A new endangered species of Mollinedia (Monimiaceae, Laurales) endemic to the Atlantic forest in the state of Espírito Santo, Brazil, supported by morphology and genome size estimation

Abstract Mollinedia (Monimiaceae) presents numerous microendemic species, and its centre of diversity is in the Brazilian Atlantic rainforest, where more than half of the species occur. The taxonomy of microendemic species can be challenging because their morphological and genetic variations can be interpreted as a response to geographic isolation rather than a circumscription for different species. In this paper, we describe Mollinedia pignalii Lírio & Pauli, a microendemic species from the Espírito Santo state, Brazil. Vegetatively, M. pignalii is similar to Mollinedia elegans Tul. and Mollinedia schottiana (Spreng.) Perkins; however, it presents the following differences: white-puberulous leaves, staminate flowers with a flat receptacle and 6 to 14 stamens, white-puberulous pistillate flowers with a cupuliform receptacle and 8 to 22 carpels, and white-puberulous drupelets. Due to the similarity between the three species, M. pignalii has been collected and deposited in herbaria under different names. Here, we describe the new species based on morphology and genome size, a comparison with similar species, and ecological comments in an integrative approach. We also provide its conservation status and an identification key for the species of Monimiaceae occurring in the state of Espírito Santo.

In the works of Perkins (1898;1900;1902;1905;1911;1927) and Perkins and Gilg (1901), 52 species of Mollinedia were described, and seven had their status changed.Although Perkins examined exsiccatae available in the large herbaria of her time, the diversity of samples from Brazil did not depict the ample morphological diversity of some of the species described by her.For this reason, many taxa described by the author are today considered as morphological variations since these species present plasticity in their development and phenotype, mainly due to their geographic distribution.To date, 36 of the species described by Perkins have been considered synonyms (Peixoto 1979;1981;Peixoto & Pereira-Moura 1996;Peixoto et al. 2001;Renner & Haussner 1997;Lírio et al. 2020a;Lorence in press).The same occurs with other authors, such as Tulasne (1855;1856;1857), who described 15 species of Mollinedia and altered the status of other eight, 10 of which are now considered synonyms (Perkins 1900;Renner & Haussner 1997;Lírio et al. 2020a;Lorence in press).
The above data show the difficulty in delimiting species in Mollinedia.Despite the notorious morphological variation in species of this genus, some characters such as the pubescence on the abaxial surface of leaves, scars left by indumenta on the adaxial surface of leaves, the colour of leaves when dried, the shape of the receptacle in staminate flowers, and the shape and pubescence of the fruits have been considered significant for the circumscription of species of the family (Lorence 1999;Whiffin & Foreman 2007;Renner & Takeuchi 2009;Lírio et al. 2015;2020a;2020b;2021;2023b;Molz & Silveira 2021).However, other techniques have been incorporated to corroborate or reject hypotheses of new taxa delimitation based on morphological data.Among these techniques, genome size (GS) estimation has proved very useful (Ohri 1998;Zonneveld 2009;Jedrzejczyk & Rewers 2018), including in Monimiaceae, in the genus Macrotorus (Lírio et al. 2020b).Thus, this study provides an opportunity to test the GS estimation method in the delimitation of Mollinedia species.
In this work, we describe Mollinedia pignalii, a new species endemic to the mountainous region of the Espírito Santo state, Brazil, based on morphological data and GS estimation.The data are discussed comparatively with Mollinedia elegans Tul. and Mollinedia schottiana (Spreng.)Perkins.We provide ecological comments, the identification key for species of Monimiaceae occurring in the state mentioned, and assess the extinction risk of the new species.
Genome Size Estimation By Flow Cytometry: Six individuals from each species were analysed in triplicate.For each sample, approximately 50 mg of young leaf tissue was macerated with 25 mg of the internal reference standard, Pisum sativum var.Ctirad (2C = 9.09pg, Dolezel et al. 1998), in 0.5 ml of cold Ebihara buffer (Ebihara et al. 2005) supplied with 0.025 µg mL -1 RNAse.Nuclei suspensions were stained by adding 12.5 µL to 1 mg mL -1 of propidium iodide solution (PI, Sigma).The analysis was performed using a FACSCanto II cytometer (Becton Dickinson, San Jose, CA, USA) kindly made available by the Microbiology and Immunology Department of IBB-UNESP (Botucatu, Brazil).The histograms were obtained with FACSDiva software based on 5,000 events, and the statistical evaluation was performed using the Flowing Software 2.5.1 (http://www.flowingsoftware.com/).The quality control of samples was based on the coefficient of variation (CV) of each measurement, which should be below 5%, and the standard deviation (SD) among 2C-values, which should be below 3%.These limits ensure that the variations observed within and among measurements are due to technical factors and should not represent intraspecific variation among individuals (Pellicer & Leitch 2014).The species were differentiated by One-Way ANOVA followed by Tukey Test in R.
