Schizopygopsis chengi

Tian, Fei, Liu, Sijia, Chen, Shengxue & Zhao, Kai, 2025, Revision of Schizopygopsis chengi Fang 1936 (Cypriniformes: Cyprinidae), with a description of a new subspecies, Zootaxa 5590 (4), pp. 481-506 : 488-491

Validation of Schizopygopsis chengi

Morphological, genetic and phylogenetic evidence supports Schizopygopsis chengi as a valid species.

Schizopygopsis chengi is distinguished from other Schizopygopsis fishes by having an outside gill raker number of the first gill arch greater than 10, an inside gill raker number of the first gill arch greater than 16 and the commencement of a ventral fin under the 4–5 th branched ray of the dorsal fin ( Table 3). Schizopygopsis chengi differs from S. malacanthus in the following characteristics: soft unbranched rays of dorsal fins with small and fewer serratures at the posteriors edge (vs. strong unbranched rays of dorsal fin with obvisous serratures in S. malacanthus ); no black dots on the dorsal fin (vs. black dots on the dorsal fin of S. malacanthus ); large black spots on the body (vs. small dots on the back of the body or above the lateral line in S. malacanthus ); and an independent distribution area in the headstream of the Dadu River in western China (vs. the distribution of S. malacanthus in the upstream of the main stream of the Yangtz River in western China) ( Fig. 2 and Tables 3–4).

The measurements of morphological characteristics are shown in Table 4. PCA revealed that 49% of the total variance was explained by the first three components, including 24%, 14% and 11% for PC1, PC2 and PC3, respectively ( Table 5). Along PC1, the specimens were clearly separated into two groups, corresponding to S. chengi and S. malacanthus ( Fig. 3a ). PC1 loaded heavily on head length, preventral length, eye diameter, and IEW/ED, which distinguished S. chengi from S. malacanthus . Morphometric analysis indicated that S. chengi had longer predorsal, prepectoral, and preventral lengths than S. malacanthus . Furthermore, PCA revealed that DK populations of Schizopygopsis chengi were separated from S. c. baoxingensis, MK and KK populations, and the latter two populations could not be partitioned by traditional morphometric data ( Fig. 3b and Table 6 ).

......continued on the next page ......continued on the next page ......continued on the next page

Character Component

PC1 PC2 PC3 PC4 PC5

0.26 0.11 0.16 -0.24 0.09

-0.23 -0.32 0.05 -0.19 -0.46

0.03 -0.16 -0.23 -0.59 0.17

0.33 -0.17 0.32 -0.16 0.18

0.24 0.14 0.11 0.10 0.08

0.24 0.38 0.49 0.59 0.66

Character PC1 PC2 PC3 PC4 PC5 BL/TL 0.12 0.12 -0.02 -0.26 0.11 BL/BH -0.05 0.26 -0.20 -0.05 -0.24 HL/BL 0.00 -0.36 -0.38 -0.02 -0.15 CPL/BL 0.05 0.05 -0.23 0.43 0.51 CPD/BL 0.16 -0.36 0.19 0.01 0.17 PDL/BL -0.24 -0.35 -0.19 -0.05 0.10 PPL/BL 0.07 -0.18 -0.29 0.17 -0.16 PEL/BL -0.28 -0.32 -0.14 0.15 -0.01 PAL/BL -0.20 -0.31 0.01 0.16 -0.08 HD/HL 0.13 0.24 0.31 0.22 -0.02 HW/HL -0.15 -0.15 0.41 0.34 0.09 ED/HL 0.47 -0.01 -0.19 0.09 -0.02

