PhytoKeys 205: 37 1-400 (2022) ab A peer-reviewed open-access journal doi: 10.3897/phytokeys.205.7682 RESEARCH ARTICLE eel h y toKey S

https:/ / Pp hyto keys -pen soft.net Launched to accelerate biodiversity research

Re-establishment of the genus Pseudalbizzia (Leguminosae, Caesalpinioideae, mimosoid clade): the New World species formerly placed in Albizia

Gabriela Aviles Peraza!, Erik J. M. Koenen?,, Ricarda Riina?, Colin E. Hughes’, Jens J. Ringelberg*, German Carnevali Ferndndez-Concha!”, Ivén Mercedes Ramirez Morillo', Lilia Lorena Can Itza', Ivan Tamayo-Cen', Jorge Humberto Ramirez Prado®, Xavier Cornejo’,

Sawai Mattapha®, Rodrigo Duno de Stefano'!

| Herbarium CICY, Centro de Investigacion Cientifica de Yucatin, A.C. (CICY), Calle 43 No. 130, Col. Chuburna de Hidalgo, 97200, Mérida, Yucatan, Mexico 2. Evolutionary Biology & Ecology, Université Libre de Bruxelles, Av. ED. Roosevelt, 50, CP 160/12, Brussels B-1050, Belgium 3 Real Jardin Botdnico, CSIC. Plaza de Murillo, 2. Madrid 28014, Spain 4 Department of Systematic and Evolutionary Botany, University of Zurich, Zollikerstrasse 107, Zurich CH-8008, Switzerland § Orchid Herbarium of Oakes Ames, Harvard University Herbaria, 22 Divinity Avenue, Cambridge, Massachusetts 02138, USA 6 Unidad Biotecnologia Centro de Investigacién Cienttfica de Yucatin, A.C. (CICY), Calle 43 No. 130, Col. Chuburna de Hidal- go, 97200, Mérida, Yucatin, Mexico 1 Herbario GUAY, Facultad de Ciencias Naturales, Universidad de Guayaquil, Avenida Juan Tanca Marengo s/n y Avenida de las Aguas Casilla 09-01-10634, Guayaquil, Ec- uador 8 Department of Biology, Faculty of Science, Udon Thani Rajabhat University, Udon, 41000 Thailand

Corresponding author: Rodrigo Duno De Stefano (roduno@cicy.mx)

Academic editor: Gwilym P. Lewis | Received 20 October 2021 | Accepted 18 March 2022 | Published 22 August 2022

Citation: Aviles Peraza G, Koenen EJM, Riina R, Hughes CE, Ringelberg JJ, Carnevali Fernandez-Concha G, Ramirez Morillo IM, Can Itza LL, Tamayo-Cen I, Ramirez Prado JH, Cornejo X, Mattapha S$, Duno de Stefano R (2022) Re-establishment of the genus Pseudalbizzia (Leguminosae, Caesalpinioideae, mimosoid clade): the New World species formerly placed in Albizia. In: Hughes CE, de Queiroz LP, Lewis GP (Eds) Advances in Legume Systematics 14. Classification of Caesalpinioideae Part 1: New generic delimitations. PhytoKeys 205: 371-400. https://doi. org/10.3897/phytokeys.205.76821

Abstract

Following recent mimosoid phylogenetic and phylogenomic studies demonstrating the non-monophyly of the genus Albizia, we present a new molecular phylogeny focused on the neotropical species in the genus, with much denser taxon sampling than previous studies. Our aims were to test the monophyly of

the neotropical section Arthrosamanea, resolve species relationships, and gain insights into the evolution

* ‘These authors made equal contributions to this paper.

Copyright Gabriela Aviles Peraza et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

372 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

of fruit morphology. We perform a Bayesian phylogenetic analysis of sequences of nuclear internal and external transcribed spacer regions and trace the evolution of fruit dehiscence and lomentiform pods. Our results find further support for the non-monophyly of the genus Albizia, and confirm the previously pro- posed segregation of Hesperalbizia, Hydrochorea, Balizia and Pseudosamanea. All species that were sampled from section Arthrosamanea form a clade that is sister to a clade composed of Jupunba, Punjuba, Balizia and Hydrochorea. We find that lomentiform fruits are independently derived from indehiscent septate fruits in both Aydrochorea and section Arthrosamanea. Our results show that morphological adaptations to hydrochory, associated with shifts into seasonally flooded habitats, have occurred several times inde- pendently in different geographic areas and different lineages within the ingoid clade. This suggests that environmental conditions have likely played a key role in the evolution of fruit types in Albizia and related genera. We resurrect the name Pseudalbizzia to accommodate the species of section Arthrosamanea, except for two species that were not sampled here but have been shown in other studies to be more closely related to other ingoid genera and we restrict the name Albizia s.s. to the species from Africa, Madagascar, Asia, Australia, and the Pacific. Twenty-one new nomenclatural combinations in Pseudalbizzia are proposed, including 16 species and 5 infraspecific varietal names. In addition to the type species Pseudalbizzia berteroana, the genus has 17 species distributed across tropical regions of the Americas, including the Caribbean. Finally, a new infrageneric classification into five sections is proposed and a distribution map of the species of Pseudalbizzia is presented.

Keywords Arthrosamanea, hydrochory, monophyly, Neotropics, phylogeny, taxonomy

Introduction

The genus Albizia Durazz. has a complicated taxonomic history but has generally been treated as a pantropical genus with 120-140 species, of which 36 are endemic to Africa, with c. 30 species in Madagascar, of which c. 24 are endemic, c. 35 species in Asia, one in Australia, and 22 in tropical America (Lewis and Rico Arce 2005; Rico Arce et al. 2008). All species are woody, forming trees of variable stature and inhabit a wide range of lowland tropical biomes (Figs 1 and 2), including rain forests, seasonally dry tropi- cal forests, and savannas, with one species, Albizia julibrissin Durazz., the type species of the genus, in subtropical and warm temperate forests in Asia. However, Albizia re- mains poorly defined; its delimitation remains one of the most challenging taxonomic problems in the legume family, and it is currently considered the main “dustbin” ge- nus in tribe Ingeae (Koenen et al. 2020). In the past, the most problematic genus of tribe Ingeae was Pithecellobium Mart., but its taxonomy has been gradually clarified (Barneby and Grimes 1996). Resolution of the taxonomic status of Albizia has lagged behind that of Pithecellobium and only really started at the end of the twentieth century. For example, several new neotropical genera have been segregated from Albizia: Bali- zia Barneby & J.W. Grimes; Hesperalbizia Barneby & J.W. Grimes, and Hydrochorea Barneby & J.W. Grimes. Barneby and Grimes (1996) also re-established the genus Pseudosamanea Harms, which previously had been treated as a synonym within Albizia (Table 1). However, at the time they were established, the monophyly of these new and re-established genera had not been tested using phylogenetic analyses of molecular data.

Re-establishment of the genus Pseudalbizzia 37D

Figure |. Morphology of Albizia s.1. showing selected members of the genera Albizia and Pseudalbizzia a-c Albizia ferruginea (Guill. & Perr.) Benth. in Congo a detail of leaf rachis and gland between terminal pinnae b detail of leaflets of a terminal pinna € seed and funiculus attached to the valve d Albizia glaber- rima (Schumach. & Thonn.) Benth. in Malawi, detail of inflorescence e Albizia anthelmintica Brongn. in Malawi, habit f Albizia adianthifolia (Schumach.) W. Wight in Congo, habit g Albizia glaberrima in Malawi, branches and inflorescences h Albizia chinensis (Osbeck) Merr. in Thailand, inflorescences i Albizia odoratissima (L. f.) Benth. in Thailand, fruits j Albizia procera (Roxb.) Benth. in Thailand, fruits k Albizia splendens Mig. in Thailand, woody fruit | Pseudalbizzia multiflora var. multiflora in Ecuador, woody fruit m, n Pseudalbizzia pistaciifolia (Willd.) E.J.M Koenen & Duno in Ecuador m habit n woody fruit. Photos: a, b David J. Harris / With permission from RBG Edinburgh ¢ Claude Boucher Chisale d-f Giinter Baumann g Jos Stevens h Natcha Sutjaritjai iI-k Prateep Panyadee I=-n Xavier Cornejo.

374 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Figure 2. Habit, flower and fruit variation in the genus Pseudalbizzia a P adinocephala pods (Hughes 1913) b P coripatensis inflorescence (Hughes 2433) € P coripatensis pods (Hughes 2433) d P inunda- ta pods (JRI Wood 26530) e P multiflora habit (Hughes 2214) f P multiflora leaves and pods (Hughes 2214) g P pistaciifolia \eaves and inflorescence (Cornejo 8426, GUAY) h P. niopoides habit (Hughes 419) i P niopoides pods (Rivera 2245) j P polycephala inflorescence (de Queiroz 15515) k P tomentosa inflores- cence (Hughes 1143) | P sinaloensis pods (Hughes 1576) m P. tomentosa habit (Hughes 1335) n P tomentosa pods (Hughes 1307). All photos by Colin Hughes except g, Xavier Cornejo.

Re-establishment of the genus Pseudalbizzia 375

Table 1. Main taxonomic changes related to Albizia, Ingeae tribe [1981-2008]. Modified from Rico Arce et al. (2008).

Nielsen (1981) Barneby and Grimes (1996) Lewis and Rico Arce (2005) Rico Arce et al. (2008) Iganci et al. 2015

Albizia Albizia Albizia Albizia Albizia Balizia Balizia Cathormion Cathormion Cathormion 4 a Hydrochorea ; Hydrochorea Hesperalbizia Hesperalbizia Pseudosamanea Pseudosamanea

* Not explicitly mentioned in the study.

Barneby and Grimes (1996) placed the remaining New World species of Albizia in their section Arthrosamanea (Britton & Rose) Barneby & J.W. Grimes. They char- acterized this section as forming a group that is homogeneous in most respects, but diverse in the late developmental stages of the fruit, which vary in: 1) fruit opening type: dehiscent, indehiscent, or breaking, 2) lateral shape: flat to conspicuously raised above the seed chambers, 3) texture and consistency of the valves: papery, chartaceous or woody, 4) longitudinal shape: straight to weakly falcate (Barneby and Grimes 1996) (Figs 1, 2 and 4). Within section Arthrosamanea, four series were proposed by Barneby and Grimes (1996): series Paniculatae with papery, plano-compressed, inertly dehis- cent pods with continuous valves (13 species); series Arthrosamanea comprising 3 spe- cies with lomentiform, plano-compressed pods, where the ripe valves crack transversely between seeds but the wiry sutures persist at maturity; series Multiflorae (2 species) characterized by lomentiform fruits only reluctantly separating into articles, the thick- textured valves and sutural keels breaking transversely under pressure; and the mono- specific series Inundatae which bears crypto-lomentiform pods, dehiscent through the sutures and with the valves differentiating into a continuous exocarp and a segmented endocarp separating into 1-seeded segments (Barneby and Grimes 1996).

Series Paniculatae is widespread across Mexico, Central and South America, oc- curring mainly in seasonally dry forests, grasslands, and less often in humid forests (in South America). All species of series Paniculatae have papyraceous, dehiscent fruits with one exception, A. berteroana (DC.) Fawc. & Rendle (the earlier combination A. berteroana (DC.) M. Gémez was invalidly published due to incorrect citation of the basionym, see Barneby and Grimes 1996), whose fruits are indehiscent and fall to the ground entire. In contrast, the other three series are distributed from Panama to South America and are most diverse in the Amazon basin (Barneby and Grimes 1996), and usually have more or less woody fruits, which are articulated and indehiscent, some dividing into monospermous segments through the grooves of the valves, considered to be an adaptation for hydrochory, i.e., seed dispersal in riparian and seasonally inun- dated forests (¢.g., A. inundata (Mart.) Barneby & J.W. Grimes, A. pistaciifolia (Willd.) Barneby & J.W. Grimes, and A. subdimidiata (Splitg.) Barneby & J.W. Grimes).

