Extinct No More: DNA Evidence Reveals ‘Lost’ Chinese Plant Was Never a Species at All

In the misty forests of Luofu Mountain in Guangdong Province, South China, a botanical mystery that had lingered for more than a century has finally been resolved. Callicarpa oligantha, a shrub known to science from a single specimen collected in 1917, was declared possibly extinct in the provincial Red List of Higher Plants, making it one of only two plant species ever given that grim designation in Guangdong. Now, after researchers rediscovered a living population at its type locality in 2024, a new study published in PhytoKeys has delivered a verdict that is as surprising as it is consequential: the ‘extinct’ species never existed as a distinct species at all. Instead, it is simply a variant of a widespread and thriving relative, Callicarpa brevipes.

The story began with the American botanist Elmer Drew Merrill, who in August 1917 collected a single individual of a beautyberry shrub in thickets along a small stream at roughly 900 meters elevation on Luofu Mountain. Merrill described the plant as a new species, Callicarpa oligantha, in 1918, noting its alliance with C. dichotoma but distinguishing it by its longer, narrower leaves, subsessile and abaxially pubescent foliage, shorter filaments, and cymes bearing very few flowers. Crucially, his description rested on a single gathering from one individual. For nearly 110 years afterward, no additional plants matching that description were ever documented, and the species drifted into taxonomic limbo, categorized as data deficient in China’s national Red List and never assessed by the IUCN.

Doubts about the species’ validity surfaced early. In his 1951 review of the Chinese species of Callicarpa, Chang considered C. oligantha closely allied to C. brevipes and suggested it might merely represent a morphological variant of that species. The problem was that no one could test the idea, because there was nothing to test: no fresh material, no living plants, no DNA. That changed when a field survey at Luofu Mountain rediscovered a population of Callicarpa closely matching Merrill’s original description, with one notable exception. Some individuals bore pubescent fruits and occasional hairs on the lower leaf surfaces, traits absent from the protologue, likely because Merrill had only ever seen a single specimen from a single plant.

Armed with newly collected material, the research team launched an integrative taxonomic reassessment combining molecular phylogenetics with statistical morphological comparisons. For the genetic work, they sampled five individuals of C. oligantha from the type locality and three individuals of C. brevipes from Guangdong, Hainan, and Vietnam, embedding them within a broad phylogenetic frame of twenty Callicarpa species, with two Australian outgroup taxa anchoring the analysis. Total genomic DNA was extracted from silica-dried leaf tissue using the CTAB method, and the team amplified and sequenced two nuclear ribosomal regions, nrITS and nrETS, together with eight chloroplast DNA markers previously shown to be effective for resolving relationships in this genus.

The analytical strategy was deliberately conservative. To reduce potential bias in phylogenetic reconstruction, the researchers applied both concatenation-based and coalescent-based methods. Maximum likelihood analyses with 5,000 ultrafast bootstrap replicates and Bayesian inference sampling five million trees were run on concatenated datasets totaling 1,087 base pairs of nuclear ribosomal DNA and 7,435 base pairs of chloroplast DNA. A multispecies coalescent approach in StarBEAST2 was used to infer species trees, accounting for incomplete lineage sorting, a process that can seriously mislead standard tree-building methods. The results were unambiguous: Callicarpa oligantha was intermingled with C. brevipes in both nuclear and plastid datasets, with strong statistical support, and the coalescent species trees recovered C. oligantha as sister to C. brevipes. In other words, the ‘extinct’ species is not an evolutionarily independent lineage but part of the variation within C. brevipes.

The morphology told the same story. The team measured six characters regarded as key for distinguishing Callicarpa species: leaf length, leaf width, petiole length, inflorescence width, inflorescence branching, and filament length, using ImageJ software on both the new collections and historical type specimens, including the isotype of C. oligantha preserved at the United States National Herbarium and the lectotype of C. brevipes at Kew. Principal component analysis of leaf morphology explained 91.4 percent of the variance in the first two axes, and flower morphology 94.6 percent, with C. oligantha and C. brevipes forming a single cluster in every analysis. T-tests found no significant differences between the two in any of the measured characters. The ranges overlap almost completely: leaf length spans 8.5 to 14.5 centimeters in C. oligantha versus 8.4 to 17.5 centimeters in C. brevipes, and the two species share the same lanceolate to narrowly lanceolate leaf blades. The distributions of the two taxa also overlap entirely across southeastern China and Vietnam.

