UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO POSGRADO EN CIENCIAS BIOLÓGICAS INSTITUTO DE BIOLOGÍA SISTEMÁTICA DIVERSIFICACIÓN DE ANTIPHYTUM DC. EX MEISN. (BORAGINACEAE): IMPLICACIONES EN LA SISTEMÁTICA DEL GÉNERO TESIS QUE PARA OPTAR POR EL GRADO DE: DOCTORA EN CIENCIAS PRESENTA: NIDIA MENDOZA DÍAZ TUTORA PRINCIPAL DE TESIS: DRA. HILDA FLORES OLVERA INSTITUTO DE BIOLOGÍA, UNAM COMITÉ TUTOR: DRA. HELGA OCHOTERENA BOOTH INSTITUTO DE BIOLOGÍA, UNAM DR. DANIEL PIÑERO DALMAU INSTITUTO DE ECOLOGÍA, UNAM TUTOR INVITADO: DR. MICHAEL J. MOORE SCIENCE CENTER, OBERLIN COLLEGE, OHIO CIUDAD UNIVERSITARIA, CD. MX., SEPTIEMBRE, 2022. UNAM – Dirección General de Bibliotecas Tesis Digitales Restricciones de uso DERECHOS RESERVADOS © PROHIBIDA SU REPRODUCCIÓN TOTAL O PARCIAL Todo el material contenido en esta tesis esta protegido por la Ley Federal del Derecho de Autor (LFDA) de los Estados Unidos Mexicanos (México). El uso de imágenes, fragmentos de videos, y demás material que sea objeto de protección de los derechos de autor, será exclusivamente para fines educativos e informativos y deberá citar la fuente donde la obtuvo mencionando el autor o autores. Cualquier uso distinto como el lucro, reproducción, edición o modificación, será perseguido y sancionado por el respectivo titular de los Derechos de Autor. UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO POSGRADO EN CIENCIAS BIOLÓGICAS INSTITUTO DE BIOLOGÍA SISTEMÁTICA DIVERSIFICACIÓN DE ANTIPHYTUM DC. EX MEISN. (BORAGINACEAE): IMPLICACIONES EN LA SISTEMÁTICA DEL GÉNERO TESIS QUE PARA OPTAR POR EL GRADO DE: DOCTORA EN CIENCIAS PRESENTA: NIDIA MENDOZA DÍAZ TUTORA PRINCIPAL DE TESIS: DRA. HILDA FLORES OLVERA INSTITUTO DE BIOLOGÍA, UNAM COMITÉ TUTOR: DRA. HELGA OCHOTERENA BOOTH INSTITUTO DE BIOLOGÍA, UNAM DR. DANIEL PIÑERO DALMAU INSTITUTO DE ECOLOGÍA, UNAM TUTOR INVITADO: DR. MICHAEL J. MOORE SCIENCE CENTER, OBERLIN COLLEGE, OHIO CIUDAD UNIVERSITARIA, CD. MX., 2022 COORDINACIÓN DEL POSGRADO EN CIENCIAS BIOLÓGICAS INSTITUTO DE BIOLOGÍA OFICIO CPCB/738/2022 ASUNTO: Oficio de Jurado M. en C. Ivonne Ramírez Wence Directora General de Administración Escolar, UNAM P r e s e n t e Me permito informar a usted que en la reunión ordinaria del Comité Académico del Posgrado en Ciencias Biológicas, celebrada el día 20 de junio de 2022 se aprobó el siguiente jurado para el examen de grado de DOCTORA EN CIENCIAS de la estudiante MENDOZA DÍAZ NIDIA con número de cuenta 407043381 con la tesis titulada “DIVERSIFICACIÓN DE ANTIPHYTUM DC. EX MEISN. (BORAGINACEAE): IMPLICACIONES EN LA SISTEMÁTICA DEL GÉNERO”, realizada bajo la dirección de la DRA. MARÍA HILDA FLORES OLVERA, quedando integrado de la siguiente manera: Sin otro particular, me es grato enviarle un cordial saludo. A T E N T A M E N T E “POR MI RAZA HABLARÁ EL ESPÍRITU” Ciudad Universitaria, Cd. Mx., a 16 de agosto de 2022 COORDINADOR DEL PROGRAMA Presidente: DR. ALFONSO OCTAVIO DELGADO SALINAS Vocal: DR. LEONARDO OSVALDO ALVARADO C RDENAS Vocal: DRA. MARTHA JUANA MARTÍNEZ GORDILLO Vocal: DR. ANDRÉS ERNESTO ORTÍZ RODRÍGUEZ Secretario: DRA. HELGA OCHOTERENA BOOTH AGRADECIMIENTOS Agradezco al Posgrado en Ciencias Biológicas de la UNAM la oportunidad de continuar mi formación académica bajo la modalidad de doctorado. Al Consejo Nacional de Ciencia y Tecnología (CONACYT) por el apoyo económico otorgado durante la realización de mi proyecto de doctorado (becario 288698). A mi tutora principal, Dra. Hilda Flores Olvera, por el apoyo y confianza brindados durante este proyecto. A los miembros de mi Comité Tutor, por su guía invaluable y pertinentes observaciones para llevar a cabo el presente trabajo: Dra. Helga Ochoterena Booth, Dr. Daniel Piñero Dalmau y Dr. Michael M. Moore. AGRADECIMIENTOS A TÍTULO PERSONAL Primeramente, a la Dra. Hilda Flores, para quien me son insuficientes las palabras en este intento de agradecer. A la Dra. Helga Ochoterena, quien ha sido gran apoyo y fuente de aprendizaje en mi trayectoria académica. Al Dr. Daniel Piñero, por ser parte de mi Comité Tutor y aportarme valiosos comentarios y palabras que me alentaron más allá de lo académico. Al Dr. Michael J. Moore, por ser parte de este proyecto y enriquecerlo con su presencia, comentarios, visión y sugerencias. A los miembros del jurado, quienes al revisar este escrito mejoraron su contenido: Dra. Helga Ochoterena, Dra. Martha Martínez Gordillo, Dr. Alfonso Delgado, Dr. Leonardo Alvarado y Dr. Andrés Ortíz Rodríguez. Al Dr. Mauricio Bonifacino, por su apoyo en Uruguay y por la fortuna de poder colaborar con él. A la Dra. María del Socorro González Elizondo, por ser la excelente académica y persona que me albergó durante mi estancia en Durango. Su inmenso apoyo me permitió recolectar en Chihuahua, Durango y Zacatecas. Al Dr. Arturo de Nova, quien me recibió amablemente en San Luis Potosí. A Marina Díaz, a quien tuve la inmensa fortuna de conocer en el tiempo y forma precisos, y quien es una de las grandes responsables de los resultados de esta tesis. Me faltan palabras para describir lo maravilloso que ha sido coincidir en esta vida contigo. A Fabián Muñoz, por su amabilidad, aprecio y apoyo para conmigo durante mi estancia en Uruguay y fuera de ahí. A Anderson Santos de Mello, por toda su ayuda, amistad y cariño en Porto Alegre y en el tiempo presente. A los curadores y personal de los herbarios que consulté durante este proyecto y que en persona o a distancia se mostraron solícitos y amables. A quienes me apoyaron en el trabajo de campo y lo hicieron una mejor experiencia por su compañía: Cecilia Martínez, Alberto Ayala, Camilo Pérez, Guillermo Sánchez, Pedro Castillo, Gamaliel Cázeres. A la M. en C. María del Rosario García Peña, por todo el apoyo gestionando los préstamos necesarios y sus respectivos permisos para realizar este proyecto. A la Dra. Lidia Cabrera, por su apoyo en el uso del Laboratorio de Sistemática Molecular de Botánica, IBUNAM. A las M. en C. Laura Márquez y Nelly López, por el gran trabajo de secuenciación que realizan. A la Biól. Susana Guzmán Gómez, por su asistencia y apoyo en el Laboratorio de Microscopía y fotografía de la diversidad II, IBUNAM. A la M. en C. Berenit Mendoza Garfias, por su apoyo para usar los recursos del Laboratorio de Microscopía y fotografía de la diversidad I, IBUNAM. Al Dr. Pedro Mercado, por su apoyo y guía para usar el Laboratorio de Citogenética, IBUNAM. A la M. en C. Leticia Torres Colín, por su invaluable ayuda y guía para realizar la técnica de acetólisis y montaje de polen en el IBUNAM. A todo el personal del IBUNAM, quienes siempre fueron amables y resolvieron todas mis peticiones y dudas. A los doctores Alejandro Torres Montúfar y Alejandro Oceguera, por darse el tiempo de leer mi escrito y aportarme su gran conocimiento del tema. Al Dr. Jim Cohen, por considerarme dentro de quienes trabajan Boraginaceae y enriquecer mi conocimiento sobre Antiphytum. Al Dr. Michael G. Simpson, por su gran conocimiento sobre Boraginaceae que aproveché para este proyecto. Al Posgrado en Ciencias Biológicas de la UNAM y a las personas que hacen funcionar todo el programa, desde lo administrativo hasta lo académico. A Rocío González Acosta, quien es una luz inmensa, y no sólo ayuda y guía en los trámites del posgrado. Al Dr. Armando Rodríguez, por su amabilidad y extraordinaria ayuda. A la UNAM y al Instituto de Biología, que han sido mi hogar durante tantos años. A los grandes amigos y compañeros del IBUNAM: Alberto, Alejandro, Andrés, Angie, Biaanni, Carmen, Ceci, Darinka, Gina, Eunice, Lilí, Luis, María, Meli, Paty. A los basquetboleros de tantos viernes: Alberto, Armando, Felipe, Jonathan, Leonardo, María, Martín, Pedro, Samuel, Toni, Víctor, y a quienes participaron alguna vez de estos maravillosos encuentros. Al equipo de voleibol que hizo tan fabulosos los lunes: Alberto, Alita, Biaanni, Ceci, Darinka, Dr. Cervantes, Fany, Georgetta, Meli, Laura (aracnóloga), Laurita (lauráceas), Omar, Roy, Víctor, ¡y todos! Al grupo de danza Butoh, y a cada persona que ha sido parte de él. A los chicos de alemán y ruso. A todos mis amigos, especialmente a Ale, Claudia, David, Darinka, Diana, Dulce, Iván, María, Miguel, Miri, Paola, Tali, Tere, Tole. A Ta, por toda su valiosa ayuda y conocimiento editorial. A mi familia. A Dino; siempre a ti y para ti. DEDICATORIA A Dino A Lori A Bru y a todos mis niños ÍNDICE RESUMEN ......................................................................................................................................... 1 ABSTRACT ....................................................................................................................................... 3 INTRODUCCIÓN GENERAL ........................................................................................................ 5 I. LA FAMILIA BORAGINACEAE JUSS. .............................................................................................. 5 II. DISTRIBUCIÓN GEOGRÁFICA DE LA FAMILIA BORAGINACEAE EN AMÉRICA .............................. 6 III. LA SUBFAMILIA ECHIOCHILOIDEAE .......................................................................................... 8 IV. EL GÉNERO ANTIPHYTUM ......................................................................................................... 10 OBJETIVOS .................................................................................................................................... 15 METODOLOGÍA GENERAL....................................................................................................... 16 PARTE I: REEVALUACIÓN TAXONÓMICA DE LAS ESPECIES DE ANTIPHYTUM ..... 18 PARTE A. ESPECIES DE NORTEAMÉRICA ..................................................................................... 18 CAPÍTULO I: Morphological and molecular evidence reveals a new species of Antiphytum (Echiochiloideae, Boraginaceae) from Guerrero, Mexico. ........................................................... 18 CAPÍTULO II: A new and unusual endemic species from the Chihuahuan Desert, Mexico: Antiphytum geoffreyi (Boraginaceae, Echiochiloideae). ............................................................... 29 PARTE B. ESPECIES DE SUDAMÉRICA .......................................................................................... 39 CAPÍTULO III: White flowers in South America: a new species of Antiphytum (Echiochiloideae, Boraginaceae) ................................................................................................... 39 CAPÍTULO IV: A taxonomic reevaluation of the South American Myosotis species described by José Arechavaleta .......................................................................................................................... 48 PARTE II: FILOGENIA Y PATRONES DE DIVERSIFICACIÓN DE ANTIPHYTUM ....... 61 CAPÍTULO V: Antiphytum: initial insights about its phylogeny, diversification and biogeographic patterns .................................................................................................................. 61 DISCUSIÓN GENERAL .............................................................................................................. 134 CONCLUSIONES GENERALES ............................................................................................... 140 REFERENCIAS BIBLIOGRÁFICAS ........................................................................................ 145 1 RESUMEN El género americano Antiphytum DC. ex Meisner (Boraginaceae, subfamilia Echiochiloideae) es mayormente diverso en México. La lectotipificación con A. cruciatum preservó el nombre sobre taxones sudamericanos disyuntos a Norteamérica, pero actualmente éstos se encuentran mal entendidos en cuanto a circunscripción taxonómica, número de especies, distribución geográfica y su relación con las especies norteamericanas. En un estudio previo con dos marcadores moleculares (ITS y ndhF–rpl32), Antiphytum cruciatum se recuperó como hermana a algunos representantes norteamericanos. No obstante, ningún análisis filogenético ha incluido un muestreo a nivel de subfamilia, ni ha representado satisfactoriamente a las especies de Antiphytum, particularmente las sudamericanas. Dado este escenario, mi proyecto de tesis se propone circunscribir a Antiphytum y a sus especies, obtener una hipótesis filogenética sobre las relaciones del género dentro de la subfamilia Echiochiloideae y entre las especies de Antiphytum y esclarecer la relación entre los linajes norteamericano y sudamericano. El alcance de estos objetivos se planteó como la base para ubicar el género en las zonas ecológicas de las regiones anfitropicales, postular una direccionalidad en la colonización de América y plantear los posibles escenarios que dieron lugar a la diversificación de Antiphytum. La primera parte de la tesis presenta la reevaluación taxonómica del género con el hallazgo de dos nuevas especies de Norteamérica, que añade una nueva forma a las ginobases de Antiphytum. La clarificación de la diversidad de Sudamérica incluye la publicación de una nueva especie uruguaya de corolas blancas, que rompe la disrupción morfológica en el color de las corolas que se asociaba fuertemente con la geografía; la transferencia y lectotipificación de una especie distribuida en Brasil y Uruguay, y una clave para todas las especies sudamericanas. Todo este estudio taxonómico apoyado en morfología, datos moleculares, trabajo de campo, los ejemplares tipo y la nomenclatura, permitió circunscribir 13 especies en el género, de las cuales nueve se distribuyen en Norteamérica y cuatro en Sudamérica. Este resultado incrementa al doble la diversidad conocida en la zona austral, actualiza el conocimiento del género en Norteamérica y delimita el género en términos morfológicos dentro de Echiochiloideae, principalmente con la combinación de los siguientes caracteres: sufrútices perennes como forma de crecimiento; corola con simetría radial y sobrepasando el cáliz; presencia de apéndices faucales; fruto no encerrado por el cáliz, sino el cáliz extendido y subtendiendo al fruto; forma de la ginobase de plana a triangular; cicatriz en la mitad proximal del eremocarpo, no rebasando la zona media de la cara ventral de éste. La segunda parte de esta tesis presenta una propuesta filogenética de Antiphytum con un muestreo de 12 de las 13 especies reconocidas en la primera parte y la inclusión de terminales de los otros géneros de Echiochiloideae. El árbol consenso obtenido mediante cinco regiones moleculares (ITS, ndhF–rpl32, trnL–F, matK y rps16) recupera a Antiphytum como monofilético y formado por tres 2 clados en una politomia: clado norteamericano (NA) de especies sin roseta basal, clado NA de especies con roseta basal y clado sudamericano (SA), lo que, aunado al hecho de que el grupo hermano, Ogastemma, es africano, genera ambigüedad en la interpretación de direccionalidad de colonización. Sin embargo, dos de los tres posibles escenarios para resolver esta politomia sugieren que las especies norteamericanas no comparten un ancestro común inmediato e implican un origen único y más probable en Norteamérica, en contraste a la necesidad de invocar dos colonizaciones independientes de esta región. Todos los escenarios implican que el subclado con especies con corolas azules es derivado. Según el análisis de los cinco marcadores, hay una diferencia en el número de cambios hacia A. charruasorum (29) con respecto a las otras especies en SA (18), lo que apuntaría a que las especies con flores azules son más recientes. En conjunto esta evidencia sugiere que el patrón anfitropical ocurrió en un único evento de Norteamérica a Sudamérica, probablemente de dispersión a larga distancia, como ha sido el caso para otras Boraginaceae. En cuanto a diversificación, se sugiere que la mayor diversidad en NA se debe a la distribución geográfica de Antiphytum en las zonas áridas de siete diferentes regiones biogeográficas de México. Además, hay una preferencia edáfica a diversos suelos restrictivos, como los yesosos (restringidas o no), calcáreos y de origen volcánico. En cambio, el linaje sudamericano se encuentra espacialmente más restringido y sólo asociado a suelos de origen volcánico o calcáreos. El clado NA sin rosetas presenta el mayor número de especies y es en el que se encuentran todas las preferencias edáficas, pero las modificaciones en la corola y la quilla ventral de los eremocarpos están asociadas a especies que crecen sobre suelos de origen volcánico, tanto en SA como NA. El género Ogastemma se recuperó como monofilético y hermano de Antiphytum en los análisis combinados, como fuera sugerido por estudios previos, aunque las características palinológicas son muy similares entre ambos. Finalmente, Antiphytum se suma a los taxa de Boraginaceae con distribución anfitropical y se provee de una clave taxonómica para todas las especies del género. 3 ABSTRACT The American genus Antiphytum DC. ex Meisner (Boraginaceae, subfamily Echiochiloideae) is mostly diverse in Mexico, but the name is preserved over the South American species after the lectotypification with A. cruciatum. The South American species are disjunct to North American ones, but they are currently misunderstood in terms of taxonomic circumscription, number, geographic distribution and their relationships with the North American taxa. In a previous study with two molecular markers (ITS and ndhF–rpl32), Antiphytum cruciatum was recovered as sister to some North American taxa. However, all the phylogenetic analyses lack a representative subfamily level sampling and do not include the whole of the known Antiphytum species, particularly from South America. Therefore, this study aims to circumscribe Antiphytum and its species; to obtain a phylogenetic hypothesis on the relationship of the genus within Echiochiloideae and among Antiphytum species; and to clarify the relationship between the North American and South American lineages. The attainment of these objectives was proposed as the basis for locating the genus in the ecological zones of amphitropical regions, postulating directionality in the colonization of America, and positing possible scenarios that led to the diversification of Antiphytum. The first part of this thesis presents the taxonomic reevaluation with the finding of two new species from North America that adds a new gynobase shape. Clarification of South American diversity includes the description of a new Uruguayan species with white corollas. This feature breaks the morphological disruption in corolla color that was strongly associated with geography. This part of the study also derived in the transfer and lectotypification of one species from Brazil and Uruguay, and provided a published taxonomic key to all species of this region. All this taxonomic work supported by morphology, molecular data, field work, type specimens and nomenclature, allowed the circumscription of 13 species, of which nine are distributed in North America and four in South America. This result doubles the known diversity in South America, updates the genus knowledge in North America and delimits Antiphytum in morphological terms within Echiochiloideae, mainly by the combination of the following characters: suffruticose perennial as growth-form; corolla radially symmetrical surpassing the calyx; presence of faucal appendages; fruit not enclosed into the calyx, rather an extended calyx subtending the fruit; gynobase shape from flat to triangular; cicatrix in proximal half of eremocarp, not exceeding the middle zone of the ventral face. The second part of this thesis presents a phylogenetic hypothesis of Antiphytum sampling 12 of the 13 recognized species and terminals of the other genera of Echiochiloideae. The consensus tree based on five markers (ITS, ndhF–rpl32, trnL–F, matK and rps16) recovers Antiphytum as monophyletic and formed by three main clades in a polytomy: the North American (NA) clade of species without basal rosette, the NA clade of species with basal 4 rosette, and the South American (SA) clade, which along with the fact that Ogastemma, the sister group, is African, generates ambiguity in directionality interpretation of colonization. However, two of the three possible scenarios for resolving this polytomy suggest that North American species do not share an immediate common ancestor and entail a single, more likely origin in North America, in contrast to the need to invoke two independent colonizations of this region. All scenarios imply that the subclade with blue-corolla species is derived. According to the analysis of the five markers, there is a difference in the number of changes toward A. charruasorum (29) relative to the other species in SA (18), which would point to the blue-corolla species being more recent. Taken together this evidence suggests that the amphitropical pattern occurred in a single event from North to South America, probably by long-distance dispersal, as has been the case for other Boraginaceae. In terms of diversification, it is suggested that NA presents a greater diversity because Antiphytum distributes in the arid zones of seven different biogeographic regions of Mexico. In addition, there is an edaphic preference to diverse restrictive soils such as gypsiferous (restricted or not), calcareous and volcanic soils. In contrast, the South American lineage is spatially more restricted and associated only with soils of volcanic or calcareous origin. The NA clade without rosettes presents the largest number of species and is the one in which all edaphic preferences are found, but modifications in the corolla and ventral keel of eremocarps are associated with species growing on soils of volcanic origin, both in SA and NA. Ogastemma was recovered as monophyletic and sister to Antiphytum in the combined analyses, as suggested by previous studies, although the palynological characteristics are very similar between the two. Finally, Antiphytum joins the Boraginaceae taxa with amphitropical distribution and a taxonomic key is provided for all species of the genus. 5 INTRODUCCIÓN GENERAL I. LA FAMILIA BORAGINACEAE JUSS. El clado Boraginales Juss. ex Bercht. & J.Presl comprende alrededor de 2750 especies de distribución cosmopolita (Luebert et al. 2016), que a su vez se resuelve en dos clados (Refulio- Rodríguez y Olmstead 2014; Weigend et al. 2014; Luebert et al. 2016): Boraginales I, compuesto de las familias Codonaceae Weigend & Hilger, Wellstediaceae Novák y Boraginaceae s. str.; y Boraginales II, con las familias Hydrophyllaceae R.Br., Namaceae Molinari, Heliotropiaceae Schrad., Cordiaceae R.Br. ex Dumort. s.l. y Ehretiaceae Mart. s.l. Boraginales I tiene a África como área ancentral, mientras que para Boraginales II es Sudamérica (Luebert et al. 2017). Dentro del clado Boraginales I, la familia Codonaceae es tempranamente divergente y, al igual que Wellstediaceae, está restringida al continente africano (Luebert et al. 2016, 2017, Weigend et al. 2016). Boraginaceae s. str. (a partir de aquí, citada como Boraginaceae), que antes fue referida como subfamilia Boraginoideae Arn. por De Candolle (1846), Bentham y Hooker (1976), Gürke (1897), Takhtajan (1980, 1997), Cronquist (1981), Johnston (todos sus estudios sobre Boraginaceae, indexados por Hilger y Zippel 2001) y Al-Shehbaz (1991), es subcosmopolita y hermana de Wellstediaceae (Cohen 2014, Weigend et al. 2014; Luebert et al. 2016). Las plantas de Boraginaceae se distinguen por ser, en su mayoría, hierbas con un indumento hirsuto a híspido, tener estilo ginobásico, ovario dos-partido que se divide en cuatro clusas e inflorescencia tipo cima escorpioidea; otras características vegetativas, florales, palinológicas y de los frutos varían entre sus integrantes (Cohen 2014). En las últimas décadas, la clasificación intrafamiliar de Boraginaceae se ha modificado con el descubrimiento, reducción y recircunscripción de sus tribus (Långstrom y Chase 2002; Långstrom y Oxelman 2003; Weigend et al. 2010, Nazaire y Hufford 2012; Weigend et al. 2013; Cohen 2014; Chacón et al. 2016). Chacón et al. (2016) redefinieron la clasificación intrafamiliar de Boraginaceae en tres subfamilias, Echiochiloideae Weigend (≡tribu Echiochileae Långstrom y Chase), Boraginoideae Arn. (integrada por las tribus Lithospermeae Dumort. y Boragineae Rchb.) y Cynoglossoideae Weigend (la más grande y problemática con ocho tribus: Asperugeae Zakirov ex Ovczinnikova; Craniospermeae DC. ex Meisn.; Cynoglosseae W.D.J.Koch; Lasiocaryeae Weigend; Myosotideae Rchb.f.; Omphalodeae Weigend; Rochelieae A.DC.; y Trichodesmeae Zakirov ex Riedl). Mientras las jerarquías más inclusivas continúan siendo ratificadas por los diferentes estudios (Långstrom y Chase 2002; Långstrom y Oxelman 2003; Weigend et al. 2010; Nazaire y 6 Hufford 2012; Weigend et al. 2013, Cohen 2014; Chacón et al. 2016), las relaciones al interior de las tribus y géneros de Boraginaceae no se han resuelto de forma definitiva. Por ejemplo, se han recuperado géneros no monofiléticos, como Cryptantha Lehmann ex G.Don. (Hasenstab-Lehman y Simpson 2012; Mabry y Simpson 2018), Cynoglossum L. (Cohen 2011; Selvi et al. 2011; Weigend et al. 2013; Cohen 2014, 2015; Chacón et al. 2016; Pourghorban et al. 2020), Lithospermum (Cohen y Davis 2009; Weigend et al. 2009), Omphalodes Moench (Weigend et al. 2013, Otero et al. 2014), Plagiobothrys (Hasenstab-Lehman y Simpson, 2012; Guilliams 2015; Simpson et al. 2017a); cuyas complejas relaciones han derivado en cambios de circunscripción y en la descripción de nuevos géneros y especies (Cohen y Davis 2009; Hansestab-Lehman y Simpson 2012; Otero et al. 2014; Cohen 2015; Mabry y Simpson 2018; Simpson et al. 2019; Simpson y Rebman 2021; Rebman y Simpson 2022). En algunos casos, las decisiones taxonómicas son consideradas prematuras por algunos autores [p. ej. la descripción de géneros segregados de Cynoglossum s.l., como Adelinia J.I.Cohen y Andersonglossum J.I.Cohen descritos por Cohen (2015), y Oncaglossum Sutorý descrito por Sutorý (2010); véase Weigend et al. (2016)]. También se han observado contradicciones en la resolución de clados a nivel de las tribus y subtribus de Boraginaceae, dependiendo del muestreo y marcadores empleados (p. ej. Weigend et al. 2013; Cohen, 2014, 2015; Chacón et al. 2016); o la falta de resolución en la base de la filogenia (p. ej. Pourghorban et al. 2020). Por ello, para continuar con el esclarecimiento de los grupos de Boraginaceae, es necesario solucionar los problemas taxonómicos a nivel de los géneros. II. DISTRIBUCIÓN GEOGRÁFICA DE LA FAMILIA BORAGINACEAE EN AMÉRICA La mayoría de los representantes de la familia Boraginaceae se encuentra principalmente en Asia y Europa. A excepción de las tribus Craniospermeae, Lasiocaryeae y Trichodesmeae (Al-Shehbaz 1991; Chacón et al. 2016; Weigend et al. 2016), en América se encuentran unos cuantos taxones de cada una de las tribus principales. La presencia de los géneros de Boraginaceae en América no es igual entre los dos más grandes subcontinentes americanos. Por ejemplo, de los géneros compartidos con el Viejo Mundo, Eritrichium Schrad. ex Gaudin (Rochelieae), Hackelia Opiz (Rochelieae), Lappula Moench (Rochelieae) y Mertensia Roth (Asperugeae) sólo se encuentran en Norteamérica (Nazaire et al. 2014; Weigend et al. 2016); Plagiobothrys s. l. (Cynoglosseae) tiene mayor diversidad en Norteamérica (Amsinckiinae Working Group 2022) y Lithospermum (el único género de la tribu Lithospermeae en América) tiene su centro de diversificación en el suroeste de EUA y México (Cohen y Davis 2012; Cohen 2022). 