The segregate genera established by Barneby and Grimes (1996) have not been uni- versally accepted. For example, in the most recent taxonomic treatment of Albizia for Mexico and Central America (Rico Arce et al. 2008), the genera Balizia, Hesperalbizia,

376 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

and Pseudosamanea were not recognized (Table 1). However, subsequent phylogenetic analyses have confirmed that these genera were rightfully segregated as distinct evolu- tionary lineages, Hesperalbizia being more closely related to Lysiloma Benth. (Duno de Stefano et al. 2021), and the closely related Balizia and Hydrochorea (the former re- duced to synonymy of the latter, Soares et al. 2022) being placed as the sister-group of Jupunba Britton & Rose (Iganci et al. 2015; Soares et al. 2021, 2022). Phylogenomic analysis of the mimosoid clade, based on DNA sequences of 964 targeted nuclear genes confirmed these findings and, furthermore, showed that the species from the Americas (i.e., sect. Arthrosamanea) form a separate lineage from the African, Madagascan and Asian species (Koenen et al. 2020). Koenen et al. (2020) also showed that Pseudosa- manea, although difficult to place in any clade, is not closely related to either Albizia s.s. or section Arthrosamanea and is perhaps most closely related to Samanea Merr. and Chloroleucon Britton & Rose ex Record.

While the data of Koenen et al. (2020) provided a robust phylogenomic back- bone for the mimosoid and ingoid clades, and clearly demonstrated the non-mono- phyly of Albizia by sampling 25 species of that genus, only three of the 19 species from Central and South American sect. Arthrosamanea were included in that study, leaving doubts about the monophyly of that section and whether it should be seg- regated under the circumscription of Barneby and Grimes (1996), or whether there are further potential segregates, given the possibility that some of these species are more closely related to other neotropical genera. That possibility was suggested by the occurrence of lomentiform fruits in some species of section Arthrosamanea as te- flected in the classification into separate series by Barneby and Grimes (1996). Simi- lar lomentiform fruits also occur in Hydrochorea (Barneby and Grimes 1996; Soares et al. 2022) and Albizia s.s. (Albizia dolichadena (Kosterm.) I.C. Nielsen, Albizia moniliformis (DC.) F. Muell., Albizia rosulata (Kosterm.) 1.C. Nielsen and Albizia umbellata (Vahl) E.J.M. Koenen), as well as a few other ingoid lineages (Barneby and Grimes 1996: 204; Koenen 2022a). For this reason, several species of neotropical Albizia have homotypic synonyms in Arthrosamanea Britton & Rose, Samanea or Cathormion Hassk., genera which previously had been recognized and defined based mainly on characters of fruit texture and dehiscence (Barneby and Grimes 1996). Barneby and Grimes (1996: 204) considered these fruit types to have arisen multiple times in parallel in different genera, and this was confirmed by subsequent phyloge- netic (Iganci et al. 2015; Soares et al. 2021) and phylogenomic studies (Koenen et al. 2020), although the neotropical lomentiform Albizia species were not included in these studies, or remained unresolved.

Here we investigate whether Albizia sect. Arthrosamanea is monophyletic and thereby provide a more rigorous basis for recognizing its evolutionary distinctiveness from Albizia s.s. as a segregate genus. We infer a new phylogeny with emphasis on the neotropical species and make use of further insights offered by the phylogenomic anal- ysis of Ringelberg et al. (2022). In addition, we use a tree topology inferred from data from the latter study to evaluate whether lomentiform fruits in Albizia sect. Arthrosa- manea are independently derived from other lineages in which this fruit type occurs.

Re-establishment of the genus Pseudalbizzia aap

Based on our phylogenetic results and the recent findings of Koenen et al. (2020) and Ringelberg et al. (2022), we update the taxonomy of neotropical Albizia by resurrect- ing the genus Pseudalbizzia Britton & Rose.

Materials and methods

We used the nuclear ribosomal External and Internal Transcribed Spacer (ETS and ITS) regions that previously been used to study sister-group relationships within tribe Ingeae (Brown et al. 2008; Iganci et al. 2015; Souza et al. 2016). Our combined dataset included 123 accessions, of which 50 are from Genbank and 73 are newly sequenced here, including 25 species of Albizia s.l. sequenced for the first time. The outgroup, Vachellia farnesiana (L.) Wight & Arn., was designated to root the tree (Table 2). The plastid trnK region was initially explored but preliminary analyses suggested it is not sufficiently phylogenetically informative and these data were excluded from this study. Fresh leaf material collected in the field plus herbarium material from the Jardin Bo- tanico Regional Roger Orellana (CICY) were used for DNA extraction. Herbarium speci- mens used in these analyses came from AAU, CICY, FCME, MA, MEXU, and MO (acro- nyms as in Thiers 2016). Additional sequences were downloaded from GenBank (Table 2). DNA from leaf fragments was obtained using the DNeasy Plant Mini Kit (QIAGEN Inc., Valencia, California) following the manufacturer's specifications. To assess concen- tration and relative quality of DNA, 3 ul of final volume plus 2 ul loading buffer were run for 30 minutes at 6 V cm’ on a 1% agarose gel prepared with 0.5x TBE. The result- ing gel was developed by immersion for 20-30 minutes in a 0.1 ug ml" ethidium bro- mide solution and later observed in a DigiDoc-It Imaging System (version 6.7.1; UVP, Inc., Cambridge, UK) transilluminator. DNA purity and concentration were quantified with a NanoDrop 2000c. Afterwards, DNA samples were standardized to 10 ng pl". PCR amplifications were performed in an Applied Biosytems Veriti 96 Well Ther- mal Cycler. Volumes of reagents and conditions for the amplifications were as follows: ITS: 30 wL of mix containing 3 pl 10x Buffer, 2.5 ul MgCl, 0.6 ul (~10 ng) primer, 4 ul Q solution, 1 pl 1.25 mM L' dNTP, 0.2 pl (1 U) TAQ polymerase, 2 ul (-10 ng) DNA, then completed to volume (approx. 16.1 ul) with ultra-pure water. PCRs were conducted under the following protocol: 94 °C x 3 min + 30 cycles (94 °C x 1 min + 60.5 °C x 1 min + 72 °C x 2 min) + 72 °C x 7 min. Primers were $3 (AACCT- GCGGAAGGATCATTG) (Kass and Wink 1997), and 26S (TAGAATTCCCCG- GTTCGCTCGCCGTTAC) (Sun et al. 1994). ETS: 30 pl of mix containing 3 ul 1OxeBufter: 2:5" al MgCl, 0.6 ul (-10 ng) primer, 4 pl Q solution, 1 pl 1.25 mM [' dNTP, 0.2 pl (1 U) TAQ polymerase, 2 ul (~10 ng) DNA, then completed to volume (approx. 16.1 ul) with ultra-pure water. PCR amplifications were conducted under the following protocol: 94 °C x 3 min + 30 cycles (94 °C x 1 min + 60.5 °C x 1 min + 72 °C x 2 min) + 72 °C x 7 min. Primers used were 18S-IGS (5°-GAGACAAGCAT- ATGACTACTGGCAGGATCAACCAG-3’) and 26S-IGS (5’-GGATTGTTCACC- CACCAATAGGGAACGTGAGCTG-3’) (Baldwin and Markos 1998).

378

Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Table 2. Voucher information of taxa included in the phylogenetic analysis with their corresponding

GenBank accession numbers.

Accessions ITS

Accessions ETS Albizia adianthifolia (Schumach.) W. Wight, MW699934, BGRO

001

Albizia amara (Roxb.) Boivin, MW699936, BGRO 003

Albizia anthelmintica Brongn., MW699937, BGRO 004

Albizia arenicola R. Vig., MW699938, R. Randrianaivo 642, MO Albizia brevifolia Schinz, MW699940, BGRO 005

Albizia glaberrima Hutch. & Dalziel, MW699943, R.E. Gereau 6203, MA

Albizia gummifera (J.E. Gmel.) C.A. Sm., MW699944, J.E. Lawes- son 5094, AAU

Albizia harveyi E. Fourn., MW699945, BGRO 006

Albizia julibrissin Durazz., MW699946, BGRO 007

Albizia adianthifolia, MW/699372, BGRO 001

Albizia amara, MW/699374, BGRO 003

Albizia anthelmintica, MW699375, BGRO 004

Albizia antunesiana Harms, MW699376, S.H.C.P. 966, MO Albizia arenicola, MW/699377, R. Randrianaivo 642, MO Albizia brevifolia, MW/699378, BGRO 005

Albizia chinensis (Osbeck) Merr., MW69379, A. Ntemi & A. Athumani 478, MO

Albizia crassiramea Lace, MW699380, K. Larsen et al. 46378, AAU Albizia ferruginea (Guill. & Perr.) Benth., MW699382, C.H. Jongkind 2098, MA

Albizia kalkora (Roxb.) Prain, MW699947, E. Bouflord 26356, MO. Albizia glaberrima, MW/699383, R.E. Gereau 6203, MA

Albizia lebbeck (L.) Benth., MW699948, C. Chan 7539, CICY Albizia petersiana (Bolle) Oliv., MW699950, BGRO 008 Albizia procera (Roxb.) Benth., MW699953, BGRO 009 Albizia retusa Benth., MW699954, K. Yasuda 1804, MO Albizia tanganyicensis Baker f., MW699956, BGRO 010 Albizia umbellata (Vahl) E. J. M. Koenen, EF638182.1

Balizia leucocalyx (Britton & Rose) Barneby & J.W. Grimes, MW699959, S. Aguilar & E Aguilar 1833, M

Lysiloma acapulcense (Kunth) Benth., MW699960, H. Gémez D. 2003, MO

Lysiloma latisiliquum (L.) Benth., MW699961, P. Sima 2287, CICY

Pseudalbizzia adinocephala (Donn. Sm.) E.J.M. Koenen & Duno, MW699935, BGRO 002

MW699958, J.L. Linares 5406, FCME

Pseudalbizzia berteroana (Balb. Ex DC.) Britton & Rose, MW699939, A. Jimenez 2113, MO

Pseudalbizzia edwallii (Hoehne) E.J.M. Koenen & Duno, MW699942, J.M. Silva & L.M. Abe 4237, MEXU Pseudalbizzia multiflora (Kunth) E.J.M. Koenen & Duno, MW699949, X. Cornejo 1922, GUAY

Pseudalbizzia pistaciifolia (Willd.) E.J.M. Koenen & Duno, MW699951, X. Cornejo 5323, GUAY

Pseudalbizzia polycephala (Benth.) E.J.M. Koenen & Duno, MW699952, L.P. Queiroz 9578, MEXU

Pseudalbizzia sinaloensis (Britton & Rose) E.J.M. Koenen & Duno, MW699955, C.E. Hughes et al. 1576, FCME

Pseudalbizzia tomentosa (Micheli) E.J.M. Koenen & Duno, MW699957, A. Dorantes et al. 165, CICY

Pseudosamanea cubana (Britton & P. Wilson) Barneby & J.W. Grimes, MW699941, GHBG 001

Pseudosamanea guachapele (Kunth) Harms, MW699962, BGRO 011 Zapoteca formosa (Kunth) H.M. Hern., MW699963, R. Duno s.n. CICY. Additional accessions (ITS): Acacia acradenia F.Muell., AF487765.1

