The Bambusoideae (bamboos), with over 1700 described species, is the third largest subfamily of the Poaceae (grasses), native to all continents except Antarctica and Europe (Clark et al., 2015; Soreng et al., 2022). Within the Bambusoideae, three tribes representing the three major lineages are recognized. The Olyreae comprise the ca. 130 species of herbaceous bamboos, which are smaller and much less lignified than their woody cousins. With the exception of one monotypic genus in New Guinea, the rest of the Olyreae is native to the Americas, where they commonly grow in the understory of tropical forests. The bulk of bamboo diversity, nearly 1600 species, is found within the two tribes of woody bamboos, the Arundinarieae (or temperate woody bamboos, ca. 580 species) and the Bambuseae (or tropical woody bamboos, ca. 1000 species). The woody bamboos share strongly lignified culms, dimorphic leaf development (culm leaves vs. branch leaves), and long flowering cycles culminating in gregarious monocarpy. Woody bamboos usually grow in forest gaps or along forest edges, but may also form bamboo-dominated forests or occupy savannas or high elevation grasslands. Bamboos are ecologically significant and often dominant elements in their habitats (Clark et al., 2015), and woody bamboos are also of paramount commercial value in many parts of the world, especially where they are grown for edible shoots or timber (Lucas, 2013). Although much progress in understanding the complex evolutionary history of bamboos has been made in the last decade (e.g., Guo et al., 2019; Chalopin et al., 2021), new species and genera continue to be described (e.g., Ye et al., 2021; Clark et al., 2022), resolving finer scale phylogenetic relationships is challenging, identification resources are often lacking or outdated, and bamboo ecology remains understudied.

Several factors may explain why the study of bamboos lags behind that of other grass groups. The complex morphology of woody bamboos requires specialized collecting techniques to gather high quality specimens, especially for those in the vegetative phase (Soderstrom and Young, 1983; Fadrique et al., 2020). The long reproductive cycles in woody bamboos (mostly 15–120 years) mean that flowering material is often lacking, and most general collectors only make specimens if flowering structures are present. This bias has led to the need for taxonomic specialization in bamboos, historically with relatively few experts active at any given time. The advent of genetic and genomic data demonstrated that woody bamboos exhibit a slower rate of molecular evolution relative to other grasses, likely due to the long flowering cycles, and confirmed that woody bamboos are uniformly polyploid (Ma et al., 2017; Chalopin et al., 2021). Several woody bamboo clades apparently have undergone recent rapid radiation (e.g., Fisher et al., 2014; Zhang et al., 2016), exacerbating the problem of resolving phylogenetic relationships. In summary, the bamboos have long been regarded as taxonomically difficult, even among grass specialists, and as noted above, many questions remain.

The three bamboo papers in this issue help to resolve some of these long-standing bamboo mysteries. Rakotonasolo et al. (2023) explore intergeneric relationships in the Hickeliinae, an Afro-Malagasy subtribe of paleotropical Bambuseae, using complete plastomes. Liu et al. (2023), using a combination of phylogenomics and morphology, resolve the status of three Chinese species originally described in Dinochloa and provide a treatment of Melocalamus (Bambuseae) in China. And in the Arundinarieae, Lv et al. (2023) test the utility of complete plastomes and nuclear ribosomal DNA (nrDNA) in resolving relationships within Fargesia, the most diverse genus in the tribe.

The Hickeliinae (9 genera, 32 species) is a small subtribe within the Bambuseae, but its morphological diversity represents a remarkable radiation primarily in Madagascar (Vorontsova et al., 2018). Analysis of individual plastid markers typically has not yielded satisfactory resolution within bamboo subtribes (Zhang et al., 2016; Hackel et al., 2018). Complete plastomes have been analyzed broadly for bamboos (Ma et al., 2014; Wysocki et al., 2015; Chalopin et al., 2021), but their utility in obtaining resolution and support within a bamboo lineage (i.e., among relatively closely related taxa) has not been adequately tested. Rakotonasolo et al. (2023), sampling 22 species (about 2/3 of the species diversity) of Hickeliinae, show convincingly that plastome phylogenetics can provide both strong resolution and support within a subtribe. This robust Hickeliinae phylogeny lays a solid foundation for ongoing work on the generic classification of the subtribe and analyses of its character evolution and biogeography.

Dinochloa and Melocalamus, two genera of paleotropical woody bamboos with a climbing or scrambling habit, resemble each other in the vegetative condition. Three species of Dinochloa were described from Hainan Island in 1940, in the vegetative state, and to this day are unknown in flower. Although their generic placement has long been in doubt, it was not possible to resolve their true identity until DNA sequencing tools and technology became available. The ddRAD-seq (double digest restriction-site associated DNA sequencing) analysis presented by Liu et al. (2023) unambiguously places these three species within Melocalamus and the vegetative morphology of the two genera is carefully compared. Best of all, Liu et al. include a beautifully illustrated treatment of all the species of Melocalamus in China, which will be useful for specialists and non-specialists alike.

Turning to the Arundinarieae, Fargesia exemplifies the challenges in studying a recently radiated bamboo lineage. Over 90 species have been described, many have long flowering cycles, the rate of molecular evolution is relatively slow, and species-level taxonomy is difficult. Standard barcode markers simply do not provide sufficient resolution (Triplett and Clark, 2010; Cai et al., 2012). The rigorous analysis of Lv et al. (2023) shows that complete plastomes and nrDNA are each more informative than standard barcodes, and overall nuclear markers performed better than plastid markers in discriminating among species of Fargesia, but challenges remain. The conflicting plastid and nuclear signals may be due to hybridization and introgression or incomplete lineage sorting, and additional exploration of the nuclear genome will be needed to resolve these questions.

Thanks to the current Bamboo Tree of Life project spearheaded by the De-Zhu Li research group, about 40 bamboo systematists around the world are tackling generic-level problems using both molecular and morphological data. Without doubt, this is the largest number of bamboo researchers that has ever been active at any one time. These three papers showcase methods that can be usefully applied to other bamboo questions, and we can look forward to satisfying solutions to more bamboo mysteries in the very near future.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (grant no. 32120103003).

Author contributions

L.G.C. wrote the manuscript.

Declaration of competing interest

I have no conflicts of interest relevant to this article.