7 Los géneros endémicos americanos Amsinckia Lehm. (Cynoglosseae), Antiphytum (Echiochiloideae), Cryptantha s.l. (Cynoglosseae), y Pectocarya DC. ex Meisn. (Cynoglosseae) tienen su mayor diversidad en el hemisferio norte del continente (Johnston 1927; Hasenstab- Lehman y Simpson 2012; Guilliams et al. 2013; Mendoza-Díaz 2015). En contraste, otros géneros son endémicos de Norteamérica, como Dasynotus I.M.Johnst. (Cynoglosseae), Mimophytum Greenm. (Omphalodeae) y Oncaglossum (Cynoglosseae) (Johnston 1948; Sutorý 2010; Holstein et al. 2016a); o de Sudamérica, como Moritzia DC. ex Meisn. (Boragineae), Selkirkia Hemsl. (Omphalodeae) y Thaumatocaryon Baill. (Boragineae) (Weigend et al. 2010, 2016; Chacón et al. 2016; Holstein et al. 2016b). La subtribu Moritziinae Weigend (integrada por Moritzia y Thaumatocaryon) es la única representante de la tribu Boragineae en el continente americano (Weigend et al. 2010); sin embargo, el registro fósil indica que durante el Mioceno Tardío y el Plioceno Temprano taxones de Boragineae también se distribuyeron en Norteamérica, y ahí la tribu alcanzó una gran diversidad, abundancia y extensión (Weigend et al. 2010). La distribución en Sudamérica tampoco es homogénea, pues los cerca de 20 géneros (Jørgensen et al. 2014; Zuloaga et al. 2021) y aproximadamente 90 especies (Johnston 1927) de Boraginaceae se distribuyen principalmente en la zona occidental del subcontinente (Argentina, Chile, Colombia, Perú y Venezuela). Únicamente tres géneros (Antiphytum, Moritzia y Thaumatocaryon) se distribuyen en el área comprendida entre Uruguay y el sur de Brasil. Respecto a los géneros compartidos entre Norte y Sudamérica, Boraginaceae presenta 19 ejemplos de distribución anfitropical en América (“American amphitropical disjunct”, AAD por sus siglas en inglés, sin incluir a Antiphytum) y es la segunda familia con mayor número de AAD después de Poaceae, con 51 ejemplos (Simpson et al. 2017b). La relación filogenética dentro de Boraginaceae explica un patrón biogeográfico en el cual Sudamérica fue colonizada a partir de elementos de origen norteamericano (p. ej. Weigend et al. 2010; Cohen y Davis 2012; Hasenstab-Lehman y Simpson 2012; Mabry y Simpson 2018; Otero et al. 2019), es decir que, a nivel de géneros, la dispersión y el establecimiento en el continente fue exclusivamente de norte a sur (Chacón et al. 2017). La presencia de taxones con disyunciones entre Norteamérica y Sudamérica tiene dos principales explicaciones potenciales de acuerdo a Raven (1963): 1) individuos de un linaje ancestral norteamericano o sudamericano atravesaron los Trópicos en uno o más pasos por medio de dispersión a larga distancia, ya sea por aves o grandes mamíferos. Si fue por mamíferos, la dispersión debió ocurrir durante el Gran Intercambio Biótico Americano (GABI por sus siglas en inglés) (Guilliams et al. 2017). 2) Congéneres o coespecíficos de Norteamérica y Sudamérica 8 llegaron a estar aislados tras la fragmentación histórica de un intervalo continuo de distribución que abarcó el área ocupada actualmente por los Trópicos Americanos (llamada hipótesis de la “fragmentación del intervalo” por Guilliams et al. 2017). La evidencia publicada en Boraginaceae sugiere que no hay datos que respalden una distribución ancestral amplia que abarcara toda la zona tropical de América (Chacón et al. 2017; Mabry y Simpson 2018; Otero et al. 2019). Asimismo, los datos publicados hasta ahora foverecen la explicación de dispersión a larga distancia vía aves y no por mamíferos (Chacón et al. 2017; Guilliams et al. 2017). III. LA SUBFAMILIA ECHIOCHILOIDEAE Desde que fuera propuesto como tribu (Långstrom y Chase 2002), el clado de la subfamilia Echiochioideae ha sido recuperado con alto soporte en las filogenias moleculares, con independencia del método aplicado por los autores (Weigend et al. 2010; Nazaire y Hufford 2012; Weigend et al. 2013; Cohen, 2014; Weigend et al. 2014; Mendoza-Díaz 2015; Chacón et al. 2016). La subfamilia Echiochiloideae es hermana del resto de Boraginaceae (Weigend et al. 2010, 2014; Nazaire y Hufford 2012; Weigend et al. 2013; Cohen 2014; Weigend et al. 2014; Chacón et al. 2016) y presenta la menor diversidad dentro de la familia con tres géneros, Antiphytum, Ogastemma Brummitt y Echiochilon Desf., y aproximadamente 30 spp. (Lönn 1999; Långström y Oxelman 2003; Mendoza-Díaz 2015). La relación filogenética al interior de Echiochiloideae se ha recuperado como (Echiochilon, (Ogastemma, Antiphytum)) aunque sin un muestreo significativo del género Antiphytum, pues en muchos trabajos se omite (p.ej. Weigend et al. 2010 y Otero et al. 2014), o sólo se incluyen hasta tres especies, ninguna de Sudamérica (Långström y Chase 2002; Långström y Oxelman 2003; Nazaire y Hufford 2012; Weigend et al. 2013; Cohen 2014; Chacón et al. 2016, 2017; Luebert et al. 2017). También la diversidad de formas de crecimiento en Echiochiloideae es menor en comparación con las otras subfamilias, que incluyen formas desde efímeras a arborescentes (Weigend et al. 2014). La subfamilia Echiochiloideae, al igual que las familias Codonaceae y Wellstediaceae, está compuesta en su mayoría por plantas desérticas sufruticosas/fruticosas (Weigend et al. 2014) que habitan en condiciones adversas o suelos empobrecidos (Lönn 1999; Setubal 2010; Mendoza-Díaz 2015). Los taxones de la subfamilia se distribuyen en zonas áridas y no tienen representantes asiáticos. Antiphytum es el único elemento americano; Ogastemma, el género hermano, se distribuye en las Islas Canarias, el norte de África y la Península Arábiga; a su vez, Echiochilon se encuentra desde 9 el Norte de África hasta Kenia, en la Península Arábiga y en Paquistán e India (Lönn 1999). Echiochiloideae tiene, pues, una disyunción anfiatlántica. Anteriormente, un taxón con distribución anfiatlántica se explicaba por un origen Gondwánico y la ruptura posterior de este continente (Raven y Axelrod 1974), de modo que la extensión del taxón en el Hemisferio Norte era producto de las migraciones desde el sur. Sin embargo, las edades de muchos grupos han resultado más recientes que la separación total de África y Sudamérica (Wilkström et al. 2001; Givnish y Renner 2004) fechada hace ca. 105 Ma. (McLoughlin 2001). En su lugar, el registro fósil y la afinidad de las floras del Hemisferio Norte se ha explicado a través de una flora ancestral denominada boreotropical (Wolfe 1975), es decir, durante el Paleogeno existió un intercambio preferencial entre Norteamérica y Eurasia de plantas tropicales facilitada por la proximidad entre ambas placas y el máximo térmico del Eoceno (Lavin y Luckow 1993). La hipótesis de la flora boreotropical (Lavin y Luckow 1993) predice que si existió un intercambio preferencial durante el Terciario Temprano a través del Atlántico Norte, los grupos de plantas y animales, con Norteamérica como su lugar de diversificación y con registro fósil en la región desde el Terciario, tienen una alta probabilidad de tener a sus grupos hermanos en los Paleotrópicos y grupos derivados en Sudamérica. Varias explicaciones biogeográficas recientes sobre grupos disyuntos han invocado la hipótesis de la flora boreotropical (Lavin y Luckow 1993; Givnish y Renner 2004; Calvillo-Canadell y Cevallos-Ferriz 2005; Smerdmark y Anderberg 2007; Erkens et al. 2009), de tal forma que muchos taxones sudamericanos tuvieron sus ancestros en Norteamérica (Cevallos-Ferriz y Ramírez 2004; Pérez-García et al. 2012). Parte de la hipótesis de la flora boreotropical implica una conexión por medio de puentes intercontinentales transatlánticos que existieron durante el Paleogeno hasta probablemente el Mioceno-Medio (Wolfe 1975; Grønlie 1979; Tiffney 1985; Sanmartín et al. 2001; Brikiatis 2014). Tiffney y Manchester (2001) consideraban a las conexiones directas de tierra como una mejor explicación que invocar varios eventos paralelos y repetidos de dispersión a larga distancia de taxones no relacionados y con una amplia variedad de mecanismos de dispersión. Sin embargo, la conexiones físicas entre América, Groenlandia y Europa llegaron a ser inusuales durante el Oligoceno y el Mioceno, y el puente terrestre noratlántico (NALB, por sus siglas en inglés) ya estaba roto por el Eoceno Tardío (Tiffney y Manchester 2001). Por ello, en los grupos que tienen tiempos de diversificación más recientes que las edades de estos puentes se explica la afinidad anfiatlántica por dispersión a larga distancia (Kadereit y Baldwin 2012). 10 La distribución de los géneros que integran a la subfamilia Echiochiloideae podría encajar dentro de la hipótesis de la flora Boreotropical. Långström y Chase (2002) plantearon como posible explicación de esta distribución un origen en Eurasia, con Antiphytum diseminándose a Norteamérica vía el Puente Thulean. Este puente conectó el oeste de Europa con el este de Norteamérica a través de Groenlandia desde el Paleoceno Tardío hasta el Eoceno Tardío –o incluso más– (Grønlie 1979; Tiffney 1985; Sanmartín et al. 2001). Para Echiochilon y Ogastemma, los mismos autores postularon una diseminación hacia el sur desde Eurasia, debido al enfriamiento que inició en el Eoceno Temprano. La publicación de Långström y Oxelman (2003) mantiene la idea del paso por el puente Thulean, pero explica el patrón actual de distribución de Echiochilon y Ogastemma por medio de una distribución amplia del ancestro probable de estos géneros, seguida de eventos de vicarianza y especiación. La edad de los nodos Echiochiloideae y Antiphytum fechados por Chacón et al. (2017) en 45 Ma y en 9.4 Ma, respectivamente, indican que el arribo de la subfamilia al continente americano fue por dispersión a larga distancia desde el oeste de Europa a Norteamérica cuando las conexiones por el NALB ya se habían interrumpido (Chacón et al. 2017). El evento de dispersión implicó cladogénesis y un cambio en el modo de dispersión de epizoocoria a endozoocoria, de acuerdo a estos mismos autores. Después de los trabajos de Lönn (1999); Långstrom y Chase (2002) y Långstrom y Oxelman (2003), la subfamilia Echiochiloideae no ha sido objeto de investigación reciente a excepción de Mendoza-Díaz (2015). En consecuencia, aunque ha habido esfuerzos de compilar la información taxonómica, morfológica, cromosómica y palinológica que describe a la familia Boraginaceae, en muchos de estos campos la subfamilia Echiochiloideae carece de información (véase Weigend et al. 2016) y en algunos aspectos morfológicos y evolutivos se puede considerar imprecisa (obs. pers.). Este proyecto propone completar la información de la subfamilia, enfocado al único y poco estudiado género americano Antiphytum, a la vez que busca responder a preguntas evolutivas y biogeográficas partiendo de una filogenia del género con marcadores moleculares. IV. EL GÉNERO ANTIPHYTUM Antiphytum es un pequeño género de plantas sufruticosas perennes cuyas especies fueron descritas mayormente en el siglo XIX (Chamisso 1829; Martens y Galeotti 1844; De Candolle 1846; Torrey 1859; Watson 1883; Rose 1890), muchas de ellas bajo distintos géneros como Anchusa L., Eritrichium, Krynitzkia Fisch. & C.A.Mey. y Lithospermum. 11 Hasta que Långström y Chase (2002) propusieron la otrora tribu Echiochileae, el género Antiphytum presentó por mucho tiempo dificultades para ser ubicado dentro de la clasificación tribal de Boraginaceae, al igual que otros géneros “problemáticos” de la familia. Antiphytum figuró dentro de la tribu Cynoglosseae (De Candolle 1846), Lithospermeae (Bentham y Hooker 1876; Baillon 1890; Gürke 1897; Johnston 1923), Eritrichieae Gürke (Johnston 1954) y finalmente, fue incluido dentro de la tribu artificial Trigonotidae Riedl subtribu Antiphytinae Riedl (Riedl 1967, 1968). La historia de Antiphytum inicia con su publicación (Meisner 1840) con cuatro especies, tres sudamericanas y una mexicana correspondiente a una iconografía de la Real Expedición Botánica a la Nueva España. El género había sido propuesto por De Candolle a partir del material iconográfico de Sessé y Mociño, responsables de la Real Expedición. No obstante, la iconografía que formalmente dio origen a Antiphytum no fue la correcta, ya que Meisner (1840) citó a “A. Mexicana, DC. ap. Moc. ic. Fl. Mex. ined.”. En la descripción póstuma de Sessé y Mociño (1887), “Anchusa mexicana Sessé & Moc.” (McVaugh 2000) tiene filotaxia alterna, lo cual no es congruente con la filotaxia opuesta de la diagnosis original de Antiphytum [“Genus jam foliis oppositis (unde nomen) insigne”, Meisner 1840, ii, p. 188; y “folia opposite (unde nomen)”, De Candolle 1846, p. 121]. Contrariamente, las especies sudamericanas sí presentaban filotaxia opuesta. De ellas, sólo “Anchusa cruciata Cham.” y “Anchusa stoechadifolia Cham.” permanecerían subsecuentemente en el género, pues el tercer elemento sudamericano enlistado (“Anchusa oppositifolia Kunth”) es en realidad Plagiobothrys linifolius (Willd. ex Lehm.) I.M.Johnst. (Holstein y Weigend 2016). En el primer tratamiento formal del género, De Candolle (1846) realizó la combinación nomenclatural para las especies sudamericanas enlistadas por Meisner (1840), Antiphytum cruciatum (Cham.) DC., A. stoechadifolium (Cham.) DC. y A. linifolium (HBK) DC., y aclaró que la iconografía correcta de Sessé y Mociño era la correspondiente a la especie Symphytum fruticosum Sessé & Moc. (Sessé y Mociño 1887). McVaugh (2000) la refiere portando el número 288 y en ella se puede observar que De Candolle escribió el nombre “Antiphytum oppositifolium” (McVaugh 2000; Mociño y Sessé 2010), aunque la combinación final fue “Antiphytum Mexicanum DC.” (De Candolle 1846). Actualmente se sabe que la especie mexicana que dio origen al género en realidad corresponde a Euploca mexicana (DC.) M.W.Frohl. & M.W.Chase, Frohlich et al. 2020). De Candolle (1846) también incorporó una nueva especie norteamericana (Antiphytum heliotropioides A. DC.) y otras dos especies de Sudamérica: A. walpersii A.DC., en realidad Plagiobothrys kunthii (Walp.) I.M.Johnst., Johnston (1923), y A. tetraquetum (Cham.) DC., en realidad Thaumatocaryon 12 tetraquetrum (Cham.) I.M. Johnst. (Johnston 1924) en calidad de dudosa, la cual permanecería por mucho tiempo dentro de Antiphytum. Bentham y Hooker (1876) consideraron que el género Antiphytum era heterogéneo y que la iconografía que lo originaba (de acuerdo a Meisner 1840) era inconsistente con la descripción morfológica del mismo. Por ello, restringieron Antiphytum a las especies sudamericanas A. cruciatum, A. stoechadifolium y A. tetraquetum con las que Fresenius (1857) trató el género en su trabajo regional Flora brasiliensis. Con esta decisión, el género por primera vez era dividido de acuerdo a un patrón geográfico. Como consecuencia de esta circunscripción, Gray (1885) incluyó a las especies norteamericanas segregadas de Antiphytum en el subgénero Amblynotus de Krynitzkia junto con otras dos especies no relacionadas y Macbride (1916) creó a su vez el género Amblynotopsis J.F.Macbride para reunir en un grupo natural a las otrora especies norteamericanas de Antiphytum. Al crear este género, Macbride (1916) no las comparó con las especies sudamericanas. Johnston (1923), tras comparar Amblynotopsis con Antiphytum, encontró insuficientes las diferencias morfológicas entre ambos géneros y los unificó bajo Antiphytum, reconociéndolo como un género con especies norteamericanas y sudamericanas de distribución disyunta, como fuera inicialmente. El trabajo de Johnston (1923) es el tratamiento más reciente publicado del género y el único trabajo del autor que trata con la totalidad del mismo. En 1927, Johnston hizo un cambio de circunscripción al excluir finalmente a Thaumatocaryon tetraquetrum (Cham.) I.M. Johnst., la cual había sido heredada desde De Candolle (1846), y reconocer a A. cruciatum y A. stoechadifolium como las únicas especies sudamericanas de Antiphytum (Johnston 1927). La suma de los trabajos en los que se incluye Antiphytum (Johnston 1923, 1924, 1925, 1927, 1957, 1966) dan cuenta de nueve especies, con las especies sudamericanas como las menos entendidas en cuanto a número y circunscripción. Más recientemente, bajo la circunscripción genérica de Johnston, Higgins y Turner (1983) describieron A. hintoniorum, especie norteamericana restringida a suelos yesosos y endémica del municipio de Galeana, en Nuevo León. La historia taxonómica de Antiphytum plantea por igual un problema nomenclatural y uno de circunscripción. Dado el origen errado del género, Johnston (1923) lectotipificó Antiphytum con la especie sudamericana A. cruciatum, pues era parte de la publicación original y correspondía con la diagnosis genérica de Meisner (1840). Johnston (1923) había considerado A. stoechadifolium un sinónimo taxonómico de A. cruciatum, por lo que esta decisión no sólo preservaba el nombre sobre las especies sudamericanas, sino que planteaba el reconocimiento de una única especie en esta 13 región. Aunque posteriormente Johnston (1927) reconocería que no eran coespecíficas, A. steochadifolium es un taxón prácticamente sin colectas fuera del material tipo. Por otro lado, el nombre Amblynotopsis es necesario para las especies norteamericanas si éstas no quedan circunscritas con las especies sudamericanas a las que está asociado el nombre Antiphytum. En algunos trabajos, Johnston (1927, 1957) se cuestionó la naturalidad de Antiphytum, pues consideró que el género aún requería ser revisado, especialmente las especies sudamericanas que él conocía principalmente por descripción (Johnston 1927). Incluso llegó a llamar verdadero Antiphytum a las especies sudamericanas (Johnston 1927) y Amblynotopsis a las especies de México (Johnston 1957), aunque en el último trabajo en que trató con el género, Johnston (1966) lo siguió considerando disyunto. Por el contrario, Brand (1929, 1931) consideró no natural la unión de las especies sudamericanas y norteamericanas de Antiphytum e inválida la lectotipificación de Johnston, por lo que no se justificaba el uso del nombre. El nombre Antiphytum empero es válido y está correctamente aplicado. La circunscripción de Antiphytum ha cambiado a lo largo de su historia taxonómica, con especies entrando y saliendo de acuerdo al concepto del autor que tratara con el género. Circunscrito tal como lo planteara Johnston (1923, 1925, 1927) al excluir a Thaumatocaryon tetraquetrum, el género es un grupo monofilético (Mendoza-Díaz 2015), de distribución disyunta, que según criterios de Simpson et al. (2017b) podría considerarse un grupo “trans-Norteamérica oriental- Sudamérica AAD”, es decir, que aunque presenta un patrón de distribución anfitropical en América, tiene taxones distribuidos en los sistemas montañosos tropicales de Norteamérica y taxones en la zona oriente de Sudamérica. Sin embargo, tal consideración geográfica requiere primero una evaluación taxonómica, tanto a nivel del género como de las especies, valorando tanto la morfología como otras fuentes de evidencia. Particularmente, la circunscripción de las especies sudamericanas requiere un estudio pormenorizado con colectas y observaciones de campo, dada la problemática de acceso al material herborizado y a la pobreza de la información bibliográfica. En el pasado, caracteres presentes en las especies sudamericanas, como la filotaxia opuesta, la inserción totalmente basal de las clusas y el color azul de la corola frente al blanco de las norteamericanas, fueron empleados para proponer la división de Antiphytum. Sin embargo, todos estos caracteres no son disruptivos porque a lo largo del género, los mismos se presentan como una gradación (Mendoza-Díaz 2015). Por otra parte, todas las colectas sudamericanas son determinadas como A. cruciatum, lo cual subestima el número de especies y oscurece la distribución del género en Sudamérica. 14 Para explicar los patrones de distribución y diversificación de Antiphytum se requiere de una filogenia bien sustentada que permita plantear hipótesis biogeográficas y esclarecer las relaciones entre sus especies y grupos hermanos, la cual tenga implicaciones directas en la sistemática del género, como la circunscripción del mismo, la delimitación de especies y el reconocimiento de caracteres morfológicos y moleculares diagnósticos o de agrupación. En el caso particular de Antiphytum, su distribución disyunta en el continente americano, aunada a su distribución amplia y diferencial en México, requiere un enfoque a nivel de subfamilia para explicar la biogeografía histórica del mismo. 15 OBJETIVOS -Circunscribir el género Antiphytum y las especies que lo integran por medio de datos morfológicos, geográficos y moleculares. -Obtener una hipótesis filogenética del género que aborde la relación con los géneros de la subfamilia Echiochiloideae y las relaciones entre las especies de Antiphytum. -Esclarecer la relación entre los linajes norteamericano y sudamericano. -Evaluar la evolución de caracteres morfológicos clave en el género por medio de la filogenia. -Plantear posibles escenarios que han dado lugar a la diversificación de Antiphytum y la direccionalidad en la colonización de América a partir de los datos disponibles. 16 METODOLOGÍA GENERAL -La revisión taxonómica preliminar de Mendoza- Díaz (2015) constituye el antecedente más reciente y directo para esta investigación que se robusteció para la ulterior circunscripción genérica y de especies de Antiphytum. Se llevó a cabo trabajo de herbario con la revisión de ejemplares de ARIZ, ASU, B, BM, CAS, CHAPA, CIIDIR, ENCB, F, FCME, G, HCIB, HUH, IBUG; IEB, K, MEXU, MO, NMC, NY, P, R, RSA-POM, TEX, UAT, UC, y US a partir de préstamos, visita a herbario o imágenes disponibles en línea (AAH, BR, G, G-DC, GH, K, UC, US, y W). -Se realizó trabajo de campo en localidades esenciales para completar el muestreo pertinente para la reconstrucción filogenética, procesamiento de polen y observación de eremocarpos, y para conocer en su medio natural a las poblaciones de taxones poco entendidos o desconocidos del género. El trabajo de campo también se diseñó para corroborar localidades que eran registros únicos o atípicos para algunas especies. Recolectas en Brasil y Uruguay complementaron el trabajo de campo realizado en México. -Se hizo una revisión de caracteres morfológicos y palinológicos diagnósticos del género y de importancia taxonómica en la delimitación de especies de Antiphytum, pero también de Ogastemma: forma de crecimiento; filotaxia; presencia o ausencia de roseta basal o de tricomas glandulares; tirsoides homocládicos o heterocládicos, surgiendo desde la base de la rama o más distalmente; número de ferófilos en cada ramificación dicasial; paraclados opuestos o alternos, formando o no cimas escorpioideas; largo de las cimas escorpioideas; presencia o ausencia de metaxifilos; flores sésiles, subsésiles o pediceladas, bracteadas o ebracteadas; corola rebasando o no el cáliz; tubo de la corola de igual o menor tamaño que los lóbulos de cáliz; presencia o ausencia de fornículos en la garganta de la corola, superficie y color de los fornículos; color de la corola; número de eremocarpos por fruto; fruto envuelto o no por el cáliz acrescente; extensión del cáliz cuando sustenta a los eremocarpos; forma de la ginobase; forma, ornamentación y número de quillas del eremocarpo; quilla ventral formando o no un estípite o una especie de giba; posición de la cicatriz del eremocarpo; tejido de la cicatriz proyectado o no, probablemente un elaiosoma; forma y tamaño de la mónada de polen; ornamentación y características de la exina; número de aberturas en la mónada. Se tomaron medidas de estructuras morfológicas para la descripción taxonómica y se fotografiaron muestras de los eremocarpos de la subfamilia en el Laboratorio de Microscopía y Fotografía de la Biodiversidad (II) del IBUNAM. Fotografías de microscopía electrónica de barrido de algunas muestras de polen procesadas durante el trabajo de laboratorio y algunos eremocarpos fueron tomadas en el Laboratorio de Microscopía Electrónica y Fotografía de la Biodiversidad (I). 17 -Se obtuvieron hipótesis filogenéticas de Antiphytum y géneros hermanos con análisis independientes y concatenados de cinco marcadores moleculares (la región nuclear ITS y cuatro de cloroplasto: ndhF–rpl32, trnL–F, rps16 y matK). La elección de los marcadores se basó en la bibliografía que contiene análisis filogenéticos para el orden Boraginales que buscaban resolver la relación desde niveles de género, tribu, subfamilia y orden (Moore y Jansen 2006; Cohen 2014; Gottschling et al. 2014; Nazaire et al. 2014; Weigend et al. 2014; Chacón et al. 2016). Se dio especial atención a buscar marcadores de evolución adecuada al nivel de género, así como a complementar los datos con secuencias depositadas en GenBank. El muestreoincluyó al menos un terminal para el género monotípico Ogastemma y algunas de las especies de Echiochilon (E. collenettei I.M.Johnst., E. fruticosum Desf., E. johnstonii Cufod., E. kotschyi (Boiss. & Hohen.) I.M.Johnst., E. longiflorum Benth. o E. persicum (Burm.f.) I.M.Johnst., dependiendo del marcador y la preservación del terminal en la matriz combinada), y dos o más individuos por especie de Antiphytum. De las 13 especies circunscritas, la sudamericana A. stoechadifolium no fue incluida en el análisis filogenético, ya que no se pudo extraer ADN del escaso material herborizado disponible. El muestreo por especie se redujo a uno enlos casos donde sólo se pudo amplificar un individuo para el marcador (matK: A. peninsulare; ndhF–rpl32: A. charruasorum, A. berroi, A. cruciatum, A. geoffreyi y A. hintoniorum). La extracción de ADN y los protocolos de amplificación de las regiones genómicas se realizaron en el Laboratorio de Biología Molecular del Instituto de Biología, UNAM. Los productos de PCR fueron limpiados y secuenciados en LaNABio (IBUNAM). Dada la interdependencia de los métodos mencionados, el trabajo fue recursivo y se retroalimentó en todo momento. Por ello, los datos moleculares fueron empleados en dos contextos: en la primera parte de reevaluación taxonómica, como otra línea de evidencia con un muestreo dirigido para el reconocimiento de una nueva especie y en la segunda parte, para obtener una hipótesis filogenética de las especies de Antiphytum en su contexto dentro de la subfamilia Echiochiloideae. La morfología y datos geográficos y ambientales permitieron la toma de decisiones taxonómicas en la primera parte de la tesis y la discusión de los patrones obtenidos en la segunda. Mayor detalle de los métodos se da en los capítulos correspondientes de la tesis. 