Acacia longifolia (Andrews) Willd., HM007655.1

Acaciella angustissima (Mill.) Britton & Rose, EF638169. 1 Balizia pedicellaris (DC.) Barneby & J.W. Grimes, JX870657.1 Calliandra dysantha Benth., JX870684.1

Albizia gummifera, MW/699384, J.E. Lawesson 5094, AAU Albizia harveyi, MW699385, BGRO 006

Albizia julibrissin, MW699387, BGRO 007

Albizia kalkora, MW/699388, E. Bouflord 26356, MO

Albizia lebbeck (L.) Benth., MW699389, C. Chan 7539, CICY Albizia lebbekioides (DC.) Benth., MW699390, H. Balslev 9333, AAU

Albizia lucidior (Steud.) 1.C. Nielsen ex H. Hara, MW699391, J.E Maxwell 95-259, MO

Albizia petersiana (Bolle) Oliv., MW699394, BGRO 008

Albizia procera, MW699396, BGRO 009 Albizia retusa, MW699397, K. Yasuda 1804, MO

Albizia sahafariensis Capuron, MW699398, R. Randrianaivo et al. 1387, MO Albizia tanganyicensis, MW699400, BGRO 010

Albizia umbellata, EF638157.1 Balizia leucocalyx, MW699403, S. Aguilar & F. Aguilar 1833, M

Balizia pedicellaris (DC.) Barneby & J.W. Grimes, MW699404, PR. House 1880, MA Havardia mexicana, MW699405, S. Foldi s.n., CICY

Hesperalbizia occidentalis, MW699406, J.G. Hernandez Oria 21, FCME Lysiloma acapulcense, MW699407, H. Gomez D. 2003, MO

Lysiloma latisiliquum, MW/699408, P. Sima 2287, CICY

Paraserianthes lophantha, MW699409, H. Balslev et al. 62450, AAU Pithecellobium diversifolium, MW699410, J.E.B. Pastore & R.M. Harley 2599 MO

Pithecellobium excelsum, MW699411, G. P. Lewis et al 2339, MO Pseudalbizzia adinocephala, MW/699373, BGRO 002 MW699402, J.L. Linares 5406, FCME

Pseudalbizzia edwallii, MW699381, J.M. Silva & L.M. Abe 4237, MEXU

Re-establishment of the genus Pseudalbizzia

Accessions ITS

Calliandra foliosa Benth., EF638181.1 Cojoba arborea (L.) Britton & Rose, JX870758.1 Cojoba undulatomarginata L. Rico, EF638187.1

Ebenopsis ebano (Berland.) Barneby & J.W. Grimes, JX870759.1 Enterolobium contortisiliquum (Vell.) Morong, EF638190.1

Enterolobium cyclocarpum (Jacq.) Griseb., EF638191.1 Enterolobium timbouva Matt., JX870760.1

Faidherbia albida (Delile) A. Chev., EU812008.1 Havardia mexicana (Rose) Britton & Rose, JX870762.1

Havardia pallens (Benth.) Britton & Rose, KF921656.1 Hesperalbizia occidentalis (Brandegee) Barneby & J.W. Grimes, EF638195. 1

Hycrochorea corymbosa (Rich.) Barneby & J.W. Grimes, JX870763.1

Jupunba trapezifolia (Vahl.) Moldenke, EF638166.1 Mariosousa coulteri (Benth.) Seigler & Ebinger, EF638198.1 Mariosousa dolichostachya (S.F. Blake) Seigler & Ebinger, EF638199.1

Paraserianthes lophantha (Willd.) 1.C. Nielsen, EF638204. 1 Pithecellobium diversifolium Benth., JX870768.1 Pithecellobium dulce (Roxb.) Benth., EF638207.1 Pithecellobium excelsum (Kunth) Mart., EF638208.1

Samanea saman (Jacq.) Merr., JX870770.1

Samanea tubulosa (Benth.) Barneby & J.W. Grimes, EF638212.1 Pseudosamanea guachapele (Kunth) Harms, JX870769.1 Senegalia berlandieri (Benth.) Britton & Rose, KY688777. 1 Sphinga acatlensis (Benth.) Barneby & J.W. Grimes, EF638214.1 Vachellia campechiana (Mill.) Seigler & Ebinger, EF638215.1 Vachellia farnesiana (L.) Wight & Arn., EF638219.1 Viguieranthus ambongensis (R. Vig.) Villiers, JX870773.1 Viguieranthus densinervus Villiers, JX870774.1

Viguieranthus megalophyllus (R. Vig.) Villiers, JX870776.1 Viguieranthus subauriculatus Villiers, JX870778.1

Zapoteca tetragona (Willd.) H.M. Hern., JX870784.1

Dy)

Accessions ETS Pseudalbizzia inundata (Mart.) E.J.M. Koenen & Duno, MW699386, H. Balslev et al. 97355, AAU Pseudalbizzia multiflora (Kunth) E.J.M. Koenen & Duno, MW699392, X. Cornejo & T. Andres 8705, GUAY Pseudalbizzia niopoides (Spruce ex Benth.) E.J.M. Koenen & Duno, MW699393, J.R. Grande 374, VEN Pseudalbizzia polycephala MW/699395, L.P. Queiroz 9578, MEXU Pseudalbizzia sinaloensis, MW699399, C.E. Hughes et al. 1576, FCME Pseudalbizzia tomentosa, MW699401, A. Dorantes et al. 165, CICY Pseudosamanea cubana (Britton & P. Wilson) Barneby & J.W. Grimes, MW699412, BJ FTGH 2000 Pseudosamanea guachapele, MW699413, BGRO 011 Samanea tubulosa (Benth.) Barneby & J.W. Grimes, MW699414, G.A. Parada & V.D. Rojas 2480, MO. Additional accessions: Acacia acradenia, EF638116.1 Acacia longifolia, EF638115.1 Acaciella angustissima EF638082.1

Pseudalbizzia adinocephala EF638144.1

Albizia kalkora EF638158.1 Albizia lebbeck EF638155.1 Albizia saponaria (Lour.) Blume, EF638085.1

Archidendropsis basaltica (F. Muell.) 1.C. Nielsen, EF638141.1 Archidendropsis thozetiana (F. Muell.) 1.C. Nielsen, EF638140.1 Calliandra dysantha EF638121.1

Calliandra foliosa EF638122.1

Cojoba arborea EF638095.1

Cojoba undulatomarginata EF638096.1

Ebenopsis confinis (Standl.) Britton & Rose, EF638100.1 Ebenopsis ebano EF638101.1

Enterolobium contortisiliquum EF638151.1

Enterolobium cyclocarpum EF638149.1

Faidherbia albida EF638163.1

Havardia pallens EF638146.1

Hesperalbizia occidentalis EF638139.1

Hycrochorea corymbosa EF638138.1

Jupunba trapezifolia (Vahl.) Moldenke, EF638110.1 Mariosousa coulteri (Benth.) Seigler & Ebinger, EF638124.1 Mariosousa dolichostachya EF638084.1

Pararchidendron pruinosum (Benth.) 1.C. Nielsen, EF638129.1 Paraserianthes toona (Bailey) I.C. Nielsen, EF638106.1 Pithecellobium dulce EF638142.1

Pseudosamanea guachapele EF638160.1

Samanea saman EF638136.1

Samanea tubulosa EF638135.1

Senegalia berlandieri EF638162.1

Sphinga acatlensis EF638145.1

Vachellia farnesiana EF638128.1

Viguieranthus ambongensis KR997873.1

Viguieranthus densinervus JX870891.1

Viguieranthus megalophyllus KR99787 1.1

Viguieranthus subauriculatus KR997076.1

Zapoteca formosa EF638 134.1

Zapoteca tetragona EF638 133.1.

380 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

The quality of the PCR products was evaluated by agarose electrophoresis (3 ul of final volume plus 2 ul of bromophenol blue, gel prepared with 0.5x TBE and 1% agarose, run at 120 volts and 25 amperes 30 min). PCR products were sequenced at Macrogen (http://www.macrogen.com/eng/) using the same amplification primers. The sequencing products were assembled and edited using the Sequencher v. 5.2.3. An initial automated alignment was conducted with MAFFT (Katoh et al. 2002) using the E-INS-i algorithm option, a 1OOPAM/k = 2 scoring matrix, a gap opening pen- alty of 1.3, and an offset value of 0.123. The alignments were visually inspected and manually edited for further improvement. The Akaike Information Criterion (AIC), implemented in jModeltest (Posada 2008) was used to select the best model of nucleo- tide substitution for each alignment. The selected models were TVM+1+G for ETS and GTR+I+G for ITS. Phylogenetic analyses were performed with MrBayes v.3.2.5 (Ronguist et al. 2012) separately for each dataset, and subsequently concatenated with each partition treated as independent and associated with its own evolutionary model. Analyses were performed using default parameters for 5 million generations. Two in- dependent threads were run. Convergence was assessed with both MrBayes and Tracer (Rambaut et al. 2018). Posterior Probabilities (PP) < 0.95 were considered weakly sup- ported whereas PP of 0.95—1.0 were deemed strongly supported (Alfaro et al. 2003).

To examine the evolution of fruit types within New World Albizia we utilize a phylogeny derived from a new analysis based on data of Ringelberg et al. (2022) for the Jupunba clade, as described in Soares et al. (2022), because it is based on a large set of 560 nuclear exons and flanking non-coding regions, and therefore shows enhanced resolution within this clade compared to the ITS + ETS phylogeny.

Results

Alignments of our combined datasets recovered by MAFFT required few manual ad- justments. The ETS sequences had 381 bp and, once aligned, 52% of the data were informative. In the case of ITS, the sequences were slightly longer, 551 bp but only 34% were informative.

None of the molecular-based analyses (ETS, ITS, and ETS+ITS) using Bayesian inference recovered the genus Albizia as monophyletic. The combined ETS + ITS phylogeny (Fig. 3) is used as the basis for discussing the results in detail. The outgroup Vachellia farnesiana, plus Acaciella angustissima (Mill.) Britton & Rose, Senegalia ber- landieri (Benth.) Britton & Rose and two species of Mariosousa Seigler & Ebinger form a paraphyletic grade subtending a fully supported clade (PP = 1) that includes all members of the tribe Ingeae as well as Acacia s.s.