Conservation Status Assessment: To assess the conservation status of the species, we followed the Categories and Criteria of IUCN Red List of Threatened Species (thereafter, IUCN Red List) (IUCN 2012;2022).We calculated the Extent of Occurrence (EOO) using the area of the minimum convex polygon and the Area of Occupancy (AOO) using a grid of 4 km 2 cells (IUCN 2022).The analyses were performed using the geospatial conservation assessment tool GeoCat (Bachmann et al. 2011).
Etymology: The specific epithet pays homage to Marc Pignal (1964-), French botanist of the Muséum National d'Histoire Naturelle, Paris, who has extensively contributed to research on the taxonomy, morphology, botanical history and conservation, including of Monimiaceae.He also works on developing the knowledge of the Brazilian flora, participating in significant expeditions, projects and scientific works in Brazil.
Diagnosis: The new species is similar to M. elegans and M. schottiana but can be distinguished by the combination of the following characters: white-puberulous leaves on the abaxial surface, staminate flowers with a flat receptacle, 6 to 14 stamens, white-puberulous pistillate flowers with a cupuliform receptacle, 8 to 22 carpels, and white-puberulous drupelets (Table 1).
Distribution and habitat: Mollinedia pignalii is found in four municipalities in the state of Espírito Santo (Fundão, Santa Leopoldina, Santa Maria de Jetibá and Santa Teresa), in Mountainous Dense Ombrophilous Forest, at altitudes of 575 to 875 m (Fig. 2).The species is distributed in the Central Corridor of the Atlantic forest (CCAF), which extends from the southern portion of Bahia throughout Espírito Santo.This corridor aims to maintain and restore the connection between forests, mainly to facilitate the genetic flow between populations and ensure the survival of several biological species.The CCAF is considered a priority region due to the high degree of vulnerability and fragmentation of its ecosystems when compared to others (Brasil 2007).
Phenology:The species is perennial with persistent and evergreen leaves.Flowers were collected from June to September and fruits in October, December and January to March.
Conservation status: Mollinedia pignalii has estimated EOO of ca.257.665 km² and an AOO equal to 40 km², thus falling into the Endangered category, under the criterion B. There are four known locations and all of them are under different ongoing threats.Habitat loss is the main threat for this species, considering that the Atlantic forest only 12.4% of its original vegetation cover (Fundação SOS Mata Atlântica & INPE 2021).Fragmentation and habitat degradation due to urban expansion, agriculture and livestock farming are also important threats.Pasturelands cover a great proportion of the municipalities where the species occurs, representing 38.6% of land use in Fundão, 22% in Santa Teresa, 17.5% in Santa Maria de Jetibá, and 17.3% in Santa Leopoldina (LAPIG 2022).In addition, the increase of droughts and fires has been negatively impacting the endemic species of the region (Fraga et al. 2019).Finally, defaunation is a relevant threat for plant species which rely on threatened fauna for pollination and/or dispersion, which is the case for many Monimiaceae species.Therefore, a continuing decline in habitat quality is expected in view of all the above mentioned factors.
The global population of M. pignalii is also considered small: restricted to less than 50 herbarium records or observations.An effort to quantify the population size based on the number of mature individuals is needed in order to confirm if it would fall into the Critically Endangered category threshold.The species occurs in four protected areas with different levels of protection and permitted uses [Área de Proteção Ambiental de Goiapaba-Açu (Fundão), Estação Biológica de Santa Lúcia (Santa Teresa), Reserva Particular do Patrimônio Natural Rancho Chapadão (Santa Leopoldina), and the integral protection conservation unit Reserva Biológica Augusto Ruschi (Santa Teresa)].Despite that, phytosociological studies have indicated that the number of individuals is decreasing.In a floristic and phytosociological study carried out at the Estação Biológica de Santa Lúcia covering sections of 3,400 m² at valley, slope and mountain top areas totaling 1.02 ha, and considering individuals with a CAP equal to or greater than 20 cm, Thomaz and Monteiro (1997) found two individuals of the species, one in the slope area (between 675 and 700 m) and another in the top of the hill (between 820 and 855 m).In a review of the study carried out in 2014, only one individual of M. pignalii was found in the same area, which demonstrates a decrease in the number of individuals with  (19º58'18.257''S, 40º31'57.277''W)small tree of 2 m, ripe purple fruits, immature green, 11 September 2013, fr., E. J.

