New Data on Boletaceae (Agaricomycetes, Basidiomycota) from Central Vietnam with Description of Two New Species and Creation of a New Combination Based on Morphological and Phylogenetic Evidence

Two new species of Boletaceae (Hortiboletus rubroreticulatus and Tylopilus aurantiovulpinus) discovered during an investigation of the mycobiota of Central Vietnam (Kon Chu Rang Nature Reserve; Ta Dung National Park; Bidoup—Nui Ba National Park; Kon Ka Kinh National Park) are described on the basis of molecular and morphological data. Illustrated descriptions of their macro- and microscopic features and discussion on similar taxa are given. Additionally, eight species which were recorded for the first time in Vietnam are listed and illustrated here. A new combination Kgaria virescens was made for one of these species. These results were confirmed by the phylogenetic analysis based on nrITS1-5.8S-ITS2, nrLSU, and tef1α regions.


Introduction
Boletoid fungi, i.e., representatives of the ordo Boletales characterized by soft putrescent pileate basidiomata with tubulate or rarely lamellate hymenophore [1], play a significant role in tropical ecosystems.As ectomycorrhizal partners of the main forest tree species, including those of Dipterocarpaceae, Fabaceae, Fagaceae, and Pinaceae, boletoid fungi improve their nutrition and growth, increase resistance to adverse environmental conditions, as well as participate in the decomposition of organic matter.Many of them are important as a source of nutrition for humans and as potential producers of biologically active substances for medical purposes.
The aim of the present study is to contribute to the knowledge on the fungal diversity of the Dalat and Kon Tum Plateaus, and Ta Dung Ridge (Central Vietnam), where large stands of montane and pedemontane forests have been preserved.While mycological research on the Dalat Plateau began in the first decade of the 20th century (see reviews [2,3]), the first information about the diversity of pileate fungi on the Kon Tum Plateau and Ta Dung Ridge appeared relatively recently [4][5][6][7][8][9][10][11][12][13].

Morphological Study
Microscopic measurements and drawings were made with an Axio Imager.A1 ZEISS equipped with Axiocam 506 color (Carl Zeiss) and an AxioScope A1 light microscope was equipped with a Zeiss AxioCam 1Cc3 digital camera with AxioVisionRel.4.6 software (CarlZeiss, Jena, Saxe-Weimar-Eisenach, Germany).Spores, basidia, and cystidia were observed in squash preparations of small parts of the lamellae or tubes in 5% KOH or 1% Congo Red in concentrated NH 4 OH.The pileipellis was examined from a radial section of the pileus in 5% KOH or 1% Congo Red in concentrated NH 4 OH.Basidiospore dimensions were based on 20 spores and cystidia and basidia dimensions on at least 10 structures per collection.Basidia were measured without sterigmata and the spores without an apiculus.Spore length to width ratios were reported as Q.The dried specimens were deposited in the Mycological Herbarium of the Komarov Botanical Institute RAS (LE) and in the Herbarium of the Joint Vietnam-Russia Tropical Science and Technology Research Centre, Hanoi (VRTC).

DNA Extraction, Amplification, and Sequencing
PCR products were obtained without a DNA purification step using the Thermo Scientific Phire Tissue Direct PCR Master Mix (Thermo Fisher Scientific, Inc., Waltham, MA, USA) standard protocol.The ribosomal ITS1-5.8S-ITS2region was amplified with the fungal specific primers ITS1F and ITS4B [34].Sequences of nrLSU-rDNA were generated using the JS1-LR5 pair of primers [35].Pairs of primers EF1-983F-EF1-1567R and EF1-B-F1-EF1-B-R were used to amplify approximately 500-700 bp of tef1α [36,37].For ITS, PCR was carried out under the following cycling parameters: initial denaturation at 98 • C for 4 min; followed by 35 cycles at 98 • C for 1 min, 52 • C for 1 min, and 72 • C for 1 min; and a final extension at 72 • C for 3 min.For nrLSU, the process was as follows: initial denaturing at 98 • C for 5 min; then 12 cycles of denaturing at 98 • C for 5 s, annealing at 67 • C for 1 min, extension at 72 • C for 1.5 min; then 35 cycles of denaturing at 98 • C for 5 s, annealing at 56 • C for 1 min, extension at 72 • C for 1.5 min; and a final extension step of 72 • C for 10 min.
For tef1a the process was as follows: initial denaturing at 98 • C for 5 min; then 8 cycles of denaturing at 98 • C for 5 s, annealing at 60 PCR products were visualized using agarose gel electrophoresis and Gel Red staining and subsequently purified with the Fermentas Genomic DNA Purification Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA).Sequencing was performed with an ABI model 3500 Genetic Analyzer (Applied Biosystems, Inc., Waltham, CA, USA).
This work was carried out using equipment from the Core Facility Centre 'Cell and Molecular Technologies in Plant Science' of the Komarov Botanical Institute (St.Petersburg, Russia) and of the Joint Vietnam-Russia Tropical Science and Technology Research Centre (Hanoi, Vietnam).Raw data were edited and assembled in MEGA X [38].Newly generated sequences have been deposited in GenBank.