18 PARTE I: REEVALUACIÓN TAXONÓMICA DE LAS ESPECIES DE ANTIPHYTUM PARTE A. ESPECIES DE NORTEAMÉRICA CAPÍTULO I: Morphological and molecular evidence reveals a new species of Antiphytum (Echiochiloideae, Boraginaceae) from Guerrero, Mexico. Nidia Mendoza-Díaz, Helga Ochoterena, Michael J. Moore & Hilda Flores-Olvera Systematic Botany (2020) 45(1): 190–199. Systematic Botany (2020), 45(1): pp. 190–199 © Copyright 2020 by the American Society of Plant Taxonomists DOI 10.1600/036364420X15801369352469 Date of publication February 25, 2020 Molecular and Morphological Evidence Reveals a New Species of Antiphytum (Echiochiloideae, Boraginaceae) from Guerrero, Mexico Nidia Mendoza-Dı́az,1 Helga Ochoterena,2 Michael J. Moore,3 and Hilda Flores-Olvera2,4 1Posgrado en Ciencias Biológicas, Instituto de Biologı́a. Universidad Nacional Autónoma de México. 3er. Circuito Exterior s/n, C.P. 04510 Ciudad de México, México 2Departamento de Botánica, Instituto de Biologı́a. Universidad Nacional Autónoma de México. Apartado Postal 70-233, C.P. 04510 Ciudad de México, México 3Department of Biology, Oberlin College, Oberlin, Ohio 44074, USA 4Author for correspondence (mahilda@ib.unam.mx) Communicating Editor: Jeffery Saarela Abstract—Molecular and morphological evidence supports a new species in the genus Antiphytum from the Sierra Madre del Sur, in the state of Guerrero, Mexico, here described asAntiphytum brevicalyx. This species is unique in the genus by possessing a calyx shorter than the corolla tube at anthesis; it is similar to A. floribundum in inflorescence arrangement, but differs from that species in lacking a basal leaf rosette and having ap- pendages on the corolla throat. According to phylogenetic analysis of nuclear ITS and plastid ndhF-rpl32 sequences representing seven of the eight North American species of the genus besides the proposed new species, accessions of A. brevicalyx form a well-defined clade within Antiphytum, corroborating its distinctive morphology. However, the analyses do not resolve the phylogenetic position of A. brevicalyx within the genus. Keywords—Mexican flora, endemism, gypsovag, vulnerable. Antiphytum DC. ex Meisn. (Boraginaceae) is the only American genus of the subfamily Echiochiloideae Weigend (Chacón et al. 2016). According to Johnston (1923, 1927) and Mendoza-Dı́az (2015), the genus includes 10–12 species of erect or prostrate subshrubs with sessile leaves, the cauline ones opposite throughout or only at the base. Some species of Antiphytum possess a true basal leaf rosette whereas others have no basal rosette, but each new branch possesses a pseudobasal rosette (terminology according to Cohen 2011). The inflorescence unit is a scorpioid cyme (but solitary flowers occur inA. hintoniorum L.C.Higgins & B.L.Turner) arranged as paraclades of heterocladic or homocladic thyrsoids (termi- nology according to Weigend et al. 2016). The flowers are perfect, pentamerous, actinomorphic, and gamopetalous, and the corolla is infundibuliform or hypocrateriform, white or blue, and may or may not possess faucal appendages. The gynobase is flat to short-pyramidal bearing lustrous, ovoid or triangular eremocarps (sensu Hilger 2014, also known as mericarps or nutlets), with the cicatrix (attachment scar) in the medial to supra-basal position in the North American species, but in a basal position in the South American taxa. Almost all of the species in the genus were described in the 19th century (Chamisso de Boncourt 1829; Martens and Galeotti 1844; de Candolle and de Candolle 1846; Torrey 1859; Watson 1883; Rose 1890), although two unusual species from northern Mexico were described more recently. Higgins and Turner (1983) described the atypical gypsum endemic A. hintoniorum and Mendoza-Dı́az et al. (2018) described A. geoffreyi N. Mend. & Flores Olv., which has an unusual gynobase shape. During ongoing phylogenetic study of the genus, a new species from the state ofGuerrero,Mexico,was found,which is here described. Materials and Methods Plant Material—As part of an ongoing revision of the genus Anti- phytum, specimens were borrowed and/or examined from AAH, ARIZ, ASU, B, BM, BR, CAS, CHAPA, CIIDIR, ENCB, FCME, G, G-DC, GH, HCIB, HUH, IEB, K, MEXU, MO, NMC, NY, P, R, RSA-POM, TEX, UAT, UC, US, and W (acronyms according to Thiers 2019). Several specimens were encountered thatwere assigned toAntiphytum based on the position of the eremocarp cicatrix and the shape of the gynobase, but do not correspond to any known species of the genus. These specimens, all from the Mexican state of Guerrero, are unique within the genus by possessing a calyx shorter than the corolla tube at anthesis, a medial position of the cicatrix on the ventral face of the eremocarp, and het- erocladic thyrsoid inflorescences with alternate hypopodia (following Weberling 1989). Based on these unusual specimens, field work was carried out near Xochipala, on the road to Filo de Caballos, Guerrero, where specimens were collected and leaf samples dried in silica gel for DNA studies. The hypothesized new species was tested with cladistic analyses using mo- lecular data. Fruit Observation—Mature fruits were studied with a confocal stereo microscope (Leica Z16 APO A, Leica Microsystems GmbH, Wetzlar, Germany), and photographed with a camera (Leica DFC490, Leica Microsystems GmbH) at the Laboratorio de Microscopı́a y Fotografı́a de la Biodiversidad (II) of the Instituto de Biologı́a, UNAM (IBUNAM). Ere- mocarps were also studied using a scanning electron microscope (Hitachi SU1510, Hitachi Ltd., Tokyo, Japan) in the Laboratorio de Microscopı́a Electrónica y Fotografı́a de la Biodiversidad (I) of IBUNAM. To prepare samples for electron microscopy, eremocarps were mounted on stubs with common double stick carbonated tape, and coated with gold to 20 mÅ for two minutes in a sputter coater (Quorum Q150OR ES, Quorum Tech- nologies Ltd, East Sussex, UK). DNAExtraction, PCRAmplification, and Sequencing—Total DNAwas isolated from herbarium specimens or fresh leaves dried in silica gel for 53 accessions representing seven North American species of Antiphytum and seven accessions of the proposed new species. DNAwas isolated using the GE Healthcare Nucleon PhytoPure kit (#RPN8511, Sigma-Aldrich, St. Louis, Missouri) or using the CTAB extraction method of Doyle and Doyle (1987) modified by the addition of 2% of polyvinylpyrrolidone (PVP) to the extraction buffer. Sequences of the nuclear internal transcribed spacer (ITS) and the plastid ndhF-rpl32 spacer were amplified using the primer pairs ITS4 and ITS5 (White et al. 1990) and the newly designed primers rpl32.4R and ndhF.40R.Bor (GAAGTACGYTTTTTTGGAACTGC; ACATAGGAACT- GTWAGTGGAAG), respectively. PCR conditions were as follows: DNA was amplified using GoTaq® Flexi DNAPolymerase (#9PIM829, Promega, Madison, Wisconsin) combined with FailSafe™ PCR 2X PreMix E (Epi- centre, Illumina, Madison, Wisconsin), or GoTaq® Colorless MasterMix (#9PIM713, Promega), adapting the manufacturer’s protocol for 12.5 ml reaction volumes (0.1 ml GoTaq plus 6.25 ml PreMix and 5.15 ml double distilled water, or 6.25 ml GoTaq and 5.25 ml nuclease free water), adding 0.25 ml of each primer (10 mM concentration) and 0.5 ml DNA template of varying concentrations. PCRs were run in a PTC-200 DNA Engine® thermocycler (Bio-Rad, Hercules, California) under the following condi- tions for both loci: initial denaturation at 94°C for 3 min, followed by 35 cycles at 94°C for 50 sec, 50°C for 50 sec, and 72°C for 1min 20 sec,with a final extension at 72°C for 7 min. PCR products were purified using ExoSAP-IT (USBCorp., Cleveland, Ohio) prior to sequencing atNeogenomics (Houston, Texas) or were sent unpurified to LaNABio (IBUNAM) for sequencing. At 190 Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) both locations, sequences were generated using an ABI 3730xL sequencer (Applied Biosystems, Waltham, Massachusetts). Sequence chromatograms were edited, assembled, and manually aligned with PhyDE 0.9971 (Müller et al. 2010); SSRs and polynucleotide regions were annotated in both individual marker matrices and the cor- responding aligned positions were used to label the base pair number(s) in the motif codifications (see below). Echium vulgare L. was included as an outgroup, usingGenBank sequences for bothmarkers fromLångström and Oxelman (2003) and Cohen and Davis (2009) (Appendix 1); no other close outgroup was available, but the inclusion of this single species suffices to test phylogenetic evidence for the proposed new species. Matrices for both loci were imported to WinClada 1.00.24 (Nixon 1999–2002) and merged to create a concatenated matrix. For descriptive comparative purposes, distances among the concate- nated sequences of the Antiphytum species were calculated as absolute nucleotide differences using PAUP 4.0a165 (Swofford 2003) (Table S1 in Supplementary Material 1; Mendoza-Dı́az et al. 2019). Phylogenetic Analysis—Gaps were coded using simple indel coding according to Simmons and Ochoterena (2000) and microstructural char- acters were coded according to Ochoterena (2009); both were added to the nucleotide concatenated matrix to build the final matrix for the analysis. The concatenated matrix was analyzed in WinClada to remove non- informative characters prior to analysis. Phylogenetic analyses were performed with the parsimony criterion using TNT 1.1 (Goloboff et al. 2000) alternating the four new technology search (NTS) strategies: sectorial searches (Goloboff 1999), ratchet (Nixon 1999), tree drifting, and tree fusing (Goloboff 1999) until the consensus was stabilized three times by a factor of 100. The most parsimonious trees were saved and analyzed in WinClada. Ambiguously supported nodes were collapsed and identical trees removed. Unambiguous character transfor- mations were plotted onto the tree using the Fitch criterion only for the potentially informative characters. Jackknife (JK) support values were calculated with TNT using NTS (excepting sectorial searches) using a 36% of character removal probability and 10,000 replicates, until the minimum length was found four times. The frequency of appearance for each node was calculated among these trees and mapped onto the consensus tree using WinClada. We considered support to be lacking when the values were below 64% (Farris et al. 1996), moderate support 65–80%, and strong support above 80%. The resulting tree was saved as an EMF file and edited with Corel Draw X5, in which microstructural characters weremanually colored and labeled according to the optimization in WinClada. The same procedure and analyses were also conducted for each locus separately (Figs. S1, S2 in Supplementary Material 1, Mendoza-Dı́az et al. 2019). Single-locus aligned matrices with labeled microstructural charac- ters (PhyDE format) and the final concatenated matrix (nucleotide and microstructural characters in nexus format) are available in the Dryad Digital Repository (Mendoza-Dı́az et al. 2019). The list of microstructural characters is also available as Supplementary Material 2 (Mendoza-Dı́az et al. 2019). Distribution Map—Maps were produced using QGIS Brighton (2.6.1) employing layers from INEGI (2017) and CONABIO (1997, 1998). The distribution of the species was obtained from specimen labels, using geographic coordinates when available. Specimens without coordinates were georeferenced using Google Earth 7.1.8.3036 (32 bit) (see paratypes). Results Sequence variation among individuals of the putative new species Antiphytum brevicalyx (average nucleotide differences between sequences5 1.3; Table S1) was similar to that in other species with restricted distributions (average 5 0.99) vs. species with broad distributions (average 5 7.6). The average sequence difference of A. brevicalyx compared to other species of Antiphytumwas similar to the average sequence differences among all species pairs in the genus (28.24 vs. 28.17). Average sequence divergence betweenA. brevicalyx and other species of Antiphytum was highest for A. caespitosum I.M.Johnston and A. paniculatum I.M.Johnston (36.01 and 35.06, respectively) and lowest (19.1) for A. peninsulare (Rose) I.M.Johnston (Table S1). The analyses of the concatenated matrix resulted in a single most parsimonious tree (two before collapsing ambiguously supported nodes) of 172 steps (Fig. 1A, where length5 L). The non-rosette forming species A. hintoniorum was sister to all remaining Antiphytum, which were recovered in two main clades, one of which was composed of species bearing a basal leaf rosette (Fig. 1A). All accessions corresponding to A. brevicalyx were retrieved in a clade with strong support (JK 99%) within the clade of species without a basal leaf rosette. Antiphytum peninsulare of Baja California was sister to the new species clade although without jackknife support (Fig. 1A). This was also the case for the ITS analysis (Fig. S1), but not for the chloroplast analysis, in which all the accessions of A. brevicalyx were paraphyletic to a clade including all the accessions of A. heliotropioides A.DC. (Fig. S2). Little or no geographic structure was found within the new species (Fig. 1B). Taxonomic Treatment AntiphytumbrevicalyxN.Mend., FloresOlv. &H. Ochoterena, sp. nov. TYPE: MEXICO. Guerrero: Municipio Eduardo Neri, 9 km (on the road to Filo de Caballos) west of Xochipala, 1635 m, 17°46041.40N, 99°46041.40W, 18 No- vember 2016, N. Mendoza-Dı́az 643 with H. Flores Olvera, H. Ochoterena, A. Torres-Montúfar & C. Martı́nez-Rojas (Holotype: MEXU! Isotypes: ENCB!, FCME!, K!, MEXU!, MO!). Antiphytum brevicalyx is the only species in the genus with the calyx shorter than the corolla tube at anthesis. It is similar to A. floribundum (Torr.) A.Gray in the inflorescence arrange- ment, but differs by the absence of a basal rosette (vs. present in A. floribundum) and the presence of appendages on the corolla throat (vs. absent in A. floribundum). Compared to A. peninsulare, the eremocarp of A. brevicalyx shares a medial position of the cicatrix as well as tuberculate and papillate ornamentation, but differs in shape (ovate in A. brevicalyx vs. triangular in A. peninsulare). Suffruticose perennial, erect, up to 1.20 m tall, covered by strigose and hispid indumenta, each trichome with several mineralized foot cells (lithocystic structure), with amber- colored and non-persistent glandular-capitate trichomes mainly on the younger parts of branches, leaves, bracts, bracteoles, and calyx. Stems numerous and broadly branched, arising from a main stem, branches alternate or opposite, non- fistulose (solid), slender (ca. 2.0 mm diam), with the most external layer exfoliating. Leaves basal, pseudobasal and cauline, sessile, blades linear-oblanceolate to oblanceolate, and lanceolate to narrowly elliptical-acuminate, margins flat or slightly thickened, entire, apices acute, nearly mucronate, bases truncate and non-connate, glandular trichomes only on the margins at the very base, surfaces concolorous, midvein visible as a groove in the adaxial face and as raised-strigose on the abaxial, secondary veins often not visible; basal leaves marcescent, not forming a rosette, up to 70.0(–80.0) 3 10.0(–15.0) mm; cauline leaves mostly alternate and only opposite at the base of the stem or the branches, 40.0–50.0(–60.0) 3 2.0–4.0(–5.0) mm, gradually smaller up- ward along the stem; pseudobasal leaves at the beginning of each branch, the newest ones 10.0 3 2.0 mm. Inflorescence ending in a flower (monotelic) with alternate paraclades, forming a terminal heterocladic determinate thyrse (thyrsoid) with dichasial branching, one side ending in a scorpioid cyme and the other side branching up to several times, first inter- node of a lateral axis (hypopodium) 40.0–60.0(–70.0) mm in length; mature scorpioid cymes up to 90 mm in length, loosely MENDOZA-DÍAZ ET AL: NEW SPECIES OF ANTIPHYTUM FROM GUERRERO, MEXICO 1912020] Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) Fig. 1. Combined parsimony tree (L 5 172) using ITS and ndhF-rpl32 sequences, including microstructural DNA characters. A. Single most parsi- monious tree highlighting Antiphytum brevicalyx. Sequences in gray are identical to the sequence of the first black-colored accession name above. Numbers above branches correspond to jackknife support above 64%. Solid boxes on branches indicate synapomorphic changes, open boxes indicate homoplasy, and colored boxes correspond to microstructural characters, as follows: blue boxes represent gap characters, and pink boxes represent simple sequence repeats (SSRs). Numbers next to the microstructural synapomorphies indicate their base pair position(s) in the aligned matrix of the respective locus. A schematic representation of the eremocarp cicatrix position is shown next to the hypothetical ancestor of each condition: SB 5 supra-basal; M 5 medial; IM 5 infra- medial. B. Geographic origin of A. brevicalyx accessions from Guerrero, Mexico. Mapped collections that are not connected to the phylogeny lack sequences. C. Flowers of A. brevicalyx showing the characteristic calyx: shorter than the corolla tube. D. A. caespitosum, the other species distributed in Guerrero, showing the basal leaf rosette characteristic of the entire clade (dotted box). SYSTEMATIC BOTANY [Volume 45192 Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) many-flowered, bracteate; floral bracts obovate (2:1) to oblong, sessile, mucronate, indumenta and midvein as on the leaves, blades 7.0–11.0 3 3.2–4.0 mm, gradually smaller upward along the cyme. Flowers perfect, sessile or subsessile (pedicel 0.2–0.5 mm in length); calyx strigose, glandular-capitate and scattered hispid trichomes on abaxial surface, accrescent in fruit, divided almost to the base, lobes asymmetric, mucronate to acuminate, scarious at the very base, two of them lanceolate Fig. 2. Antiphytum brevicalyx. A. General appearance of a young plant showing detail of the abaxial surface of a basal leaf (from Mendoza-Dı́az et al. 644C). B. General appearance of an older individual showing detail of the adaxial surface of the cauline leaves (fromMendoza-Dı́az et al. 643A). C. Portion of cyme inflorescence (fromMendoza-Dı́az et al. 644A). D. Flower (fromMendoza-Dı́az et al. 644A). E. Open corolla (fromMendoza-Dı́az et al. 644A). F. Fruit with three eremocarps removed showing the attachment point at the gynobase (from Mendoza-Dı́az et al. 643A). G. Eremocarp in ventral and lateral view (from Mendoza-Dı́az et al. 643A). MENDOZA-DÍAZ ET AL: NEW SPECIES OF ANTIPHYTUM FROM GUERRERO, MEXICO 1932020] Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) and three obovate, ca. 2.0–2.2 3 0.3–0.4 mm, the bigger one obovate and resembling a floral bract, 2.53 1.0mm, all shorter than the corolla tube at anthesis, up to 4.3mm in length in fruit; corolla hypocrateriform (rotate), white, with five yellow pa- pillose faucal appendages on the throat opposite to the corolla lobes, tube widely surpassing the calyx, 2.1–3.3 mm in length, 1.3 mm diam at the base, limb (4.0–) 4.25–5.25 mm diam, lobes oriented nearly perpendicular to the tube, imbricate, rounded, sinuate at apex, ca. (1.0–) 1.2–1.4 3 (1.0–) 1.2–1.4 mm, abaxially strigose at the middle; stamens Fig. 3. Antiphytum brevicalyx in the field (Mendoza-Dı́az et al. 643, 644). A. Young individual with flowers and associated vegetation. B. Mature individual. C. Basal leaves in a young individual. D. Individual showing the habitat and soil. E. Thyrsoid view from above. F. Partial inflorescence. G. Portion of inflorescence. H. Flower at anthesis showing the faucal appendages. Photos by Helga Ochoterena. SYSTEMATIC BOTANY [Volume 45194 Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) inserted, alternate to the corolla lobes, adnate at ca. 1.7 mm from the base of the corolla tube; filaments up to 0.5 mm in length, slender; anthers oblong, slightly lanceolate, dorsi- fixed, ca. 0.8–0.93 0.2–0.3 mm; ovary with four ovules, style gynobasic, 0.7–0.8mm in length at anthesis, stigma bilobed at the apex of the style (terminal). Fruit dry, covered by a persistent calyx, usually with 2–3 mature eremocarps; ere- mocarps lustrous, ovate and apically acuminate in ventral view, 1.8–1.9 3 1.3 mm, dorsally convex, tuberculate and papillate except at the apex of the tubercles; cicatrix at medial position on the ventral face, ovate, ca. 0.65 mm in length; one ventral keel (apical) straight from the apex of the eremocarp to the cicatrix apex not forming a stipe, without lateral (basal) keels; gynobase deltate-pyramidal, with four areoles corre- sponding to the cicatrix of each eremocarp, style plus stigmas 1.7(–2.0) mm, surpassing the eremocarps. Figures 2, 3. Fig. 4. Distribution map of A. brevicalyx within Guerrero, showing the two municipios (Eduardo Neri and Tixtla de Guerrero) in which the species has been recorded, and vegetation types. The inset map shows the locations of the two municipios within Guererro and the location of Guerrero in Mexico. Table 1. Main morphological characters that distinguish Antiphytum brevicalyx from morphologically similar species of the genus. The diagnostic feature of the new species is highlighted in bold. A. brevicalyx A. floribundum A. heliotropioides A. peninsulare Glandular-capitate trichomes Present on vegetative parts Absent Absent Present on vegetative parts Basal leaf rosette Absent Present Absent Absent Main axis of the inflorescence with alternate hypopodia Present Present Absent Absent Corolla diameter (mm) 4.0–5.25 5.0–7.0 4.0–6.0 10.0–12.0 Relationship of calyx/corolla tube length at anthesis All calyx lobes shorter Calyx lobes equal to longer Calyx lobes equal to longer Calyx lobes equal to longer Faucal appendages Present Absent Present Present Eremocarp shape Ovate Ovate Ovate Triangular Eremocarp ornamentation Tuberculate and papillate Tuberculate Tuberculate and papillate Tuberculate and papillate Eremocarp cicatrix position Medial Supra-basal Infra-medial Medial MENDOZA-DÍAZ ET AL: NEW SPECIES OF ANTIPHYTUM FROM GUERRERO, MEXICO 1952020] Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) Etymology—The specific epithet refers to the main mor- phological character that distinguishesA. brevicalyx from other Antiphytum; it is the only taxon known in the genus in which the calyx is shorter than the corolla tube at anthesis (Figs. 1C, 2C–D, 3F–G). Phenology—Flowering and fruiting from May to November. Distribution and Habitat—Antiphytum brevicalyx is only known from the central region of Guerrero, in the municipios of Eduardo Neri and Tixtla de Guerrero (Fig. 4). It occurs at Fig. 5. Comparison of A. brevicalyx to the morphologically most similar species of Antiphytum. A. Distribution. B–D. Comparison among trichomes; note the capitate- glandular trichomes in C and D. E–G. Eremocarps; note the medial position of the cicatrix in F and G. H–J. Pericarp ornamentation of tubercles and papillae. B, E, H: A. heliotropioides (B, E, Mendoza-Dı́az et al. 629, H, Torres-Hernández & Hernández 802). C, F, I: A. brevicalyx (Mendoza-Dı́az et al. 643A). D, G, J: A. peninsulare (Wiggins 17328). SYSTEMATIC BOTANY [Volume 45196 Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) elevations ranging from 1080 to 1900m, on limestone soils and probably also on gypsum, in deciduous forest with Brahea, Bursera, Juniperus, and Quercus. Conservation Status—Antiphytum brevicalyx is known only from seven subpopulations in the state of Guerrero, Mexico (Fig. 4). As a whole, the extent of occurrence is less than 261 km2 and some subpopulations occur among areas of rainy season agriculture (Fig. 4). Hence, habitat loss is the principal threat for the subpopulations. According to the IUCN (2012), this species meets the criteria B1ab(iii) for the category of Vulnerable (VU). Paratypes—Mexico.—GUERRERO: Municipio Eduardo Neri, 9 km (on the road to Filo de Caballos) west of Xochipala, 1635 m, 17°46041.40N, 99°46041.40W, 18 November 2016,Mendoza-Dı́az et al. 644 (MEXU!, FCME!, ENCB!, K!); La Yesera, 12.3 km N de Zumpango del Rı́o, 1420 m, 17°420230N, 99°300040W, 25 October 1994, Jiménez-Ramı́rez & Luna-Flores 971 (FCME!); cerca de carretera, desv. a Xochipala, 1080 m, [17°47027.70N, 99°370080W,62 km], 26 October 1990, Peralta &Villegas 127 (FCME!); 14 km SW de Xochipala, carretera a Filo de Caballos, 1900 m, [17°45052.60N, 99°42017.20W, 62 km], 17 October 1983, Soto-Nú~nez 5626 (MEXU!); along the road to Filo de Caballo, just above Xochipala at km 22, below La Laguna, 1600 m, 17°480N, 99°360W, 21 May 1987, Miller & Goreti 2826 (MEXU!, MO!); 7 km O Carrizalillo, 1600 m, 17°510340N, 99°400170W, 09 September 1993, Martı́nez-Gordillo 798 (FCME!, MEXU!). Municipio Tixtla de Guerrero, Summit of mountains northeast of Chilpancingo on road to Chilapa, 6000 ft, [17°340140N, 99°26025.10W,62 km], 19August 1948,Moore Jr. & Wood, Jr. 4672 (GH!). AdditionalSpecimensExamined for theMorphologicalStudy—Antiphytum caespitosum.Mexico.—GUERRERO: Municipio Tlapa, on El Salado bridge, 8 km NofTlapa, road toHuamuxtitlan, 990m, 16November 1982,Mart́ınez et al. 2680 (MEXU!). Antiphytum heliotropioides. Mexico.—NUEVO LEÓN: Municipio Rayones, road Galeana-Rayones, 6 km N of Los Nogales and 17 km N of Galeana, 1360 m, 24°55043.70N, 100°03051.10W, 24 September 2016, Mendoza- Dı́az et al. 629 (MEXU!). Antiphytum peninsulare. Mexico.