The genus Goniogryllus Chopard, 1936 belongs to Gryllinae, Gryllidae, Orthoptera. It was established by Chopard in 1936, with G. punctatus Chopard, 1936 from Fujian, China as the type species. Prior to this study, there were 21 species reported from Japan, Assam, and China with 19 species from China (Chopard 1936; Bey-Bienko 1956; Chopard 1969; Ichikawa 1987; Wu and Wang 1992; Chen and Zheng 1995a, 1995b, 1996; Xie and Zheng 2003; Xie and Ou 2005; Xie 2005; Xie et al. 2006; Cigliano et al. 2023). Compared with other Gryllinae species, Goniogryllus species are characterized by yellow stripes extending from occiput, along the eyes to the lateral ocelli; no wings or scaly wings; no tympanum on the fore tibia (Chopard 1969; Wu and Wang 1992; Xie et al. 2006).

Goniogryllus is primarily identified based on morphological features, such as the presence or absence of pubescence on the thorax and abdomen, yellow stripes on the head with or without bifurcation, the color of the maxillary palpus, hind femur or pronotum, and the number of dorsal spurs on the hind tibia (Chopard 1936; Wu and Wang 1992; Chen and Zheng 1995b, 1996). However, some species possess similar morphological features with only minor differences. For example, there is no significant morphological difference between G. cirilinearis Xie, 2005 and G. atripalpulus Chen & Zheng, 1996 (both species have four yellow-brown spots on the 2th to 8th segments of the abdomen). The only difference mentioned by Xie (2005) is that G. cirilinearis has two short reddish-brown stripes on its occiput, whereas G. atripalpulus does not. Additionally, over half of the species (11 species) have been published with only holotype specimens, which is higher than the 21.53% reported in other insects (Deng et al. 2019). Consequently, further revision of the taxonomy of Goniogryllus is needed.

The genus Qingryllus Chen & Zheng, 1995 also belongs to Gryllinae, and its type species, Q. striofemorus Chen & Zheng, 1995, is collected from Xunyangba and Huoditang, Ningshan, Shaanxi, which are also the collecting sites for the holotype of G. ovalatus Chen & Zheng, 1996 and G. atripalpulus (Chen and Zheng 1995a, 1996). These two genera share a similar appearance, both possessing a pair of yellow-brown longitudinal stripes on each side of the head, and lacking a tympanum on the fore tibia. The main difference lies in the fully developed forewings and hindwings of Qingryllus, while those of Goniogryllus are either underdeveloped or apterous (Chen and Zheng 1995a; Liu et al. 2017).

He (2018) compared sequences of the COI gene of G. ovalatus and Q. striofemorus, and found them to be identical. He (2018) suggested that Qingryllus is a junior synonym of Goniogryllus, but did not make further revisions. Yang et al. (2019) resurrected Qingryllus as a valid genus, because they considered that species of “two genera had different habitus and habitats”. However, they did not provide additional evidence.

The holotype of Callogryllus yunnanus Wu & Zheng, 1992 was collected from Yunnan, China. It bears a resemblance to the above two genera. The unique feature of this species is the presence of a pair of short forewing with irregular veins in males. Wu and Zheng (1992) placed this species in the genus Callogryllus Sjöstedt, 1910, likely due to its morphological similarities to Callogryllus species from India. However, they did not study the morphology of male genitalia. In fact, C. yunnanus in females lack wings, which is very similar to Goniogryllus species, and males are also very similar if its forewing are not considered as its features.

To clarify the relationship among partial Goniogryllus, Qingryllus and C. yunnanus, and to revise partial species in the genus Goniogryllus, we reconstructed a phylogenetic tree based on molecular markers (COI gene) and estimated the number of molecular operational taxonomic units (MOTUs) by three methods, Automatic Barcode Gap Discovery (ABGD), Assemble Species by Automatic Partitioning (ASAP), and Bayesian Poisson Tree Processes (bPTP). We also combined it with morphological classification to evaluate these morphological features.

The identification of Goniogryllus is primarily based on its morphological characteristics (Chopard 1936; Wu and Wang 1992; Chen and Zheng 1995b, 1996). In this study, we examined specimens from various provinces in China including Chongqing, Fujian, Gansu, Hubei, Shaanxi, Sichuan, Yunnan and Zhejiang (Fig. 2). Male genitalia showed that there are only four different types in our study, which indicated a diversity of species lower than current recognized. However, if identification is based on a combination of pubescence, color and wings, they can be classified as eleven different species. Therefore, morphological features in this genus display inconsistent.

The COI gene, also known as DNA barcoding, has proven to be a valuable tool for identifying species, including crickets (Dong et al. 2018; Tan et al. 2021; Li et al. 2021). Additionally, the use of COI gene is encouraged in the publishing of new species (Liu et al. 2018; Tanga et al. 2018; Chen et al. 2019; Li et al. 2021; Liu and He 2022). The single-gene can be used to distinguish different species well. Kim et al. (2022) have suggested that the COI gene is useful for identifying species in the genus Teleogryllus Chopard, 1961, but they note that detailed morphological investigations are essential for DNA barcoding before undertaking any molecular study. Hence, in this study, we utilized the COI gene to evaluate the morphological features.