Alignment and Phylogenetic Analyses
For this study, 12 nrITS, 9 tef1α, and 10 nrLSU sequences were newly generated.In addition, 38 nrITS, 20 tef1α, and 43 nrLSU sequences, including outgroups, were retrieved from the GenBank database (www.ncbi.nlm.nih.gov//genbank/,accessed on 5 February 2024), using the BLASTn application (https://blast.ncbi.nlm.nih.gov/Blast.cgiaccessed on 5 February 2024).The information on all these sequences is presented in Table 1.Three datasets were analyzed: tef1α, nrITS, and nrLSU.DNA sequences were aligned with the MAFFT v.7.110 web tool [73], using the G-INS-i option, and then manually modified where necessary in MEGA X [38].To determine the phylogenetic positions of the studied collections, phylogenetic reconstructions were performed for three datasets using Bayesian Inference (BI) and Maximum Likelihood (ML) analysis.BI was performed using MrBayes 3.2.1 [74], under a GTR model for tef1α and the nrLSU dataset and a GTR + G model for the nrITS dataset.The analyses were run with two parallel searches, four chains, five million generations for ITS, and four million generations for LSU and tef1α, starting with a random tree.The trees were sampled every 100 generations.The runs were terminated and the average standard deviations of split frequencies were 0.003879, 0.018290, and 0.002999, respectively.Tracer v1.7.2 [75] was used to check for the convergence of MCMC analyses and to obtain estimates of the posterior distribution of parameter values.ML analyses were run on the IQ-TREE web server [76].Bootstrap values were obtained using UFBoot [77] with 1000 bootstrap replicates and 1000 maximum iterations.The phylogenetic trees were visualized in FigTree v1.4.4 and edited in Adobe Illustrator CS4.Posterior probability (PP) ≥ 0.95 and bootstrap support (BS) ≥ 65% values are considered significant.
To identify and to reveal new species, we followed both a morphological and phylogenetic species concept, referring to the examples for fungi in Taylor et al. [78].Monophyletic clades are recognized as phylogenetic species when they are supported by the received phylogenetic trees and when ITS sequence similarities are less than 97%.