The most relevant clade from the perspective of this study is highly supported (PP = 0.96) and includes all members of the genus Albizia and a few other genera of tribe Ingeae (clade A). The genus Albizia, as currently circumscribed, is non-monophyletic with species placed in two separate, strongly supported clades (Fig. 3). As in Koenen et al. (2020), species of Albizia sect. Arthrosamanea are placed in clade B (PP = 0.96),

Re-establishment of the genus Pseudalbizzia 381

Vachellia famesiana Acaciella angustissima

a7 Senegalia berlandieri 1 +— Mariosousa coulteri Mariosousa dolichostachya

i —— Archidendropsis basaltica 79 5 Archidendropsis thozetiana 1;— Paraserianthes tooana 1 Paraserianthes tooana b 1 96 Paraserianthes lophantha 4 Acacia acradenia Acacia longifolia __ 1 Ebenopsis confinis 1, Ebenopsis ebano 1 _- Ebenopsis ebano b 4, Sphinga acatlensis phinga acatlensis b 1 65 1 Harvardia pallens _ 1) Harvardia mexicana 65 Harvardia mexicana b 1 Pithecellobium dulce 86 rr - Pithecellobium diversifolium : Pithecellobium excelsum 7 Faidherbia albida 1 Cojoba arborea 4) Cojoba arboreab 1 Cojoba undulatomarginata

Pararchidendron pruinosum

Lysioma spi i Lysiloma latisiliguum b 84 1, Hesperalbizia occidentalis 1, Hesperalbizia occidentalis b op Hesperalbizia occidentalis c — Viguiorenthus subaumculatus Viguieranthus megalophyllus 99 a4 iguieranthus ambongensis

W iguieranthus densinervus Thailentadopsis nitida ' 20 j - Calliandra dysantha 69 Calliandra foliolosa i Zapoteca formosa ar Zapoteca tetragona — Jupunba trapezifolia ST | 1) Balizia pedicellaris

; Getina Reaicellaris va J b B li . H d h rochorea corymbosa 7 C Lite fisrecharea compose, upunba, Balizia, Hydrochorea . — Balizia leucocalyx Balizia leucocalyx 98 Balizia leucocalyx c od r ; B ip— Pseudalbizzia edwallii 1 Pseudalbizzia edwallii b 4 prcudabbizzia pol) tenho b i r. 'seudalbizzia polycephala P d Ib Pseudalbizzia tornentosa D F : lg pcelcenh lomentosal seu a ] ZZi a seudalbizzia sinaloensis ao =e Pseudalbizzia sinaloensis b 199 Pseudalbizzia adinocephala New World Al bizi a Pseudalbizzia adinocephala b 1 ip Pseudalbizzia adinocephalac. i k - Pps falbizzia adinocephala d - Pseudalbizzia berteroana | Pseudalbizzia inundata ; A 96] Pseudalbizzia subdimidiata var. minor 89 1p— Pseudalbizzia multiflora Pseudalbizzia multiflora b .79'— Pseudalbizzia multiflorac Pseudalbizzia niopoides Pseudalbizzia niopoides b 1, Pseudalbizzia pistaciifolia Pseudalbizzia pistacilfolia b

Pseudosamanea cubana 96 Pseudosamanea cubana

Pseud ibana b CR ip RT Pseudosamanea = Eneroiooum cfepeamum ; 1 n E 96 Enterolobium confortisfiquum E n ter ol obi um .96" Enterolobium timbouva ip Samanea saman

' 1 Samanea saman b Samanea

ip Samanea tubulosa Samanea tubulosa b F Albiziaumbellata 1— Albizia chinensis \— Albizia chinensis b

1-— Albizia retusa 8 u Albizia retusa b A | b i i; Albizia kalkora I Zz. | a Albizia kalkora b

Abid heise ae (O Id World )

Albizia julibrissin b Albizia procera b

a Albizia lucidor 80 75L Albizia procera 1 Albizia crassiramea, Albizia crassiramea b

Albizia saponaria_ Albizia lebbeckioides Albizia lebbeckioides b Albizia lebbeck

= Albizia petersiana Albizia sahafariensis | 1 Albizia gummifera 1) Albizia glaberrima | Albizia glaberrima b Albizia brevifolia b Albizia brevifolia 2 Albizia amara 80° Albiziaamarab ip Albizia anthelmintica Albizia anthelmintica b ‘ Albizia harveyi

Albizia adianthifolia

AS 78

Albizia harveyi b i;— Albizia antunesiana Albizia antunesiana b

= Albizia ferruginea 1p Albizia tanganyicensis Albizia tanganyicensis b

06

Figure 3. Phylogeny of the ingoid clade (sensu Koenen et al. 2020), i.e., the traditionally recognized tribes Ingeae + Acaciaeae (excl. Vachellia). Phylogram derived from Bayesian analysis in MrBayes of the combined ETS and ITS data for Albizia and related genera. Main clades are labeled A=F (see text). Pos-

terior support values are indicated above branches.

which equates to the Jupunba clade of Koenen et al. (2020). All species that were sam- pled from this section are included in this clade (Fig. 3 clade D), which received full support (PP = 1), and as in Koenen et al. (2020) this section is sister to clade C (PP = 1) comprising the genera Jupunba, Balizia and Hydrochorea.

‘usus0y yi Aq ‘wpqgdoony ‘[ pue vuvtinogsvg vqundnf ‘vyvuisivu paLoqgs04p] JO ‘saysny UTjOD Aq ‘szuvyaotpad vizyvg pur vsojuamoy <f sapiodoiu gy ‘vLoyfiynul ¢f ‘vIVp -UNUL VIZZIGIVPNAST JO SCIOYY “‘VaLogr04pE] pure vizyvg ul suotusueN paypered sepruns pur vrzz1q7vpnasy Ul syeuqey poyepunut A]jeuOsvas UT SULMOIS saTdods YITM porelsos -se suNJy snojoysorpAy WIo;MUIWOT 07 APuanbasgns pur syMsy WdsTYySpuT a1eIdas 0} syNIy possaiduros-ouryd ‘Aroded ATTeMssoue wosy suoTsueN ATeUOTNIOAD oy) SUT -MOYS (ZZOZ) ‘Je 19 Bioqjesuny wos vrep soz wy) (7ZOZ) ‘Te 39 soreog Aq sisATeue 901 satoods TYY,[ SV Ue Wo, UMEIPoI opepD equndn{ ay) Jo AudsoyAY J *p aaNBI4 QOUBOSIYSP JEINOIIJO} K¥

soajoie pepees-| oul Buiyesedes dueoopue eu} Ajuo ‘Juewojo}dAuo

SIINJJ WUOJUSWO| 0} UONISUe A

Spaes Usamjaq ajejdas ‘pod juadsiyspul 0} uoNisuey of

~~,

Z89z Buopieg ByE;nounpediBuog; equndnr uf €98 loueb) eouNquoy equndnr ? gge@ iourby xAjeooso1w equndnr 6e8 loueby eyAydoona; equndnr ; 009} UoWIS ByoJIzedey eEqundnr 98'°0 160 Srlor bat ByEINWIWIOD aed ‘ Lg loueb] eueinoqueg equndnr od es €zp seB10g esoyia equndnr 86:0 62 0 ‘ Serz Bu] eAyoR)sAyoeg equndnr ‘ll so Loz joueb] eeu) equndnr 7 L8pZ Bw esojvawely equndnpr ; LZPLSZ WOW ejae) equndnr €49 nepeuog e7eayyo09 equndnp 130 Plog) sa1Barq eyoysoddo equndnr ’ Ozelg luouez moage equnanp ’ egg LEUNG ByONUa!dse equndnr ’ Z9961 OAR sysuediu equndn, ZS19 Jeyuo, BipyyAyddxo equndnr SLZL Bpesend epodoip: equndnar Bez OUleqlY WyopsBue) equndnr z9g loueb| zesoquiA109 bocca Siena GS9 Napeuog | esoquiAsoo PasoyoospAH £9G WOW BreuBew BaoyooipAH 969S¢ NAIL yasuGBu06 easoyooipAH 619 BBuel\y winjejoyoyinbygo uojWoYyeED . BL9E UOIYBIeg wWinyoyIquioYys UOIWOYeED 96'0 MLLS WHO Aouds BIzeg 028 loUueb suebayja elzyeg MH Mezss1 zouenp sueyesiped eyzyeg 66} JeIINBy xAjeo0one) B/zyeg 1zog Biajino7 ejuapesoew equndnr 9Z9E SaIeOS ZoueWIP evoyswaoe eqniund gees Ajeq yselayeo eqniung Gps seqoosy sjuUewYya egniung

ZS29 souopojed yolyny eqniung OLZ Bilausy "AOUQUIOD B7EIPILWIPGNS “eA B]eIpIIPgns eizziq7epnasd L PE SUOD "AOUQLUOD JOUIW “IRA Be IpIWIPgns eizziqepnasd 2/0 OfG9Z POONA ‘“AOU'QUUOD ByEpuNUT BIzzIq/epNasy ee ZS9l Sima] Reale Brejaduqes6 eizziqgepnesg - O60E SeYBNH ‘aou'quios Bioyninw BIZzIqQ~epNasg LEZ BUdUIIA, “AOUQWIOD EypUezap eizzigjepnasg BIZZ. IQ] Bpnesd L09| UOWIS ‘AOU"qUIOD seplodoyu eB/zziqiepnasg 7m (efeydasourpe ejzziqepnasg =) sep) seyBny eayAYydouex elziqiy ‘ OZOL SeyBNH ‘Aouquios B/eydesdoulpe eizziqiepnesg * 9/5} SeyuGnyY ‘Aouquiod s/suaojeuis e/zziqrepnasg ’ 879 SOYBN} ‘Aou'quiod Psojuawo] eizziqepnasd d 929 SO]UBS-[PANS "AOU'qUIOD BUeIeying elzziqepnasd ZizZ OSBLWIE "“AOU'QUUOD I//eEMpa BIZZIq/EDNasd serz seyBny ‘aou'quwios sjsuajeduoa eizziqepnasg

Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

dnoi6jno

382

Re-establishment of the genus Pseudalbizzia 383

Within sect. Arthrosamanea, three clades are well supported, one compris- ing Albizia polycephala (Benth.) Killip ex Record and Albizia edwallii (Hoehne) Barneby & J.W. Grimes of ser. Paniculatae, a second clade comprising species of ser. Paniculatae endemic to Mexico, Central America, and the Caribbean, and a third clade that includes Albizia niopoides (Benth.) Burkart (also ser. Paniculatae) and the species from the other three series. The phylogeny of Ringelberg et al. (2022) present- ed here in Fig. 4 based on a new analysis of the Jupunba clade accessions from Soares et al. (2022), has greater resolution within sect. Arthrosamanea and shows that ser. Paniculatae forms a well-resolved paraphyletic grade in which the other three series are nested. The other two non-monospecific series also appear to be non-monophyl- etic, with the monospecific ser. nundatae nested inside ser. Arthrosamanea and these together in turn nested in ser. Multiflorae, although support for the paraphyly of ser. Multiflorae is only 0.72 pp.

The Old World species of Albizia form a monophyletic group (PP = 1) placed in clade E (PP = 0.96) (Fig. 3), with the genera Enterolobium Mart., Samanea, and Pseudosamanea. Within clade E Pseudosamanea (PP = 0.96) is sister to clade F which includes Enterolobium, Samanea and Old World Albizia. All these clades have high support (PP = 1). These analyses also support the transfer of Cathormion umbellatum Kosterm., which is placed in the Old World Albizia clade (Fig. 3), to Albizia, as pro- posed by Koenen et al. (2020).

Our analyses also confirm that the monotypic genus Hesperalbizia: H. occidentalis (Brandegee) Barneby & J.W. Grimes is sister to Lysiloma, in the Cojoba clade (sensu Koenen et al. 2020), unrelated to either New World Albizia (clade D) or Old World Albizia (clade F), as previously shown by Duno de Stefano et al. (2021). Furthermore, Pseudosamanea guachapele (Kunth) Harms (previously Albizia guachapele (Kunth) Dugand in Rico Arce et al. (2008)) is also unrelated to New World Albizia but is instead a member of clade E, sister to clade F which includes Enterolobium, Samanea and Old World Albizia.

In both Albizia sect. Arthrosamanea and the closely related Balizia and Hydro- chorea, these phylogenies suggest that lomentiform fruits were independently derived from indehiscent fruits that are septate between the seeds, as species with the latter fruit type form paraphyletic grades to the lomentiform species in both cases (Fig. 4). In turn, these indehiscent septate fruits are nested within paraphyletic assemblages of species with fruits that dehisce along one or both sutures in both groups. Inter- estingly, in both cases, a single species with crypto-lomentiform fruits is found, but it is not clear whether these were derived from the same ancestral fruit type or not. In Hydrochorea this crypto-lomentiform species appears as an intermediate between indehiscent and lomentiform species, while in Albizia sect. Arthrosamanea the crypto- lomentiform-fruited A. inundata appears to be derived from a lomentiform-fruited ancestor. Another difference between these two groups is that the follicular dehiscence of Balizia pedicellaris (DC.) Barneby & J.W. Grimes fruits appears to be secondarily derived from indehiscent fruits, but we note that similar dehiscence is also found in a few species of Jupunba.