—BAJA CALIFORNIA SUR: Gulf of California, volcanic hillside near northwest end of Isla SanMarcos, 27°150N, 112°070W, 28 March 1962, Wiggins 17328 (MEXU!). Discussion Antiphytum has a preference for impoverished soils, such as limestone and gypsum in North America. Except for A. hintoniorum, the only gypsum endemic species of the genus, and A. floribundum, which is not known to occur on gypsum (Mendoza-Dı́az 2015), all remaining species grow on and off gypsum (such species are called gypsovags; Moore et al. 2014). Althoughwe only observed populations growing in calcareous soils, A. brevicalyx also seems to be a gypsovag, because the collection Jiménez-Ramı́rez & Luna-Flores 971 (FCME) grew on gypsum soils at “La Yesera” (Spanish for “gypsum place”). Before the description of A. brevicalyx, only A. caespitosumwas known from Guerrero, near the boundary with the state of Oaxaca. Thediscovery of this newspecieswaspossible due to the systematic floristic exploration of the state of Guerrero, which began in 1979 with the Flora of Guerrero project at Facultad de Ciencias, UNAM (Flores-Olvera et al. 2010). Earlier collections in the state,mainly gatheredby foreignbotanists, are also important for the knowledge of the flora of Guerrero, and such collections are also known for the new species (see paratypes). Although A. brevicalyx and A. caespitosum are both distrib- uted in Guerrero, they are allopatric, morphologically distinct, and not phylogenetically sister. Moreover, A. brevicalyx lacks a basal leaf rosette and reaches 1.2 m in height, whereas A. caespitosum reaches a height of less than 1 m. Morphological features of the inflorescence of A. brevicalyx aremost similar toA. floribundum, fromwhich it differs by the presence of appendages on the corolla throat and the lack of leaves arranged in a basal rosette (Table 1). In addition, A. floribundum is distributed in western Mexico, principally along the Sierra Madre Occidental. A sister relationship between A. brevicalyx and A. peninsulare was also supported by morphological characters (Table 1; Figs. 1, 5). Detailed observations show similarities in the presence of capitate glandular trichomes, the medial position of the eremocarp cicatrix and the ornamentation of the fruit, which is also shared with A. heliotropioides, a species without capitate glandular trichomes and with a cicatrix in an infra-medial position (Fig. 5B–J; Table 1). Despite these similarities, A. brevicalyx is disjunct geographically from both species (Fig. 5A). In short, both morphological and molecular data strongly support the recognition ofA. brevicalyx as a new species, which is unique in the genus by possessing a calyx shorter than the corolla tube at anthesis, emphasized by the specific epithet of the name. In spite of the lack of jackknife support and the conflicting results among the nuclear and chloroplast loci, the combined molecular analysis recovered all the accessions ofA. brevicayx in a clade. The phylogenetic relationships amongAntiphytum species and the exact position of A. brevicalyx inside the genus require further studywith a broader sampling of loci and taxa, including additional outgroups and all species ofAntiphytum, including the South American ones (Mendoza-Dı́az et al. in prep.). Acknowledgments We thank the directors and curators of the herbaria cited in this paper. We recognize the facilities provided by Martha Juana Martı́nez at FCME; she and several other colleagues of the Facultad de Ciencias, UNAM, have collected invaluable specimens for the Flora de Guerrero project, which are also important to the knowledge of the Mexican flora in general. The authors thank the following colleagues at IBUNAM: Berenit Mendoza- Garfias for assistance with the scanning microscope, Susana Guzmán- Gómez for help with the stereoscopic microscope, Lidia Cabrera for use of the facilities at Laboratorio de Biologı́a Molecular, IBUNAM, and Laura Márquez and Nelly López at the sequencing service LaNaBio. We espe- cially thank Alejandro Torres-Montúfar and Cecilia Rojas-Martı́nez for fieldwork help and company,Marı́aAguilar-Morales for support to get one stereoscopic photograph, and Albino Luna for the botanical illustration. The first author is a recipient of a doctoral fellowship (CONACYT grant 288698). We also thank the US National Science Foundation for support (DEB 1054539 to MJM). Author Contributions Nidia Mendoza-Dı́az conducted the research as part of the D.Sc. thesis guided byHilda Flores-Olvera. Helga Ochoterena guided the phylogenetic study. All coauthors conducted field work. Michael J. Moore provided lab space and support for parts of the molecular study. The manuscript was written with the collaboration of all the authors. Literature Cited Chacón, J., F. Luebert, H. H. Hilger, S. Ovchinnikova, F. Selvi, L. Cecchi, C. M. Guilliams, K. Hasenstab-Lehman, K. Sutorý, M. G. Simpson, and M. Weigend. 2016. The borage family (Boraginaceae s.str.): A revised infrafamilial classification based on new phylogenetic evidence, with em- phasis on the placement of some enigmatic genera. Taxon 65: 523–546. Chamisso de Boncourt, L. C. A. 1829. De Plantis in expeditione speculatoria Romanzoffiana observatis disserere perguntAd. deChamisso et Died. de Schlechtendal. Boragineae Juss. Linnaea. Ein Journal für die Botanik in ihrem ganzen Umfange 4: 435–496. Cohen, J. I. 2011. A phylogenetic analysis of morphological and molecular characters of Lithospermum L. (Boraginaceae) and related taxa: Evo- lutionary relationships and character evolution. Cladistics 27: 559–580. Cohen, J. I. 2014. A phylogenetic analysis of morphological and molecular characters of Boraginaceae: Evolutionary relationships, taxonomy, and patterns of character evolution. Cladistics 30: 139–169. MENDOZA-DÍAZ ET AL: NEW SPECIES OF ANTIPHYTUM FROM GUERRERO, MEXICO 1972020] Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) Cohen, J. I. and J. I. Davis. 2009. Nomenclatural changes in Lithospermum (Boraginaceae) and related taxa following a reassessment of phylo- genetic relationships. Brittonia 61: 101–111. CONABIO. 1997. Provincias biogeográficas deMexico, escala 1:4000000. Comisión Nacional para el Conocimiento yUso de la Biodiversidad deMexico. http:// conabio.gob.mx/informacion/metadata/gis/rbiog4mgw.xml?_xsl5/ db/metadata/xsl/fgdc_html.xsl&_indent5no (last accessed March 2018). CONABIO. 1998. Uso de suelo y vegetación de INEGI agrupado por CONABIO, escala 1:1000000. Modificado de: Instituto Nacional de Estadı́stica, Geografı́a e Informática (INEGI). Comisión Nacional para el Conocimiento y Uso de la Biodiversidad de Mexico. http:// www.conabio.gob.mx/informacion/metadata/gis/usva1mgw. xml?_xsl5/db/metadata/xsl/fgdc_html.xsl&_indent5no (last accessed March 2018). de Candolle, A. P. and A. L. P. P. de Candolle. 1846. Borragineae. Pp. 117–178 in Prodromus Systematis Naturalis Regni Vegetablis, Sive, Enumeratio Contracta Ordinum Generum Specierumque Plantarum Huc Usque Cognitarium, Juxta Methodi Naturalis, Normas Digesta, vol. 10, ed. A. P. de Candolle. Paris: Victoris Masson. Doyle, J. J. and J. L. Doyle. 1987. CTAB DNA extraction in plants. Phy- tochemical Bulletin 19: 11–15. Farris, J. S., V. A. Albert, M. Källersjö, D. Lipscomb, and A. G. Kluge. 1996. Parsimony jackknifing outperforms neighbor-joining.Cladistics 12: 99–124. Flores-Olvera, H., J. J. Morrone, L. Soto-González, R. Chávez, M. F. Rico- Bernal, L. E. Eguiarte, and M. Martı́nez-Ramos. 2010. Biologı́a. Pp. 279–323 in La UNAM por Mexico, vol. 1, eds. L. M. Chehaibar Náder, J. Franco López, J. A. Garcı́a-Sáinz, and A. Mayer. Mexico: Uni- versidad Nacional Autónoma de Mexico. Goloboff, P. A. 1999. Analyzing large data sets in reasonable times: So- lutions for composite optima. Cladistics 15: 415–428. Goloboff, P. A., S. Farris, and K. Nixon. 2000. TNT (tree analysis using new technology) (BETA) v. 1.1. Tucumán, Argentina: Published by the authors. Higgins, L. and B. L. Turner. 1983. Antiphytum hintoniorum (Boraginaceae), a bizarre new gypsophile fromNorth-Mexico. The SouthwesternNaturalist 28: 457–458. Hilger, H. H. 2014. Ontogeny, morphology, and systematic significance of glochidiate and winged fruits of Cynoglosseae and Eritrichieae (Boraginaceae). Plant Diversity and Evolution 131: 167–214. IUCN. 2012. IUCN red list categories and criteria version 3.1. Ed. 2. Gland, Switzerland and Cambridge, UK: IUCN. INEGI. 2017. Datos vectoriales del Marco Geoestadı́stico Nacional. Instituto Nacional de Estadı́stica, Geografı́a e Informática. Available from: http:// www.beta.inegi.org.mx/app/biblioteca/ficha.html?upc5889463171829 (last accessed February 2018). Johnston, I.M. 1923. Studies in the Boraginaceae.Contributions from the Gray Herbarium of the Harvard University 68: 43–80. Johnston, I. M. 1927. Studies in the Boraginaceae VI: A revision of the South American Boraginoideae. Contributions from the Gray Herbarium of Harvard University 78: 3–118. Långström, E. and B. Oxelman. 2003. Phylogeny of Echiochilon (Echiochileae, Boraginaceae) basedon ITSsequencesandmorphology.Taxon52: 725–735. Martens, M. and H. Galeotti. 1844. Enumeratio synoptica plantarum phaner- ogamicarum ab Henrico Galeotti in regionibus Mexicanus collectarum. Bulletin de l’AcademieRoyale des Sciences et Belles-lettres de Bruxelles 11: 319–340. Mendoza-Dı́az, N. 2015. Sistemática de Antiphytum DC. ex Meisn. (Bor- aginaceae). M.S. thesis. Ciudad de Mexico: UNAM. Mendoza-Dı́az, N., H. Flores-Olvera, M. G. Simpson, and M. J. Moore. 2018. A new and unusual endemic species from the Chihuahuan Desert, Mexico: Antiphytum geoffreyi (Boraginaceae, Echiochiloideae). Phytotaxa 367: 275–283. Mendoza-Dı́az, N., H. Ochoterena, M. J. Moore, and H. Flores-Olvera. 2019. Data from: Molecular and morphological evidence reveals a new species of Antiphytum (Echiochiloideae, Boraginaceae) from Guerrero, Mexico. Dryad Digital Repository. https://doi.org/10.5061/dryad.r07202b. Moore, M. J., J. F. Mota, N. A. Douglas, H. Flores-Olvera, and H. Ochoterena. 2014. The ecology, assembly, and evolution of gyp- sophile floras. Pp. 97–128 in Plant Ecology and Evolution in Harsh Environments, eds. N. Rajakaruna, R. S. Boyd, and T. B. Harris. New York: Nova Science Publishers. Müller, J., K. Müller, C. Neinhuis, and D. Quandt. 2010. PhyDE- Phylogenetic Data Editor v. 0.9971. http://www.phyde.de/ (last accessed February 2018). Nixon, K. C. 1999. The Parsimony Ratchet, a new method for rapid par- simony analysis. Cladistics 15: 407–414. Nixon, K. C. 1999–2002. WinClada (BETA) v. 1.00.24-IMG BETA. Ithaca, New York: published by the author. Ochoterena, H. 2009. Homology in coding and non-coding DNA sequences: A parsimony perspective. Plant Systematics and Evolution 282: 151–168. Rose, J. N. 1890. List of plants collected by Dr. Edward Palmer in Lower California and Western Mexico in 1890. Contributions from the United States National Herbarium 1: 63–90. Simmons, M. P. and H. Ochoterena. 2000. Gaps as characters in sequence- based phylogenetic analyses. Systematic Biology 49: 369–381. Swofford, D. 2003. PAUP* Phylogenetic analysis using parsimony (*and other methods), v. 4. Sunderland: Sinauer Associates. Thiers, B. 2019 [continuously updated]. Index Herbariorum: A global di- rectory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium. http://sweetgum.nybg.org/science/ ih/ (last accessed March 2018). Torrey, J. 1859. General botany. Pp. 27–236 in Report on the United States and Mexican Boundary Survey,Made Under the Direction of the Secretary of the Interior, vol. 2, part 1, ed. W. H. Emory. Washington, DC: Cornelius Wendell. Watson, S. 1883. Contributions to American botany. Proceedings of the American Academy of Arts and Sciences 18: 96–196. Weigend, M., F. Selvi, D. C. Thomas, and H. H. Hilger. 2016. Boraginaceae. Pp. 41–102 in The Families and Genera of Vascular Plants, vol. 14, eds. J. W. Kadereit and V. Bittrich. Cham, Switzerland: Springer In- ternational Publishing. Weberling, F. 1989. Morphology of Flowers and Inflorescences. Cambridge, UK: Cambridge University Press. White, T. J., T. Bruns, S. Lee, and J. Taylor. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. Pp. 315–322 in PCR Protocols: A Guide to Methods and Applications, eds.M.A. Innis, D.H.Gelfand, J. J. Sninsky, and T. J.White. SanDiego: Academic Press. Appendix 1. List of accessions used in the morphological and mo- lecular phylogenetic analyses. Asterisks (*) indicate lack of morphological observations. The following information is included for each accession: species, geographic origin, collector(s) and collection number, herbarium, and GenBank accession numbers for ITS and ndhF-rpl32, respectively. An em-dash indicates missing data while boldface indicates sequences ob- tained from outside sources, which are correspondingly cited. Outgroup: Echium vulgare L. Great Britain, cult., *Chase 6061 (K), AJ555896, Långström and Oxelman (2003),—; Cornell Plantations, *Cohen 212, —, FJ827364, Cohen and Davis (2009). Antiphytum brevicalyx N. Mend., Flores Olv. & H. Ochoterena. Mexico.—GUERRERO: Jiménez-Ramı́rez & Luna-Flores 971 (FCME), MN265414, MN313511; Martı́nez-Gordillo 798 (MEXU), MN265415, MN313571; Mendoza-Dı́az et al. 643F (MEXU), MN265416, MN313512; Mendoza-Dı́az et al. 643G (MEXU), MN265417, MN313513; Mendoza-Dı́az et al. 644A (MEXU), MN265418, MN313514; Mendoza-Dı́az et al. 644B (MEXU), MN265419, MN313515; Mendoza-Dı́az et al. 644D (MEXU), MN265420,MN313516.Antiphytumcaespitosum I.M.Johnston.Mexico.— OAXACA: Mendoza-Dı́az et al. 581 (MEXU), MN265421, MN313517; Mendoza- Dı́az et al. 583b (MEXU), MN265422, MN313518; Mendoza-Dı́az et al. 583d2 (MEXU),MN265423,MN313519;Mendoza-Dı́az et al. 603e (MEXU),MN265424, MN313520;Mendoza-Dı́az et al. 612a (MEXU),MN265425,MN313521;Mendoza- Dı́az et al. 612d (MEXU), MN265426, MN313522; Mendoza-Dı́az et al. 613e (MEXU),MN265427,MN313523;Mendoza-Dı́az et al. 614a (MEXU),MN265428, MN313524;Mendoza-Dı́az et al. 618b (MEXU), MN265429, MN313525; Rangel- Landa 1400, MEXU, MN265431, MN313527.—PUEBLA: Medina-Lemos & Mart́ınez-Salas 5715 (MEXU), MN265430, MN313526. Antiphytum flo- ribundum (Torr.) A.Gray. Mexico.—MICHOACÁN: Pérez-Cálix 2508 (MEXU), MN265433, MN313529; *Cohen 227 (—), KF287948, Cohen (2014), JF488879, Cohen and Davis (2009).—JALISCO: Calónico & Flores 3822, MEXU, MN265434, MN313530 USA. —TEXAS: Lott et al. 5353 (MEXU), MN265432, MN313528. Antiphytum heliotropioides A.DC. Mexico.—GUANAJUATO: Rzedowski 52626 (MEXU),MN265440,MN313536.—HIDALGO:Mendoza-Dı́az et al. 599b (MEXU), MN265444, MN313540; Mendoza-Dı́az et al. 599d (MEXU), MN265445, MN313541;Mendoza-Dı́az et al. 601a (MEXU),MN265446,MN313542.—NUEVO LEÓN: Moore et al. 1332 (MEXU), MN265442, MN313538; Moore et al. 1363 (MEXU), MN265443, MN313539. —OAXACA: Calzada 21801 (MEXU), MN265438,MN313534;Calzada 22146 (MEXU),MN265439,MN313535; Salinas & Mart́ınez-Correa 7929 (MEXU), MN265435, MN313531; Torres-Hernández & Hernández 802 (MEXU), MN265436, MN313532. —QUERÉTARO: Carranza 3781 (IEB), MN265437, MN313533; Mendoza-Dı́az et al. 623a (MEXU), MN265447, MN313543; Mendoza-Dı́az et al. 623b (MEXU), MN265448, MN313544. —SAN SYSTEMATIC BOTANY [Volume 45198 Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) LUIS POTOSÍ: Torres-Coĺın & Terrazas 17191 (MEXU), MN265449, MN313545. Antiphytum hintoniorum L.C.Higgins & B.L.Turner. Mexico.—NUEVO LEÓN: Hinton et al. 18321 (MEXU),MN265450,MN313546;Hinton et al. 19125 (MEXU), MN265451, MN313547; Moore et al. 1308 (MEXU), MN265452, MN313548; Moore et al. 1334 (MEXU),MN265453,MN313549; Patterson et al. 7475 (MEXU), MN265454, MN313550; Antiphytum paniculatum I.M.Johnston. Mex- ico.—PUEBLA: Mendoza-Dı́az et al. 619a (MEXU), MN265455, MN313551; Mendoza-Dı́az et al. 619b (MEXU), MN265456, MN313552; Mendoza-Dı́az et al. 620a (MEXU), MN265457, MN313553; Mendoza-Dı́az et al. 621a (MEXU), MN265458, MN313554; Mendoza-Dı́az et al. 621e (MEXU), MN265459, MN313555.Antiphytum parryi S.Watson.Mexico.—OAXACA:Garcı́a-Mendoza et al. 2515 (MEXU), MN265460, MN313556; Mendoza-Dı́az et al. 579 (MEXU), MN265464, MN313560; Mendoza-Dı́az et al. 579_2 (MEXU), MN265465, MN313561; Mendoza-Dı́az et al. 579a (MEXU), MN265466, MN313562; Men- doza-Dı́az et al. 579b (MEXU), MN265467, MN313563;Mendoza-Dı́az et al. 579c (MEXU), MN265468, MN313564; Mendoza-Dı́az et al. 610a (MEXU), MN265471, MN313567. —HIDALGO: Mendoza-Dı́az et al. 596 (MEXU), MN265469, MN313565; Mendoza-Dı́az et al. 598 (MEXU), MN265470, MN313566.—PUEBLA:Rosas et al. 2950 (MEXU),MN265461,MN313557;Náder et al. 54 (MEXU), MN265462, MN313558. —QUERÉTARO: Rzedowski 53929 (MEXU), MN265463, MN313559 . Antiphytum peninsulare (Rose) I.- M.Johnston. Mexico. —BAJA CALIFORNIA SUR: Valov 2004046 (HCIB), MN265472, MN313568; Valov 200517 (MEXU), MN265473, MN313569; León 7736 (HCIB), MN265474, MN313570. MENDOZA-DÍAZ ET AL: NEW SPECIES OF ANTIPHYTUM FROM GUERRERO, MEXICO 1992020] Downloaded From: https://bioone.org/journals/Systematic-Botany on 27 Feb 2020 Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM) 29 PARTE I: REEVALUACIÓN TAXONÓMICA DE LAS ESPECIES DE ANTIPHYTUM PARTE A. ESPECIES DE NORTEAMÉRICA CAPÍTULO II: A new and unusual endemic species from the Chihuahuan Desert, Mexico: Antiphytum geoffreyi (Boraginaceae, Echiochiloideae). Nidia Mendoza-Díaz, Hilda Flores-Olvera, Michael G. Simpson & Michael J. Moore Phytotaxa (2018) 367(3): 275–283. Phytotaxa 367 (3): 275–283 http://www.mapress.com/j/pt/ Copyright © 2018 Magnolia Press Article PHYTOTAXA ISSN 1179-3155 (print edition) ISSN 1179-3163 (online edition) Accepted by Federico Luebert: 1 Aug. 2018; published: 6 Sept. 2018 https://doi.org/10.11646/phytotaxa.367.3.7 275 A new and unusual endemic species from the Chihuahuan Desert, Mexico: Antiphytum geoffreyi (Boraginaceae, Echiochiloideae) NIDIA MENDOZA-DÍAZ1,2, HILDA FLORES-OLVERA2,5, MICHAEL G. SIMPSON3 & MICHAEL J. MOORE4 1 Posgrado en Ciencias Biológicas, Instituto de Biología. Universidad Nacional Autónoma de México. 3er. Circuito Exterior s/n, Coyoacán, 04510, Ciudad de México, México. 2 Departamento de Botánica, Instituto de Biología, Universidad Nacional Autónoma de México, Ap. Postal 70-233, Coyoacán, 04510, Ciudad de México, México. 3 Department of Biology, San Diego State University, San Diego, California 92182, U. S. A. 4 Biology Department, Oberlin College, Oberlin, Ohio 44074, U. S. A. 5 Author for correspondence: mahilda@ib.unam.mx Abstract A new and unusual species of Antiphytum, here named A. geoffreyi, is described and illustrated. This new species is assigned to the genus Antiphytum because it possesses eremocarps directly attached to a pyramidal gynobase, which forms a ventral cicatrix on the eremocarp. However, it is unique within Antiphytum in having multiple inflorescence branches arising nearly at the base of the plant, a triangular-pyramidal gynobase, and the proportionally longest eremocarp cicatrix of the genus. Some of these characteristics are reminiscent of Ogastemma, the sister genus of Antiphytum, making A. geoffreyi of potential great significance in the systematics of Boraginaceae subfam. Echiochiloideae. Among species of Antiphytum, A. geoffreyi is most similar to A. peninsulare, with which it shares a triangular-shaped eremocarp, but it differs from this species in possessing an infra-medial cicatrix and a smaller corolla limb diameter. The new species is known from only two collections in Coahuila, Mexico, both made by the Hinton family, who have discovered many new species in their extensive explorations of the country and provided important specimens to the country’s flora. Keywords: endangered species, fruit, gynobase, North America Introduction Antiphytum DC. ex Meisner (1840: 280) (Boraginaceae, Echiochiloideae Weigend; see Chacon et al. 2016) is an American genus with nine recognised species, following Johnston (1923, 1927) who last revised the genus. Antiphytum is characterised by having eremocarps (terminology sensu Hilger 2014; also termed mericarps or nutlets) directly attached to a flat or depressed-pyramidal gynobase, forming a ventral or basal attachment scar on the eremocarp, the scar termed cicatrix by Hilger (2014). Most species are North American, with seven species ranging from the southwestern USA to southern Mexico, but two species are endemic to South America, in Brazil and Uruguay (Johnston 1927). The most recent species described in the genus is Antiphytum hintoniorum Higgins & Turner (1983: 457), which is endemic to gypsum soils in Nuevo Leon, Mexico. This species was discovered by James Hinton and named in honor of the Hinton family, who for three generations have explored the botanical richness of Mexico, leading to the discovery and description of several hundred new species (Hinton & Rzedowski 1972). George B. Hinton, who was the first botanist of the family, concentrated his fieldwork in the states of Guerrero, Michoacan, and Mexico (Rico et al. 2008). His son James Hinton collected in the mountains of Nuevo Leon, Coahuila, Michoacan, Guerrero, and Oaxaca (Hinton 2008), and his grandson George S. Hinton has continued the family tradition through his extensive exploration of northern Mexico. Their collections are available at the G. B. Hinton herbarium (GBH). In this manuscript we describe yet another important taxonomic novelty, Antiphytum geoffreyi, collected by George S. Hinton in Coahuila, Mexico. This species represents an important new addition to the flora of the Chihuahuan Desert and, due to its unusual characters, is important for understanding the systematics of the genus and the subfamily. This new species is assigned to the genus Anthiphytum by having eremocarps directly attached to a pyramidal gynobase, forming a ventral cicatrix on the eremocarp. MENDOZA-DÍAZ ET AL.276 • Phytotaxa 367 (3) © 2018 Magnolia Press Material and Methods Herbarium specimens (acronyms according to Thiers continuously updated):—The revision of Boraginaceae at GBH revealed two collections of an undescribed species of Antiphytum of the state of Coahuila, Mexico. These specimens were collected by George S. Hinton, who provided duplicates as donation to MEXU. Images provided by Hinton from GBH specimens were also studied. To properly describe the characteristic inflorescence of the species, we follow Weberling (1989) and Weigend et al. (2016). This terminology is defined as part of the description when necessary. Fruit observations:—Mature eremocarps and gynobases from specimens of the new species, as well as from the morphologically similar Antiphytum peninsulare (Rose) Johnston (1923: 51), and Ogastemma pusillum (Coss. & Durieu ex Bonnet & Barrate) Brummit (1982: 680), were observed with a confocal stereo microscope (Leica Z16 APO A, Leica Microsystems GmbH, Wetzlar, Germany), and photographed with a camera (Leica DFC490, Leica Microsystems GmbH) at Laboratorio de Microscopía y Fotografía de la Biodiversidad (II) of Instituto de Biología, Universidad Nacional Autónoma de México (UNAM). Also, SEM photographs were taken (Hitachi SU1510, Hitachi Ltd., Tokyo, Japan) at Laboratorio de Microscopía Electrónica y Fotografía de la Biodiversidad (I). The image files were cleaned in Adobe Photoshop CS5 12.0 (Adobe Systems Incorporated, San Jose, California). Distribution map:—A distribution map was produced using QGIS Brighton (2.6.1) employing layers from Instituto Nacional de Estadística y Geografía (2017). Taxonomy Antiphytum geoffreyi N. Mend. & Flores Olv., sp. nov. (Figs. 1–2). Type:—MEXICO. Coahuila: Municipality Francisco I. Madero, West side of Sierra de Tlahualilo, 1275 m, 26° 36′ 46.8″ N, 103° 21′ 13″ W, 21 April 2015. Hinton et al. 29613 (Holotype: GBH!, Isotypes: ANSM, CIIDIR, GBH, MEXU!, TEX). Diagnosis:—Differs from all other species of the genus by possessing a terminal inflorescence arising at the bottom of each branch rather than from the distal part, by possessing a triangular-pyramidal rather than a deltate-pyramidal or flat gynobase, and by possessing the proportionally longest eremocarp cicatrix in the genus. Suffruticose perennial, erect (or decumbent?), up to 20 cm tall, covered by strigose and hispid indumenta with a multicellular lithocystic structure, and hyaline glandular-capitate trichomes on branches, leaves, and other leafy organs. Stems of numerous branches arising from almost the base of a short main stem, branches alternate or opposite, non-fistulose, slender, with white exfoliating bark on the older parts. Leaves cauline and pseudobasal, without a basal rosette, sessile, blades linear-oblanceolate to linear-oblong, margins flat, entire, apices acute, bases truncate connate, surfaces concolorous, veins not visible; cauline leaves opposite, alternate (?) when subtending inflorescences, 8.0-10.0 × 1.0 mm, gradually smaller upward along the stem; pseudobasal leaves at the base of some branches, the newest ones 3.0 × 0.5 mm. Inflorescence arising at the bottom of each branch, ending in a flower (monotelic) with alternate paraclades, forming a terminal homocladic determinate thyrse (thyrsoid) with dichasial branching, one side a scorpioid cyme and the other side branching many times; first internode of a lateral axis (hypopodium) 4.3–8.2 mm in length; mature scorpioid cymes up to 133 mm in length, many-flowered, bracteate; floral bracts oblanceolate to linear-oblong, sessile, strigose, with glandular capitate-trichomes at margins, veins not visible, blades 6.0–6.5 × 1.7 mm, gradually smaller upward along the cyme. Flowers perfect, sessile or subsessile (pedicel ca. 0.7 mm in length); calyx strigose on both surfaces, glandular-capitate and hispid trichomes at margins, accrescent in fruit, divided almost to the base, lobes asymmetric, acute, scarious at the base, three linear-oblong and two oblanceolate, ca. 2.2–3.3 × 0.4–1.2 mm; corolla hypocrateriform (rotate), white, with five papillose faucal appendages of color unknown on the throat opposite to corolla lobes, tube not surpassing the calyx, 1.4–1.5 mm in length, 0.8 mm in diameter at the base, limb 4.7(–5.0) mm in diameter, lobes oriented nearly perpendicular to the tube, imbricate, rounded, sinuate at apex, ca. (2.0–) 2.1–2.4 × (1.9–) 2.0–2.2 mm, abaxially strigose at the middle; stamens inserted, alternate to corolla lobes, adnate at ca. 0.6 mm from the base of the corolla tube; filaments up to 0.4 mm in length, slender; anthers oblong, slightly lanceolate, dorsifixed, ca. 0.6–0.7 × 0.3–0.4(–0.5) mm; ovary with four ovules, style gynobasic, 0.7–0.8 mm in length at anthesis, stigma bilobed to near the apex of the style (subterminal). Fruit dry, covered by a persistent calyx, usually with 3–4 developed eremocarps; eremocarps lustrous, triangular and apically acute in ventral view, 1.6–1.7 × 1.3–1.4 mm, dorsally convex, tuberculate and papillate except to the tubercles apex, cicatrix at infra-medial position on the ventral face, triangular, ca. 0.9 mm in length, three keels straight, a ventral (apical) not forming a stipe, and two lateral (basal) HERBARIUM of G. B. HINTON 29613 Boraginaceae Cryptantha mexicana (Brandegee) I.M. Johnst. Determined by G.S. Hinton HERBARIUM of G. B. HINTON No. 29613 Collected by Hinton et al. 2015.04.21 West side of Sierra de Tlahualilo 1275 m. Francisco 1, Madero, Coahuila, México. Rocky outcrop on gypsum hill. Fl. white. 20 cm. Colony. no Dd > Sa 10) 2 Dd | pd - | Le O) e > ¡o O | ¡0 A NEW SPECIES OF ANTIPHyTUM Phytotaxa 367 (3) © 2018 Magnolia Press • 277 FIGURE 1. Holotype of Antiphytum geoffreyi sp. nov. Photo provided by George S. Hinton from the GBH specimen. XA JS OS HA ATA YO es a a MA DAA AA == O Y ==,» 3 — — -—-.