The genus Myrcia (Myrtaceae, the guava family) is one of the largest angiosperm tree genera, with c. 800 species and a native range extending from southern Mexico to tropical Central and South America (POWO, 2024). The state of Rio de Janeiro is one of the smallest in Brazil (c. 44,000 km²) and one of the most densely populated (c. 16 million inhabitants; IBGE, 2024). It is a key region for Myrcia, harbouring 113 (28%) of the 402 species known to occur in Brazil (Santos et al., Reference Santos, Amorim, Burton, Fernandes, Gaem and Lourenço2020) in just 0.5% of the country.

Species of Myrcia are primarily pollinated by bees and produce edible fruits that are a significant food source for frugivores in the Atlantic Forest (Pizo, Reference Pizo, Levey, Silva and Galetti2000; Gressler et al., Reference Gressler, Pizo and Morellato2006; Staggemeier et al., Reference Staggemeier, Cazetta and Morellato2017), thus forming a key ecological component of this phytogeographic domain (Murray-Smith, Reference Murray-Smith, Brummitt, Oliveira-Filho, Bachman, Moat, Nic Lughadha and Lucas2009; Lucas & Bünger, Reference Lucas and Bünger2015). Despite their ecological importance, dozens of Myrcia species are known only from a small number of specimens and have not been recorded in the wild for a long time, qualifying them as so-called ‘lost species’ (Long & Rodriguez, Reference Long and Rodríguez2022). Of the 254 species of Myrcia recorded in the Brazilian Atlantic Forest (Santos et al., Reference Santos, Amorim, Burton, Fernandes, Gaem and Lourenço2020), 86 are known from ≤ 15 herbarium specimens (T. Fernandes, unpubl. data, 2020). This situation leads to an acute lack of primary data to support subsequent studies on ecology, evolution and conservation (see Rivers et al., Reference Rivers, Taylor, Brummitt, Meagher, Roberts and Lughadha2011; Goodwin et al., Reference Goodwin, Muñoz-Rodríguez, Harris, Wells, Wood, Filer and Scotland2020). Plant conservation science relies heavily on occurrence data from herbarium records, and gaps in taxonomic information in these collections can have serious implications for conservation planning and prioritization of conservation interventions (Colli-Silva et al., Reference Colli-Silva, Reginato, Cabral, Forzza, Pirani and Vasconcelos2020).

We rediscovered two endemic species of Myrcia during field expeditions in 2021–2022 in forest remnants in the state of Rio de Janeiro. Both Myrcia colpodes (Kiaerskou, Reference Kiaerskou and Warming1893) and Myrcia rubiginosa (Cambessèdes, Reference Cambessèdes, Saint-Hilaire, Jussieu and Cambessèdes1832) were known to have restricted geographical distributions and were last recorded in 1958 and 2005, respectively. They were previously known from only a few specimens (fertile branches) collected in the state of Rio de Janeiro, mostly in the 19th century.

Following the rediscoveries, populations were revisited according to their phenology, with visits scheduled to coincide with flowering and fruiting periods. The visits were made until November 2024 to gather information on population size and to collect fertile branches for deposit in scientific collections (collector T. Fernandes, field numbers 918, 925, 934, 943, 1007, 1088, 1264, 1332 and 1385 for M. rubiginosa, and 1006, 1013, 1060 and 1484 for M. colpodes). These specimens were deposited at the herbarium of the Rio de Janeiro Botanical Garden (RB) with duplicates distributed to herbaria located in Brazil (MBM, MBML, NIT, RBR, RFFP, SPF), France (P), the UK (K) and USA (NY, US; herbarium acronyms follow Thiers, Reference Thiers2024). In addition, five saplings of each species were collected and transplanted to a nursery in the Rio de Janeiro Botanical Garden, where they are currently being cultivated in the institution’s arboretum for ex situ conservation purposes. Following a careful examination of the herbarium specimens derived from our collections, we produced a comprehensive taxonomic treatment of the rediscovered taxa (Fernandes et al., Reference Fernandes, Lucas, Vasconcelos and Braga2025), providing updated morphological descriptions along with notes on their taxonomy, distribution, habitat, and phenology.

Myrcia colpodes (Plate 1a–d) was first described from specimens collected in 1862 from a forest fragment (Morro da Viração) in the municipality of Niterói, currently a protected area under municipal management (Parque Natural Municipal de Niterói; Fig. 1). The species was subsequently collected from unspecified localities in 1867 and 1880. The last collection was in 1958 in the Tijuca Massif in the city of Rio de Janeiro, most of which is currently a protected area under federal management (Parque Nacional da Tijuca; Fig. 1). We were unable to locate M. colpodes in either of these areas despite our targeted search efforts. However, in 2022, 64 years after the last collection, we rediscovered two populations of five mature trees each in forest fragments 30–40 km to the east in the municipality of Maricá, in a municipally administered protected area (Refúgio de Vida Silvestre Municipal das Serras de Maricá; Fig. 1). Despite being a protected area, we found no physical delimitation, signage or evidence of protective management. In addition to its proximity to farms and the tourist beach of Sacristia, a further threat in this location is the occurrence of the invasive Mauritius hemp Furcraea foetida (Asparagaceae).