Phylogenetic Analysis
Over the course of data analysis, it became clear that constructing multigene trees using an incomplete dataset for each species can lead to errors that are not obvious in the results of such analysis.Therefore, in cases where the tree topology differs for different markers, we prefer to analyze each marker separately.The overall topologies of the BI and ML trees were nearly identical for all datasets.Therefore, we present a single phylogram for each marker.
The full tef1α dataset contained 50 sequences with 660 characters (gaps included).Gyroporus cyanescens and Gyrodon lividus were selected as outgroups.Thirteen sequences were newly generated.The dataset also includes 37 other sequences retrieved from the GenBank NCBI database-the closest possible representatives of the Boletaceae.
The full nrITS dataset contained 52 sequences with 783 characters (gaps included).Gyroporus cyanescens and G. castaneus were selected as the outgroup.Besides the 16 newly generated sequences, the dataset includes 36 other sequences retrieved from the GenBank NCBI database-the closest possible representatives of the Boletaceae genera.
The full nrLSU dataset contained 58 sequences with 891 characters (gaps included).Gyrodon lividus and Gyroporus cyanescens were chosen as an outgroup.The dataset contained 13 newly generated sequences, as well as 45 representatives of the main subdivisions of the Boletaceae.
The results of the phylogenetic analyses are presented in Figure 1  Hortiboletus rubroreticulatus.New species is nested within the /Hortiboletus clade with high support in tef1α and ITS trees, but its position is relatively independent in the LSU tree.The latter can be explained by the low resolution of the LSU marker.
Tylopilus aurantiovulpinus.New species is nested within the /Tylopillus ballouicomplex clade with high support values in all trees.
Cyanoboletus flavocontextus.Our molecular phylogenetic analyses indicate the new collections and the holotype of the Cyanoboletus flavocontextus group together with a strong statistical support in all three trees.However, while in the ITS tree, these sequences are almost identical; in tef1α and LSU trees, they slightly differ but not enough to consider them as separate species.
Kgaria virescens (=Tylopilus virescens; =T.callainus).In the phylogenetic analysis, the species is represented by our two sequences.The specimens were collected in the Kon Chu Rang Nature Reserve with an interval of 7 years-in 2016 and 2023.Despite the existing morphological differences, these sequences are almost identical.They are also identical to the ITS and LSU sequences of the holotype specimen of Tylopilus callainus, which is recognized as a synonym of T. virescens [79].However, there is no correlation between our tef1α sequences and the only tef1α sequence of Tylopilus callainus presented in GenBank (MG365904) and cited in relevant publications [61,64].The sequence MG365904 nests in a clade with representatives of the T. balloui complex.It is also present in the multigene tree in the work devoted to Tylopilus of China [61].Since this sequence is not from the holotype and it is rather short, while our ITS and LSU sequences in double repetition coincide with the ones from the holotype, we suspect that an error, perhaps some kind of contamination, could have crept in here.Our data show that the sequences of this species do not form a common clade with other representatives of the genus Tylopilus but occupy a position isolated from them.In the tef1α tree, they are grouped together with Porphyrellus and Kgaria with high statistical support.There are no ITS data on the genus Kgaria in the GenBank.In this case, the grouping of this species with representatives of Xerocomoidea raises some doubts.This may be due to the "long branch attraction" in the absence of close sequences.In the LSU tree, our sequences grouped with the /Kgaria-clade but without significant support.Based on these results, the taxonomic position of Tylopilus virescens should be considered uncertain but close to the genus Kgaria.Until the problem is finally resolved, we propose a new combination Kgaria virescens.
Parvixerocomus pseudoaokii.The sequences in our collection and the holotype of this species are almost identical in the analysis based on the LSU dataset.In the ITS tree, it is identical to the specimen from Thailand (KX017306).
As a result of phylogenetic analysis, it was possible to confirm the identity of sequences obtained from specimens of the genus Phylloporus with the corresponding holotypes of the following species: Phylloporus hainanensis (ITS, LSU), Ph. microsquamus (ITS, LSU), Ph. nigrisquamus (LSU from holotype, ITS from reference specimen from Thailand), and Ph.subbacillisporus (ITS, tef1α).Tylopilus rubrotinctus.Sequences of our collection and the holotype of this species are identical in the analysis based on the tef1α dataset, and they are nested within the /Tylopillus balloui-complex clade.

