in the Korean population. In the figures of the endosomal sclerites provided by Tamanini (1962), bifurcated sclerites were observed in both Heterotoma species. Considering that intraspecific variation in the branching of endosomal sclerites can occur in other species of Orthotylinae (Oh et al. 2022), the above differences are possibly due to variation. It is concluded that the Korean material examined pertains to H. planicornis.

Our study confirmed that the Korean H. planicornis uses Spiraea japonica as a breeding host. This plant species is native to Japan but has been widely planted in Europe since its introduction in the mid-19^th century (Trees and Shrubs Online 2025). In Korea, cultivated S. japonica plants are mostly imported from Japan and European countries such as Germany and the Netherlands. Although various cultivars of S. japonica are cultivated in Japan, no species of the genus Heterotoma have been reported from the country or from any other locality in East Asia. Considering that all species of the genus Heterotoma are native to Europe, it is highly likely that individuals detected in Korea were introduced from populations in Europe, presumably on their host plants. Since eggs of Heterotoma species can overwinter and be transported with host saplings (Kment and Bryja 2006), the population might have been imported as eggs.

Individuals used in the analysis were collected from the same plant, and no significant differences were observed in their external characteristics. An analysis of COI sequences from H. planicornis collected in this study and from taxonomically related species revealed that the Korean population comprises two distinct populations, exhibiting a 1.80–1.98% sequence difference (Table 4). This was also reflected in the NJ tree, where one population clustered with the North American population, while another branched into a separate node. However, due to the unavailability of European specimens, the relationship between Korean and European populations could not be ascertained. With sufficient sequence data on H. planicornis from diverse geographic regions, it could be possible to trace the origin and introduction pathway of the Korean population.

In species delimitation studies of Miridae using COI sequences, genetic distances less than 2% are largely considered conspecific (Jung and Lee 2012; Kim and Jung 2018). A recent DNA barcoding study by Kim and Jung (2018) on 273 individuals from 123 Miridae species revealed intraspecific variation ranging from 0 to 2.8%, with a mean variation of only 0.2%. Furthermore, in their analysis of 27 individuals from 14 Orthotylinae species, maximum intraspecific variation did not exceed 2%, while interspecific variation ranged from 1.6% to 27.5%. However, as currently only limited molecular data are available for populations of H. planicornis, further research is necessary.