——- -== AA O MENDOZA-DÍAZ ET AL.278 • Phytotaxa 367 (3) © 2018 Magnolia Press FIGURE 2. Antiphytum geoffreyi: a, general appearance, b, fragment of the scorpioid cyme, c, flower in lateral view, d, flower showing the corolla limb, e, open flower, f, detailed of faucal appendage, g, fruit, h, fruit showing the gynobase and eremocarp insertion, i, eremocarp. (Illustration by Albino Luna from Hinton et al. 29613 & 29132 in MEXU and GBH). A NEW SPECIES OF ANTIPHyTUM Phytotaxa 367 (3) © 2018 Magnolia Press • 279 from the vertices of the eremocarp to the corresponding cicatrix vertices; gynobase triangular-pyramidal, with four areoles corresponding to the cicatrix of each eremocarp, style plus stigmas 1.8 (–1.9) mm, surpassing the eremocarps. Additional specimens examined (paratypes):—MEXICO. Coahuila: Municipality Parras, South of Parras, 1326 m, 25° 18′15.4″ N, 102° 13′ 42.5″ W, 30 April 2010. Hinton et al. 29132 (ANSM, CIIDIR, GBH!, MEXU!, TEX). Additional specimens examined (other species):—Antiphytum cruciatum (Chamisso) DC. in Candolle & Candolle (1846: 121). BRASIL. Rio Grande do Sul, Porto Alegre: Belem novo (Hindenburgshöhe) near Porto Alegre, November 1926. Jüngers 254 (B!, GH!). Itapoan, Granja Neugebauer p. Itapoan, 11 October 1950. Rambo 48956 (B!, US!). Sao Sepe, 02 October 1988. Wasum et al. 4538 (MO!, NY!). URUGUAY. Cerro Largo: Río Negro y Palleros, December 1937. Rosengurtt 2346 (GH!). Antiphytum floribundum (Torrey) Gray (1874:55). MEXICO. Coahuila: Municipality Ocampo, Sierra de Hechiceros, vicinity of Rancho El Tule, near Chihuahuan boundary, 17–19 September 1940. Johnston & Muller 1330 (GH!). Antiphytum heliotropioides A. DC. in Candolle & Candolle (1846: 122). MEXICO. Coahuila: Municipality Muzquiz, ca. 130 road km NW of Muzquiz on Coa Hwy 2A (Muzquiz-Boquillas), then SW of 2A on tunnel road to La Encantada basin, Sa. Buenavista of Sa. La Encantada, 1650 m, 28° 34′ N, 102° 30′ W, 03 June 1992. Nesom & Mayfield 7421 (MEXU!). Municipality General Cepeda, in upper reaches of Canada el Aguirreno, from 0–1 km W of old Barita mine, ca. 4–5 air km from mouth of canyon, ca. 11 air km SE of Las Coloradas, on NW side of Sierra La Paila; Las Coloradas 1:50 000, 26° 06′15″ N, 101° 35′ 30″ W, 05 April 1990. Bridges et al. 13074 (MEXU!). Municipality Arteaga, El Chorro, 1800 m, July–August 1942. Lyonnet 3681 (MEXU!). Antiphytum peninsulare. MEXICO. Baja California Sur: Gulf of California, volcanic hillside near north west end of Isla San Marcos, 27° 15′ N, 112° 07′ W, 28 March 1962. Wiggins 17328 (MEXU!). Municipality Mulege, N side of river, botton of sandy/gravelly arroyo, 5 m, 26.8998 N, 111.9762 W, 18 January 2005. Valov 2005017 (MEXU!). Municipality Mulege, N side of river; bottom of sandy/gravelly arroyo, 5 m, 26° 59.990′ N, 111° 58.570′ W, 04 February 2004. Valov 2004046 (HCIB!). Municipality Santa Rosalia, along the wash emptying at Punta El Aterrizaje, just N of airstrip, ca. 5.6 km N of Santa Rosalia, along gravel road turning N from Mex Hwy 1, 5 m, 27° 23.6′ N, 112° 18.2′ W, 28 February 1995. Porter & Machen 11056 (HCIB!). Municipality Santa Rosalia, Isla San Marcos, 15 m, 27° 14′ N, 112° 06′ W, 27 October 1995. León de la Luz 7736 (HCIB!). Ogastemma pusillum. KUWAIT. Wadi Al-Batin, 12 km N Al-Salmi border station with Saudi Arabia, 280 m, 17 April 1990. Boulos & Cope 17614 (K!). Shu´aib Al- Batin, 25 km E Al-Shagayah, 280 m, 17 April 1990. Boulos & Cope 17685 (BM!, K!). Medairah halophytes project north of city in National Park, 05 March 1990. Hepper 8000 (K!). 35 km north-east of Jahra along Jahra-Al Subbiyah road, 29° 36′ N, 47° 56′ E, 30 March 1995. Mathew 2633 (K!). SAUDI ARABIA. 27-48 N 47-18 E, 27 February 1970. Mandaville Jr. 2720 (BM!). Additional specimens examined (type specimens):—Antiphytum. Type: Anchusa cruciata Chamisso (1829:438) (designated by Johnston 1923):—BRASIL. Brasilia Meridionalis, Sellow s/n (UC https://plants.jstor.org/ stable/10.5555/al.ap.specimen.uc493522). Antiphytum peninsulare. Type:—MEXICO. Baja California Sur, Santa Rosalia, 23 February to 3 March 1889. Palmer 203 (GH!). Ogastemma pusillum. Type:—ALGERIA. Biskra, sur les collines incultes, 10 May 1853, Balansa 1035 (E https://plants.jstor.org/stable/10.5555/al.ap.specimen.e00115613). Etymology:—The epithet honours Geoffrey Hinton, who accompanied his father George S. Hinton in the discovery of the new species in Sierra Tlahualilo, Coahuila. Phenology:—Flowers appear at the beginning of April, perhaps earlier, with fruits appearing at the end of April. Whether the species can also flower in the late summer during the rainy season is unknown. Distribution and habitat:—Antiphytum geoffreyi is known from only two collections by George S. Hinton, in disjunct localities in the municipalities of Francisco I. Madero in the Sierra Tlahualilo, and Parras, Coahuila (Fig. 3). It occurs at elevations of 1,275 m and 1,326 m, in a steep rockslide and a rocky outcrop on a gypsum hill. This new species is a gypsovag (signifying that it grows on and off gypsum), as are many other North American species of the genus, such as A. caespitosum Johnston (1923: 52), A. heliotropioides, A. parryi Watson (1883: 122), and A. peninsulare. So far as is known, A. geoffreyi is endemic to Coahuila and hence represents the third known Antiphytum in the state, along with A. heliotropioides and A. floribundum, the latter only known in Coahuila from one collection (Johnston & Muller 1330, GH!) close to the border with Chihuahua. Conservation status—The new species is known only from two disjunct subpopulations in the state of Coahuila, Mexico (Fig. 3) and the estimated area of occupancy is geographically limited to less than 500 km2. According to the IUCN (2012), it meets the criteria under B2ab(iii) for the category of Endangered (EN). Discussion:—A number of unusual traits characterize Antiphytum geoffreyi, most prominently involving the fruit. Fruit characters are useful for the classification and taxonomy of Boraginaceae (Weigend et al. 2016). When the eremocarps are released, the part of the fruit that remains on the plant is termed a gynobase (Hilger 2014), which in Antiphytum is either flat or depressed-pyramidal (Johnston 1927). Within the genus, the length/width ratio of the MENDOZA-DÍAZ ET AL.280 • Phytotaxa 367 (3) © 2018 Magnolia Press gynobase can be classified as deltate (1:1) to shallowly triangular (1:2) to totally flat (pers. obs.). The shape of the gynobase is linked with the eremocarp orientation (Weigend et al. 2016), and is therefore also associated with the position and shape of the eremocarp cicatrix. Antiphytum geoffreyi possesses a triangular-pyramidal gynobase (2:1; see Table 1), which represents a new gynobase shape for the genus (Fig. 4). The new species also possesses the longest cicatrix, relative to the eremocarp (cicatrix length/ventral eremocarp face length) in the genus (Table 1). The unusual gynobase shape in A. geoffreyi is closer to the almost narrowly triangular shape present in the sister taxon of Antiphytum, the monotypic Ogastemma pusillum (Längstrom & Oxelman 2003, Cohen 2014), which has a relatively long triangular cicatrix (Fig. 4) that extends from the middle to the infrabasal part of the eremocarp (Table 1). FIGURE 3. Map of the known distribution of Antiphytum geoffreyi. The areas shaded in grey indicate the municipalities where specimens have been collected, dots the location of the two individual collections. The inflorescences of Antiphytum geoffreyi are also similar to those of Ogastemma. Specifically, A. geoffreyi is the only species in the genus with multiple branches bearing inflorescences that arise nearly at the base of the plant, similar to Ogastemma pusillum. Despite the similarities of A. geoffreyi with Ogastemma pusillum, we consider the new species to be part of Antiphytum due to shared floral characters, such as the calyx exceeding the corolla in Ogastemma compared to the reverse in Antiphytum. Other species of Antiphytum, such as A. heliotropioides, A. hintoniorum, and A. peninsulare, have principal stems branching multiple times with inflorescences distal from the base. As with A. geoffreyi, all of these species lack leaves arranged in basal rosettes. Those species with basal rosettes, such as A. caespitosum, have distal inflorescences bearing well-defined unbranching flowering stems. Within Antiphytum, A. geoffreyi is most similar to A. peninsulare. Both species share white colouration on old branches, the presence of capitate glandular trichomes, a triangular-shaped eremocarp and similar pericarp ornamentation (Fig. 5). Nevertheless, Antiphytum geoffreyi differs in several key characters from A. peninsulare, including the basal (vs. distal) position of inflorescences, a corolla limb diameter of 4.7–5.0 mm (vs. 10.0–12.0 mm), and an infra-medial (vs. medial) and relatively long eremocarp cicatrix. The unusual characters of Antiphytum geoffreyi and the similarity of some of its characters to Ogastemma make it a potentially important species for the systematics of Antiphytum and the subfamily Echiochiloideae as a whole. Future molecular phylogenetic studies may elucidate its position in the complex. This distinctive new species is also an important addition to the Chihuahuan Desert flora, a region still in need of exploration as shown by the collections of the Hinton family. A NEW SPECIES OF ANTIPHyTUM Phytotaxa 367 (3) © 2018 Magnolia Press • 281 FIGURE 4. Comparison of eremocarps in oblique (A–C) and ventral views (D–F), with their respective gynobases (G–I). A, D & G, Ogastemma pusillum (Boulos &. Cope 17614, K), B, E & H, Antiphytum peninsulare (Wiggins 17328, MEXU), C, F & I, A. geoffreyi (Hinton et al. 29613, MEXU). FIGURE 5. Pericarp ornamentation. A, Ogastemma pusillum (Boulos &. Cope 17614, K), B, Antiphytum peninsulare (Wiggins 17328, MEXU), C, A. geoffreyi (Hinton et al. 29613, MEXU). Note larger tubercles and smaller papillae in B and C. MENDOZA-DÍAZ ET AL.282 • Phytotaxa 367 (3) © 2018 Magnolia Press TABLE 1. Differences among Antiphytum geoffreyi and morphologically similar taxa. Antiphytum geoffreyi A. peninsulare Ogastemma pusillum Corolla diameter (mm) 4.7–5.0 10.0–12.0 2.0 Eremocarp cicatrix position infra-medial medial extending from the middle to the infrabasal part Gynobase (length: width) triangular (2:1) deltate (1:1) almost narrowly triangular (5:2) Ratio of cicatrix length/eremocarp ventral face length 50% 30–40% 70% Acknowledgements The authors especially thank George S. Hinton for allowing the authors to consult the GBH herbarium, and for providing duplicates to MEXU, photographs and additional information for the new species. We also thank the following staff at the Instituto de Biología, UNAM, for their assistance in this project: Helga Ochoterena for revising and improving the manuscript and for the help in making the figures, María del Rosario Peña for handling loans and sampling permits from several herbaria, Susana Guzmán for assisting with the digital photographs of reproductive structures, Berenit Mendoza-Garfias for handling the SEM, and Albino Luna for the illustration. Special thanks are due to Jim Cohen for informing the first author of the existence of this new species. We express our gratitude to GBH, K, and MEXU for supplying the specimens and the permits to remove fruits for the study, and the herbaria B, BM, GH, HCIB, K, MO, NY, UC, and US for the loans seen as additional material examined. The first author is a recipient of a doctoral fellowship from CONACYT (grant 288698). References Brummitt, R.K. (1982) Ogastemma, a new name for Megastoma (Boraginaceae). Kew Bulletin 26 (4): 679–680. https://doi.org/10.2307/4117908 Candolle, A.P. de & Candolle, A.L.P.P. de (1846) Borragineae. In: Candolle, A.P. (Ed.) Prodromus systematis naturalis regni vegetablis, sive, Enumeratio contracta ordinum generum specierumque plantarum huc usque cognitarium, juxta methodi naturalis, normas digesta, Pars Decima. Victoris Masson, Paris, pp. 1–178, 587–588. Chacón, J., Luebert, F., Hilger, H.H., Ovcinnikova, S., Selvi, F., Cecchi, L., Guilliams, C.M., Hasenstab-Lehman, K., Sutorý, K., Simpson, M.G. & Weigend, M. (2016) The borage family (Boraginaceae s.str.): A revised infrafamilial classification based on new phylogenetic evidence, with emphasis on the placement of some enigmatic genera. Taxon 65: 523–546. https://doi.org/10.12705/653.6 Chamisso de Boncourt, L.C.A. (1829) De plantis in expeditione speculatoria Romanzoffia observatis disserere pergunt Ad. De Chamisso et Died. de Schlechtendal. Boragineae Juss. Linnaea. Ein Journal für die Botanik in ihrem ganzan Umfange 4: 435–496. Cohen, J.I. (2014) A phylogenetic analysis of morphological and molecular characters of Boraginaceae: evolutionary relationships, taxonomy, and patterns of character evolution. Cladistics 30 (2): 139–169. https://doi.org/10.1111/cla.12036 Gray, A. (1874) Contributions to the Botany of North America: Notes on Boraginaceae. Proceedings of the American Academy of Arts and Sciences 10: 48–62. Higgins, L. & Turner, B.L. (1983) Antiphytum hintoniorum (Boraginaceae), a Bizarre New Gypsophile from North-Mexico. The Southwestern Naturalist 28 (4): 457–458. https://doi.org/10.2307/3670827 Hilger, H.H. (2014) Ontogeny, morphology, and systematic significance of glochidiate and winged fruits of Cynoglosseae and Eritrichieae (Boraginaceae). Plant Diversity and Evolution 131 (3): 167–214. https://doi.org/10.1127/1869-6155/2014/0131-0080 Hinton, J. & Rzedowski, J. (1972) George B. Hinton, collector of plants in southwestern Mexico. Journal of the Arnold Arboretum 53 (2): 141–181. Hinton, G.S. (2008) Semblanza: James Hinton (1915-2006). Acta Botánica Mexicana 83: 1–11. https://doi.org/10.21829/abm83.2008.1056 A NEW SPECIES OF ANTIPHyTUM Phytotaxa 367 (3) © 2018 Magnolia Press • 283 Instituto Nacional de Estadística y Geografía (2017) Datos vectoriales del Marco Geoestadístico Nacional. Available from: http://www. beta.inegi.org.mx/app/biblioteca/ficha.html?upc=889463171829 (accessed 8 February 2018) IUCN (2012) IUCN Red List Categories and Criteria, version 3.1, Second edition. Gland, Switzerland and Cambridge, UK, iv + 32 pp. Johnston, I.M. (1923) Studies in the Boraginaceae. Contributions from the Gray Herbarium of the Harvard University 68: 43–80. Johnston, I.M. (1927) Studies in the Boraginaceae VI: A revision of the South American Boraginoideae. Contributions from the Gray Herbarium of Harvard University 78: 3–118. Långström, E. & Oxelman, B. (2003) Phylogeny of Echiochilon (Echiochileae, Boraginaceae) based on ITS sequences and morphology. Taxon 52 (4): 725–735. https://doi.org/10.2307/3647347 Meisner, C.F. (1840) Plantarum vascularium genera. Libraria Weidmannia, Leipzig, 442 pp. Rico, M.L., Rodríguez, A. & Moreno-Gutiérrez, E. (2008) Las colecciones de leguminosas efectuadas por George B. Hinton, depositadas en el herbario de Royal Botanic Gardens, Kew. Acta Botánica Mexicana 84: 73–92. https://doi.org/10.21829/abm84.2008.1067 Thiers, B. (2018, continuously updated) Index Herbariorum: A global directory for public herbaria and associated staff. New York Botanical Garden´s Virtual Herbarium. Available from: http://sweetgum.nybg.org/science/ih/ (accesed January 2018) Watson, S. (1883) Contributions to American Botany. Proceedings of the American Academy of Arts and Sciences 18: 96–196. https://doi.org/10.2307/20013481 Weberling, F. (1989) Morphology of flowers and inflorescences. Cambridge University Press, Cambridge, 405 pp. Weigend, M., Selvi, F., Thomas, D.C. & Hilger, H.H. (2016) Boraginaceae. In: Kadereit, J.W. & Bittrich, V. (Eds.) The families and genera of vascular plants Vol. 14. Flowering plants:Eudicots; Aquifoliales, Boraginales, Bruniales, Dipsacales, Escalloniales, Garryales, Paracryphiales, Solanales (except Convolvulaceae), Icacinaceae, Metteniusaceae, Vahliaceae. Springer International, Cham, pp. 41–102. https://doi.org/10.1007/978-3-319-28534-4_5 39 PARTE I: REEVALUACIÓN TAXONÓMICA DE LAS ESPECIES DE ANTIPHYTUM PARTE B. ESPECIES DE SUDAMÉRICA CAPÍTULO III: White flowers in South America: a new species of Antiphytum (Echiochiloideae, Boraginaceae) Nidia Mendoza-Díaz, Marina Díaz, Patricia Brussa, Fabián Muñoz, José M. Bonifacino & Hilda Flores-Olvera Systematic Botany (2020) 45(4): 913–920. Systematic Botany (2020), 45(4): pp. 913–920 © Copyright 2020 by the American Society of Plant Taxonomists DOI 10.1600/036364420X16033963649291 Date of publication December 8, 2020 White Flowers in South America: A New Species of Antiphytum (Echiochiloideae, Boraginaceae) Nidia Mendoza-Dı́az,1 Marina Dı́az,2 Patricia Brussa,3 Fabián Mu~noz,2 José M. Bonifacino,4 and Hilda Flores-Olvera5,6 1Posgrado en Ciencias Biológicas, Instituto de Biologı́a. Universidad Nacional Autónoma de México, 3er. Circuito Exterior s/n, C.P. 04510 Ciudad de México, México 2Museo y Jardı́n Botánico de Montevideo, Av. 19 de Abril 1181, C.P. 11700 Montevideo, Uruguay 3Grupo Biodiversidad y Ecologı́a de la Conservación, Instituto de Ecologı́a y Ciencias Ambientales, Facultad de Ciencias, Universidad de la República, Iguá 4225, C.P. 11400 Montevideo, Uruguay 4Laboratorio de Botánica, Facultad de Agronomı́a, Universidad de la República, Av. Garzón 780, C.P. 12900 Montevideo, Uruguay 5Departamento de Botánica, Instituto de Biologı́a. Universidad Nacional Autónoma de México, Apartado Postal 70-233, C.P. 04510 Ciudad de México, México 6Author for correspondence (mahilda@ib.unam.mx) Communicating Editor: Chuck Bell Abstract—Antiphytum charruasorum, a new endemic species fromwestern Uruguay, is described. This is the only South American species in the genus with white corolla and an infra-medial cicatrix on the ventral face of the eremocarp, both features shared with the North American species of Antiphytum, in sharp contrast with the blue corolla and the basal cicatrix at the end of a stipe-like prolongation that characterizes all South American species in the genus.Antiphytum charruasorum increases the diversity of the genus in South America, as well as the endemism and distribution range into Uruguay. Moreover, this new species adds novel features in Antiphytum such as the irregularly dichasial inflorescences, the hirsute faucal appendages, the eremocarps with a tissue plug from the gynobase, and the habitat where the species has been reported. Keywords—Endemism, grasslands, North America, Uruguay. Antiphytum DC. ex Meisn. (Boraginaceae, Echiochiloideae Weigend; see Chacón et al. 2016) is an American disjunct genus with 11 species, nine in North America, two of them recently described to Mexico (Mendoza-Dı́az et al. 2018, 2020), and two endemic to South America (Johnston 1927). The known South American species of Antiphytum are subshrubs restricted to Uruguay and adjacent Brazil (Johnston 1927; Mendoza-Dı́az 2015),which differ from theirNorthAmerican relatives by having blue corollas, opposite leaves throughout, and flat gynobases forming a basal cicatrix (attachment scar) on the nutlets (eremo- carps sensuHilger 2014) that are basally attached by a short stipe- like prolongation to the gynobase (Johnston 1923). In contrast, the North American species have white corollas, mostly alternate leaves, and eremocarps directly attached to a short-pyramidal gynobase that leaves a medial to suprabasal cicatrix on the ere- mocarp ventral face (Johnston 1923; Mendoza-Dı́az 2015). During fieldwork carried out in Paysandú Department in western Uruguay we discovered a hitherto unknown species ofAntiphytum.Our objective is to describe this as a new species for which we propose Antiphytum charruasorum; we present illustrations and offer comments on its morphology and on the geographically and morphologically comparable species within the genus. Materials and Methods PlantMaterial and Fruit Observation—We discovered the new species in December 2016 in El Pucará plantation forest farm, in Uruguay´s Paysandú Department, as part of the diversity evaluation ran by the UPM Forestal Oriental company. Additional collections were later obtained in 2017 and 2018 in the same known locality in order to have a broader sampling of themorphological variation. During 2019we discovered a new population in Los Cobres farm, Rı́o Negro Department, Uruguay. We obtained herbarium specimens and silica gel dried material for phyloge- netic studies (Mendoza-Dı́az in preparation). Mature and immature fruits of the new species and of morphologically related species were studied with a confocal stereo microscope (Leica Z16 APO A, Leica Microsystems GmbH, Wetzlar, Germany), and photo- graphed with an attached camera (Leica DFC490, Leica Microsystems GmbH) at the Laboratorio deMicroscopı́a y Fotografı́a de la Biodiversidad (II) of the Instituto de Biologı́a of Universidad Nacional Autónoma de México (IBUNAM). Fruit samples were also mounted on stubs with common double stick carbon tape, and coated with gold to 20 mÅ for two minutes in a sputter coater (Quorum Q150OR ES, Quorum Technologies Ltd, East Sussex, UK) to be studied with a scanning electron microscope (Hitachi SU1510, Hitachi Ltd., Tokyo, Japan) at the Laboratorio de Microscopı́a Electrónica y Fotografı́a de la Biodiversidad (I) of IBUNAM. The new species was compared to the known species of the genus. We are using morphological operational criteria for species delimitation. The taxonomic hypothesis is corroborated using geographical, morphological, and ecological data that “break out” of the taxonomic circle (DeSalle et al. 2005). Distribution Map—A distribution map was produced using QGIS Brighton (2.6.1) employing layers from Agencia de Gobierno Electrónico y Sociedad de la Información y del Conocimiento (AGESIC 2019), Dirección Nacional de Medio Ambiente (DINAMA 2019), Dirección General de Recursos Naturales (DGRN 2019), Infraestructura de Datos Espaciales del Ministerio de Transporte y Obras Públicas (IDE MTOP 2019), Instituto Nacional de Estadı́sticas (INE 2011), and the collections coordinates. Taxonomic Treatment Antiphytum charruasorum N.Mend. & Marina Dı́az, sp. nov. TYPE: URUGUAY. Paysandú: Municipio Quebracho, UPM Forestal Oriental “El Pucará” plantation forest farm, 50 m, 31°57002.080S, 57°54048.060W, 15November 2017, F.Mu~noz with P. Brussa, C. Brussa and M. Dı́az (Holotype: MVJB! MVJB30748; Isotype: MVJB! MVJB30749). Antiphytum charruasorum is unique in the genus by irregu- larly dichasial inflorescences, by possessing hirsute-pubescent corolla faucal appendages, and the eremocarp is detached with a tissue plug from the gynobase, leaving a hole in the areola after detachment. Suffrutex erect, 15–18(–20) cm tall, stems (mainly 3–4) arising from a subterraneous stem, branches opposite, solid (non-fistulose), slender (ca. 1.5mmdiameter), later exfoliating, 913 densely strigose, each trichome with several mineralized foot cells (lithocystic structure), eglandular. Leaves basal, pseu- dobasal, and cauline, decussate, sessile, blades linear to nar- rowly elliptic, bases truncate, connate-sheathing, apex acute, mucronate, margins flat, entire, surfaces strigose concolorous, midvein not visible on the adaxial side, visible and raised on the abaxial side, secondary veins not visible; basal leaves marcescent, not forming a rosette, up to 40.0 3 2.0–2.8 mm; cauline leaves 18.0–38.0 3 1.0–2.0 mm, gradually smaller to- wards the apex; pseudobasal leaves at the beginning of each Fig. 1. Antiphytum charruasorum. A. General appearance. B. Lateral view of flower. C. Superior view of flower. D. Open corolla. E. Lateral view of immature fruit. F. Lateral view of fruit. G. Mature fruit with several eremocarps removed. H. Eremocarp in ventral view. I. Eremocarp in lateral view. Drawing by Javier Lage. SYSTEMATIC BOTANY [Volume 45914 branch, the youngest ones ca. 5.0 3 0.5 mm. Inflorescences ending in a flower (monotelic) subtending by opposite para- clades, forming a terminal heterocladic determinate thyrse (thyrsoid) with dichasial branching, in some cases twice dichasial, and sometimes one side as a new branch with dichasial branching at the distal nodes, first internode of a lateral axis 14.0–15.0 mm in length; bracts subtending each terminal dichasial flower 2, 23.5–33.03 1.7–2.0 mm, gradually smaller upward along the cyme, linear to narrowly elliptical, sessile, apex acute, nearly mucronate, strigose. Flowers Fig. 2. Antiphytum charruasorum in the field (Brussa et al. MVJB30747;Mu~noz et al. MVJB30748;Mu~noz and Dı́az MVJB31567, MVJB31669;Mendoza-Dı́az 649 et al.). A. Habitat. B. Soil detail; individual of A. charruasorum near a water body marked with a white arrow. C. Individual showing the subterranean stem. D. View from above of the general appearance. E. Portion of the inflorescence bearing flowers and immature fruits. F. Flower. Photos A, B, C by Nidia Mendoza-Dı́az; E by Patricia Brussa; D, F by Fabián Mu~noz. MENDOZA-DÍAZ ET AL.: A NEW SOUTH AMERICAN SPECIES OF ANTIPHYTUM 9152020] perfect, pedicel (–1.5)2.0–3.5 mm in length; calyx strigose on both sides, with scattered hispid trichomes on abaxial surface and numerous hispid trichomes adaxially, accrescent, divided almost to the base, lobes 3.0–4.0 3 1.2–1.4 mm, triangular, narrowly asymmetric, acute, all longer or equal than the co- rolla tube at anthesis, up to 5.6 mm in length in fruit; corolla hypocrateriform (rotate), white, with five yellow hirsute- pubescent faucal appendages on the throat in the sinuses of the corolla lobes, tube 2.7–3.0 mm in length, 1.0 mm diameter at the base, not surpassing the calyx, yellow, limb 7.5–8.0 mm in diameter, lobes 2.5–3.03 2.5 mm oriented perpendicular to the tube, imbricate, oblong, rounded sinuate at apex, abaxially strigose at the middle; stamens included, alternate to the co- rolla lobes, adnate at ca. 1.8 mm from the base of the corolla tube; filaments up to 0.5mm in length, slender; anthers oblong, sub-lanceolate, dorsifixed, ca. 0.8 mm long; style gynobasic, 1.5–2.0 mm in length at anthesis, stigma bilobed at the apex of the style (terminal), included, ovules 4. Fruit dry, subtending by a spreading calyx, 4 mature eremocarps; eremocarps 2.2–2.5 3 1.8–2.0 mm, ovate and apically acute-acuminate in ventral view, dorsally convex, lustrous, striate-tuberculate and papillate under the external layer; cicatrix ca. 1.0 mm long, at infra-medial position on the ventral face, ovate to triangular, with a tissue projection without pericarp of ca. 1.0 mm in length resembling a stipe; one ventral keel (apical) straight from the apex of the eremocarp to the cicatrix apex, not forming a stipe, lateral (basal) keels absent; gynobase very shallowly triangular (ca. 1:3), with four big and sunken areoles corresponding to the cicatrix projection of each eremocarp, style plus stigmas 1.8–2.0 mm long, not surpassing the ere- mocarps. Figures 1, 2. Etymology—The specific epithet honors the Charrúas, an indigenous community of Uruguay that inhabited the land before Spanish colonization. Nowadays,Charrúas is used as an adjective that defines Uruguayans’ spirit of courage, fight, and persistence. Phenology—Flowers and fruits were observed in October, November, and December, although probably the flowering period starts earlier. Distribution and Habitat—Antiphytum charruasorum is only known from two disjunct localities from the West Sedi- mentary Basin ecoregion (Brazeiro et al. 2015) in Uruguay, in the Paysandú and Rı́o Negro Departments at less than 50 km east of Rı́o Uruguay (Fig. 3). Populations are very small and individuals occur in sparse grasslands in well drained cal- careous soils (José F. Ramos, UPM senior soil specialist, pers. comm.; División de Suelos y Aguas 2002). Both locations have soils developing over sedimentary parent rock deposited during a desertic continental environment from Mercedes Formation (Bossi 1966; Bossi et al. 1975). The first populationwas found in El Pucará plantation forest farm, Paysandú Department, municipio Quebracho, in grasslands over flood plains together with Eragrostis lugens Nees, Bouteloua megapotamica (Spreng.) Kuntze, and Adesmia incana var. grisea (Hook. f.) Burkart (Fig. 2). The new species occurs at elevations of 50 m amid grasses, forming clusters of scattered individuals, in a narrow strip of land of 10 m wide, along small slopes. The second population occurs in the Rı́o Negro Department, Los Cobres farm, near El Ombú, 100 km south from the first locality, on rocky hillsides with steep slopes (. 