384 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Discussion

This study addresses the non-monophyly of the genus A/bizia and our results provide important insights into the evolutionary history of the neotropical species placed in sect. Arthrosamanea, with implications for their taxonomic classification. We show that sect. Arthrosamanea, with expanded taxon sampling relative to Koenen et al. (2020), and as also shown by the study of Ringelberg et al. (2022), is monophyletic with only two exceptions: Albizia leonardii Barneby & J.W. Grimes which is placed among taxa of the ‘senegalioid grade’ (Ringelberg et al. 2022; Terra et al. 2022) and Albizia carbon- aria Britton that is more closely related to Pseudosamanea (Koenen 2022b; Ringelberg et al. 2022).

The geographically-based splitting of a large genus in tribe Ingeae, such as Albizia, which occupies a pantropical distribution, is not unprecedented, nor unexpected, es- pecially given the lack of pantropical monographic synthesis or geographically widely sampled phylogenies for the mimosoid clade. For example, the genus Pithecellobium, once the largest genus of tribe Ingeae, has been progressively divided during the last 50 years into multiple genera (see Brown et al. 2008 for a general history of the tribe). Another example is the genus Calliandra Benth. for which the New World species of Calliandra. ser. Laetevirentes were segregated into Zapoteca H.M. Hern. (Hernan- dez 1986), and almost all Old World species allocated progressively to other segre- gated genera: Viguieranthus Villiers (Villiers 2002), Thailentadopsis Kosterm. (Lewis and Schrire 2003), Sanjappa E.R. Souza & M.V. Krishnaraj (Souza et al. 2016) and Afrocalliandra E.R. Souza & L.P. Queiroz (Souza et al. 2013). All these taxonomic rearrangements were supported by morpho-anatomical and molecular phylogenetic analyses. Molecular data have also demonstrated that Abarema is polyphyletic (Iganci et al. 2015), prompting reinstatement of the genera Jupunba and Punjuba Britton & Rose (Soares et al. 2021). Finally, neither Zygia P. Browne nor Marmaroxylon Killip are monophyletic, although a new generic classification for those genera has yet to be proposed (Ferm et al. 2019). The non-monophyly of Albizia documented here and elsewhere (Koenen et al. 2020; Ringelberg et al. 2022) is thus not a surprise and reflects the state of flux surrounding generic delimitation in mimosoids, especially within the ingoid clade.

Some of the taxonomic proposals of Barneby and Grimes (1996) relative to the American segregates of Albizia s.l. are confirmed by our results. The genera Balizia and Hydrochorea form part of the Jupunba clade (sensu Koenen et al. 2020) (Figs 3 and 4) (Iganci et al. 2015), although neither Hydrochorea nor Balizia are monophyletic in our phylogeny (Figs 3 and 4, see Soares et al. 2022). Two neotropical species, included in Albizia by Rico Arce et al. (2008) are also placed outside New World Albizia (Fig. 3): Hesperalbizia occidentalis is closely related to Lysiloma, in agreement with previous results (Iganci et al. 2015; Duno de Stefano et al. 2021); similarly, Pseudosamanea guachapele, is also placed outside Albizia in our phylogeny, emerging, as expected, to- gether with the other species Pseudosamanea cubana (Britton & Rose) Barneby & J.W. Grimes, although relationships within this clade are unresolved (Fig. 3).

Re-establishment of the genus Pseudalbizzia 385

Here we show that the dehiscent, papery, plano-compressed fruit type is ances- tral within Albizia sect. Arthrosamanea (Fig. 4) and is associated with species growing predominantly in seasonally dry tropical forest and woodland, with successive shifts to septate indehiscent fruits and then lomentiform fruits with hydrochorous seed dis- persal associated with species growing in seasonally inundated varzea forest, riverine habitats and low-lying margins of palm and white-sand savannas (Fig. 4). Interestingly, in the sister group of Albizia sect. Arthrosamanea, the mainly neotropical clade com- posed. of Jupunba, Punjuba, Balizia, and Hydrochorea, a similar parallel evolutionary transition in fruit types is apparent. In Jupunba and Punjuba, fruits are always dehis- cent, while a transition to septate indehiscent fruits occurred in Balizia, an exception being Balizia pedicellaris which has follicular dehiscence and a newly described species with crypto-lomentiform fruits (Fig. 4, and Soares et al. 2022). Nested within the paraphyletic Balizia is a clade comprising the genus Hydrochorea plus two African spe- cies of Cathormion, all species of which have indehiscent lomentiform fruits adapted for hydrochory and are found in riparian or other periodically flooded habitats in the Amazon basin, West Africa and the Congo basin (Fig. 4, and Soares et al. 2022).

Barneby and Grimes (1996) pointed out that a radiation of species with similarly heterogeneous fruit types to that seen in section Arthrosamanea occurs in Madagascan Albizia s.s. and that the association between lomentiform fruits, hydrochorous seed dis- persal, and seasonally flooded habitats is also apparent in Old World Albizia s.s. For example, Albizia dolichadena, A. moniliformis, A. rosulata, and A. umbellata from Aus- tralasia also have lomentiform fruits and are distributed near streams or in riparian and swamp forests (Rico Arce et al. 2008). Furthermore, as indicated above and pointed out by Barneby and Grimes (1996), similar transitions to lomentiform fruits have occurred in parallel in several other lineages across the ingoid clade, including Cathormion altissi- mum (Hook.f.) Hutch. & Dandy (sometimes referred to as Albizia altissima Hook.f; Koenen 2022a) and Senegalia rostrata (Humb. & Bonpl. ex Willd.) Seigler & Ebinger (syn. Dugandia rostrata (Humb. & Bonpl. ex Willd.) Britton & Killip, syn. Manganaroa articulata Speg.; Barneby and Grimes 1996: 204) in all cases apparently also closely as- sociated with riparian and/or periodically inundated habitats. These repeated parallel derivations of similar, but not strictly homologous fruit types attest to the high evolv- ability of the mimosoid fruit more generally. In the light of phylogenetic evidence, it is now clear that these evolutionarily highly labile morphological adaptations of the fruit related to seed dispersal syndrome do not provide reliable characters for generic delimitation, supporting inclusion of the species that were placed in ser. Arthrosamanea, ser. Inundatae and ser. Multiflorae within Albizia sect. Arthrosamanea by Barneby and Grimes (1996), i.e., the clade of New World Albizia that is recovered in our analysis.

Taxonomic treatment

There are two validly published generic names — Pseudalbizzia of Britton and Rose (1928) and Arthrosamanea of Britton and Killip (1936) — that could be applied to the New World clade of Albizia. In accordance with Principle III of the International

386 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Code of Nomenclature (Turland et al. 2018), we reinstate Pseudalbizzia, the earlier name associated with this clade, and provide the corresponding new combinations for its constituent species.

Pseudalbizzia Britton & Rose, N. Am. Fl. 23: 48. 1928.

Type. Pseudalbizzia berteroana Britton & Rose.

Arthrosamanea Britton & Rose, in Britton & Killip, Ann. New York Acad. Sci. 35: 128, 1936. Albizia section Arthrosamanea (Britton & Rose) Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 206. 1996. Type: Arthrosamanea pistaciifolia Brit- ton & Rose.

Description. Unarmed trees with sympodial growth, up to 30 m, rarely small treelets of c. 3 m, microphyllidious to macrophyllidious; trunk 35—120(—150) cm dbh; young stems and all leaves and inflorescence-axes more or less densely tomentellous to pilosulous; stipules puberulent to glabrous, deltate, narrowly triangular, triangular- ovate, narrowly ovate, or narrowly lanceolate, veinless or faintly 3-veined, falling early to tardily, perhaps sometimes obsolete and/or lacking on mature leaves. Leaves bipin- nate, not sensitive, (1-)2-15(-19) pairs of pinnae; leaflets (2—)16—52(—63) pairs per pinna; a nectary immediately below first pair of pinnae, near or well below mid-petiole, sometimes lacking or reduced to a minute pore, round, elliptic or vertically elongate, either shallow-cupular or almost plane, thick-rimmed, sometimes immersed in peti- olar groove or even obsolete, much smaller nectaries at some distal pinnae, at the tip of most pinnae, and between 1-2 furthest pairs of leaflets; leaflets gently decrescent toward each end of the rachis or toward the base of the rachis or sub-equilong, the first pair of leaflets often reduced to paraphyllidia, sometimes minute, sometimes absent or perhaps falling early, the blades of the remaining leaflets elliptic, elliptic-ovate, oblong- elliptic, narrowly oblong-elliptic, lance-oblong to linear-lanceolate, base obliquely truncate to shallowly semi-cordate, apex deltately subacute, deltately acute to subacute, obtuse or apiculate, the larger ones (1.5—)2—4(—6) times as long as wide, margin strong- ly to slightly revolute; venation generally palmate, of 2—4(—5) veins from the pulvinule, the nearly straight main vein a little forwardly displaced and giving rise on each side to 2-13 major secondary veins, the inner of 2(—3) posterior primary veins incurved- ascending to anastomose slightly beyond mid-blade, the outer posterior vein and some- times a faint anterior one very short and weak, all venation immersed on upper face. Inflorescence primary axis up to 30 cm long; peduncles (1—)2—8(—10) per node of the capitulate or corymbose-umbellate inflorescence, capitula 8—26(—40)-flowered; bracts heteromorphic or homomorphic, ovate, oblong-obovate or spatulate, linear-spatulate, falling early or persistent, sessile or shortly pedicellate, the flowers moderately to strong- ly dimorphic, the terminal ones generally longer. Flowers 5-merous, rarely 6-merous, glabrous to densely pubescent externally. Peripheral flowers: calyx campanulate, turbi- nate, turbinate-campanulate or narrowly campanulate, sessile or short pedicellate, lobes very short, depressed-deltate, ovate or triangular, glabrous or puberulent; corolla nar- rowly trumpet-shaped, erect or recurved, lobes ovate to lance-ovate; androecium with

Re-establishment of the genus Pseudalbizzia 387

9—-30(-32) stamens, up to 20 mm long, united at the base forming a clear stemono- zone, the staminal tube as long or longer than the stemonozone; ovary sessile or shortly stipitate, slenderly ellipsoid, conical at apex, glabrous or pubescent; style a little longer than the stamens, slightly dilated at the stigma. Terminal flowers: sessile or almost so, calyx shallowly campanulate to broadly campanulate, corolla tubular; androecium with 16—38(—42) stamens, 8.5—11.5(—13) mm long, united at the base forming a clear stemonozone, staminal tube equalling or longer than the stemonozone. Fruits solitary, or rarely 2-4 per capitulum, sessile, subsessile or cuneately contracted at base into a short pseudo-stipe, the body linear, linear-elliptic, narrowly elliptic-oblong, straight or nearly straight, sometimes decurved, plano-compressed, apex rounded but minutely apiculate to obtuse, (8—)13(—15)-seeded; valves papery, coriaceous, or grossly ligneous, olivaceous, castaneous, fuscous-greenish, or brown becoming tan-brown, closely trans- verse venulose, minutely puberulous, tomentulose, glabrescent to glabrous, framed by straight sutures or dilated, sometimes 3-angulate but not winged, transversely or hori- zontally, dehiscence tardy to very tardy, inert, through both sutures or dehiscence 0, in the latter, the pod crypto-lomentiform, incipiently lomentiform or lomentiform, then the whole fruit long persistent on the tree, commonly falling entire and breaking on the ground into 8—12 individually indehiscent segments, funicle apically sigmoid or ribbon-like (not sigmoid), lentiform; seeds obliquely ascending or straight, disciform, oblong-ellipsoid, elliptic, strongly compressed, the translucent, brownish or greyish testa produced as a peripheral wing, adherent to the embryo, which does not fill the testa-cavity, the pleurogram small, inversely U-shaped or U-shaped.

