A robust phylogenetic framework is important for resolving the circumscription of the genus
Sinojackia and for investigating evolutionary patterns and processes. Our ML phylogeny received strong support at most nodes of the tree.
Sinojackia was well-supported as a monophyletic group and sister to
Pterostyrax in the current study, in agreement with previous phylogenetic analyses [
35,
36]. Two main clades were strongly supported in our results, but the systematic placement of
S. oblongicarpa leads to the contradictions between our results and previous research [
32]. Yao et al. (2008) utilized several barcodes, including ITS,
psbA-trnH, and seven microsatellite loci, to assess the phylogenetic relationship within
Sinojackia, supported
S. oblongicarpa were sister to
S. microcarpa and formed one clade with
S. sarocarpa and
S. huangmeiensis. However, all branch nodes received extremely low bootstrap rate within
Sinojackia (< 50%). The topological structure of our phylogenetic tree is consistent with previous studies based on plastomes [
20,
35,
36,
38,
39], and result in well-resolved and highly-supported phylogenies. We are convinced that our results have better interpretation ability.
S. oblongicarpa differs from
S. sarcocarpa in its shrubby habit, smaller flowers (1.2-1.5cm long) and oblong fruits [
27].
S. huangmeiensis can be distinguished from
S. xylocarpa by its smaller flowers, broad ovate petals, smaller and grey-brown fruits with a papillate short beak [
23]. Previous research implied that the weight of
S. oblongicarpa is related to environmental competition, and the size of the flowers and the shape of the fruits are within the variation range of
S. sarcocarpa; therefore,
S. oblongicarpa was retreated as a synonymy of
S. sarcocarpa [
14]. Subsequently,
S. rehderiana and
S. huangmeiensis were retreated as a synonym of
S. xylocarpa based on morphological characters [
15]. This was inconsistent with our phylogenetic tree, in which
S. oblongicarpa was close to (
S. xylocarpa +
S. rehderiana), and formed distant phylogenetic relationships with
S. sarcocarpa. Evidence from rDNA-ITS phylogenetic tree also demonstrated that
S. oblongicarpa and
S. sarcocarpa were not sister species [
37]. Therefore, we suggest that
S. oblongicarpa should be treated as an independent species. In terms of the taxonomy of
S. huangmeiensis, both our results and previous molecular phylogenetic analyses reveal that
S. huangmeiensis form a distant phylogenetic relationship with
S. xylocarpa [
20]. The taxonomic relationship between
S. rehderiana and
S. xylocarpa remains a challenge. Hybridization tests were successful in ex situ collections between
S. rehderiana and
S. xylocarpa [
81], providing that gene flow may weaken the boundary between these two species. In general, geographical isolation may hinder the occurrence of interspecific hybridization events. The distribution of
Sinojackia is highly fragmented, enhancing the difficulty of gene flow. Therefore, some scholars have speculated that
Sinojackia may have a wide distribution range in the past, habitat destruction and self-propagation constraints may result in fragmented distribution pattern that seem as today [
32]. Thus, extensive taxon sampling and molecular characterization are essential in further taxonomic studies between
S. rehderiana and
S. xylocarpa.