New Species
Hortiboletus Simonini, Vizzini & Gelardi, Index Fungorum 244: 1 (2015).Hortiboletus was established to accommodate Boletus rubellus Krombh.and allied species with smooth spores as opposed to the closely related genus Xerocomellus that includes species with striate spores.The species were also characterized by the common occurrence of small red dots in the context at the stipe base, the average spore quotient (Qm) lower than 2.5, and its habitat in the park, gardens, and other ruderal places.With the inclusion of several new species, the genus concept expanded, including species without red dots, with much longer and narrower spores, and with a wider ecological range.According to Wu et al. [41], this genus differs from other genera of Boletaceae by a combination of the dry and subtomentose pileus usually with a wine-red tinge, a yellow-brown to red-brown stipe without a purplish tinge, and a palisadoderm pileipellis composed of vertically arranged, more or less broadened, and often incrusted hyphal elements, with the terminal cells reaching almost the same level.Fourteen species are listed in Index Fungorum (https://www.indexfungorum.org,accessed on 10 February 2024).One species (Hortiboletus rupicapreus Svetash., A.V. Alexandrova, O.V. Morozova  Diagnosis.The species is recognized by a small xerocomoid basidiomata with the pileus covered with a red squamulose reticulum on a whitish background, by elongate fusiform hymenial cystidia and by the long and narrow smooth basidiospores with a Q around three. Basidiomata small-sized, xerocomoid.Pileus 10-25 mm diam; hemispherical to convex when young, then broadly convex to applanate; surface dry or slightly viscid when moist; fibrillose-tomentose to finely squamulose; firstly, almost completely unevenly pinkish-red, greyish-red or brownish-red (9A8, 9B8, 10B5-6, 10C5-6, 11A5-6, 11D6-8), and whitish near the pileus margin; then with the red squamulose reticuloid patterns on the whitish background; context up to 5 mm thick, light yellow, greyish yellow (4B8, 4C6-8), unchanging in color.Hymenophore tubulate, adnate to sinuate, sometimes depressed around the stipe; surface pale yellow at a young age (2A3), greyish-yellow with age (2B3-4), unchanging when exposed; tubes concolorous with hymenophore surface, 3-6 mm long; pores angular or subround, 1-2 mm diam.Stipe 20-45 × 3-6 mm, central, cylindrical, equal or slightly enlarged downward and tapering towards the base; solid; longitudinally fibrillose; yellowish-brown, brownish red (9C5), basal part yellowish-brown (5D4-5); context greyish red, brownish-vinaceous (9E5-6, 12C4-6, 12D4-6), light yellow in upper part, unchanging in color when exposed.Basal mycelium yellowish.Odor and taste are not recorded.
Habitat and distribution.Solitary and in small groups on soil in tropical mountain evergreen forests.Currently known only from Vietnam.
Additional specimens examined.Notes: Hortiboletus rubroreticulatus species occupies an isolated position in the genus Hortiboletus, primarily due to the peculiarities of the pileipellis, in which small scales are unevenly arranged into a reticulate pattern on a white background.Some similarities can be observed between new species and species that are characterized by the areolate, cracked at the maturity surface of the pileus, when the scales diverge with age and also show a white background (H.subpaludosus, H. amygdalinus, H. kohistanensis, H. napaeus, and H. rufosquamosus).However, H. amygdalinus lacks reddish tinges in the color of the basidiomata and possesses broader basidiospores [41].H. rufosquamosus also has broader spores [42].H. napaeus has shorter spores [51].H. kohistanensis has large, up to 7.5 cm basidiomata [53].
Another feature unusual for this genus is the slightly sticky surface of the pileus when wet.Perhaps this feature is overestimated since both finds were made during the rainy season.
From a phylogenetic point of view, the position of the species in the tree is rather ambiguous.In the ITS and tef1α trees, this species is grouped together with other members of the genus with strong statistical support, albeit in a basal position.But they do not group together with enough support in the LSU tree.Perhaps as more data accumulate, the true taxonomic position of this species will become clearer.
Tylopilus P. Karst., Revue mycol., Toulouse 3(9): 16 (1881).In a broad sense, Tylopilus turned out to be polyphelytic, and based on molecular genetic evidence, in recent years, a significant number of new genera were separated from it.According to the latest generalizations [41,61], the genus Tylopilus, in a strict sense, can be recognized by its bitter taste in most species; white to pallid or yellowish context without discoloration or staining red to reddish brown, or blue when hurt; white to cream then pinkish to pink or yellowish on the hymenophoral surface; cutis, trichoderm, hymeniderm, or palisadoderm pileipellis; and smooth basidiospores.In total, 150 of the Tylopilus species are accepted in the Index Fungorum (https://www.indexfungorum.org,accessed on 10 February 2024), and 7 species have been reported in Vietnam to date.
Holotype.Vietnam, Kon Tum Plateau, Gia Lai Province, K'Bang district, Son Lang Commune, Kon Chu Rang Nature Reserve, N 14.49170 Diagnosis.Among other species of the Tylopilus balloui-complex with bright orange or brownish-orange basidiomata and rather small phaseoliform to ovoid basidiospores, Tylopilus aurantiovulpinus stands out due to the decurrent on the stipe boletinoid hymenophore, consisting of large angled tubes.
Basidiomata small to medium-sized.Pileus 3-9 cm in diam., at first hemispherical then plano-convex and finally broadly pulvinate-flattened, sometimes slightly depressed at center when mature, regularly to unevenly shaped; margin steady to faintly wavylobed, initially curved downwards and finally completely plane to uplifted or even slightly revolute; surface dry or subviscid when wet, matte, finely subtomentose, not cracked; variable in color, ranging from orange-yellow (4B7-8), greyish-orange (5B6-7), brownishyellow and orange (5C6-8), golden brown (5D7) to cinnamon and light brown (6D6-8), sometimes with bleached or darkened spots and areas.Context cream or light yellow (4A4-5) to pale orange (5A2-3), not discolored when bruised or turns slightly yellowbrown.Hymenophore boletinoid, adnate when young, distinctly decurrent to stipe when mature; surface firstly cream or pale orange (5A1-3), becoming salmon (6A2) with a pinkish tinge when mature, without discoloration when bruised or only slowly becoming pale dirty-brownish.Pores boletinoid, very large, subangular, and radially elongated, with concolorous or yellow edge, up to 3.5 mm wide and 5.5 mm in length; tubes 5-7 mm long.Stipe 5-8 × 1-2 cm, subcylindrical, slightly clavate or tapering towards the base, solid, firm, dry, pale orange (5A3) in upper part and light to deep orange (5A4-6) or golden yellow and greyish orange (5B6-7) in the central part, surface smooth to very finely fibrillose, context concolorous to it in the pileus, cream, light yellow to pale orange, unchanging; basal mycelium white.Odor indistinct, mild taste.Spore print yellowish to pinkish.
Habitat and distribution.Solitary and in small groups on soil in tropical mountain evergreen forests.Currently known only from Vietnam.
Notes: Morphologically this species belongs to the Tylopilus balloui-group characterized by an orange-red color of the pileus, an orange-yellow stipe, a yellowish-pinkish hymenophore, bean-shaped, rounded spores with a length-to-width ratio of less than two, and a pleuro-and cheilocystidia fusiform shape.
The first species of this group Tylopilus balloui (Peck) Singer was described by C.H. Peck in 1912 (as a Boletus) from North America [80].The generic recognition of T. balloui was ambiguous as an extension of either Gyrodon, Rubinoboletus and Gyroporus [81]; however, phylogenetic studies confidently place it in a core Tylopilus clade [63,66,[82][83][84].This species was reported from various regions of the tropical and subtropical areas, and it was considered that T. balloui sensu lato has a very wide distribution with some morphological variability for each region.
Molecular genetic methods have made it possible to begin to distinguish individual species within the complex.So, six species from Central America have already been Among these species, Tylopilus aurantiovulpinus stands out due to its hymenophore of the boletinoid type, with very large angular and radially elongated pores slightly deccurent to the stipe.Only a yet-undescribed species from Australia has a hymenophore of a similar type [88].Other species of Tylopilus balloui complex are characterized by an adnate hymenophore, depressed around the stipe apex, and occasionally decurrent with a tooth, and with round or angular pores of a smaller diameter.
Tylopilus pseudoballoui is the species most similar to T. aurantiovulpinus in morphology, but it differs by the smaller pores of the hymenophore.T. aurantiacus, T. rubrotinctus, and T. leucomycelinus have a brighter color with a predominance of red shades.Additionally, there are differences in spore size: T. aurantiacus have slightly smaller and more rounded spores, and T. leucomycelinus have slightly larger and more rounded spores.T. fuscatus, T. griseolus, and T. griseiviridus are distinguished by the presence of gray shades in the color of the pileus and stem.T. oradivensis, T. pseudoleucomycelinus, and T. pygmaeus have noticeably smaller basidiomata.