The presence or absence of pubescence on the thorax and abdomen is a classification feature of G. octospinatus and G. pubescens Wu & Wang, 1992 (Chen and Zheng 1995b). In our results, G. chongqingensis (Fig. 3F, G), G. punctatus (Fig. 3H), G. yunnanus (Fig. 3K), and G. emeicus (Fig. 3I) are always glabrous. Male G. glaber is also glabrous (Fig. 3J). It is more complex in G. ovalatus: pronotum are always pubescent; individuals without wings or with scaly wings are pubescent on mesonotum and metanotum, the abdomen of females is pubescent, while that of males is pubescent only on the first three segments; for individuals with fully developed wings (= Qingryllussyn.), they are always glabrous on mesonotum, metanotum and abdomen (Fig. 3A–C). The pubescence of G. atripalpulus is similar to G. ovalatus without wings or with scaly wings, but the pronotum of G. atripalpulus is glabrous (Fig. 3D, E). Thus, the presence or absence of pubescence on the thorax and abdomen is an important feature (interspecific difference) as in Table 2.

Yellow stripes on the head with or without bifurcation is an important feature mentioned by Wu and Wang (1992), Chen and Zheng (1995b), and their opinion is supported by our results. The yellow stripes of G. ovalatus (Fig. 4A–C), G. chongqingensis (Fig. 4D, E), G. emeicus (Fig. 4F), G. yunnanus (Fig. 4H) and G. glaber (Fig. 4G) do not bifurcate, while those of G. atripalpulus (Fig. 4I, J) and G. punctatus (Fig. 4K) always bifurcate. Thus, yellow stripes on the head are another specific feature.

The color of the maxillary palpus is another classification feature (Wu and Wang 1992; Chen and Zheng 1995a, 1996; Xie 2005). For example, G. atripalpulus is similar to G. bistriatus Wu & Wang, 1992, but they are different in: the maxillary palpus of G. atripalpulus is black, while there are yellow stripes on the apex of the 4^th joint and the dorsal side of the 5^th joint of G. bistriatus (Wu and Wang 1992). Our results are consistent with theirs: there are yellow stripes on the apex of the 4^th joint and the dorsal side of the 5^th joint of maxillary palpus of G. ovalatus (Fig. 5A–C), G. chongqingensis (Fig. 5D, E) and G. emeicus (Fig. 5F). There are yellow stripes on the dorsal side of the 5^th joint of G. yunnanus (Fig. 5G), while the maxillary palpus of G. atripalpulus (Fig. 5H, I), G. punctatus (Fig. 5J) and G. glaber (Fig. 5K) are totally black. Thus, the color of the maxillary palpus can be a classification feature.

Most studies describe the male genitalia of most Goniogryllus species (Wu and Wang 1992; Chen and Zheng 1995a, 1995b; Xie and Ou 2005; Xie et al. 2006), because the structure is important in modern cricket classification (Randell 1964; He et al. 2009; Gorochov 2015; Tan et al. 2021; Ma et al. 2021). However, the male genitalia of Goniogryllus shows minor variations in different species. In our study, Goniogryllus are very similar in genitalia except G. glaber (Fig. 6A) and G. yunnanus (Fig. 6B). There is no process on the inside of the apical lobes of epiphallus of G. atripalpulus (Fig. 6E, F). On the other side, there are processes on the inside of apical lobes of epiphallus of G. ovalatus (Fig. 6C, D), G. chongqingensis (Fig. 6G), G. punctatus (Fig. 6H), and G. emeicus (Fig. 6I).

The color of the pronotum has been mentioned as a classification feature by Wu and Wang (1992), Chen and Zheng (1995a, 1996), and Xie and Zheng (2003). For example, G. chongqingensis and G. octospinatus are distinguished by the color of the pronotum (Chen and Zheng 1995b). However, our results reveal that there is diversity in the color of pronotum in G. ovalatus, G. chongqingensis, G. punctatus, and G. atripalpulus. The pronotum of G. ovalatus (Fig. 7A–C), G. chongqingensis (Fig. 7F, G), G. punctatus (Fig. 7H) and G. atripalpulus (Fig. 7D, E) can be black, or with various yellow spots or stripes on both sides. On the other hand, the color and pattern of the pronotum remain consistent in G. emeicus (Fig. 7I), G. glaber (Fig. 7J) and G. yunnanus (Fig. 7K). Therefore, this feature should be used cautiously in the classification of at least the aforementioned species. Additionally, one distinguishing features between G. cheni Xie & Zheng, 2003 and G. atripalpulus is the presence of yellowish-brown transverse stripes near the anterior margin of the pronotum in G. cheni, while they are absent in G. atripalpulus (Xie and Zheng 2003). Thus, further verification is still required to validate the previous classification based on this morphological feature.

