The results of phylogenetic analysis, based on the whole chloroplast genome and morphological study support the transfer of a long ignored bamboo species, Sasa gracilis, to the recently established genus, Sinosasa, in this study. Morphologically, this species differs from all the other known Sinosasa species by having very short (2–3 mm) foliage leaf inner ligules, which is unusual in this genus. A revised description of its morphology and colour photos are also provided.

bamboo, phylogeny, Sinosasa, taxonomy

Sinosasa L.C.Chia ex N.H.Xia, Q.M.Qin & Y.H.Tong was recently segregated from Sasa Makino and Shibata (1901) to accommodate some species previously placed in Sasa subg. Sasa from China, based on morphological and phylogenetic evidence (Qin et al. 2021). This genus differs from Sasa in having raceme-like (vs. panicle-like) synflorescences, two to three (vs. four to ten) florets per spikelet with a rudimentary terminal floret, three (vs. six) stamens and two (vs. three) stigmas per floret, wavy (vs. usually flat) foliage leaf blades when dry and relatively long (> 1 cm) (vs. short) foliage leaf inner ligules (Qin et al. 2021). Up to now, Sinosasa contains seven species endemic to subtropical areas of China and usually found growing along the river valley or in moist areas under evergreen broad-leaved forests at elevations of 700–1200 m (Qin et al. 2021).

Sasa gracilis B.M.Yang (1990) was described based on the only collection B. M. Yang 06774 from Shangmuyuan, Jiangyong County, Hunan Province, China. After its publication, it is only recognised in ‘Bamboos of Hunan’ (Yang 1993), edited by the author of this name, ‘Iconographia Bambusoidearum Sinicarum’ (Yi et al. 2008) and its English version ‘Illustrated Flora of Bambusoideae in China’ (Shi et al. 2022). However, because of the narrow circulation of the publication Acta Scientiarum Naturalium Universtis Normalis Hunanensis (later the name was changed to Journal of Natural Science of Hunan Normal University) at that time in China (Deng and Xia 2014), this species was ignored by the widely distributed monographs, such as ‘Flora Reipublicae Popularis Sinicae’, ‘Flora of China’, ‘World Checklist of Bamboos and Rattans’ (Hu 1996; Wang and Stapleton 2006; Vorontsova et al. 2016), the well-known database GrassBase-The Online World Grass Flora (Clayton et al. 2016) and some important websites like http://www.ipni.org, http://www.tropicos.org and http://www.theplantlist.org. In the protologue, this species was described to possess a suite of vegetative characters, such as solitary branches at each branching node, strongly raised supranodal ridges and wavy foliage leaf blades when dry, which fit well with the circumscription of Sinosasa. However, this species has very short foliage leaf inner ligules that are only 2–3 mm long, while all hitherto known Sinosasa species typically have more than 1 cm long inner ligules. Therefore, the taxonomic position of Sasa gracilis needs a further study.

The specimens of Sasa gracilis were collected from its type locality during a field trip in September 2022. Fresh foliage leaves were deposited in silica gel for DNA extraction. Type specimens of Sasa gracilis deposited in the Herbarium of Hunan Normal University (HNNU) were examined. Observations and measurements were taken using a magnifier and a ruler with a scale of 0.5 mm. Some minor characters like the indumentum were observed with a stereomicroscope (Mshot MZ101). The morphological terms follow McClure (1966) and Beentje (2016). Herbarium acronyms follow Thiers (2022, continuously updated).

To study the phylogenetic position of Sasa gracilis within the tribe Arundinarieae, the whole chloroplast genomes were used for building the phylogenetic tree. A total of 24 representatives belonging to all the five subtribes of the tribe Arundinarieae (Zhang et al. 2020a) were sampled and Bambusa multiplex (Lour.) Raeusch. ex Schult. f. from the tribe Bambuseae was used as outgroup. All the sampled taxa, as well as their voucher information and GenBank accession numbers, are listed in Table 1.

The chloroplast genome size of Sasa gracilis is 140,013 bp and those of all the samples ranged from 139,394 bp (Bambusa multiplex) to 140,064 bp (Gaoligongshania megalothyrsa (Hand.-Mazz.) D.Z.Li, Hsueh & N.H.Xia) with an alignment of 144,169 bp. The data matrix was characterised by sequence divergence with 3,688 variable sites (2.56%), including 773 parsimony informative sites (0.54%) and 2,915 singleton variable sites (2.02%). The phylogenetic trees, generated by the ML and BI methods, were generally consistent in topology, so only the ML tree was shown with nodal support values from both methods labelled on each node (Fig. 1). As shown in the phylogenetic tree, Sasa gracilis is distantly related to Sasa veitchii Rehder (= Sasa albomarginata (Miq.) Makino & Shibata, the type of Sasa), but forms a monophyletic clade with three Sinosasa species with strong nodal support (BS = 100% and PP = 1.00).

Figure 1. 

Phylogenetic tree reconstruction for Sasa gracilis, based on plastid genome dataset with Maximum Likelihood and Bayesian analyses. Bootstrap values and posterior probabilities are indicated at each node.

Our phylogenetic analysis and previous studies of Zeng et al. (2010), Zhang et al. (2012), Guo et al. (2021) and Qin et al. (2021) all demonstrated that Sinosasa is monophyletic. Molecular evidence, based on plastid genomic data, further confirmed that Sasa gracilis should be a member of Sinosasa rather than Sasa. Morphologically, the characters of this species also match well with those of Sinosasa species as mentioned above. The short inner ligule (2–3 mm) of Sasa gracilis can easily differentiate this species from all the other Sinosasa species. Consequently, the previous circumscription of Sinosasa on the length of foliage leaf inner ligule should be modified. In other words, Sinosasa does not always have long (typically > 1 cm) foliage leaf inner ligules as we knew previously. If this character is excluded, Sasa gracilis is somewhat similar to two other Sinosasa species also with culm leaf auricles, viz. Sinosasa magninoda (T.H.Wen & G.L.Liao) N.H.Xia, Q.M.Qin & X.R.Zheng and Sinosasa guangxiensis (C.D.Chu & C.S.Chao) N.H.Xia, Q.M.Qin & X.R.Zheng. It further differs from Sinosasa guangxiensis by the glabrous (vs. densely strigose) abaxially mid- and upper part of the culm leaf sheath with short (0.5–1 mm vs. 1–3 mm) ligules and the glabrous (vs. puberulent) foliage leaf sheath and from Sinosasa magninoda by having culm leaf with shorter (0.5–1 mm vs. 1–3 mm) ligules, larger (2–4 × 1–2 mm vs. 1–1.5 × 1 mm) auricles and (vs. absent) oral setae and foliage leaf with more (7–12 vs. 1–4) and longer (8–20 mm vs. 5–10 mm) oral setae. A more detailed comparison amongst the three species is provided in Table 2. Based on the above evidence, it is concluded that Sasa gracilis represents a distinct species of Sinosasa and a new combination of it under Sinosasa should be made.

We thank Xing-Min Zhou for his assistance in the fieldwork. This research was funded by the Guangdong Basic and Applied Basic Research Foundation (grant no. 2021A1515011302) and National Natural Science Foundation of China (grant no. 32270227).