Notes. The genus forms a group that is homogeneous in most respects, but diverse in the late developmental stages of the fruit, including: 1) fruit opening type: dehis- cent, indehiscent, or irregularly breaking, 2) lateral shape: flat to conspicuously raised over the seed chambers, 3) texture and consistency of the valves: papery, chartaceous to woody (Barneby and Grimes 1996). Figs 1, 2 and 4.

Pseudalbizzia (clade D) is the sister group of the /upunba-Punjuba-Balizia-Hydro- chorea clade (Fig. 3). Jupunba and Punjuba are markedly different morphologically, having spirally twisted dehiscent fruits with a red or ochre endocarp, reminiscent of the fruits of several other genera in tribe Ingeae (e.g., some Pithecellobium species, and some species of Archidendron F. Muell. and Cojoba Britton & Rose). The red or red- brown testa of the seeds of Jupunba and Punjuba are very distinctive, and are never black, and the embryo is nearly always aniline-blue due to the presence of delphinidin (an anthocyanidin). Punjuba is furthermore distinguished by its spicate inflorescences, which are not seen in Pseudalbizzia. Balizia has ligneous, indehiscent or tardily dehis- cent pods, their seeds being released sometimes only after decay of the valves on the floor of terra firme forest, whereas in Hydrochorea the fruits are lomentiform, adapted to dispersal by water. ‘The fruits of Hydrochorea recall some species of Pseudalbizzia adapted to similar riparian habitats. However, the species of Pseudalbizzia are markedly different in form of inflorescence, leaflet-venation, and shape of the ovary.

Two species previously placed in Albizia from the New World which were not included in our phylogenetic analysis, Albizia carbonaria and A. leonardii, have

since been shown to be placed outside the New World Albizia clade (Ringelberg

388 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

et al. 2022; Koenen 2022b; Terra et al. 2022). Two other species, also not sampled here, nor by Ringelberg et al. (2022), are here tentatively included in Pseudalbizzia: Albizia barinensis L. Cardenas and Albizia buntingii Barneby & J.W. Grimes (see be- low for discussion about the placement of these species). The genus Pseudalbizzia was published in the Flora of North America (Britton and Rose 1928) and included just a single species, P berteroana. The original description of Pseudalbizzia closely matches Albizia and no characters distinguishing the two genera were discussed by Britton and Rose (1928). The generic name Arthrosamanea was also published by Britton & Rose, again with a single species, A. pistaciifolia (Willd.) Britton & Rose, in an account of the Mimosaceae and Caesalpiniaceae of Colombia (Britton and Killip 1936), but again no differences between the genus and Albizia or Pseudalbizzia were mentioned. Pseudalbizzia as circumscribed here comprises 17 species and 5 varieties ranging in distribution from northwestern Mexico to northern Argentina and including the Greater Antilles (Figs 5 and 6). Full synonymy, detailed species descriptions, geograph- ical distributions, representative samples of all species and keys for their identifica- tion can be found (under the name Albizia) in Barneby and Grimes (1996), Linares (2005) and Rico Arce et al. (2008). Finally, we propose a new sectional classification of

Section Pseudalbizzia @ P. adinocephala

@ P. berteroana

© P. sinaloensis

< P. tomentosa

| Incertae sedis | > P. buntingii

Section Paniculata @ P. barinensis

e P. coripatensis © P. edwallii

© P, polycephala

Section Uninervia . A P. burkartiana |

100°W 80°w

Figure 5. Distribution map of Pseudalbizzia sections Paniculata, Pseudalbizzia, Uninervia and Pseudalbi-

zzia buntingii (incertae sedis), as per the legend.

Re-establishment of the genus Pseudalbizzia 389

Section Pterocarpa © P. niopoides

Section Arthrosamanea ® P. decandra P. glabripetala P. inundata P. multiflora P. pistaciifolia P. subdimidiata

Figure 6. Distribution map of Pseudalbizzia sections Arthrosamanea and Pterocarpa, as per the legend.

Pseudalbizzia to account for the non-monophyly of the series of Barneby and Grimes (1996), based on the phylogenies (Figs 3 and 4) which sampled nearly all species. A

key to the sections is provided.

Key to the sections of the genus Pseudalbizzia

1 Leaflets with a single vein from the pulvinule....... sect. Uninervia = Peaflets with.5=5 veins throm thie pulvinle -.1.2.. s..nensbrereRaaressaecednaacitsrnnsbaest 2 2 Fruits with a narrowly winged margin, seeds oblique, foliage microphyllidi-

US occur cen cer Be ance ccancea eens cow oatinoatoucaney ou tee Hac qe carers cau megN me edeee gs sect. Pterocarpa = Fruit margins not winged, or if winged, then foliage macrophyllidious and

SRCCASIS TANS DE, cot swrasa aerate mv ienwcesen annbaneah auccr merely aunties wm cbr eeahat 3 3 Fruits indehiscent and septate or lomentiform............. sect. Arthrosamanea — Fruits dehiscent, plano-compressed, valves papery, not septate....... eee 4 4 Micro- to mesophyllidious foliage, distributed in South America.................

i Ve Re Sn, SA) | Sei le tO ee sect. Paniculata = Macro- or microphyllidious foliage, distributed in Mexico, Central America Tre CATED WEAN. 6s.) sis eescthas barcba uss Mine oestee noeadtcteasetaeds sect. Pseudalbizzia

390 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Pseudalbizzia sect. Paniculatae (Benth.) E.J.M. Koenen & Duno, stat. nov. and sect. nov. urn:lsid:ipni.org:names:77303802-1

Pithecellobium sect. Samanea ser. Paniculatae Benth. pro parte, London J. Bot. 3: 219. 1844.

Albizia sect. Arthrosamanea ser. Paniculatae (Benth.) Barneby & J.W. Grimes pro parte, Mem. New York Bot. Gard. 74(1): 208. 1996. Type species (designated by Barneby and Grimes, Mem. New York Bot. Gard. 74(1): 208. 1996.): Pithecellobium polycephalum Benth. = Pseudalbizzia polycephala (Benth.) E.J.M. Koenen & Duno.

Pithecellobium sect. Samanea set. Parviflorae [sic] Benth. pro parte, Trans. Linn. Soc. London 30: 591 (exclus. sp. 77). 1875 & in Martius, Fl. Bras. 15(2): 445. 1876. Type species (designated by Barneby and Grimes, Mem. New York Bot. Gard. 74(1): 208. 1996.): Pithecellobium polycephalum Benth. = Pseudalbizzia polycephala (Benth.) E.J.M. Koenen & Duno.

Type. Pithecellobium polycephalum Benth. = Pseudalbizzia polycephala (Benth.) E.J.M. Koenen & Duno.

Notes. Micro- to mesophyllidious trees with paniculate compound inflorescences of efoliate pseudoracemes and dehiscent plano-compressed papery fruits. Four species of humid, semi-deciduous and seasonally dry tropical and extratropical forests and

woodland in South America (Fig. 5).

Pseudalbizzia barinensis (L. Cardenas) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303803-1

Basionym. Albizia barinensis L. Cardenas, Ernstia 21: 5, f. sn. 1983.

Type. Venezuela. Barinas, muy cerca de Punta de Piedra, 3 Apr 1976, L. Cardenas de Guevara 2273 (holotype: MY; isotypes: BM!, F! [F0093839F], K! [K000527984], NY! [NY00001781], RB! [RB00539860], US! [US00385615], VEN).

Notes. This species has not been included in any phylogenetic analysis, but its foli- age, efoliate pseudoracemes and plano-compressed papery fruits leave little doubt that it should be placed in Pseudalbizzia. It is here included in section Paniculata based on these characters and its South American distribution.

Pseudalbizzia coripatensis (Rusby) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303804-1

Basionym. Pithecellobium coripatense Rusby, Bull. New York Bot. Gard. 4: 349. 1907.

Type. Bolivia. La Paz, Sur Yungas, at Coripata, 6 May 1894, M. Bang 2176 (holotype: NY! [NY00334642]; isotypes: BM! [BM000952433], G-2! [G00364414, G00364429], GH-2! [GH00064010, GH00064011], M! [M0218258], K! [K000527985], MINN, MO! [MO-954213], US).

Re-establishment of the genus Pseudalbizzia 39]

Pseudalbizzia edwallii (Hoehne) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303805-1

Basionym. Pithecellobium edwallii Hoehne, Bol. Inst. Brasil. Sci. 2: 243. 1926. Type. Brazil, Sao Paulo, G. Edwall 5608 (lectotype: SP, designated by Barneby and Grimes, Mem. New York Bot. Gard. 74(1): 209. 1996).

Pseudalbizzia polycephala (Benth.) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303806-1

Basionym. Pithecellobium polycephalum Benth., London J. Bot. 3: 219. 1844.

Type. Brazil. Rio de Janeiro, /.B.E. Pohl 1420 (lectotype: K! (herb. Bentham) [K000528000], designated by Barneby and Grimes, Mem. New York Bot. Gard. 74(1): 208. 1996).

Pseudalbizzia sect. Uninervia E.J.M. Koenen & Duno, sect. nov. urn:lsid:ipni.org:names:77303807-1

Type. Albizia burkartiana Barneby & J.W. Grimes = Pseudalbizzia burkartiana (Barneby & J.W. Grimes) E.J.M. Koenen & Duno.

Notes. Microphyllidious trees with the inflorescences of section Paniculata, but with a single vein from the pulvinule at the base of the leaflets. A single, narrowly endemic species in Parana pine woodland and the Southern Mata Atlantica of Brazil

(Fig. 5).

Pseudalbizzia burkartiana (Barneby & J.W. Grimes) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303808-1

Basionym. Albizia burkartiana Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 211-212. 1996.

Type. Brazil. Santa Catarina, Capinzal, on upper Rio Uruguai, 700 m, 21 Dec 1973, PR. Reitz & R. M. Klein 14359 (holotype: NY! [NY00001783]; isotype: US! [US00811452]).

Notes. In the protologue the fruits were not described as these were not known at that time. This rare, locally endemic species has since been collected in fruit (Stival-Santos 678, BR), and we here provide a description of these. Fruits sessile but with a narrow pseudo-stipitate base, dehiscent along both slightly thickened sutures, the valves plano-compressed, papery in texture, light brown with finely prominent transverse veins, 6.5-12 x 1.2-1.6 cm, 7—12-seeded when well- fertilized.

392 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Pseudalbizzia sect. Pseudalbizzia. Type as for the genus.

Notes. ‘Irees with micro- or macrophyllidious foliage, inflorescences composed of efoliate pseudoracemes arising singly from a leaf axil or sometimes the capitula solitary or paired in the leaf axils, or the pseudoracemes combined into a terminal panicle, fruits plano-com- pressed with papery valves, dehiscent along both sutures or more rarely indehiscent (in P berteroana), sometimes with a winged margin, seeds straight. Four species predominantly of seasonally dry tropical forests in Mexico, Central America and the Caribbean (Fig. 5).

Pseudalbizzia adinocephala (Donn. Sm.) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303809- 1

Albizia xerophytica J. Linares, syn. nov., Revista Mex. Biodiversidad 76: 7. 2005. Type: Honduras. El Paraiso, Municipio Morocelf, orillas de Quebrada Grande c. 3.9 km al NE de Moroceli por el camino hacia El Plan. 2002. /.L. Linares et al. 5674 (holotype: MEXU! [MEXU01160777]; isotype: EAP).