Fig 1. Location of historical and current occurrence records of Myrcia colpodes and M. rubiginosa. The inset map shows the location of the main map in south-eastern Brazil. ES, Espírito Santo; MG, Minas Gerais; RJ, Rio de Janeiro; SP, São Paulo.

Plate 1 (a–d) Myrcia colpodes and (e–h) Myrcia rubiginosa in their natural habitat: (a), (e) trunk; (b), (f) young shoots; (c), (g) flowering branch with a flower highlighted; (d), (h) fruiting branch with a fruit highlighted. Photos: (a,b,d-h) Thiago Fernandes, (c) Carlos Ferreira.

Myrcia colpodes is categorized as Critically Endangered on the IUCN Red List (Canteiro & Lucas, Reference Canteiro and Lucas2019a). Given our findings regarding threats and population size, with an extent of occurrence (EOO) of only 75 km² and severely fragmented subpopulations, we recommend the categorization (IUCN, 2012, 2024) as Critically Endangered should be maintained based on criteria B1abiii; i.e. with EOO < 100 km² (B1), severely fragmented subpopulations (a) and a continuing decline inferred (b) in the area, extent and/or quality of the habitat (iii). Our updated assessment with additional data will be formalized by submission to IUCN.

Myrcia rubiginosa (Plate 1e–h) was first collected between 1816 and 1821 in the Tijuca massif (Fig. 1) but we were unable to locate it there, despite our targeted search efforts. However, in 2021 we found two populations in a protected area under state management (Parque Estadual da Serra da Tiririca, Fig. 1) in the municipality of Niterói. One population, located in Alto Mourão, comprises 12 mature individuals; the other, in Córrego dos Colibris, consists of a single mature individual. Following our rediscovery, we searched recent herbaria collections and found three additional samples of M. rubiginosa among misidentified plant material collected in this protected area in 1982, 2001 and 2005. Two further populations were subsequently identified in the Parque Natural Municipal de Niterói (Fig. 1), each comprising a single mature individual. Previously undocumented, the species’ ripe fruits are purple berries with a sweet, whitish pulp (Plate 1).

Myrcia rubiginosa is categorized as Data Deficient on the IUCN Red List (Canteiro & Lucas, Reference Canteiro and Lucas2019b) and the data presented here provide important new information. Despite its occurrence within protected areas, the species faces threats from habitat fragmentation, urban development and wildfires (SEA & INEA, 2015; SMARHS, 2021). Its EOO is 50 km², with severely fragmented subpopulations. Based on our findings, we reassess the extinction risk (IUCN, 2012, 2024) for M. rubiginosa as Endangered based on criteria B1abiii; i.e. with EOO < 5,000 km² (B1), severely fragmented subpopulations (a) and a continuing decline inferred (b) in the area, extent and/or quality of the habitat (iii). Our new assessment will be formalized by submission to IUCN.

To ensure that the data and recommendations from this research are used to develop and implement public policies by the Brazilian government, we will provide the relevant information to the Brazilian National Center for Conservation of Flora (CNCFlora), which is the entity responsible for publishing the official list of endangered and protected species under national legislation. In addition, we call on the Brazilian government, scientific institutions and civil society to act to conserve these rediscovered but rare species. Specifically, we recommend the following six actions: (1) Further targeted collection efforts should be made to search for additional populations of M. colpodes and M. rubiginosa in other potential areas of occurrence. (2) Species population dynamics and methods of pollination, seed dispersal and germination need to be examined to provide fundamental data for in situ conservation planning. (3) Effective management, continuous monitoring and strict law enforcement should be implemented in protected areas to ensure they fulfill their conservation purpose. (4) Visitors to protected areas such as national, natural and municipal parks, and residents in adjacent areas, need to be made aware of the presence of rare and threatened plants, drawing attention to their uniqueness as endemic species and explaining their role in maintaining the forest dynamics and associated ecosystem services. (5) Information material should be produced and disseminated widely to raise awareness of the rare and threatened species occurring in forest remnants, and to foster pride and encourage a sense of stewardship and responsibility amongst local communities. People should be encouraged to adopt roles as protectors and conservators of these species, for example by anonymously reporting ongoing or imminent environmental crimes to the relevant authorities. (6) Seeds of these rediscovered taxa should be collected and distributed amongst private and public horticultural establishments and botanical gardens for ex situ conservation, following the Global Strategy for Plant Conservation, Target 8, which aims to minimize the impacts of climate change on biodiversity and build resilience (CBD, 2024). Together, these efforts can help to ensure the persistence of these unique plants in the Atlantic Forest.