Discussion
Some studies show that the number of known angiosperm species is likely lower than we imagine, and the reasons might be the deficit of chromosomal data in some plants' groups, the lack of taxonomic revisions and fieldwork.Even in well-studied areas, new species are often found: e.g.Macrotorus genuflexus, from the Biological Reserve of Poço das Antas, in the state of Rio de Janeiro, and Eleocharis pedrovianae C.S. Nunes, R.Trevis.& A.Gil (Cyperaceae), from Serra dos Carajás, in the state of Pará (Nunes et al. 2016); therefore, the rarity of some species need to be considered as well (Soltis et al. 2007;Cheek et al. 2020;Christenhusz & Byng 2016).Mollinedia pignalii is a new species with a small distribution range, and shares morphological similarities with M. elegans and M. schottiana regarding its vegetative traits.Despite these similarities, the new species can be differentiated by a set of characteristics such as leaf pubescence, staminate and pistillate flower and fruit traits, and the genome size estimation (Table 1).This indicates that taxonomic studies and field collections focused on this group are still needed.Finally, it is worth noting that many species are described based solely on morphological approaches, often neglecting the evaluation of molecular and cytogenetic data, which are very important for accurately determining new species and solving species complexes.
Since genome size estimation can be related to the number or size of chromosomes, this technique becomes particularly relevant for species description in groups with large ploidy variations, such as those in the Monimiaceae family (n = 19, 20, 22, 39, 40-42, c. 44, c. 48, 50, 72, c. 83, 90;Morawetz 1986;Rohwer 1993;Renner & Chanderbali 2000;Oginuma & Tobe 2006;Lírio et al. 2022).By providing a rapid estimation of genome size, this method allows for comparisons between species and provides indications of species' ploidy levels, which are otherwise confirmed through chromosomal number determination.genome size estimation proved to be a valuable tool in distinguishing species in the family before, as demonstrated in a previous study on the genus Macrotorus (Lírio et al. 2020b).Macrotorus has two species: the known tetraploid species Macrotorus utriculatus (Mart.ex Tul.) Perkins (2n= 80, 2C=5.545pg) and the recently described diploid species M. genuflexus (2n= 40, 2C = 2.644 pg) (Lírio et al. 2020b).
Genome size estimation also has been employed to assign plant species to different families, demonstrating its reliability and applicability across various taxonomic groups (Ohri 1998;Zonneveld 2009;Jedrzejczyk & Rewers 2018).In our study, we were able to differentiate the three Mollinedia species based on their genome size.However, M. schottiana exhibited a large standard deviation (SD) of 8% with a low coefficient of variation (CV).This significant variation suggests that the observed SD is likely a result of biological factors rather than methodological issues.As currently circumscribed, the taxonomic entity of M. schottiana may be polymorphic and could potentially encompass more than one species.Furthermore, such intraspecific genome size variation could be associated with the wide geographic distribution of M. schottiana, as previously suggested in different species (Biémont 2008;Moraes et al. 2022).
Regarding the geographic distribution and conservation status, although M. pignalii occurs in four protected areas with different levels of protection and permitted uses, the species is restricted to regions of Mountainous Dense Ombrophilous Forest in only four municipalities of the state of Espírito Santo, distributed in fragmented areas due to deforestation, agriculture and livestock.Mollinedia ruschii, a recently described species, has a similar geographic distribution, occurring only in conservation units (Estação Biológica de Santa Lúcia and Reserva Biológica Augusto Ruschi).These data demonstrate the fundamental role of conservation units in the survival of species with restricted distribution (Lírio et al. 2021;2023a).
The present study demonstrates the importance of integrating different research techniques to increase our knowledge of biodiversity, and also highlights the importance of field and herbarium work for the advancement of taxonomic understanding, reaffirming the fundamental role of conservation units in the protection of species that are endemic to small areas, ensuring their survival, especially in light of the unbridled environmental degradation that increases every year, transforming the conservation of these organisms into a race against time.

Figure 2 .
Figure 2. Geographic distribution map of Mollinedia pignalii.Orange dots represent occurrence records and blue polygons represent Conservation Units.

Figure 3 .
Figure 3. Genome size analysis of Mollinedia pignalii, Mollinedia elegans, and Mollinedia schottiana by Flow Cytometry.(A) Flow cytometry histograms illustrating the analysis of M. pignalii (top, in red), M. elegans (middle, in blue), and M. schottiana (bottom, in green).The G1 peaks of both Pisum sativum (a reference) and the samples are indicated.(B) The violin plot showcases the differences in genome size (Y-axis) among the three species (X-axis): M. pignalii (red), M. elegans (blue), and M. schottiana (green).The height of each violin corresponds to the internal standard deviation, while the width indicates the internal coefficient of variation.Each violin includes an internal box plot representing the interquartile range, spanning from the lower to the upper quartile, indicating 50% of the data.Additionally, a line within the violin represents the mean 2C value.