Sequencing of the Cyt b gene generated 36 haplotypes, including 24 in S. chengi and 12 in S. malacanthus . Additionally, we included 2 haplotypes of S. c. baoxingensis and S. m. malacanthus obtained from public databases (Liu, et al. 2015). Therefore, 26 and 14 haplotypes of S. chengi and S. malacanthus , respectively, were subjected to phylogenetic analysis. The haplotype network demonstrated the clear separation between S. chengi and S. malacanthus without any common haplotypes ( Fig. 4a ). The BI and ML trees yielded identical topologies, revealing that S. chengi did not form a sister lineage with S. malacanthus ( Fig. 4b ). Schizopygopsis malacanthus was grouped with other Schizopygopsis fish and Heizensteinia microcephalus , while S. chengi formed an independent lineage with strong support. Within S. chengi , samples from the DK River formed a monophyletic group, and the MK and KK populations were grouped together without phylogenetic differentiation. Two haplotypes of S. c. baoxingensis formed a monophyletic group in S. c. chengi ( Fig. 4b ).

Based on Cyt b haplotypes, the interspecific genetic distance among these species ranged from 0.019 (between S. pylzovi and S. microcephalus ) to 0.091 (between S. anteroventris and G. potanini ). The genetic distance between S. chengi and S. malacanthus was 0.076, which was greater than that of the other 13 comparisons ( Table 7). The genetic differentiations (F st) ranged from 0.654 (between S. malacanthus and S. microcephalus ) to 0.966 (between S. kialingensis and S. chengi ). The genetic differentiation between S. chengi and S. malacanthus was 0.834, which was greater than that between S. malacanthus-S. kialingensis , S. malacanthus-H. microcephalus , S. malacanthus-S. anteroventris , and S. malacanthus-S. pylzovi . ( Table 7). The genetic distance was 0.022 between the DK and MK populations and reached 0.024 between the DK and KK populations ( Table S1). The genetic distances within S. chengi were even greater than that between H. microcephalus and S. pylzovi (0.019) as well as those between H. microcephalus and S. kessleri (0.018). The F st values were 0.473 and 0.487 between DK populations and and S. c. chengi as well as between DK populations and S. s. baoxingensis, suggesting genetic differentiation among the geographic populations ( Table S1).Additionally, we found that the nucleotide and haplotype diversities were greater in the MK/KK populations than in the DK populations, which might be associated with the more restricted range of the DK population ( Table S2).

Species delimitation was carried out using bPTP and ASAP, both of which decisively supported that S. chengi and S. malacanthus represented two distinct species ( Fig. 5 ). Based on the bPTP model, 14 species were defined, with Bayesian support values ranging from 0.93 to 1.00. Schizopygopsis chengi and S. malacanthus were delimited as two species, which was supported by the posterior probabilities of 0.98 and 0.97 for the two lineages. ASAP identified the 14 best partitions based on pairwise genetic distance using Cyt b sequences ( Table S3). According to the lowest score of 2.0, the ASAP method delimited all the samples into 13 species, and S. chengi and S. malacanthus were separated into two taxonomic classifications ( Fig. 5 ).

It has three subspecies,i.e. Schizopygopsis chengi chengi , Schizopygopsis chengi baoxingensis and Schizopygopsis chengi duokeheensis , subsp. nov.

Figures

FIGURE 2. Pictures of S. chengi and S. malacanthus. a. The lateral view of S. chengi. b. Dorsal fin of S. chengi. c. The lateral view of S. malacanthus. d. Dorsal fin of S. malacanthus.

FIGURE 3. PCA analysis PCA analysis for S. chengi circles).

FIGURE 4. Population genetic analyses. a Haplotype network. Green, red, yellow and blue circles indicated haplotypes from MK, KK, BX, DK of S. chengi, and purple circles indicated haplotypes identified in S. malacanthus. b.The maximum likelihood tree. ML and BI trees demonstrated the same topology. The posterior probabilities and Bayesian supported values were labeled on the branches. The green and blue squares indicated samples from S. chengi and S. malacanthus.

FIGURE 5. The species delineation. a. Branch in blue was defined as a single species by bPTP analysis, and branch in red was rejected as a species by bPTP analysis. The numbers on branches indicated Bayesian support values. b. Squares in colors indicated all samples. The ASAP scores, the number of subsets, and the corresponding rank for were recorded on the top. The square with the same color is defined as a species. The top 10 partition results were listed.