6%) over calcareous sandstone (División de Suelos y Aguas 2002). This second population extends towards a Fig. 3. Distribution map of A. charruasorum (white dots) in Uruguay. A. Paysandú and Rı́o Negro Departments where the species occurs; note in light gray the respective municipality and land section of the collections. B. Municipio Quebracho in Paysandú Department; in darker gray the plantation forest farm El Pucará is highlighted. C. Part of the 11va Seccional Policial El Águila in Rı́o Negro Department in light gray; in darker gray the farm Los Cobres where the collection was done is highlighted. SYSTEMATIC BOTANY [Volume 45916 Fig. 4. Characteristic fertile structures of A. charruasorum. A. Mature gynobase of A. charruasorum in lateral view. B. Oblique view from above of the mature gynobase of A. charruasorum; note the sunken areolas. C. Detail of a faucal appendage showing the dense hirsute indumenta. D. Immature fruit of A. charruasorum, note the very shallowly triangular gynobase with the full areola. E. Mature flat gynobase of A. cruciatum. F. Mature deltate gynobase of A. heliotropioides. G. Comparison of A. charruasorum immature eremocarp with broken projection (left; Mendoza-Dı́az 649 et al.), with respect to its mor- phologically and geographically comparable species A. cruciatum (in the middle, note the clear stipe in lateral view; Mendoza-Dı́az 651 and Santos de Mello) and A. heliotropioides (right; Mendoza-Dı́az 623 et al.). H. Mature eremocarp of A. charruasorum (Mendoza-Dı́az 648 et al.) in lateral, ventral, and basal views. Note the length of the scar projection. I. Pericarp ornamentation seen in stereoscopic microscopy; note the papillas. J. Pericarp ornamentation seen in SEM. MENDOZA-DÍAZ ET AL.: A NEW SOUTH AMERICAN SPECIES OF ANTIPHYTUM 9172020] watercourse in a grassland characterized by Bouteloua mega- potamica. No other Antiphytum populations are reported from Paysandú nor Rı́o Negro Departments. Adding to the mor- phological singularity of this discovery, the habitat of A. charruasorum is unusual for the entire genus, since Anti- phytum species occur mainly in xeric habitats. Conservation Status—Antiphytum charruasorum is only known from two localities, both in farms near Eucalyptus plantations. According to the IUCN (2012), this species meets the criteria B2ab (iii, v) and we recommend that if a formal assessment were performed that it would be categorized as Critically Endangered (CR). Identification Key to the South American Species of Antiphytum 1. Inflorescences with only the terminal flower of each dichasial branching; corollas white; eremocarp cicatrix infra-ventral. . . . . . A. charruasorum 1. Inflorescences with flowers in scorpioid cymes; corollas blue; eremocarp cicatrix basal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Leaves 15.0–30.0 3 3.0–6.0 mm, oblanceolate; eremocarps 1.3–1.4 3 1.0–1.1 mm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. cruciatum 2. Leaves 25.0–50.0 3 0.15–1.0 mm, linear; eremocarps 1.8–2.0 3 1.4–1.6 mm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. stoechadifolium Discussion The discovery of Antiphytum charruasorum breaks down the taxonomicknowledgeof thegenus inSouthAmericadue to its very shallowly triangular gynobase, the absence of a stipe, the cicatrix on the ventral face of the eremocarp in infra-medial position, and the white corollas. Prior to the discovery ofA. charruasorum, therewas a clear morphological pattern for the North and South American species of Antiphytum (Johnston 1923; Mendoza-Dı́az 2015). The SouthAmerican speciesweredistinguishableby the combinationof opposite leaves, flat gynobases, basal eremocarp cicatrix at the end of a stipe-like projection of the ventral keel, and blue corollas. The North American species were characterized by leaves mostly al- ternate, short-pyramidal gynobases, ventral cicatrixwithout a stipe- like projection, and white corollas. Since the gynobase and the eremocarp shape are basic in the taxonomy of Boraginaceae (Weigend et al. 2016), the differences between the North American and South American species of Antiphytum were used to hint (Johnston 1927) or even propose their recognition as two separate genera (Brand 1929, 1931), but without “fundamental characters which set off the Mexican species from the true Antiphyta of South America” (Johnston 1927, p. 12). Johnston (1923) already knew that opposite leaves occur in North American species at least basally on the stem, but opposite leaves along the stemwere described in the NorthAmerican speciesA. hintoniorumL.C.Higgins andB.L.Turner (Higgins andTurner 1983).Moreover, flat gynobases sometimesare observed inA. floribundum (Torr.) A.Gray (Brand 1929; NMD pers. obs.) from western Mexico and southern US. Thenewspecies ismorphologically closer to theNorthAmerican species A. heliotropioidesA.DC. and A. geoffreyiN.Mend. and Flores Olv. with which it shares infra-medial position of the cicatrix and theabsenceof abasal leaf rosette.Nevertheless, both species and the known South American species, A. cruciatum (Cham.) DC. and A. stoechadifolium (Cham.) DC., have the flowers arranged in scorpioid cymes as the basic unit of the inflorescences. In sharp contrast, A. charruasorum only bears the terminal flower of each dichasial branching. The presence of trichomes on the faucal ap- pendages and the tissue projection of the cicatrix, with the corre- sponding sunken areolas in the gynobase, are features not previously reported from Antiphytum. Moreover, the entire fruit and individual eremocarps reach the largest size in thegenus (Fig. 4; Table 1). Eremocarps have a peg-shaped projection of gynobase tissue that is common inalmostallBoragineaeand insomeLithodora Table 1. Main morphological characters to distinguishAntiphytum charruasorum from the geographically1 andmorphologically2 comparable species of the genus. A. charruasorum A. cruciatum1 A. geoffreyi2 A. heliotropioides2 A. stoechadifolium1 Cauline leaf blades (mm) Linear to narrowly elliptic (18.0–38.0 3 1.0–2.0) Oblanceolate (15.0–30.0 3 3.0–6.0 mm) Linear-oblanceolate to linear-oblong (8.0–10.0 3 1.0) Linear-oblanceolate to oblanceolate (10.0–60.0 3 1.5–8.0) Linear (25.0–50.0 3 0.15–1.0) Flower arrangement Only the terminal flower of each dichasial branching Scorpioid cymes Scorpioid cymes Scorpioid cymes Scorpioid cymes Corolla diameter (mm) 7.5–8.0 4.0–5.0 4.7–5.0 4.0–6.0 5.0–7.0 Corolla color White Blue White White Blue Faucal appendage surface Densely hirsute Papillose Papillose Papillose Papillose Eremocarp ventral keel Not forming a stipe Forming a stipe Not forming a stipe Not forming a stipe Forming a stipe Eremocarp cicatrix position Infra-ventral Basal Infra-ventral Infra-ventral Basal Cicatrix tissue Forming a long projection Not forming a projection Not forming a projection Not forming a projection Not forming a projection Eremocarp size (mm) 2.2–2.5 3 1.8–2.0 1.3–1.4 3 1.0–1.1 1.6–1.7 3 1.3–1.4 (–1.3)1.5–1.7 3 (–0.9)1.1–1.2 1.8–2.0 3 1.4–1.6 Gynobase shape (length:width) Very shallowly triangular (ca. 1:3) Flat Triangular (2:1) Deltate (ca. 1:1) Flat Areola tissue Forming part of the eremocarp projection Remaining in the gynobase Remaining in the gynobase Remaining in the gynobase Remaining in the gynobase Areolas Sunken in mature fruits Solid in mature fruits Solid in mature fruits Solid in mature fruits Solid in mature fruits SYSTEMATIC BOTANY [Volume 45918 Griseb. and Myosotis L. (Hilger 2014; Weigend et al. 2016). De- tachment of the mature eremocarp occurs when an abscission or separation zone forms between this structure and the gynobase (Roth 1977), while removal from the immature fruit breaks off this “plug tissue” and makes evident the very shallowly triangular gynobase shape as well as the area and position of the cicatrix (Fig. 4). At least in Boragineae and Myosotis, the peg-shaped plug is a food body serving in myrmecochory (elaiosome) (Hilger 2014; Weigend et al. 2016). Further studies are needed to better under- stand the role and development of this structure inA. charruasorum where ants’ visits have been documented (Fig. 2F). The habitat and distribution of Antiphytum charruasorum are novelties for the genus since the new species grows near to a water body on wet soils with a high content of calcium car- bonate. The remaining species grow in dry places and some on calcareous soils in North America. No other localities of Antiphytum in Paysandú nor Rı́o Negro Departments have been reported, which are nearer to the Argentina border than to southern Brazil and northernUruguay.A. charruasorum thus represents an isolated outpost of the genus (Fig. 3). In short, the new species represents an addition to Anti- phytum that adds novel characters to the genus and removes themorphological distinction of fertile structures (corolla color and eremocarp cicatrix position) between the South andNorth American species, expands the range of ecological conditions under which Antiphytum has been reported, as well as its distribution area and diversity in South America, and adds a new endemic species to Uruguay. Paratypes—Uruguay. —PAYSANDÚ: Municipio Quebracho, Forestal Oriental “El Pucará” plantation forest farm, 50 m, 31°57003.50S, 57°54048.40W, 14 December 2016, Brussa et al. (MVJB! 30747); 48 m, 31°57003.70S, 57°54048.30W, 07 December 2017, Mendoza-Dı́az 648 et al. (MEXU!, MVFA!); 50 m, 31°57002.90S, 57°54046.20W, 07 December 2017, Mendoza-D ı́az 649 et al. (MEXU!, MVFA!); 50 m, 31°57003.330S, 57°54047.880W, 18 October 2018,Mu~noz and Dı́az (MVJB! MVJB 31567). RÍO NEGRO: 11va Seccional Policial, “Los Cobres” farm, “El Ombú” place, 31 m, 32°53045.140S, 57°31047.730W, 19 October 2019, Mu~noz and Dı́az (MVJB! 31669). Additional Specimens Examined (Other Species)—Antiphytum cru- ciatum. Brazil. —RIO GRANDE DO SUL: Porto Alegre, Morro S~ao Pedro, 173 m, 30°10030.70S, 51°06042.90W, 14 December 2017, Mendoza-Dı́az 651 and Santos de Mello (MEXU!, MVFA!). Antiphytum floribundum. Mexico. —MICHOACÁN: Municipio Morelia, La Alberca de Teremendo de los Reyes, 2194 m, 19°48011.10N, 101°26058.80W, 09 Oct 2013, Contreras J. 52 (MEXU!). Antiphytum geoffreyi. Mexico. —COAHUILA: Municipio Francisco I. Madero, west side of Sierra de Tlahualilo, 1275 m, 26°36046.80N, 103°210130W, 21 April 2015, Hinton 29613 et al. (MEXU!). Antiphytum heliotropioides. Mexico. —QUERÉTARO: Municipio Landa de Matamoros, 43 km (on road) east of Landa de Matamoros, on the road Jalpan de Serra- Xilitla, 1570 m, 21°16022.70N, 99°12019.40W, 13 September 2014, Mendoza- Dı́az 623 et al. (MEXU!).—NUEVO LEÓN: Municipio Rayones, road Galeana- Rayones, 6 km N of Los Nogales and 17 km N of Galeana, 1360 m, 24°55043.70N, 100°03051.10W, 24 September 2016, Mendoza-Dı́az 629 et al. (MEXU!). Antiphytum hintoniorum. Mexico. —NUEVO LEÓN: Municipio Rayones, road Galeana-Rayones, 6 km N of Los Nogales and 17 km N of Galeana, 1348–1360 m, 24°55038.70N, 100°03052.20W, 24 September 2016, Mendoza-Dı́az 630 et al. (MEXU!). Acknowledgments This paper arose as collaboration between the staff of MVFA, MVJB in Uruguay and IBUNAM, Mexico. MD, FM, and PB thank Ing. Agr. Iván Grela, Manager of UPM Sustainable Plantations of Paysandú, for allowing fieldwork in “El Pucará” plantation forest farm and for his great interest on this study. MD, FM, and PB also thank Ing. Agr. José Ramos, UPM’s senior soil specialist in Uruguay, for the complete soil report of “El Pucará” plantation forest farm, and Gustavo Mazziotto and family, owners of “Los Cobres” farm for allowing us to conduct fieldwork at their property. Special thanks go to Federico Haretche, curator of MVJB, for his support while studying MVJB collections. Special thanks to Carlos Brussa who was part of the team that discovered the species, and who offered valuable comments, fieldwork assistance, and photographs. We also thank to Javier Lage for the line drawings. We thank Berenit Mendoza-Garfias and Susana Guzmán-Gómez from IBUNAM for their assistance with the SEM and stereoscopic microscope respectively. NMD especially thanks Anderson Santos de Mello, from Universidade Federal do Rio Grande do Sul, for his invaluable help and assistance during fieldwork in Brazil. NMD is a re- cipient of a doctoral fellowship (CONACYT grant 288698) and received support from Posgrado en Ciencias Biológicas, UNAM (PAEP 2017) to fund part of the travel to Uruguay and Porto Alegre, Brazil. Author Contributions Nidia Mendoza-Dı́az led manuscript preparation. Marina Dı́az and Fabián Mu~noz contributed to the manuscript writing. Patricia Brussa helped with fieldwork, map elaboration and description. José M. Bonifacino and Hilda Flores-Olvera contributed to manuscript preparation. Literature Cited AGESIC. 2019. Catálogo de Datos Abiertos. República Oriental de Uru- guay. https://catalogodatos.gub.uy/dataset/direccion-nacional-de- catastro-shapes-del-parcelario-rural-y-urbano/resource/fea63a36- 3016-4492-b2d1-92099467f5a2 (accessed 05 Nov 2019). Bossi, J. 1966.Geologı́a delUruguay.Montevideo:DepartamentodePublicaciones de la Universidad de la República. Bossi, J., L. Ferrando, A. Fernández, G. Elizalde, H. Morales, J. Ledesma, E. Carballo, E. Medina, I. Ford, and J. Monta~na. 1975. Carta geológica del Uruguay 1/1000000. Montevideo: Dirección de Suelos y Fertilizantes, M.A.P. Brand, A. 1929. Verweisung des Gattungsnamens Antiphytum in die Synonymie. Feddes Repertorium Specierum Novarum 27: 145–149. Brand, A. 1931. Borraginaceae-Borraginoideae-Cryptantheae. Das Pflan- zenreich Regni Vegetabilis Conspectus 97: 1–236. Brazeiro, A., D. Panario, A. Soutullo, O. Gutiérrez, A. Segura, and P. Mai. 2015. Identificación y delimitación de eco-regiones de Uruguay. Pp. 46–59 in Eco-regiones de Uruguay: Biodiversidad, Presiones y Con- servación. Aportes a la Estrategia Nacional de Biodiversidad. Montevideo: Facultad de Ciencias, CIEDUR, VS-Uruguay, SZU. Chacón, J., F. Luebert, H. H. Hilger, S. Ovchinnikova, F. Selvi, L. Cecchi, C. M. Guilliams, K. Hasenstab-Lehman, K. Sutorý, M. G. Simpson, and M. Weigend. 2016. The borage family (Boraginaceae s.str.): A revised infrafamilial classification based on new phylogenetic evi- dence, with emphasis on the placement of some enigmatic genera. Taxon 65: 523–546. DeSalle, R., M. G. Egan, and M. Sidall. 2005. The unholy trinity: Tax- onomy, species delimitation and barcoding. Philosophical Transac- tions of the Royal Society of London. Series B, Biological Sciences 360: 1905–1916. DGRN. 2019. Biblioteca Digital. Carta de Suelos. Ministerio de Ganaderı́a, Agricultura y Pesca.DirecciónGeneral deRecursosNaturalesRenovables. República Oriental de Uruguay. http://www.mgap.gub.uy/unidad- organizativa/direccion-general-de-recursos-naturales/tramites-y-servicios/ biblioteca-digital/carta-de-suelos (accessed 05 Nov 2019). DINAMA. 2019. Geoservicios. Ministerio de Vivienda, Ordenamiento Territorial y Medio Ambiente. República Oriental de Uruguay. https://www.dinama.gub.uy/geoservicios/ (accessed 05 Nov 2019). División de Suelos y Aguas. 2002. Carta de Reconocimiento de Suelos. Departamento de Rı́o Negro. Ministerio de Ganaderı́a, Agricultura y Pesca. Dirección General de Recursos Naturales Renovables. República Oriental de Uruguay. http://www.mgap.gub.uy/sites/ default/files/multimedia/1634_1419250023cartarn_0.pdf (accessed 27 Oct 2019). Higgins, L. and B. L. Turner. 1983. Antiphytum hintoniorum (Boraginaceae), a bizarre new gypsophile fromNorth-Mexico. The SouthwesternNaturalist 28: 457–458. Hilger, H. H. 2014. Ontogeny, morphology, and systematic significance of glochidiate and winged fruits of Cynoglosseae and Eritrichieae (Boraginaceae). Plant Diversity and Evolution 131: 167–214. IDEMTOP. 2019. GeoPortal. https://geoportal.mtop.gub.uy/geoservicios (accessed 05 Nov 2019). INE. 2011. Mapas Vectoriales A~no 2011. República Oriental de Uruguay. http://ine.gub.uy/mapas-vectoriales (accessed 05 Nov 2019). IUCN. 2012. IUCN red list categories and criteria, version 3.1. Ed. 2. Gland, Switzerland and Cambridge, UK: IUCN. MENDOZA-DÍAZ ET AL.: A NEW SOUTH AMERICAN SPECIES OF ANTIPHYTUM 9192020] Johnston, I.M. 1923. Studies in the Boraginaceae.Contributions from the Gray Herbarium of the Harvard University 68: 43–80. Johnston, I. M. 1927. Studies in the Boraginaceae VI: A revision of the South American Boraginoideae. Contributions from the Gray Herbarium of Harvard University 78: 3–118. Mendoza-Dı́az, N. 2015. Sistemática de Antiphytum DC. ex Meisn. (Bor- aginaceae). M.S. thesis. Ciudad de México: UNAM. Mendoza-Dı́az, N., H. Flores-Olvera, M. G. Simpson, and M. J. Moore. 2018. A new and unusual endemic species from the Chihuahuan Desert, Mexico: Antiphytum geoffreyi (Boraginaceae, Echiochiloideae). Phytotaxa 367: 275–283. Mendoza-Dı́az,N.,H.Ochoterena,M. J.Moore, andH. Flores-Olvera. 2020. Molecular and morphological evidence reveals a new species of Antiphytum (Echiochiloideae, Boraginaceae) from Guerrero, Mexico. Systematic Botany 45: 190–199. Roth, I. 1977. Fruits of Angiosperms. Berlin-Stuttgart: Gebruder Borntraeger. Weigend, M., F. Selvi, D. C. Thomas, and H. H. Hilger. 2016. Boraginaceae. Pp. 41–102 in The Families and Genera of Vascular Plants, vol. 14, eds. J. W. Kadereit and V. Bittrich. Cham, Switzerland: Springer International Publishing. SYSTEMATIC BOTANY [Volume 45920 48 PARTE I: REEVALUACIÓN TAXONÓMICA DE LAS ESPECIES DE ANTIPHYTUM PARTE A. ESPECIES DE SUDAMÉRICA CAPÍTULO IV: A taxonomic reevaluation of the South American Myosotis species described by José Arechavaleta Nidia Mendoza-Díaz, José M. Bonifacino, Marina Díaz & Hilda Flores-Olvera Novon (2022) 30: 80–91. VERSION OF RECORD FIRST PUBLISHED ONLINE ON 19 AUGUST 2022. doi: 10.3417/2022743 NOVON 30: 80–91. A Taxonomic Reevaluation of the South American Myosotis Species Described by José Arechavaleta Nidia Mendoza-Díaz,1 José M. Bonifacino,2 Marina Díaz,2 and Hilda Flores-Olvera3* 1 Posgrado en Ciencias Biológicas, Instituto de Biología, Universidad Nacional Autónoma de México, 3er. Circuito Exterior s/n, C.P. 04510, Ciudad de México, Mexico. 2 Laboratorio de Botánica, Facultad de Agronomía, Universidad de la República, Av. Garzón 780, Montevideo, C.P. 12900, Uruguay. 3 Departamento de Botánica, Instituto de Biología, Universidad Nacional Autónoma de México, Apartado Postal 70-233, C.P. 04510, Ciudad de México, Mexico. * Author for correspondence: mahilda@ib.unam.mx ABSTRACT. In Flora Uruguaya, José Arechavaleta de- scribed two Uruguayan species of Myosotis L. (Borag- inaceae) that I. M. Johnston later placed within Antiphy- tum DC. ex Meisn. Johnston considered M. uruguayensis Arechav. to be a doubtful species that was probably in Antiphytum, apparently nearest to A. cruciatum (Cham.) DC., whereas he treated M. berroi Arechav. as a syn- onym of A. stoechadifolium (Cham.) DC. We studied the type specimens and protologues of these two names in Myosotis. We conclude that M. berroi belongs to An- tiphytum but is not conspecific with A. stoechadifolium based on eremocarp features, such as pericarp orna- mentation and size, together with shorter leaves. We provide an updated circumscription and distribution, publish the new combination A. berroi (Arechav.) N. Mend. & Flores Olv., and clarify the type. A taxonomic key for the South American species of Antiphytum is provided. Myosotis uruguayensis does not belong to Antiphytum due to its shallowly divided calyx entirely covering the immature eremocarps; it retains its uncer- tain taxonomic status due to the scanty and immature original material and inadequate protologue. Key words: Antiphytum, Boraginaceae, Brazil, Myo- sotis, new combination, Uruguay. The genus Antiphytum DC. ex Meisn. (Boraginaceae subfamily Echiochiloideae; see Chacón et al., 2016) was published by Meisner (1840), containing two South American species previously described by Chamisso (1829) as Anchusa cruciata Cham. and Anchusa stoechadifolia Cham. Both species were transferred to Antiphytum by de Candolle (1846), i.e., Antiphytum cruciatum (Cham.) DC. and Antiphytum stoechadifo- lium (Cham.) DC. These species were later treated as heterotypic synonyms by Johnston (1923), who lecto- typified the genus with Antiphytum cruciatum. In a subsequent work, Johnston (1927) accepted both spe- cies as the only two well-supported South American species of Antiphytum. At the same time, Johnston (1927) also cited for the first time in the literature the two species described by José Arechavaleta, a Spanish botanist based in Uruguay, under Myosotis L. in the Flora Uruguaya (Arechavaleta, 1898–1911): M. berroi Arechav. and M. uruguayensis Arechav. Johnston (1927) treated the former as a heterotypic synonym of Anti- phytum stoechadifolium and the latter as probably a species of Antiphytum, apparently nearest to Antiphy- tum cruciatum. However, he did not transfer the name to Antiphytum. In the meantime, between Arechavale- ta’s and Johnston’s publications, no work dealing with Boraginaceae referred to Flora Uruguaya nor the Myo- sotis species there described. As part of the taxonomic revision of the genus Anti- phytum currently underway by the lead author, we re- viewed the treatment of the Boraginaceae in the Flora Uruguaya (Arechavaleta, 1898–1911) with a focus on the two species of Myosotis as possible species of Anti- phytum. In parallel, we revisited Johnston’s hypothesis on Arechavaleta’s species. We searched for the types in Uruguayan and international herbaria where Arecha- valeta’s specimens may be, according to Stafleu and Cowan (1976–1988) in the second edition of Taxo- nomic Literature (TL-2), checked the protologues, and examined specimens to clarify the taxonomic affinities of these names. TAXONOMIC HISTORY Flora Uruguaya (Arechavaleta, 1898–1911) is one of the first botanical works published in South Amer- ica together with Flora de Chile (Reiche, 1896–1911). In Flora Uruguaya, Boraginaceae s. str. is only repre- sented by Myosotis, for which Arechavaleta described Volume 30 2022 Mendoza-Díaz et al. 81 Reevaluation of the Myosotis species of Arechavaleta M. berroi and M. uruguayensis (Arechavaleta, 1911a, 1911b). Myosotis berroi was named to honor Mariano B. Berro, a botanist who collected plants of the species in the department of Rivera in northern Uruguay. The collec- tion information on Berro 4975 (MVFA and MVM; ac- ronyms according to Thiers, 2022) perfectly matches the information in the protologue: “Vive en campos de Rivera en los llamados Cerros del Gobierno, pedregosos áridos. Florece en Diciembre. Hallado por el Sr. Berro a quien se lo dedico” (Arechavaleta, 1911a: 70). As annotated on the specimen at MVFA (Fig. 1), this final epithet was chosen after Arechavaleta initially named the specimen in sched. as “Myosotis rupestris n. sp. non Pall. nec Willd.” Myosotis rupestris had been used already by Pallas (published in Georgi, 1775) and even by Willdenow (Ledebour, 1847) as a later homonym. Arechavaleta later wrote a second label with the name that would eventually be validly published, along with the publication information (Fig. 1). Myosotis uruguayensis was also collected in northern Uruguay in the department of Tacuarembó in sand-clay soils (Arechavaleta, 1911a). It seems that the collec- tion was not made by Arechavaleta himself, although the protologue lacks a collector’s name and number. The name Pantaleón Pintos appears on the label of the only located original material (MVM; Fig. 2). Pan- taleón Pintos likely is the collector of this specimen, as in the case of Bambusa tacuara Arechav. as recognized in the protologue (Arechavaleta, 1897). No information regarding a specific locality is given either in Flora Uruguaya or on the label of the specimen; “Campos de Tacuarembó” is reported, but Tacuarembó covers a very large area. As in the case of M. berroi, the final epithet “uruguayensis” was chosen after two unavail- able epithets were considered, as the label shows. Before the publication of Flora Uruguaya, Meisner (1840) had published the genus Antiphytum, and de Candolle (1846) had formalized it. However, neither Antiphytum nor the two species originally described in Anchusa by Chamisso (1829) were mentioned in the Flora Uruguaya. Arechavaleta may not have been aware of the presence of Antiphytum in Uruguay, or he may have considered that the morphological features defining the genus did not match the Uruguayan spec- imens known to him. Johnston (1927) had access to Arechavaleta’s Flora Uruguaya when he published the revision of the South American “Boraginoideae.” However, Johnston did not have access to specimens, so his understanding of the two Myosotis species came from Arechavaleta’s de- scriptions alone. Based on the description and the type locality, Johnston (1927) considered M. berroi a hetero- typic synonym of Antiphytum stoechadifolium. Accord- ing to Johnston (1927), the German botanist Friedrich Sellow explored northern Uruguay during the early part of 1823. Because Sellow’s collections were used by Chamisso to describe Anchusa stoechadifolia (the basi- onym of Antiphytum stoechadifolium), Johnston wrote that, “it is, hence, not impossible that the types of Anchusa stoechadifolia and Myosotis Berroi were both obtained in northern Uruguay and perhaps in the same region” (Johnston, 1927: 13–14). On the other hand, Johnston (1927) considered M. uruguayensis a doubt- ful species of Antiphytum. He wrote that it is “a fruti- cose plant probably of this genus [Antiphytum] and apparently nearest to A. cruciatum but differing in its alternate leaves and tubular calyx. I have seen no ma- terial of this species” (Johnston, 1927: 14). Johnston’s description of Antiphytum cruciatum was thus adapted from the original description only (Johnston, 1927: 12). None of Johnston’s subsequent publications include South American species of Antiphytum or Uruguayan species of Myosotis. In an unpublished letter to Uru- guayan botanist Bernardo Rosengurtt dated 1938 (held at MVFA), Johnston considered M. berroi to be a third species of the genus Antiphytum based on observation of the specimen Rosengurtt 2346. We studied this spec- imen on loan from GH (see Additional specimens ex- amined). Although this specimen actually corresponds to A. cruciatum, Johnston annotated it as “Antiphytum Berroi (Arech.) Johnston” (Fig. 3B), this being the first time that this combination was used, although he never formally published it. No mention of M. uruguayensis was made in the above-mentioned letter. The taxonomic status of Myosotis species described by Arechavaleta therefore remains yet unresolved. Al- though both species were validly published, and the names are cited in current floristic works such as Catálogo de las Plantas Vasculares del Cono Sur (Zulo- aga et al., 2021), their circumscriptions have not been updated and the type specimens have not yet been examined. MATERIALS AND METHODS Literature and herbarium specimens of South Amer- ican species of Antiphytum, including protologues and types, were located and examined. To locate original material of the species of Myosotis described by Are- chavaleta (1898–1911) in Flora Uruguaya, we first searched Uruguayan herbaria MVFA and MVM for historic collections of Mariano Berro and José Arecha- valeta, who were both based in Uruguay and collected throughout the country in the late 1800s and early 1900s. We found one specimen of M. berroi each at MVFA and MVM, and one specimen of M. uruguay- ensis at MVM. Based on information in TL-2 (Stafleu & Cowan, 1976–1988), we then sought information from the following herbaria to determine whether they housed hi] o pa o 17] 19] = pr <= El E > 2 7 o Hd hora, RECHAVA AO | Pus de a Repática ,4) Cags AS y Mio sols DE 2 E | des, eh 2 ch sept zo < da rro. Note the 82 Novon Figure 1. Lectotype of Antiphytum berroi (Arechav.) N. Mend. & Flores Olv. at MVFA collected by Mariano Be . ote label with the basionym Myosotis berroi Arechav. and the data of publication annotated by José Arechavaleta. o Le) ho] < EN pd 1] 1 0) 3 < o [oh 3 E D 2 g E =. 