Wu and Wang (1992) mentioned that G. bomicus Wu & Wang, 1992 could be distinguished from other species by the color of hind femur. Besides, it is one of the morphological features for distinguishing G. cheni from G. atripalpulus mentioned by Xie and Zheng, 2003 (a yellow longitudinal stripe only at the apex of the hind femur of G. cheni, while a yellow longitudinal stripe from one third to the apex of the hind femur of G. atripalpulus). However, this may be caused by limited samples. In our results, the color of the hind femur is varied. For example, the hind femur of G. ovalatus can be black or yellow, or with a yellow longitudinal stripe on the superior margin with varied lengths (Fig. 8A–C). It is in the same situation for G. atripalpulus (Fig. 8D, E), G. chongqingensis (Fig. 8F, G), G. punctatus (Fig. 8H), G. yunnanus (Fig. 8K), and G. emeicus (Fig. 8I). The color of the hind femur of G. glaber is yellow with black patterns (Fig. 8J). Thus, we treat the color of the hind femur as one of the intraspecific variations.

Wu and Wang (1992), Chen and Zheng (1995b, 1996), Xie and Zheng (2003), and Xie (2005) point out that the number of dorsal spurs on the hind tibia is a classification feature. There are four outer spurs and four inner spurs of G. octospinatus, and three outer spurs and four inner spurs of G. pubescens (Chen and Zheng 1995b). Our results indicate that the number of dorsal spurs on the hind tibia can be varied in G. ovalatus, G. punctatus, G. yunnanus, and G. glaber. G. ovalatus possesses three or four outer dorsal spurs and four or five inner ones (Fig. 9A–C). G. punctatus possesses three outer ones and four or five inner ones (Fig. 9H). G. glaber possesses three or four outer ones and three inner ones (Fig. 9J). G. yunnanus possesses three outer ones and three or four inner ones (Fig. 9K). The number of dorsal spurs on hind tibia is stable in G. atripalpulus (Fig. 9D, E), G. chongqingensis (Fig. 9F, G), and G. emeicus (Fig. 9I) as 4/4, 4/4, 4/4 (outer/inner), respectively. Thus, the classification depending on the number of dorsal spurs on the hind tibia is questionable in some species. For example, one of the distinguishing features between G. atripalpulus and G. sexflorus Xie & Zheng, 2003 is that the former possesses four outer dorsal spurs and four inner ones, but the latter possesses three outer dorsal spurs and four inner ones (Xie and Zheng 2003).

Our molecular analysis revealed that the specimens in this study can be classified into seven species (six species according to the bPTP method), which supports the classification based on the presence or absence of pubescence on the thorax and abdomen, yellow stripes on the head with or without bifurcation, the color of the maxillary palpus, and the shape of male genitalia. However, in Hubei, Shaanxi, and Sichuan, we encountered some “species” that appear different if we rely solely on the color of the hind femur, the color of the pronotum, and the number of dorsal spurs on the hind tibia for identification. When we examined the presence of pubescence on the thorax and abdomen, yellow stripes on the head with or without bifurcation, the color of the maxillary palpus and the shape of male genitalia, these seemingly different “species” showed no discernible variation. Furthermore, our molecular results indicate that they actually belong to the species. Therefore, we consider these features as intraspecific variations.

The presence of specific structures in the male genitalia serves as a mechanism to prevent mating between different species (hybridization), and genital morphology is commonly used for identification purposes in crickets. However, the genitalia of Goniogryllus exhibit only minor differences. Nevertheless, molecular analysis and parts of morphological features provide support for classifying them as distinct species. This may be the first example that genitalia cannot be used for identification in crickets as we know. Our hypothesis is that Goniogryllus species inhabiting high-altitude regions (Chopard 1969; Wu and Wang 1992; Chen and Zheng 1995a, 1995b, 1996; Xie and Zheng 2003; Xie and Ou 2005; Xie 2005; Xie et al. 2006) face geographical barriers that result in reduced selective pressure on their genitalia. Consequently, male genitalia offer limited assistance in distinguishing between species.

G. yunnanuscomb. nov. (= C. yunnanus) possesses forewings with irregular veins (Fig. 3K), which may lead to confusion regarding its generic classification. However, based on its genitalia morphology and molecular results, it is evident that this species belongs to the genus Goniogryllus. In future study, it would be intriguing to explore the origin and speciation of species in this genus by incorporating additional genes or conducting phylogenomic analyses.

In conclusion, some morphological features, such as the presence or absence of pubescence on the thorax and abdomen, yellow stripes on the head with or without bifurcation, the color of the maxillary palpus and the shape of male genitalia are interspecific differences, and these can be used as classification features. However, the color of the hind femur and pronotum, and the number of dorsal spurs on hind tibia show intraspecific variations in some species. Therefore, the classification based solely on these morphological features requires verification, particularly for the aforementioned species with only intraspecific variations when published. We recommend that future studies consider utilizing these interspecific differences, increase the sample size, or incorporate molecular data for more robust classification.