Basionym. Pithecellobium adinocephalum Donn. Sm., Bot. Gaz. Crawfordsville. 57: 419. 1914.

Type. Costa Rica. San José, Ad fundum La Verbena prope Alajuelita, 100 m, Aug 1894, A. Tonduz 8932 (US-3); Dec 1894 (lectotype: A. Tonduz 9077 [US-212774]}; isolectotypes: BR-3! [BRO000005189519, BRO000005189182, BRO000005189847], G! [G00364416], designated by Barneby and Grimes, Mem. New York Bot. Gard. 74(1): 218. 1996).

Notes. Albizia xerophytica was described from material from dry forest habitats in southern Honduras based on minor differences in leaf and fruit morphology, but we do not consider these to be significantly different from the range of variation that is observed in P adinocephala and prefer the broader concept of the species as de- scribed in Barneby and Grimes (1996: 218-220). The difference in habitat (i.e., lower rainfall regions) also appears to be minor, as some specimens from wetter sites have been identified as A. xerophytica (see map in Rico Arce et al. 2008) while specimens of P adinocephala have been collected across the full range of drier and wetter sites. Finally, the distribution of A. xerophytica is entirely enclosed by the much wider range of P adinocephala.

Pseudalbizzia berteroana (Balb. ex DC.) Britton & Rose, N. Amer. Fl. 23: 48. 1928.

Basionym. Acacia berteroana Balb. ex DC., Prodr. 2: 470. 1825.

Re-establishment of the genus Pseudalbizzia 393

Type. Republica Dominicana, Sto. Domingo, C\L.G. Bertero, herb. Balbis s.n., 1821 (holotype: G; isotype: M! [M0218254]).

Pseudalbizzia sinaloensis (Britton & Rose) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:773038 10-1

Basionym. Albizia sinaloénsis in Britton & Rose, N. Amer. Fl. 23(1): 45. 1928.

Type. Mexico. Sinaloa, vicinity of Fuerte, 26 March 1910, /.N. Rose, PC. Standley & Russell 13559 (holotype: NY! [NY00001775]; isotype: US! [US00000483)]).

Pseudalbizzia tomentosa (M. Micheli) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:7730381 1-1

Basionym. Pithecellobium tomentosum M. Micheli, Mém. Soc. Phys. Genéve 34: 285, te 28.1903.

Type. Mexico. Michoacan, rives de l’Espiritu Santo, 600 m, 19 April 1898 [E. Langlassé\ 107 (G): Zilmatango, 30 m, aout 1898, n 280 (G). (lectotype: FE. Langlassé 107 G-385667!; isolectotypes: K! [K000082098], NY (fragm.)! [NY00001777], des- ignated by Standley, Contr. U.S. Natl. Herb. 23: 396. 1922).

Pseudalbizzia tomentosa var. nayaritensis (Britton & Rose) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:773038 12-1

Basionym. A/bizzia nayaritensis Britton & Rose, N. Amer. Fl. 23: 47. 1928.

Type. Mexico. Nayarit; San Blas, La Palma, 20 m, 1923, /. Gonzdlez Ortega 90N (holotype: US! [US00918691]; isotypes: K! [K000082100], NY-2! [NY00001768, NY00001769)).

Pseudalbizzia tomentosa vat. purpusii (Britton & Rose) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:773038 13-1

Basionym. A/bizzia purpusii Britton & Rose, N. Amer. Fl. 23: 45. 1928.

Type. Mexico. Veracruz, Rancho Remudadero, 19°15'N, 96°34'W, April 1922, C.A. Purpus 8723 (holotype: NY! [NY00001773]; isotypes: GH! [GH00069252], MO! [MO-120564], UC! [UC214372], US! [US00000479]).

394 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Pseudalbizzia tomentosa var. tomentosa

Pseudalbizzia sect. Pterocarpa E.J.M. Koenen & Duno, sect. nov. urn:lsid:ipni.org:names:773038 14-1

Type. Pithecellobium niopoides Spruce ex Benth. = Pseudalbizzia niopoides (Spruce ex Benth.) E.J.M. Koenen & Duno.

Notes. Microphyllidious trees with the inflorescence usually composed of axil- lary efoliate pseudoracemes, sometimes a partly or wholly terminal panicle (but not surpassing the foliage), the fruit with a narrowly winged margin and seeds oblique. A single widespread species found in deciduous seasonally dry forests, gallery forest, and evergreen forests in Mexico, Central and South America (Fig. 6).

Pseudalbizzia niopoides (Spruce ex Benth.) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:773038 15-1

Basionym. Pithecellobium niopoides Spruce ex Benth., Trans. Linn. Soc. London 30: BOT. L675:

Type. Brazil, Para, Santarem, Nov 1851, R. Spruce 1088, Herb. Bentham (holotype: K! [K000528013]).

Pseudalbizzia niopoides vat. colombiana (Britton) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:773038 16-1

Albizia niopoides var. colombiana (Britton) Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 222. 1996.

Basionym. Albizzia colombiana Britton, in Britton & Killip, Ann. New York Acad. Sci, Sows, 1936:

Type. Colombia. Magdalena, near Bonda, Santa Marta, 3 August 1899, H.H. Smith 38 (holotype: NY! [NY00001784]; isotypes: BR! [BRO000005111176], E! [E00313853], K! [K000527990], NY!, U-2! [U0003354, U1253389]).

Pseudalbizzia niopoides var. niopoides Pseudalbizzia sect. Arthrosamanea (Britton & Rose) E.J.M. Koenen & Duno, comb. nov.

urn:lsid:ipni.org:names:77303817-1

Arthrosamanea Britton & Rose, Ann. New York Acad. Sci. 35: 128, pro gen. 1936, sensu stricto. Albizia sect. Arthrosamanea (Britton & Rose) Barneby & J.W. Grimes

Re-establishment of the genus Pseudalbizzia 395

pro parte, Mem. New York Bot. Gard. 74(1): 206. 1996. Type species: Arthrosama- nea pistaciifolia (Willd.) Britton & Rose = Mimosa pistaciifolia Willd. = Pseudalbiz- zia pistaciifolia (Willd.) E.J.M. Koenen & Duno.

Albizia sect. Arthrosamanea ser. Multiflorae Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 234. 1996.

Albizia sect. Arthrosamanea ser. Inundatae Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 238. 1996.

Notes. Micro- or macrophyllidious trees, usually the efoliate pseudoracemes arising singly and only rarely arranged in panicles, fruits indehiscent and septate, or lomenti- form, one species crypto-lomentiform. Six species of usually humid, often seasonally inundated forest or riparian habitats in South America (Fig. 6).

Pseudalbizzia decandra (Ducke) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:773038 18-1

Basionym. Pithecellobium decandrum Ducke, Arch. Jard. Bot. Rio de Janeiro 5: 121. 1930.

Type. Brazil. Pard, habitat in silvis non inundatis civitatis Para circa Obidos, A. Ducke (Herb. Amaz. Mus. Para 15.724, et H.J.B.R. 10.174) et loco Serra do Dedal ad lacum Faro, A. Ducke (H.J.B.R. 20.198), ubi florebat Januario 1927, A. Ducke (lectotype: A. Ducke 10174 RBI; isolectotypes: G! [G00364418], K-2!: [K000527990, K000527998], U-2! [U0003349, U0003350], designated by Barneby and Grimes, Mem. New York Bot. Gard. 74(1): 234. 1996.).

Pseudalbizzia glabripetala (H.S. Irwin) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303819-1

Basionym. Pithecellobium glabripetalum H.S. Irwin, in Mem. New York Bot. Gard. L5G19:4109. 1966,

Type. Guyana. Orealla, Corantyne River, Oct 1879, G.S. Jenman 364 (holotype: NY! [NY00334664]; isotypes: BM!, P!).

Pseudalbizzia inundata (Mart.) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303820-1

Basionym. Acacia inundata Matt., Spix & Mart. in Reise Bras. 1: 555. 1823. Type. Brazil. Minas Gerais, Rio Sao Francisco, 1818, C.EP von Martius 1659 (holotype: M! [M0218478]; isotypes: K! [K000797598], NY!).

396 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Pseudalbizzia multiflora (Kunth) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303821-1

Basionym. Acacia multiflora Kunth, Nov. Gen. Sp. (quarto ed.) 6: 277-278. 1823. Type. Peru. Cajamarca, Prov. Jaén, San Felipe, 980 m, Aime Bonpland & EW.H.A. von Humboldt 3562 (holotype: P! [P00679365)).

Pseudalbizzia multiflora var. multiflora

Pseudalbizzia multiflora var. sagasteguii (Barneby & J.W. Grimes) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303822-1

Basionym. Albizia multiflora var. sagasteguii Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 237-238. 1996.

Type. Peru. Cajamarca, Prov. Contumaza, in a quebrada near San Benito, A. Sagdstegui 15410 (holotype: F! [F0042945F]; isotypes: MO! [MO-149743], NY!, US! [US00624358]).

Pseudalbizzia pistaciifolia (Willd.) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303823-1

Basionym. Mimosa pistaciaefolia [sic] Willd., Sp. Pl. 4: 1028. 1806. Type. Venezuela. Caracas. F Bredemeyer s.n., herb. Willdenow (holotype: B).

Pseudalbizzia subdimidiata (Splitg.) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303824- 1

Albizia subdimidiata (Splitg.) Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 234. 1996.

Basionym. Acacia subdimidiata Splitg. Tijdschr. Natuurl. Gesch. Physiol. 9: 112 (1842). Type. Suriname. “ad ripas fluminis Surinami superioris”, 27 April 1838. Splitger- ber 917 (holotype: L [L0018505]).

Pseudalbizzia subdimidiata vat. minor (Barneby & J.W. Grimes) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303825-1

Basionym. Albizia subdimidiata var. minor Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 234. 1996.

Re-establishment of the genus Pseudalbizzia 397

Type. Guyana. Basin of Essequibo river, Kuyaliwak Falls, 1 Jan 1937, A.C. Smith 2156 (holotype: NY! [NY00001790]; isotypes: A! [A00069262], G! [G00364427], K! [K000528004], PB, U! [U0003358]).

Pseudalbizzia subdimidiata var. subdimidiata

Incertae sedis

Pseudalbizzia buntingii (Barneby & J.W. Grimes) E.J.M. Koenen & Duno, comb. nov. urn:lsid:ipni.org:names:77303826-1

Basionym. Albizia buntingii Barneby & J.W. Grimes, Mem. New York Bot. Gard. 74(1): 223. 1996.

Type. Venezuela. Zulia, alrededores de Casigua El Cubo, 100 m, al este del em- palme de la via hacia Casigua con la carretera Machiques-La Fria, 25 Feb 1985, G.S. Bunting 13370 (holotype: NY! [NY00001782]).

Notes. Fruits of this species are unknown and the species is only known from the type locality (Fig. 5), but it is similar in leaf and inflorescence morphology to several South American species of Pseudalbizzia, as described in the protologue. Especially the efoliate pseudoracemes point to this species most likely being cor- rectly accommodated in Pseudalbizzia. Collection of fruits and/or inclusion of the species in phylogenetic studies is needed to confirm its generic and sectional placements.

Non-native species

Some cultivated and sometimes naturalized Old World Albizia species are found in the New World, including: A. procera (Roxb.) Benth., A. julibrissin, A. lebbeck (L.) Benth., and A. chinensis (Osbeck) Merr. For these species, Barneby and Grimes (1996) proposed Albizia section Albizia, now considered as Albizia s.s.