Based on this congruent molecular and morphological evidence, the authors formally propose synonymizing C. oligantha under C. brevipes, following Article 11 of the International Code of Nomenclature for Algae, Fungi, and Plants, the Shenzhen Code. As part of the taxonomic treatment, they also designated a lectotype for C. brevipes, a Champion specimen at Kew that matches the original description, untangling a complicated nomenclatural history that stretches back to Hooker’s 1825 misidentification of a living Chinese plant as Callicarpa longifolia. The rediscovered population, once a candidate for urgent protection as a last-of-its-kind survivor, is now understood to be a population of a species assessed as least concern, with stable populations and no major threats across its range in Guangdong, Guangxi, Hainan, Hong Kong, and Vietnam.

The implications reach far beyond one shrub. The study joins a growing body of evidence showing how shaky taxonomy can distort conservation priorities. The authors point to Fagus chienii, a beech designated as a key protected species in 2016 despite being known from a single 1935 collection, only for molecular work to reveal it is conspecific with F. hayatae. They also cite Torreya grandis var. jiulongshanensis, once listed as a Category II protected plant, which phylogenetic study showed to be a natural hybrid rather than a distinct lineage. Perhaps most striking is the clover fern Marsilea azorica, managed as critically endangered under the Bern convention and the European Union’s habitats directive, until researchers demonstrated it was conspecific with an invasive species and recommended its removal from every conservation list it occupied.

For Callicarpa oligantha, the practical consequences were immediate and concrete. Local forestry authorities had been likely to include the species in the list of key protected wild plants, directing substantial conservation effort toward a name rather than a genuinely unique lineage. The new study recommends excluding C. oligantha from extinction lists, freeing those resources for species facing real threats. The authors note that the only other plant assessed as probably extinct in Guangdong, the mycoheterotroph Corsiopsis chinensis, remains a genuinely enigmatic case, known from an unusual specimen collected in 1999 and not observed in field surveys since.

The broader lesson is one that conservation biology keeps relearning: accurate species delimitation is the foundation on which every red list entry, protected area decision, and recovery plan ultimately rests. When species boundaries are drawn from single specimens or superficial resemblance, limited resources can flow toward phantom extinctions while genuinely endangered lineages go unprotected. The researchers emphasize that robust taxonomic frameworks, ideally built on integrative, multidisciplinary evidence combining DNA sequences, morphology, and fieldwork, must be established before conservation action is taken. They also call for population genomic studies with expanded sampling across the full distribution of C. brevipes to reveal patterns of intraspecific genetic diversity that single-locus phylogenies cannot capture. A century-old enigma has been closed, an extinction has been un-declared, and a small purple-fruited shrub on a mountain in South China has become an unlikely poster child for the quiet, essential science of taxonomy.

Subject of Research: Integrative taxonomic reassessment of the supposedly extinct plant Callicarpa oligantha and its synonymy with Callicarpa brevipes

Article Title: Rediscovery and reexamination of a declared 'extinct' plant in South China: The species delimitation and conservation implications of Callicarpa oligantha ( Lamiaceae )

Article References: Cai, H., Yang, Q., Xu, Y., Tu, T., & Ma, Z. (2026). Rediscovery and reexamination of a declared 'extinct' plant in South China: The species delimitation and conservation implications of Callicarpa oligantha (Lamiaceae). PhytoKeys, 278, 157-176. https://doi.org/10.3897/phytokeys.278.203604

Image Credits: AI Generated

DOI: 10.3897/phytokeys.278.203604

Keywords: Callicarpa oligantha, Callicarpa brevipes, taxonomy, species delimitation, rediscovery, extinction, conservation, phylogenetics, Guangdong, Lamiaceae, biodiversity, red list