Discussion
The morphological examination of the specimens as well as phylogenetic analysis made it possible to reveal two new species for science and raise the question of the taxonomic position of the species known as Tylopilus virescens.In all three cases, the taxonomic position of the species in question is ambiguous.The new Hortiboletus rubroreticulatus occupies a basal position in the /Hortiboletus clade in the tef1α and the ITS trees (together with Hortiboletus rupicapreus), in both cases with high statistical support.At the same time, the LSU analysis-which, however, has a lower resolution-does not indicate that the new species belongs to the genus Hortiboletus.The heterogeneous structure of the pileipellis, which is not a true polysadoderm, since the hyphae of the pileipellis have different lengths, is also not entirely characteristic for the genus.However, until more data are accumulated, we propose to consider this new species in the genus Hortiboletus.
An Important distinguishing featu"e of'the Tylopilus balloui species complex is the absence of the bitter taste which is characteristic for typical species of this genus.Taking into account the structural features of the hymenophore discussed above, the lack of a pronounced pink tint in the color of the spores, and also the unstable, near basal position of the species of this group on phylogenetic trees, it can be assumed that this group could also represent a phylogenetic line separate from the Tylopilus core.But this issue requires further study.
The taxonomic position of Tylopilus virescens, for which we proposed a new combination Kgaria virescens, is also controversial and requires the accumulation of additional data, especially on species with a similar type of autooxidation.
As a conclusion, we cannot help but notice that the area of tropical forests currently is drastically declining.Therefore, in Vietnam, in recent decades, much attention has been paid to the conservation of natural landscapes and the protection of biodiversity.There are powerful programs, and the network of natural parks and reserves is expanding.This prevents the continued decline in the area of tropical forests, along with which the species associated with them also disappear, often without even receiving a name.In Central Vietnam, the greatest diversity of fungi of the Boletaceae family is found in middle-mountain evergreen broad-leaved forests.They require special attention, study, and careful handling.Studying the fungal diversity of typical tropical forests adds to the knowledge on the macromycetes biota of Vietnam and may be useful in developing conservation measures for these valuable natural areas and the species inhabiting them.
(tef1α), Figure 2 (nrITS), and Figure 3 (nrLSU).The phylogenetic trees demonstrated rather different topology and different support values of clades.Based on the obtained topologies, the taxonomic positions of the newly generated sequences are discussed below.