Author contributions

GAP, RR, GCFC, IVM, and RDD designed the study. GAP, LLCI, ELC, RDD con- tributed labwork. RDD, GCFC, IRM contributed data by supervising students in the lab. EIMK, RDD, XC, SM and CEH contributed taxonomic knowledge, JR contrib- uted species distribution data and the maps. GAP, RDD, ITC, JRP, and RR undertook the phylogenetic analyses. EIMK, RDD, RR, GCFC, CEH and JR contributed to

writing the manuscript.

398 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Acknowledgements

We thank the National Council of Science and Technology (CONACYT) for financial support through project 81799 and the Centro de Investigacion Cientifica de Yucatan A.C. (CICY). The mobility scholarship program for students of CONACYT support- ed the research stay by GAP at the Real Jardin Botanico-CSIC under the supervision of RR, and EJMK and JR were supported by the Swiss National Science Founda- tion (Early.Postdoc.Mobility fellowship P2ZHP3_199693 to EJMK, and grants 310003A_156140 and 31003A_182453/1 to CH). The visit to Aarhus University (Denmark) as well as Chiang Mai University (Thailand) was supported by a grant from the Carlsberg Foundation under the Flora of Thailand project (Dr. Henrik Balslev). We thank curators of the herbaria AAU, CICY, FCMEMEXU, MO, MA, and MEXU, who provided material for this study. All molecular work was done in the Laboratory of Molecular Markers (Unidad de Recursos Naturales). We remain especially grateful to Nestor Raigoza, Matilde Margarita Ortiz Garcia, Verénica Limones Briones, and Silvia Hernandez for laboratory support. Finally, we thank Gwilym Lewis for comments and editorial input. An earlier version of this article was part of the Master thesis of GAP.

References

Alfaro ME, Zoller S, Lutzoni F (2003) Bayes or bootstrap? A simulation study comparing the performance of Bayesian Markov chain Monte Carlo sampling and bootstrapping in as- sessing phylogenetic confidence. Molecular Biology and Evolution 20: 255-266. https:// doi.org/10.1093/molbev/msg028

Baldwin BG, Markos S (1998) Phylogenetic Utility of the External Transcribed Spacer (ETS) of 18S—26S rDNA: Congruence of ETS and ITS trees of Calycadenia (Compositae). Molecu- lar Phylogenetics and Evolution 10: 449-463. https://doi.org/10.1006/mpev.1998.0545

Barneby RC, Grimes JW (1996) Silk tree, Guanacaste, Monkey’s earring: A generic system for the synandrous Mimoseae of the Americas. Part I. Abarema, Albizia and allies. Memoirs of the New York Botanical Garden 74: 1-292.

Britton NL, Killip EP (1936) Mimosacez and Caesalpiniacez of Colombia. Annals of the New York Academy of Sciences 35: 101—228. https://doi.org/10.1111/j.1749-6632.1933.tb55366.x

Britton NL, Rose JN (1928) North American Flora. Part I. (Rosales). Mimosaceae. 23: 1-194.

Brown GK, Murphy DJ, Miller JT, Ladiges PY (2008) Acacia s.s. and its relationship among tropical legumes, Tribe Ingeae (Leguminosae: Mimosoideae). Systematic Botany 33: 739- 751. https://doi.org/10.1600/036364408786500136

Duno de Stefano R, Tun Tun C, Lopez Contreras JE, Carnevali Fernandez-Concha G, Leopar- di Verde CL, Ramirez-Prado JH, Can Itza LL, Tamayo Cen I (2021) Phylogeny of Lysiloma (Fabaceae), a genus restricted to Megamexico with outliers in the West Indies and Florida. Acta Botanica Mexicana 128: e1728. https://doi.org/10.21829/abm128.2021.1782

Ferm J, Korall PB, Lewis GP, Stahl B (2019) Phylogeny of the Neotropical legume genera Zygia and Marmaroxylon and close relatives. Taxon 68: 661-672. https://doi.org/10.1002/ tax.1 2117

Re-establishment of the genus Pseudalbizzia 399

Hernandez HM (1986) Zapoteca: A new genus of Neotropical Mimosoideae. Annals of the Missouri Botanical Garden 73: 755-763. https://doi.org/10.2307/2399204

Iganci JR, Soares MV, Guerra E, Morim MP (2015) A preliminary molecular phylogeny of the Abarema alliance (Leguminosae) and implications for taxonomic rearrangement. International Journal of Plant Sciences 177: 34—43. https://doi.org/10.1086/684078

Kass E, Wink M (1997) Molecular phylogeny and phylogeography of Lupinus (Leguminosae) inferred from nucleotide sequences of the rbc L gene and ITS 1 + 2 regions of rDNA. Plant Systematics and Evolution 208: 139-167. https://doi.org/10.1007/BF00985439

Katoh K, Misawa K, Kuma K, Miyata T (2002) MAFFT: A novel method for rapid multiple sequence alignment based on Fourier transform. Nucleic Acids Research 30: 3059-3066. https://doi.org/10.1093/nar/gkt436

Koenen EJM (2022a) On the taxonomic afhnity of Albizia carbonaria Britton (Leguminos- ae, Caesalpinioideae-mimosoid clade). In: Hughes CE, de Queiroz LP, Lewis GP (Eds) Advances in Legume Systematics 14. Classification of Caesalpinioideae Part 1: New gener- ic delimitations. PhytoKeys 205: 363-370. https://doi.org/10.3897/phytokeys.205.82288

Koenen EJM (2022b) Osodendron gen. nov. (Leguminosae, Caesalpinioideae), a new genus of mimosoid legumes of tropical Africa. In: Hughes CE, de Queiroz LP, Lewis GP (Eds) Ad- vances in Legume Systematics 14. Classification of Caesalpinioideae Part 1: New generic delimitations. PhytoKeys 205: 453-470. https://doi.org/10.3897/phytokeys.205.8282 1

Koenen EJM, Kidner C, de Souza E, Simon ME, Iganci JR, Nicholls JA, Brown GK, Queiroz LP de, Luckow M, Lewis GP, Pennington T, Hughes CE (2020) Hybrid capture of 964 nuclear genes resolves evolutionary relationships in the mimosoid legumes and reveals the polytomous origins of a large pantropical radiation. American Journal of Botany 107(12): 1710-1735. https://doi.org/10.1002/ajb2.1568

Lewis GP, Rico Arce L (2005) Tribe Ingeae. In: Lewis G, Schrire B, Mackinder B, Lock M (Eds) Legumes of the World. Royal Botanic Gardens, Kew, Richmond, 193-231.

Lewis GP, Schrire BD (2003) Thailentadopsis Kostermans (Leguminosae: Mimosoideae: Ingeae) resurrected. Kew Bulletin 58(2): 491-494. https://doi.org/10.2307/4120634

Linares L (2005) Especie nueva de Albizia (Leguminosae: Mimosoideae) de Centroamérica. Revista Mexicana de Biodiversidad 76: 7—10. https://doi.org/10.22201/ib.20078706e.2005.001.361

Nielsen IC (1981) Tribe 5. Ingeae. In: Polhill RM, Raven PH (Eds) Advances in Legume Systematics. Part 1, Royal Botanic Gardens, Kew, Richmond, 173-190.

Posada D (2008) jModelTest: Phylogenetic model averaging. Molecular Biology and Evolution 25: 1253-1256. https://doi.org/10.1093/molbev/msn083

Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA (2018) Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Systematic Biology 67(5): 901-904. https://doi. org/10.1093/sysbio/syy032

Rico Arce ML, Gale SL, Maxted N (2008) A taxonomic study of Albizia (Leguminosae: Mimosoideae: Ingeae) in Mexico and Central America. Anales del Jardin Botanico de Madrid 65: 255-305. https://doi.org/10.3989/ajbm.2008.v65.i2.294

Ringelberg JJ, Koenen EJM, Iganci JR, de Queiroz LP, Murphy DJ, Gaudeul M, Bruneau A, Luckow M, Lewis GP, Hughes CE (2022) Phylogenomic analysis of 997 nuclear genes reveals the need for extensive generic re-delimitation in Caesalpinioideae (Leguminosae).

In: Hughes CE, de Queiroz LP, Lewis GP (Eds) Advances in Legume Systematics 14.

400 Gabriela Aviles Peraza et al. / PhytoKeys 205: 371-400 (2022)

Classification of Caesalpinioideae Part 1: New generic delimitations. PhytoKeys 205: 3-58. https://doi.org/10.3897/phytokeys.205.85866

Ronquist E, Teslenko M, van der Mark P, Ayres DL, Darling A, Hohna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61: 539-542. https://doi.org/10.1093/sysbio/sys029

Soares MVB, Guerra E, Morim MP, Iganci JRV (2021) Reinstatement and recircumscription of Jupunba and Punjuba (Fabaceae) based on phylogenetic evidence. Botanical Journal of the Linnean Society 196(4): 456-479. https://doi.org/10.1093/botlinnean/boab007

Soares MVB, Koenen EJM, Iganci JRV, Morim MP (2022) A new generic circumscription of Hydrochorea (Leguminosae, Caesalpinioideae, mimosoid clade) with an amphi-Atlantic dis- tribution. In: Hughes CE, de Queiroz LP, Lewis GP (Eds) Advances in Legume Systematics 14. Classification of Caesalpinioideae Part 1: New generic delimitations. PhytoKeys 205: 401-438. https://doi.org/10.3897/phytokeys.205.82775

Souza ER de, Lewis GP, Forest F, Schnadelbach AS, van den Berg C, Queiroz LP de (2013) Phylogeny of Calliandra (Leguminosae: Mimosoideae) based on nuclear and plastid mo- lecular markers. Taxon 62: 1201-1220. https://doi.org/10.12705/626.2

Souza ER de, Krishnaraj MV, Queiroz LP de (2016) Sanjappa, a new genus in the tribe Ingeae (Leguminosae: Mimosoideae) from India. Rheedea 26: 1-12.

Sun Y, Skinner DZ, Liang GH, Hulbert SH (1994) Phylogenetic analysis of Sorghum and related taxa using internal transcribed spacers of nuclear ribosomal DNA. Theoretical and Applied Genetics 89: 26-36. https://doi.org/10.1007/BF00226978

Terra V, Ringelberg JJ, Maslin B, Koenen EJM, Ebinger J, Seigler D, Hughes CE (2022) Dilemmas in generic delimitation of Senegalia and allies (Caesalpinioideae, mimosoid clade): how to reconcile phylogenomic evidence with morphology and taxonomy? In: Hughes CE, de Queiroz LP, Lewis GP (Eds) Advances in Legume Systematics 14. Clas- sification of Caesalpinioideae Part 1: New generic delimitations. PhytoKeys 205: 261-278. https://doi.org/10.3897/phytokeys.205.79378

Thiers BM (2016) Index Herbariorum: A global directory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium, New York. http://sweetgum.nybg.org/ih/

Turland NJ, Wiersema JH, Barrie FR, Greute W, Hawksworth DL, Herendeen PS, Knapp S, Kusber WH, Li DZ, Marhold K, May TW, McNeill J, Monro AM, Prado J, Price MJ, Smith GF [Eds] (2018) International Code of Nomenclature for algae, fungi, and plants (Shenzhen Code) adopted by the Nineteenth International Botanical Congress Shenzhen, China, July 2017. Regnum Vegetabile 159. Koeltz Botanical Books, Glashiitten. https:// doi.org/10.12705/Code.2018

Villiers JF (2002) Viguieranthus. In: Du Puy D, Labat J-N, Rabevohitra R, Villiers J-F, Bosser J, Moat J (Eds) The Leguminosae of Madagascar. Royal Botanic Gardens, Kew, Richmond, 271-285.