Based on consulting original literature, the examination of specimens, and field investigations, Didymostigma trichanthera is shown to be conspecific with Henckelia anachoreta. Therefore, Didymostigma trichanthera is formally treated as a synonym of Henckelia anachoreta here.

China, Guangdong, Didymostigma trichanthera, Henckelia anachoreta, taxonomy

The genus Didymostigma W.T. Wang (1984) was once considered to be a monotypic genus, having only one species, D. obtusum (C.B. Clarke) W.T. Wang (1984). Subsequently, two new taxa, D. leiophyllum D. Fang & X.H. Lu (Fang et al. 1994) and D. trichanthera C.X. Ye & X.G. Shi (2005) were discovered and described. The type species of this genus, D. obtusum, is widely distributed from eastern Guangdong to southern Fujian, China (Wang et al. 1998; Li and Wang 2004; Wei et al. 2010). The other two species, however, are regarded as narrowly endemic species and have only been found at their type localities (Fang et al. 1994; Ye and Shi 2005).

Didymostigma trichanthera was simply described based on a single collection (Chuang-Xing Ye 5960) from Nankunshan, Guangdong Province, China. Ye and Shi (2005) thought that D. trichanthera mainly differs from D. obtusum by its lanose fertile anthers, hairy filaments, and the unhidden pistil to the corolla tube. The species status of D. trichanthera was once doubted by Wei et al. (2010). They find the calyx of D. trichanthera is completely different from the type species D. obtusum, but they did not carry out further study on this questionable species. After consulting original literature about this genus, type species, and this so-called new species (Wang 1984; Ye and Shi 2005), carrying out field investigations in Nankunshan, and conducting examinations of type materials of D. trichanthera, we are convinced that D. trichanthera is not a species belonging to the genus Didymostigma. Our detailed morphological comparisons find that this so-called Didymostigma species is actually conspecific with Henckelia anachoreta (H.F. Hance) D.J. Middleton & Mich. Möller (Weber et al. 2011). Consequently, it is essential to reduce Didymostigma trichathera to a synonym of Henckelia anachoreta.

We performed detailed comparisons of type materials between Didymostigma trichanthera and the type species of Didymostigma (D. obtusum), and also between D. trichanthera and Henckelia anachoreta. The study of specimens was conducted in IBK, IBSC and SYS. We also checked high-resolution digital images of the specimens in A (https://huh.harvard.edu/), BM (http://data.nhm.ac.uk/), E (http://www.rbge.org.uk/), K (https://www.kew.org/), P (https://science.mnhn.fr/institution/mnhn/search), TI (http://umdb.um.u-tokyo.ac.jp/Dshokubu/Tshokubu.htm) and WU (http://herbarium.univie.ac.at/index.htm) by their web service, as well as via online databases, such as the Chinese Virtual Herbarium (http://www.cvh.ac.cn/), JSTOR Global Plants (http://plants.jstor.org/) and Specimens Database of Native Plants in Taiwan (http://www.hast.biodiv.tw/Specimens/specimenQueryC.aspx). Moreover, detailed morphological studies of both H. anachoreta and Didymostigma obtusum were undertaken based on plants from natural populations at Nankunshan. Some other field observations of these two species were carried out in Guangxi, Guangdong, Fujian of China.

We did not find any other Didymostigma species in Nankunshan in our field work, except for D. obtusum (Fig. 1B). However, Henckelia anachoreta, a species which is similar to Didymostigma obtusum in its vegetative characteristics to some extent, can be easily found here (Fig. 1A). In fact, Nankunshan has been intensively botanized in recent years, because of its high biodiversity (Chen et al. 2017; Xu et al. 2017), and these studies also did not find D. trichanthera. As we know, the natural environment of Nankunshan has been well protected since it was listed as a national forestry park in 1993. Nevertheless, D. trichanthera has not been found or collected again since it was described in 2005. It is difficult to believe that D. trichanthera, if indeed a well characterized species, is represented only by its type collection.

Figure 1. 

Living plants of Henckelia anachoreta (A, C, D) and Didymostigma obtusum (B, E) at Nankunshan. A, B habitat C side view of flower D, E opened corolla showing pistil and stamens. Scale bar: 1 cm. Arrow indicates calyx and stigma.

Unfortunately, the holotype of D. trichanthera cannot be found in SYS. According to the isotype (Fig. 2B) preserved in A and the description made by Ye and Shi (2005), we can clearly find that the calyx of D. trichanthera is 5-lobed to near middle, with an obvious calyx tube, and the lobes are triangular (Fig. 2B). In contrast, the representative calyx of Didymostigma is 5-parted to near base, and the lobes are lanceolate-linear (Fig. 1E; Wang 1984). Additionally, the typical stigma of Didymostigma is only ca. 1 mm long (Fig. 1E; Wang 1984). Therefore, although the detailed characteristics of the stigma of D. trichanthera have not been clearly observed in the isotype, the description (3–4 mm long) of it in the protologue (Ye and Shi 2005) indicates that it is not a typical character of Didymostigma. However, it seems that these two important characters of D. trichanthera are well matched with Henckelia Spreng. Our detailed comparisons find that there is no obvious difference between the isotype of Didymostigma trichanthera (Fig. 2B) and the holotype of Henckelia anachoreta (Fig. 2A), and between the images of Didymostigma trichanthera given by Ye and Shi (2005) and Henckelia anachoreta photo by us at Nankunshan (Fig. 1A, C). In fact, most of the descriptions of Didymostigma trichanthera in the protologue, especially the diagnosis characters (lanose fertile anthers, hairy filaments and the unhidden pistil to the corolla tube), are exactly matched with Henckelia anachoreta.

Figure 2. 

A Holotype of Henckelia anachoreta (F.H. Hance 11377, BM-000041739!) and B isotype of Didymostigma trichanthera (Chuang-Xing Ye 5960, A-00135544!). Arrow indicates calyx.

All the reasons mentioned above have prompted us to carefully consider that Didymostigma trichanthera is conspecific with Henckelia anachoreta. As a result, the taxonomic treatment of Didymostigma trichanthera needs to be made here.

This work was supported by grants from the NSFC-Guangdong Natural Science Foundation Joint Project (U1501211), Science Research Foundation of Guangxi Academy of Sciences (no. 2017YJJ23022), Guangxi Natural Science Foundation Program (2015GXNSFBB139004 & 2017GXNSFAA198006) and the National Natural Science Foundation of China (31860047). We thank Winston J. Goretsky (Calgary, Alberta, Canada) for reviewing and editing this work.