9 == AS Esa | | | Herbario DE J. ARECHAVAL y Plantas de la República del Uruguay. | | musas ES | Myoscts frame. bel, Y oh a | ID. de ii E o y o O RAR hata Frclg | 5 ZA Volume 30 2022 Mendoza-Díaz et al. 83 Reevaluation of the Myosotis species of Arechavaleta Figure 2. Original material of Myosotis uruguayensis Arechav. at MVM collected by Pantaleón Pintos. 84 Novon original material from these species: B, BAF, CORD, F, G, K, L, LD, M, MIN, NY, PH, R, WRSL, and ZT. An- swers from most of the herbaria indicate that there is no more additional material out of Uruguay, but GH holds a phototype of the MVM specimen of M. berroi. As we found collections referable to M. berroi that were made in southern Brazil, the ICN herbarium was also checked for additional collections to upgrade taxonomic and geographic information of the species. Fieldwork was conducted in the departments of Tacuarembó and Rivera in Uruguay. Morphological characters, particularly from the fruit and the inflorescence (following Weberling, 1989) were used to differentiate and to describe the species: ere- mocarp number and size, pericarp ornamentation, cleft of the calyx lobes, kind of diaspora. Material from Myosotis berroi was available from type specimens and recent collections, including our own collections (see Additional specimens examined). Since there are no duplicates or additional collections of M. uruguayensis, it was not possible to make direct observation of the fruits. Mature eremocarps (nutlets or mericarps, termi- nology sensu Hilger, 2014) from our collected speci- mens of M. berroi and one specimen of Antiphytum stoechadifolium (Marchesi 1420) were observed with a confocal stereo microscope (Leica Z16 APO A, Leica Microsystems GmbH, Wetzlar, Germany), and photo- graphed with a camera (Leica DFC490, Leica Microsys- tems GmbH) at Laboratorio de Microscopía y Fo- tografía de la Biodiversidad II of Instituto de Biología, Universidad Nacional Autónoma de México (UNAM). SEM photographs were also taken (Hitachi SU1510, Hitachi Ltd., Tokyo, Japan) at Laboratorio de Micro- scopía Electrónica y Fotografía de la Biodiversidad I. A distribution map was made using QGIS (2.6.1- Brighton) (QGIS Development Team, 2014) and the GIS layers from official websites of Argentina (Instituto Geográfico Nacional de la República Argentina, 2020), Brazil (Fundação Estadual de Proteção Ambiental Henrique Luiz Roessler, 2020; Núcleo de Economía Regional e Urbana da Universidade de São Paulo, 2020; Serviço Geológico do Brasil, 2020), and Uru- guay (Infraestructura de Datos Espaciales del Ministe- rio de Transporte y Obras Públicas de Uruguay, 2020; Instituto Nacional de Estadística, 2020; Ministerio de Industria, Energía y Minería de Uruguay, 2020). For the delimitation of Serra Geral Formation, Brazil, we followed Wildner et al. (2004). Figure 3. Reference specimens. —A. Specimen of Antiphytum berroi (Arechav.) N. Mend. & Flores Olv. collected by Karner Hagelund in Rio Grande do Sul, Brazil (MO). Note the mention of his own herbarium. —B. Specimen of Antiphytum cruciatum (Cham.) DC. collected in Uruguay and annotated as “Antiphytum Berroi” by I. M. Johnston (GH). Volume 30 2022 Mendoza-Díaz et al. 85 Reevaluation of the Myosotis species of Arechavaleta The designation of types follows McNeill (2014) and Article 9 of the International Code of Nomenclature for algae, fungi, and plants (Turland et al., 2018). RESULTS We propose transferring Myosotis berroi to the genus Antiphytum, which would make it the fourth species native to South America, and the third with blue corol- las (see Mendoza Díaz et al., 2020). Features from the eremocarp, along with the cleft of the calyx lobes al- most to the base in the flower, and the extended calyx in the fruit that does not enclose the four eremocarps, are characteristic of Antiphytum and exclude A. berroi (Arechav.) N. Mend. & Flores Olv. from genera of Morit- ziinae. Antiphytum berroi comprises a well-defined species that differs from A. stoechadifolium by the er- emocarp ornamentation and leaf length. From the two syntypes found, the specimen at MVFA (M. B. Berro 4975) is designated as the lectotype. By contrast, based on the short protologue descrip- tion and the type found in MVM, we conclude that Myosotis uruguayensis does not belong in Antiphytum due to its shallowly divided calyx and the eremocarps that remain enclosed in the accrescent calyx. Myosotis uruguayensis, although valid, remains with uncertain taxonomic status since there are not enough features to reliably assign it to any genus of Boraginaceae. After our search to find duplicates, only the specimen at MVM is known. In the face of the uncertainty of the original gathering, it seems unnecessary to designate a lectotype according to McNeill (2014). TAXONOMIC TREATMENT Antiphytum berroi (Arechav.) N. Mend. & Flores Olv., comb. nov. Basionym: Myosotis berroi Are- chav., Anales Mus. Nac. Montevideo ser. 2, 1(3): 69. 1911. TYPE: Uruguay. Dpto. Rivera: “Vive en campos de Rivera en los llamados Cerros del Gobierno, pedregosos áridos,” M. B. Berro 4975 (lectotype, designated here, MVFA!; isolectotype, MVM!, MVM photo at GH!). Figure 4C–H. Suffrutex, erect, 25–35(–40) cm tall; stems (usually 3) broadly branching, arising from a main slender stem (ca. 2–3 mm diam.); branches opposite, solid (non- fistulose), slender (ca. 1 mm diam.), later exfoliating, densely strigose, each trichome with several mineral- ized foot cells (lithocystic structures). Leaves pseudo- basal and cauline, the cauline opposite, without a basal rosette, sessile; lamina linear-lanceolate or linear- oblong to linear-oblanceolate, in cauline leaves (10–)14.6– 20.9(–30) × 1–1.6(–2.3) mm, becoming gradually smaller on the branch distally, in youngest pseudo- basal leaves ca. 5 × 0.4 mm; base truncate, connate- sheathing; apex acute, mucronate; margins thickened, entire; surfaces strigose, with trichomes more densely and finely distributed on abaxial surface, concolorous; midvein not visible on adaxial surface, but groovelike at the very base, visible and raised on abaxial surface; secondary veins not visible. Inflorescences ending in a flower (monotelic) forming a terminal homocladic determinate thyrse (thyrsoid); most distal node with paired scorpioid cymes (boragoids) up to 170 mm when mature, loosely many-flowered, bracteate, with well- developed terminal flower arising from a dichasial division; previous nodes (until 3) often with opposite boragoids as paraclades of the thyrsoid, but sometimes one side on the distal nodes as a new flower branch; most proximal internode of a lateral axis 13–26 mm; bracts subtending the inflorescence and inflorescence nodes opposite, sessile, triangular-lanceolate, the larg- est 11.5 × 2.5–3 mm, becoming gradually smaller dis- tally, base truncate, apex acute, mucronate, margins hispid, indumenta and midvein as on leaves; flower- subtending bracts sessile, triangular-lanceolate, the largest 4.5 × 1.5–2 mm, becoming gradually smaller toward the apex of the cyme, apex mucronate, margins hispid, indumenta and midvein as on the leaves. Flow- ers monoclinous, pedicellate; pedicel 1.2–1.5 mm; calyx accrescent, divided ± to the base, strigose on both surfaces, scarious at the very base, calyx lobes ± symmetric, lanceolate, 2.2–3.5 × 0.5–1.5 mm, longer than or equal to corolla tube at anthesis, up to 4 mm in  fruit, apex slightly acuminate, mucronate in fruit, trichomes more densely distributed on abaxial surface, few amber-colored and hyaline glandular-capitate at margins, more at base; corolla hypocrateriform (rotate), blue, with 5 white trapezoid papillose faucal append- ages on throat in sinuses of corolla lobes, tube light blue, 2.2–2.7 mm, 1 mm wide at base, limb purple- blue, 6 mm diam., lobes oriented perpendicular to tube, imbricate, oblong, 2.1–2.3 × 1.6–2.2 mm, apex rounded-sinuate, abaxial surface densely pilose at mid- dle; stamens included, alternate to corolla lobes, adnate to corolla tube ca. 1.2 mm from base of tube; filaments up to 0.5 mm, slender; anthers dorsifixed, oblong, sub- lanceolate, ca. 1.3 mm; style gynobasic, 1.7–2.2 mm at anthesis; stigma bilobed, branches short (subterminal), included; ovules 4. Fruit dry, subtended by a persistent extended calyx with acuminate lobes apically recurved; eremocarps usually 4, rarely 3 by abortion, 1.6–1.9 × 1.3–1.5 mm, ovate and apically acute in ventral view, dorsally convex, lustrous, striate; cicatrix at basal po- sition under a stipelike projection, ovate to elliptic, ca. 1 mm; 1 ventral keel (apical) straight from apex of ere- mocarp to base, forming a stipe; lateral (basal) keels absent; gynobase flat to slightly raised (length:width ratio ca. 1:6), with 4 areoles corresponding to the cica- trix of each eremocarp; style plus stigmas 2.4 mm, sur- passing the eremocarps. 86 Novon Figure 4. Antiphytum berroi (Arechav.) N. Mend. & Flores Olv. —A. Habitat; note the grassland near to deciduous forest. —B. Habitat close up, with sedimentary rocks in foreground. —C. Habit. —D. Linear leaves. —E. Portion of scorpioid cyme and flower. —F. Fruit. —G. Front view of one eremocarp (Mendoza-Díaz et al. 650A). —H. Eremocarp ornamentation of A. ber- roi (Mendoza-Díaz et al. 650B). I. Eremocarp ornamentation of A. stoechadifolium (Cham.) DC. (Marchesi 1420). Photographs: A, B, F, G by Nidia Mendoza-Díaz; C–E by Camilo Pérez; H, I by Berenit Mendoza-Garfias. Volume 30 2022 Mendoza-Díaz et al. 87 Reevaluation of the Myosotis species of Arechavaleta Distribution. Antiphytum berroi is native to South America, occurring in southern Brazil (Rio Grande do Sul only) and northern Uruguay (adjacent departments Tacuarembó and Rivera) in disjunct localities (Fig. 5). A third locality in Rivera Department, i.e., the rocky habitat of the type locality of “Cerros del Gobierno” (Berro 4975; MVFA, MVM), no longer exists as the hill has been transformed into a quarry (Díaz, pers. obs.; Fig. 5, labeled triangle). The distribution of the species in southern Brazil is supported by Hagelund’s collec- tions from Rio Grande do Sul, mainly from the ICN herbarium, but the precise localities remain unknown (see Notes). Habitat and underlying geology. According to Uru- guayan records from the departments of Rivera and Tacuarembó, Antiphytum berroi is found in grasslands near a riverside subtropical forest formed in the can- yons (“quebradas”) with herbaceous species such as Baccharis trimera (Less.) DC., Crocanthemum brasil- iensis Spach., Plantago myosuros Lam., Schlechten- dalia luzulifolia Less., and species of Andropogon L., Aristida L., Borreria G. Mey., and Chascolytrum Desv. The area falls into what Lezama et al. (2019) classified as Eryngium horridum [Malme]–Juncus capillaceus [Lam.] community, a densely vegetated grassland, dom- inated by mesophytic species, encompassing stands with high plant cover values (near 100%) that occupy medium and deep soils of the Eastern Hills, North Eastern Sedimentary Basin, and the South Central re- gions. We found and collected scattered individuals from one small population (Mendoza-Díaz et al. 650, to be sheltered in MEXU and MVFA) at ca. 200 m.s.m., mainly on a sedimentary rock wall in one locality in the department of Tacuarembó whose substrate comprised 2%–10% rocky outcrops and 1%–10% stones (accord- ing to GIS layers provided by Dirección General de Recursos Naturales Renovables, 2002). The under- lying geology of these areas comprises basic lavas of tholeiitic basalt type with flow structure and intercala- tion of aeolian sandstones (Preciozzi et al., 1985). The localities of occurrence in Uruguay are part of the Ar- apey Formation (Upper Jurassic–Lower Cretaceous) of igneous volcanic origin that alternates on the east side with sandstones of sedimentary origin from Tacuarembó Formation. The Arapey Formation is an extension of the Serra Geral Formation in Brazil (Walther, 1911). Phenology. The main flowering and fruiting period is from September to December. Figure 5. Distribution map of Antiphytum berroi (Arechav.) N. Mend. & Flores Olv. in both Uruguay (Tacuarembó and Rivera Departments) and Brazil (Rio Grande do Sul State, not indicated). ▲ “Cerros del Gobierno” locality [now destroyed]; ● other localities. Note the distribution of the Uruguayan localities over the Serra Geral–Arapey Formation and the distant Municipio Arroio dos Ratos. 88 Novon Notes. After examining the type material and re- viewing the protologue, we confirmed that the species Myosotis berroi belongs to the genus Antiphytum, a conclusion supported by the nature of the calyx almost entirely divided to the base, the flat gynobase, the ere- mocarp with its stipelike projection and basal cicatrix (scar), the bracteate flowers arranged in scorpioid cymes, the blue corolla, the open extended calyx sub- tending the four eremocarps, and the opposite leaves throughout the length of the stem. The calyx lobe fea- tures and the kind of diaspora exclude this species from any genera of Moritziinae, which include Thau- matocaryon Baill. and Moritzia DC. ex Meisn., whose species are rosette herbs with fruits that consist of a single, ventrally and dorsally keeled nutlet enclosed into and dispersed with the calyx (Chacón et al., 2016). Antiphytum berroi is morphologically most similar to the South American species A. cruciatum and A. stoechad- ifolium. The blue corolla alone sets it apart from the North American species and the recently described Uruguayan species A. charruasorum N. Mend & Mar. Díaz (Mendoza-Díaz et al., 2020), all of which have white corollas. Antiphytum berroi is easily distinguishable from A. cruciatum by its linear leaves that are 0.4–1.6 mm wide (vs. oblanceolate leaves ca. 3–4 mm wide), but also by the eremocarp size and ornamentation (1.3–1.4 × 1–1.1 mm and rugose in A. cruciatum). The general vegetative appearance is similar to that of A. stoechad- ifolium, which also bears linear leaves, but the leaves in A. stoechadifolium are considerably longer (ca. 5 cm long) than those of A. berroi (1–2[–3] cm long). There- fore, the length of the leaves gives a lax appearance to the internodes in A. berroi, in contrast to A. stoechadi- folium, whose internodes are covered with leaves. Berro’s specimens (Berro 4975, MVFA and MVM) have a peri- carp ornamentation of the eremocarp that was described as “somewhat rugose” (Arechavaleta, 1911a: 69, “un tanto rugosa”) but is rather more or less striate as shown in the original illustration (Arechavaleta, 1911a: fig. 6F), and also in the SEM and stereoscope photo- graphs (Fig. 4H, I). In marked contrast with this, Chamisso (1829: 440) described the pericarp ornamen- tation of A. stoechadifolium as “tuberculato-granulatae,” a description later ratified by Johnston (1927) and by our own observations (Fig. 4H, I). Although the de- scriptions of vegetative characters of both species by Chamisso (1829) and Arechavaleta (1911a, 1911b) are apparently similar, we found enough evidence in leaves and eremocarps to recognize A. berroi as a dif- ferent species within Antiphytum, not conspecific with A. stoechadifolium, which seems to be allopatric. Regarding the known distribution, specimens from Rio Grande do Sul are geographically unreliable. The collector of these specimens, Karner Hagelund (1913– 1988), was a researcher and collector who had his own herbarium in “Fazenda Faxinal,” in the municipality of Arroio dos Ratos, Rio Grande do Sul (Eisinger, 1987), a farm with a local traditional agrosilvopastoral system (faxinal) that is performed in some areas of southern Brazil (Antoneli et al., 2020). In 1988, Hage- lund donated to the ICN herbarium his collections from areas of Rio Grande do Sul currently used for farming, especially from the northern region Alto Uru- guai e das Missões, whose flora prior to their devastation is well represented in Hagelund’s collection (Catálogo REMAM, 2013). Jesuit missions were located in west- ern areas of Rio Grande do Sul, following the Uruguay River, along the Serra Geral Formation. This basaltic formation extends southward into Uruguay, where it is named Arapey Formation (Walther, 1911) and in which the Uruguayan populations of Antiphytum berroi are located. Hagelund’s collections of A. berroi are labeled using the herbarium locality as the only geographic reference, “Fazenda Faxinal, Arroio dos Ratos, RGS [Rio Grande do Sul]” (Fig. 3A), which is far away from the known Uruguayan localities (Fig. 5); hence, there is doubt whether Arroio dos Ratos is actually the local- ity of distribution of A. berroi in Brazil. Therefore, new surveys in the Brazilian state of Rio Grande do Sul are warranted to relocate historical populations and find new populations of A. berroi. Such surveys should be conducted in areas where the Jesuit missions were based, as well as at additional localities in the Serra Geral and Arroio dos Ratos, to establish the current distribution and status of A. berroi in Brazil. Additional specimens examined. BRAZIL. Rio Grande do Sul: Arroio dos Ratos, Fazenda Faxinal, 5 Sep. 1974, Hagelund 8161 (ICN-162311); 18 Oct. 1976, Hagelund 10541 (ICN-153403); 19 Oct. 1980, Hagelund 13468 (ICN- 153402, MO-3641208); 10 Oct. 1982, Hagelund 14046 (ICN-162310). URUGUAY. Dpto. Rivera: 2 Nov. 1926, Fe- lippone 5224 (K-H2002/02368); Arroyo Gajo del Lunarejo, 17 Sep. 1995, Bonifacino s.n. (MVFA-25051). Dpto. Tac- uarembó: Cd. Tacuarembó, Cerros Chatos, 31°37′16.4″S, 56°02′33.3″W, 215 m, 9 Dec. 2017, Mendoza-Díaz et al. 650 (MEXU, MVFA). Additional specimens examined (other species). Antiphy- tum cruciatum (Cham.) DC.: BRAZIL. Rio Grande do Sul: Arroio dos Ratos, Fazenda Faxinal, 5 Oct. 1979, Hagelund 12846 (ICN); Canguçú, estrada Amaral Ferrador-Canguçú, 8 Oct. 1977, Miotto 558 (ICN), entre Amaral Ferrador e Can- guçú, 8 Dec. 1977, Fleig 751 (ICN); Camaquã, Boa Vista, Sep. 1985, Sobral et al. 4165 (ICN); Caçapava do Sul, Pedra do Segredo, 29 Nov. 2002, Kinupp & Irgang 2504 (ICN), BR 290, 31 Oct. 2010, Silva Filho 1603 (ICN); Guaíba, Centro Agronômico de Guaíba, 10 Oct. 1974, Schultz s.n. (ICN); Lavras do Sul, Mina Volta Grande, 5 Oct. 1984, Sobral 3074 (ICN); Piratini, Ponte Piratini, 15 Nov. 1975, Arzivenco s.n. (ICN); Porto Alegre, morro São Pedro, 30 Sep. 1956 (ICN), morro São Pedro, Econsciéncia Espaço de Conservação, 20 Oct. 2005, Setubal 492 (ICN); Rio Pardo, na beira da RS-7, Km. 32, 4 Oct. 1972, Lindeman et al. s.n. (ICN); Santana da Volume 30 2022 Mendoza-Díaz et al. 89 Reevaluation of the Myosotis species of Arechavaleta Boa Vista, 2 Nov. 1995, Jarenkow & Sobral 2777 (ICN); Toca do Tigre, perto de Itapoan, 27 Sep. 1950, Rambo 48861 (ICN); Torres, Butiazal, 26 Feb. 1985, Hagelund 15492 (ICN), Buti- azal, 2 Oct. 1975, Irgang s.n. (ICN), Butiazal, 12 Oct. 1970 (ICN); Vacaria, BR 116, Km. 50, 6 Oct. 1985, Miotto 1072 (ICN); Viamão, Itapuã, Sep. 1983, Sobral 2237 (ICN); Viamópolis, 14 Nov. 1969, Korner s.n. (ICN). URUGUAY. Dpto. Cerro Largo: Río Negro y Palleros, Dec. 1937, Rosen- gurtt 2346 (GH). Antiphytum stoechadifolium (Cham.) DC.: BRAZIL (?). Brasilia meridionali, Sellow s.n. (holotype, GH [barcode] 00032786 fragment! and image!; isotype GH [barcode] 00032785!). URUGUAY. Dpto. Maldonado: Sierra de las Ánimas, 26 Dec. 1965, Marchesi 1420 (MVFA). KEY TO THE SOUTH AMERICAN SPECIES OF ANTIPHYTUM 1. Inflorescences with flowers not arranged in scorpioid cymes, but only the terminal flower in each dichasial branch; corollas white; eremocarp cicatrix infra-ventral . . . . . . . . . . . . . . . . . . . . . . . . . . . A. charruasorum N. Mend. & Mar. Díaz 1′. Inflorescences with flowers arranged in scorpioid cymes, therefore more than the terminal flower in each dichasial branch; corollas blue; eremocarp cicatrix basal. 2. Leaves 3–4(–6) mm wide, oblanceolate; eremocarps 1.3–1.4 × 1–1.1 mm . . . . . . . . . . . . A. cruciatum (Cham.) DC. 2′. Leaves < 3 mm wide, linear-lanceolate or linear-oblong to linear-oblanceolate; eremocarps 1.6–2 × 1.3–1.6 mm. 3. Pericarp ornamentation ± striate; leaves up to 3 cm long . . . . . . . . A. berroi (Arechav.) N. Mend. & Flores Olv. 3′. Pericarp ornamentation tuberculate-granulate; leaves ca. 5 cm long . . . . . . . . . A. stoechadifolium (Cham.) DC. UNCERTAIN TAXONOMIC STATUS Myosotis uruguayensis Arechav., Anales Mus. Nac. Montevideo ser. 2, 1(3): 68. 1911. TYPE: Uru- guay. Dpto. Tacuarembó: “Vive en Tacuarembó, en terrenos arcillo-arenosos. Florece por noviembre y diciembre,” P. Pintos s.n. (type, MVM!). Distribution and habitat. Myosotis uruguayensis is known from only the type collection and reportedly oc- curs in Tacuarembó, Uruguay, on sandy-clay soils. Phenology. Flowering is in November and Decem- ber (Arechavaleta, 1911a). Notes. According to the original description and illustrations of the calyx and corolla (Arechavaleta, 1911a: 68 and fig. 4), Myosotis uruguayensis bears al- ternate linear leaves (ca. 4–5 × 0.3 cm), violet corollas, pubescent corolla faucal appendages (“escamillas trap- ezoides circundadas de pelos gruesos dactiloides”), and a shallowly divided calyx (“lacinias breves”). We were not able to fully observe all these features on the type specimen because of the immature and scanty na- ture of the original material, which prevented direct examination of the necessary fruit features to identify this specimen, but what we observed allowed us to as- sociate the material with the protologue. The short description of the fruit allows placement of the taxon under Boraginaceae, but it lacks charac- ters to place it confidently within any genus since it only mentions three to four ovoid eremocarps (“aque- nios”) with an acute apex, without an accompanying illustration. Even with the poor original material and short description, we can exclude Myosotis uruguay- ensis from Antiphytum because the calyx is not divided almost to the base and the eremocarps are not free from the calyx, as is characteristic of that genus. Myosotis uruguayensis seems closer to the Moritziinae subtribe (see Weigend et al., 2010: fig. 1), whose genera Morit- zia and Thaumatocaryon have no records from Uru- guay. Species in Moritziinae may have alternate leaves, a shallowly divided calyx, and pubescent corolla faucal appendages. However, Thaumatocaryon and Moritzia produce only one eremocarp due to abortion of the other three (Weigend et al., 2010, 2016), a key feature that may exclude Myosotis uruguayensis from these genera. In addition, Moritzia species have ebracteate inflores- cences and uncinate trichomes in the calyx, unlike the original material of Myosotis uruguayensis; hence, we strongly doubt its affiliation to this genus. Furthermore, in both genera the species have conspicuous basal leaves forming a rosette. Since Arechavaleta was not the collector of the specimen, the absence of the basal rosette in both the description and the collection can be expected. Myosotis uruguayensis could also be part of Myosotis, as diagnosed by the length of the calyx cleft and the number of eremocarps. Myosotis has exotic representatives in Uruguay, such as Myosotis latifolia Poir. and Myosotis verna Nutt. (Zuloaga et al., 2021). Unfortunately, an exact identification is not possible, even at the generic level. Except for A. berroi, which occurs in Tacuarembó, and recent records of A. cru- ciatum (Patricia Brussa, pers. comm.), no other native Boraginaceae have been reported from the type local- ity of Myosotis uruguayensis nor from neighboring areas. Therefore, Myosotis uruguayensis, which is a validly published name, remains as a name with uncertain sta- tus. Finding new populations of this taxon will be the only reliable way to know its identity by examining key fruit characteristics, such as pericarp ornamentation and cicatrix position. However, the lack of information on the locality of collection will be a challenge due to the large area of Tacuarembó. Typification notes. The MVM specimen is currently the only known original material, but following the rec- 90 Novon ommendation in McNeill (2014), lectotypification is not required. Acknowledgments. This paper arose from a collab- oration between staff from MVFA in Uruguay and the Instituto de Biología, UNAM, in Mexico. Nidia Mendoza- Díaz and Hilda Flores-Olvera would like to especially thank Daniel Piñero, who provided the initial informa- tion to contact staff at the Universidad de la República, Uruguay. All authors express our gratitude to GH, K, and MO herbaria for loaning specimens examined and cited in this paper, and to the curators and staff at ICN for allowing examination of the material of Boragina- ceae. We appreciate the support from the curators of B, BAF, CORD, F, G, K, L, LD, M, MIN, MVM, NY, PH, R, WRSL, and ZT in our search for Arechavaleta’s original material, especially in this pandemic time: Robert Vogt (B); Marcelo Wagner and Hernán Bach (BAF); Gloria Barboza (CORD); Richard Ree and Christine Niezgoda (F); Fred Stauffer and Laurence Loze (G); Alan Paton (K); Rob van den Berg and Roxali Bijmoer (L); Ulf Araup and Arne Thell (LD); Andreas Fleischmann (M); Tim Whitfeld (MIN); Meica Valdivia and Andrés González (MVM); Amy Weiss and Matthew Pace (NY); Elana Benamy (PH); Nílber Silva (R); Ewa Lenard and Krzysztof Świerkosz (WRSL); and Alessia Guggisberg (ZT). We thank the following staff at Instituto de Biología, UNAM: Berenit Mendoza-Garfias for assistance with the scanning microscope, Susana Guzmán-Gómez for help with the stereoscopic microscope, and Ma. del Rosario García-Peña for assistance with the specimen loans. The first author especially thanks Anderson San- tos de Mello for the chance and help to visit Porto Alegre and the ICN herbarium, to Fabián Muñoz and Carlos Brussa for providing valuable information for clarification about the types, and to Camilo Pérez for his assistance with fieldwork and for the photographs that he kindly shared for this project. The first author thanks Posgrado en Ciencias Biológicas, UNAM, for a doctoral fellowship (Consejo Nacional de Ciencia y Tecnología [CONACYT] grant 288698) and for support (Programa de Apoyo a los Estudios de Posgrado [PAEP] 2017) to fund part of the trip to Uruguay and Porto Alegre, Brazil. Literature Cited Antoneli, V., M. Pulido Fernández, T. de Oliveira, J. Lozano- Parra, J. A. Bednarz, M. Vrahnakis & R. García-Marín. 