Figure 1 .
Figure 1.Phylogenetic tree derived from Bayesian Inference (BI), based on tef1α data.Values of the posterior probability (PP ≥ 0.95) from the BI and of the bootstrap support (BS ≥ 65) from the ML analysis are given to the left of the nodes.The scale bar represents the number of nucleotide changes per site.New species and new records are in bold.T-holotype.New species are marked with a red circle.

Figure 1 .
Figure 1.Phylogenetic tree derived from Bayesian Inference (BI), based on tef1α data.Values of the posterior probability (PP ≥ 0.95) from the BI and of the bootstrap support (BS ≥ 65) from the ML analysis are given to the left of the nodes.The scale bar represents the number of nucleotide changes per site.New species and new records are in bold.T-holotype.New species are marked with a red circle.

Figure 2 .
Figure 2. Phylogenetic tree derived from Bayesian Inference (BI), based on nrITS1-5.8S-ITS2region data.Values of the posterior probability (PP ≥ 0.95) from the BI and of the bootstrap support (BS ≥ 65) from the ML analysis are given to the left of the nodes.The scale bar represents the number of nucleotide changes per site.New species and new records are in bold.T-holotype.New species are marked with a red circle.

Figure 2 .
Figure 2. Phylogenetic tree derived from Bayesian Inference (BI), based on nrITS1-5.8S-ITS2region data.Values of the posterior probability (PP ≥ 0.95) from the BI and of the bootstrap support (BS ≥ 65) from the ML analysis are given to the left of the nodes.The scale bar represents the number of nucleotide changes per site.New species and new records are in bold.T-holotype.New species are marked with a red circle.

Figure 3 .
Figure 3. Phylogenetic tree derived from Bayesian Inference (BI), based on nrLSU data.Values of the posterior probability (PP ≥ 0.95) from the BI and of the bootstrap support (BS ≥ 65) from the ML analysis are given to the left of the nodes.The scale bar represents the number of nucleotide changes per site.New species and new records for are in bold.T-holotype.New species are marked with a red circle.

Figure 3 .
Figure 3. Phylogenetic tree derived from Bayesian Inference (BI), based on nrLSU data.Values of the posterior probability (PP ≥ 0.95) from the BI and of the bootstrap support (BS ≥ 65) from the ML analysis are given to the left of the nodes.The scale bar represents the number of nucleotide changes per site.New species and new records for are in bold.T-holotype.New species are marked with a red circle.
• C for 40 s, extension at 72 • C for 2 min; then 36 cycles of denaturing at 98 • C for 5 s, annealing at 53 • C for 1.5 min, extension at 72 • C for 2 min; and a final extension step at 72 • C for 10 min.

Table 1 .
Specimens and GenBank accession numbers of DNA sequences used in the molecular analyses (Newly generated sequences are in bold).