2020. Partial grazing exclusion as strategy to reduce land degradation in the traditional Brazilian Faxinal System: Field data and farmers’ perceptions. Sustainability 12: 7456. Arechavaleta, J. 1897. Las gramíneas uruguayas. Anales Mus. Nac. Montevideo 1(6): 29–581. Arechavaleta, J. 1898–1911. Flora Uruguaya. Museo de His- toria Natural de Montevideo, Montevideo. Arechavaleta, J. 1911a. Vegetación uruguaya. Varias especies nuevas. Flora Uruguaya. Anales Mus. Nac. Montevideo ser. 2, 1: 59–83. Arechavaleta, J. 1911b. Flora Uruguaya Tomo IV, entrega 3. Anales Mus. Nac. Montevideo 7: 129–224. de Candolle, A. P. 1846. Borragineae. Pp. 117–178 in A. L. P. P. de Candolle (editor), Prodromus Systematis Naturalis Regni Vegetablis. Pars Decima. Victoris Masson, Paris. Catálogo REMAM. 2013. Catálogo Virtual da Rede de Mu- seus e Acervos Museológicos da UFRGS. . Chacón, J., F. Luebert, H. Hilger, S. Ovchinnikova, F. Selvi, L. Cecchi, C. M. Guilliams, et al. 2016. The borage family (Boraginaceae s. str.): A revised infrafamilial classification based on new phylogenetic evidence, with emphasis on the placement of some enigmatic genera. Taxon 65: 523–546. Chamisso, L. K. A. 1829. De plantis in expeditione specula- toria Romanzoffiana observatis disserere pergunt Adelber- tus de Chamisso et Diedericus de Schlechtendal. Borag- ineae Juss. Linnaea 4: 435–496. Dirección General de Recursos Naturales Renovables. 2002. Carta de suelos del Departamento Tacuarembó 1/100.000. División de Suelos y Aguas-Departamento de Estudios Básicos de Suelos. Ministerio de Ganadería, Agricultura y Pesca. Dirección General de Recursos Naturales Renov- ables. República Oriental de Uruguay. Eisinger, S. M. 1987. O gênero Indigofera L. (Leguminosae- Papilionoideae-Indigofereae) no Rio Grande do Sul-Brasil. Acta Bot. Brasil. 1: 123–140. Fundação Estadual de Proteção Ambiental Henrique Luiz Roessler. 2020. Biblioteca Digital. Secretaria do Meio Am- biente e Infraestructura, Brazil. . Georgi, J. G. 1775. Bemerkungen einer Reise im Russischen Reich im Jahre 1772, Vol. 1. Gedruckt bey der Kayserl. Akademie der Wissenschaften, St. Petersburg. Hilger, H. H. 2014. Ontogeny, morphology, and systematic significance of glochidiate and winged fruits of Cynoglos- seae and Eritrichieae (Boraginaceae). Pl. Diversity Evol. 131: 167–214. Infraestructura de Datos Espaciales del Ministerio de Trans- porte y Obras Públicas de Uruguay. 2020. GeoPortal. Min- isterio de Transporte y Obras Públicas. República Oriental de Uruguay. . Instituto Geográfico Nacional de la República Argentina. 2020. Capas SIG. Ministerio de Defensa, República Argen- tina. . Instituto Nacional de Estadística. 2020. Mapas Vectoriales Año 2011. Gobierno de Uruguay. . Johnston, I. M. 1923. Studies in the Boraginaceae. Contr. Gray Herb. 68: 43–80. Johnston, I. M. 1927. Studies in the Boraginaceae VI: A re- vision of the South American Boraginoideae. Contr. Gray Herb. 78: 3–118. Ledebour, C. F. von. 1847. Flora Rossica sive Enumeratio Plantarum in Totius Imperii Rossici Provinciis Europaeis, Asiaticis, et Americanis Hucusque Observatarum, Vol. 3. Corolliflorae, pt. 1. Schweizerbart, Stuttgart. Lezama, F., M. Pereira, A. Altesor & J. M. Paruelo. 2019. Grasslands of Uruguay: Classification based on vegetation plots. Phytocoenologia 49: 211–229. McNeill, J. 2014. Holotype specimens and type citations: General issues. Taxon 63: 1112–1113. Meisner, C. F. 1840. Plantarum Vascularium Genera. Libraria Weidmannia, Leipzig. Volume 30 2022 Mendoza-Díaz et al. 91 Reevaluation of the Myosotis species of Arechavaleta ISSN 1055-3177 (PRINT); ISSN 1945-6174 (ONLINE) Mendoza-Díaz, N., M. Díaz, P. Brussa, F. Muñoz, J. M. Boni- facino & H. Flores-Olvera. 2020. White flowers in South America: A new species of Antiphytum (Echiochiloideae, Boraginaceae). Syst. Bot. 45: 913–920. Ministerio de Industria, Energía y Minería de Uruguay. 2020. Geo Portal DINAMIGE. Gobierno de Uruguay. . Núcleo de Economía Regional e Urbana da Universidade de São Paulo. 2020. Dados e Programas. Universidade de São Paulo, Brazil. . Preciozzi, F., J. Spoturno, W. Heinzen & P. Rossi. 1985. Me- moria explicativa de la carta geológica del Uruguay a la escala 1:500,000. República Oriental del Uruguay, Direc- ción Nacional de Minería y Geología. Ministerio de Indu- stria y Energía, Montevideo. QGIS Development Team. 2014. QGIS Geographic Information System, Version 2.6.1-Brighton. Open Source Geospatial Foundation. . Reiche, K. F. 1896–1911. Flora de Chile. Imprenta Cer- vantes, Santiago de Chile. Serviço Geológico do Brasil. 2020. Dados, Informações e Pro- ductos do Serviço Geológico do Brasil. . Stafleu, F. A. & R. S. Cowan. 1976–1988. Taxonomic Litera- ture: A Selective Guide to Botanical Publications and Col- lections with Dates, Commentaries and Types, ed. 2 (TL-2). Bohn, Scheltema & Holkema, Utrecht. . Thiers, B. 2022 [continuously updated]. Index Herbariorum: A Global Directory of Public Herbaria and Associated Staff. New York Botanical Garden’s Virtual Herbarium. . Turland, N. J., J. H. Wiersema, F. R. Barrie, W. Greuter, D. L. Hawksworth, P. S. Herendeen, S. Knapp, et al. (editors). 2018. International Code of Nomenclature for algae, fungi, and plants (Shenzhen Code). Regnum Veg. 159. . Walther, K. 1911. Über permotriassiche Sandsteine und Erup- tivdecken aus der Norden der Republik Uruguay. Neues Jahrb. Mineral. Geol. 31: 575–609. Weberling, F. 1989. Morphology of Flowers and Inflores- cences. Cambridge University Press, Cambridge. Weigend, M., M. Gottschling, F. Selvi & H. H. Hilger. 2010. Fossil and extant Western Hemisphere Boragineae, and the polyphyly of “Trigonotideae” Riedl (Boraginaceae: Boragi- noideae). Syst. Bot. 35: 409–419. Weigend, M., F. Selvi, D. C. Thomas & H. H. Hilger. 2016. Boraginaceae. Pp. 41–102 in J. W. Kadereit & V. Bittrich (editors), The Families and Genera of Vascular Plants, Vol. 14. Flowering Plants. Eudicots: Aquifoliales, Boraginales, Bruniales, Dipsacales, Escalloniales, Garryales, Paracry- phiales, Solanales (except Convolvulaceae), Icacinaceae, Metteniusaceae, Vahliaceae. Springer International Pub- lishing, Cham. Wildner, W., V. O. Filho & L. E. Giffoni. 2004. Excursão virtual aos Aparados da Serra-RS/SC. Aspectos geológicos e turísti- cos Cânions do Itaimbezinho e Fortaleza. CPRM Serviço Geológico do Brasil. Zuloaga, F. O., O. Morrone & M. J. Belgrano. 2021 [onward]. Catálogo de las Plantas Vasculares del Cono Sur. Instituto de Botánica Darwinion. San Isidro. Buenos Aires, Argen- tina. . 61 PARTE II: FILOGENIA Y PATRONES DE DIVERSIFICACIÓN DE ANTIPHYTUM CAPÍTULO V: Antiphytum: initial insights about its phylogeny, diversification and biogeographic patterns Nidia Mendoza-Díaz, Helga Ochoterena, Michael J. Moore, Daniel Piñero & Hilda Flores-Olvera Borrador 62 Prepared to Plant Systematics and Evolution Original research article Mendoza-Díaz et al.: Initial insights to Antiphytum Antiphytum (Boraginaceae, Echiochiloideae): initial insights on its phylogeny, diversification, and biogeographic patterns Nidia Mendoza-Díaz,1 Helga Ochoterena,2 Michael J. Moore,3 Daniel Piñero,4 and Hilda Flores- Olvera2,5. 1Posgrado en Ciencias Biológicas, Instituto de Biología. Universidad Nacional Autónoma de México. 3er. Circuito Exterior s/n, C.P. 04510 Ciudad de México, México. 2Departamento de Botánica, Instituto de Biología. Universidad Nacional Autónoma de México. Apartado Postal 70-233, C.P. 04510 Ciudad de México, México. 3Department of Biology, Oberlin College, Oberlin, Ohio 44074, USA. 4Departamento de Ecología Evolutiva, Instituto de Ecología. Universidad Nacional Autónoma de México. Apartado Postal 70-275, C.P. 04510 Ciudad de México, México. 5Author for correspondence (mahilda@ib.unam.mx) 63 Abstract Antiphytum (Boraginaceae) has been an elusive genus whose disjunct endemic distribution in North and South America and apparently unlike morphology has suggested a non-natural group. Recent works on Boraginales included no more than two or even no species of Antiphytum, lacking South American samples. In this study, we aimed to get a phylogeny of the entire genus with a broad sampling that includes all the known species from both North and South America derived from taxonomic revision. We used five molecular markers (nuclear ITS and cpDNA regions ndhF–rpl32, trnL–F, matK, and rps16) to carried out parsimony, maximum likelihood and Bayesian analyses. Antiphytum was recovered as a monophyletic member of Echiochiloideae with Ogastemma closer related. The backbone of the phylogeny had no resolution, but we recovered three clades within Antiphytum: a South American (SA) clade; a North-American (NA) clade with species bearing basal-rosettes, and a North American one with species lacking basal-rosettes. Each clade has morphological and geographical patterns that could be related to edaphic factors, tectonic events or changes associated to dispersal that we discuss as possible explanations for the similar diversity in each. The direction of the colonization of America is proposed north to south, with a disjunction by a single event and probably long-distance dispersal. According to this proposal and additional results, corolla color changed from white to blue after the colonization of South America. We discussed our data in light of possible process that explained the patterns found, following the recent knowledge in Boraginaceae and groups with similar patterns. Keywords: American amphitropical disjunction, Boraginales, circumscription, phylogeny, morphology, taxonomy 64 Introduction Antiphytum DC. ex Meisn. (subfamily Echiochiloideae, see Chacón et al., 2016, Boraginaceae) is an American genus with a disjunct distribution, whose messy taxonomic history and apparently unlike morphology between the North and South American species, has suggested a non-natural group. Antiphytum has a complicated taxonomic history, which begins with its publication in Meisner's Genera (1840) without the corresponding nomenclatural changes over the four species that were included in the new genus: “Anchusa oppositifolia HBK [correct name Plagiobothrys linifolius (Willd. ex Lehm.) I.M.Johnst., according to Holstein and Weigend 2016]; A. cruciata et stoechadifolia Cham. [correct name Antiphytum cruciatum (Cham.) DC., lectotype of the genus (Johnston 1923) and Antiphytum stoechadifolium (Cham.) DC., respectively]; and A. Mexicana, DC. ap. Moc. ic. Fl. Mex. ined.”. This first publication of the genus comprised two parts, a brief description of the new genus in a diagnostic table (Meisner 1840, i, p. 280), and the “Commentarius”, with the previously listed four species (here transcribed in the same order and typing of the original work) and its bibliography (Meisner 1840, ii, p. 188). Except for the Sessé and Mociño species from Mexico, the new genus Antiphytum was composed mainly by South American species, but only Anchusa cruciata and A. stoechadifolia further kept into Antiphytum through the time. On the other hand, it is very likely that the species that originated the new genus was erroneously cited by Meisner, because according to the posthumous publication of Sessé and Mociño (1887), Anchusa mexicana Sessé and Moc. bears alternate phyllotaxy, which does not coincide with the intention expressed in the name Antiphytum (“Genus jam foliis oppositis (unde nomen) insigne”, Meisner 1840, ii, p. 188; and “folia opposite (unde nomen)”, De Candolle 1846, p. 121). Rather the Sessé and Mociño's iconography no. 288 (Mociño and Sessé 2010), corresponding to Symphytum fruticosum Sessé and Moc. (Sessé and Mociño 1887; see De Candolle 1846, p. 121), must be the correct one and whose new combination Antiphytum Mexicanum DC. [correct name Euploca mexicana (DC.) M.W.Frohl. and M.W.Chase, according to Frohlich et al. 2020] was given by De Candolle (1846) in the first formal synthesis of the genus. De Candolle (1846) made the nomenclatural combinations of the South American Meisner's Antiphytum species, added an additional North American species (A. heliotropioides A.DC.), and two South American ones, from which “A. tetraquetum (Cham.) DC.” [actually Thaumatocaryon 65 tetraquetum I.M.Johnst., Johnston 1924] was considered as a doubtful species, but later remained in Antiphytum during a large time. Fresenius' Flora of Brazil (Fresenius 1857) treated Antiphytum comprised by A. cruciatum, A. tetraquetum, and A. stoechadifolium, which would be the circumscription proposed by Bentham and Hooker (1876) for a segregated genus Antiphytum. The North American species published, either as Antiphytum or within other genera, such as Eritrichium Schrad. ex Gaudin, Krynitzkia Fisch. and C.A.Mey. or Lithospermum L. (see Martens and Galeotti 1844; Torrey 1859; Watson 1883; Rose 1890) were included later within the genus Amblynotopsis J.F.Macbride (Macbride 1916). Johnston (1923) rejected the segregation of Antiphytum and added three additional North American species. He suggested that corolla color (blue for the South American taxa vs white for the North American ones) and phyllotaxy (through opposite vs mainly alternate, respectively) were scarcely generic differences to validate the segregation. Johnston's contribution (1923) is a cornerstone on Antiphytum taxonomy, from which derives our nowadays concept of the genus and the lectotypification that validates the name and join it to the South American species, although currently, North America is the richest area. In posterior works, Johnston changed the genus circumscription, getting out A. tetraquetum for the first time since De Candolle's work (1846) and recognized only two South American species, A. cruciatum and A. stoechadifolium (Johnston 1927). He even questioned himself the genus concept (Johnston 1927, 1957), considering that it was still subject to revision, naming “true Antiphytum” to the South American species and applicable the generic name Amblynotopsis to the North American ones. In the last work dealing with the genus Johnston (1966) kept Antiphytum as a disjunct genus, opposing to Brand (1929, 1931) that proposed that Antiphytum must not even to exist, mainly due to its mess origin. No molecular work has tested the monophyly of Antiphytum. The efforts focused from order to subfamily level have included only one or two (e.g., Längstrom and Chase 2002; Längstrom and Oxelman 2003; Nazaire and Hufford 2012; Weigend et al. 2013, 2014; Cohen 2014; Chacón et al. 2016; Luebert et al. 2016) or even no Antiphytum species (e.g., Weigend et al. 2010), and no studies have included species from South America. Since Längstrom and Chase (2002) proposed tribe Echiochileae, currently subfamily Echiochiloideae (Chacón et al. 2016), this clade has been subsequently retrieved with high support and includes Antiphytum, Ogastemma Brummitt and Echiochilon Desf., being Antiphytum and Ogastemma sister groups (e.g., Längstrom and Oxelman 2003; Weigend et al. 2013; Cohen 2014; Chacón et al. 2016). Whereas the position of Antiphytum within the subfamily seems robust, the relationships among the species of the unique American genus of Echiochiloideae are poorly known. Ogastemma is monotypic and its unique species O. 66 pusillum (Coss. & Durieu ex Bonnet & Barratte) Brummitt is distributed in North Africa and Arabian Peninsula and recent records had been done in Iraq (Mousa and Shahatha 2021). Echiochilon, mainly also from North Africa and Arabian Peninsula, was the subject of study in Längstrom and Oxelmanˈs (2003) work, although the phylogeny was built only with ITS as molecular marker and morphology. Figure 1 shows distribution of subfamily Echiochiloideae based in Lönn (1999) and our observations. Overall, Echiochiloideae requires a taxonomic and phylogenetic update, with a sampling that incorporates Antiphytum, the genus more poorly sampled until now and that it is the focus of the present work. With the recent discovery of new species from North and South America (Higgins and Turner 1983; Mendoza-Díaz et al. 2018, 2020a,b), questions involving monophyly, circumscription and biogeographic history of Antiphytum need to be encompassed. In order to cover these objectives, we sampled the currently known species according to the taxonomic updated work carried on by the first author for the molecular phylogenetic reconstruction, and morphological features were observed from herbaria collections and material collected during fieldwork. We presented the first insights over the phylogeny, diversification and geographic patterns of Antiphytum based on a broad morphological and geographic revision plus the retrieved topologies from molecular phylogenetic analyses. Material and methods Morphological study, field work and terminal selection As a first step, taxa that once were included under the genus, but recent studies indicated them out of Antiphytum were excluded. The morphological study was based on the revision of protologues, bibliography and herbarium specimens. Herbarium specimens (including types) were examined directly or through loans from ARIZ, ASU, B, BM, CAS, CHAPA, CIIDIR, ENCB, F, FCME, G, HCIB, HUH, IBUG; IEB, K, MEXU, MO, NMC, NY, P, R, RSA-POM, TEX, UAT, UC, and US (Thiers continuously updated). Online specimens from AAH, BR, G, G-DC, GH, K, UC, US, and W were also revised. In order to broaden the sampling, especially for taxa with old records, few specimens, or major problems with its geographic distribution and taxonomic circumscription, field work was carried on different localities of Mexico, Uruguay, and Brazil. A morphological concept of the genus was achieved through this taxonomic revision, which allowed the recognition of Antiphytum among Boraginaceae by the combination of the following characters: suffruticose perennial as growth-form; corolla radially symmetrical surpassing the calyx; 67 presence of faucal appendages; fruit not enclosed into the calyx, rather an extended calyx subtending the fruit; gynobase shape from flat to triangular; cicatrix in proximal half of eremocarp, not exceeding the middle zone of the ventral face. Subsequently, every species was defined by diagnostic traits. As a consequence of these preliminary steps, three new species were described (Mendoza-Díaz et al. 2018, 2020a,b) and one species resurrected that derived in a new combination (Mendoza-Díaz et al. 2022). Updating Johnston's works (1923, 1927, 1966), that recognize nine Antiphytum species with only two of South America (including as heterotypic synonyms two Brand's species described as Amblynotopsis (Brand 1926): A. ehrenbergii Brand and A. humilis Brand), a total of 13 species, four from South America, were achieved (Table 1). Based on this taxonomic revision, we recruited samples for palynological observation and molecular analysis for almost all the resulting 13 species to test with phylogenetic reconstruction, in an illumination reciprocal process. For the remaining genera of Echiochiloideae subfamily, herbaria specimens were revised for Ogastemma pusillum and some Echiochilon species, such as E. collenettei I.M.Johnst., E. fruticosum Desf., E. johnstonii Cufod., E. longiflorum Benth., and E. persicum (Burm.f.) I.M.Johnst. Palynological study and fruits observation Whole flowers, sampled from fresh and herbarium collections of Ogastemma pusillum and eight Antiphytum species (A. berroi, A. brevicalyx, A.caespitosum, A. cruciatum, A. heliotropioides, A.hintoniorum, A. floribundum, and A. peninsulare) were macerated while hydrating in a water bath. Macerated material was dehydrated with acetic acid to concentrate it through centrifugation, and acetolyzed by Erdtman (1960) method with modifications from Laboratorio de Paleopalinología of Instituto de Geología, UNAM (Martínez-Hernández personal comm.). Pollen was mounted in glycerinated gelatin and some samples deposited in the palynological collection of Instituto de Geología, UNAM. Samples were analyzed in an optic microscope and photographed with a Zeiss AxioCam. Mature eremocarps (nutlets o mericarps, terminology sensu Hilger 2014) from Ogastemma pusillum, some Echiochilon species, and all the Antiphytum species (Table 1) samples were observed with a confocal stereo microscope (Leica Z16 APO A, Leica Microsystems GmbH, Wetzlar, Germany), and photographed with a camera (Leica DFC490, Leica Microsystems GmbH) at Laboratorio de Microscopía y Fotografía de la Biodiversidad (II) of Instituto de Biología, 68 UNAM. SEM photographs were also taken (Hitachi SU1510, Hitachi Ltd., Tokyo, Japan) at Laboratorio de Microscropía Electrónica y Fotografía de la Biodiversidad (I) after fruits samples were critical-point dried and mounted on stubs with common double stick carbonated tape, and coated with gold to 20 μÅ during two minutes in a sputter coater (Quorum Q150OR ES). DNA extraction, PCR amplification and sequencing Total DNA was isolated from herbarium specimens or field-collected leaves dried in silica gel using the CTAB extraction method of Doyle and Doyle (1987) modified by the addition of 2% of polyvinylpyrrolidone (PVP) to the extraction buffer. Several individuals from 19 taxa of Echiochiloideae (the sequences will be uploaded to GenBank) were newly sequenced for five molecular markers: nuclear internal transcribed spacer (ITS) using the primer pairs ITS4 and ITS5 (White et al. 1990), and the plastid markers for chloroplast ndhF–rpl32 intergenic spacer (newly designated, repl32.4R GAAGTACGYTTTTTTGGAACTGC; ndhF.40R.Bor. ACATAGGAACTGTWAGTGGAAG); trnL–F intergenic spacer (pairs C and F, Taberlet 1991); rps16 intron (pairs F and R2, Oxelman et al. 1997), and matK gene (partial maturase gene, pairs 2F and 3R, Sang et al. 1997). Final marker selection was based on the bibliography on Boraginales for genus (e.g., Moore and Jansen 2006; Cohen and Davis 2012; Nazaire et al. 2014), order (e.g., Weigend et al. 2014; Luebert et al. 2016, Chacón et al. 2016), or subtribe level (e.g., Pourghorban et al. 2020), and amplification success. Previously ITS and ndhF–rpl32 matrix from Mendoza-Díaz et al. (2020a) was expanded to include newly generated sequences (60 for ITS, 28 for ndhF–rpl32, 111 for trnL–F, 112 for rps16, and 65 for matK). Also, to the data matrix of every separate marker we added available GenBank sequences of both Echiochiloideae and other Boraginaceae subfamilies to function as outgroup (68 for ITS, 41 for ndhF–rpl32, 211 for trnL–F, 174 for rps16, and 128 for matK). PCR conditions were the same as Mendoza-Díaz et al. (2020a), as well the amplification programs for ITS and ndhF–rpl32. Both trnL–F region and rps16 intron were amplified following Moore and Jansen (2006) program for rps16. Initially, matK partial gene was difficult to amplify for North American species of Antiphytum, but it was successful when it was followed the program of Nazaire and Hufford (2014) for the marker. PCR products were sent unpurified to LaNABio (IBUNAM) for sequencing with an ABI 3730xL sequencer (Applied Biosystems, Waltham, Massachusetts). 69 Sequence assembly, alignment, coding of microstructural characters and phylogenetic reconstruction All sequence chromatograms were edited, assembled, and manually aligned and annotated with PhyDE 0.9971 (Müller et al. 2010). Regions of mutational hot spots were removed prior to phylogenetic analyses. Only for the parsimony analyses (MP), microstructural characters (gaps, simple sequence repeats or SSRs, and inversions) were added at the end of every marker matrix. Each microstructural character was named including the locus, type of character, and the corresponding aligned positions. Gaps were coded using simple indel coding according to Simmons and Ochoterena (2000) and microstructural characters were coded according to Ochoterena (2009). Resulting matrices did not match each other either in the number or identity of the terminals, even for the Echiochiloideae taxa, because the sampling was not equalled successfully amplified for all the markers. Concatenated matrices of microstructural character and nucleotide for every individual marker were analysed under MP criterion using TNT 1.1 (Goloboff et al. 2000) alternating the four new technology search (NTS) strategies: sectorial searches (Goloboff 1999), ratchet (Nixon 1999), tree drifting and tree fusing (Goloboff 1999) until the consensus was stabilized three times by a factor of 100. The most parsimonious trees were saved and analyzed in WinClada (not shown). Ambiguously supported nodes were collapsed and identical trees removed. Jackknife (JK) support values were calculated with TNT using NTS (excepting sectorial searches) using a 36% of character removal probability and 10,000 replicates, until the minimum length was found four times. The frequency of appearance for each node was calculated among these trees and mapped on to the consensus tree using WinClada. We considered support to be lacking when the values were below 64% (Farris et al. 1996). Separate analyses were performed to evaluate the position and monophyly of the putative species of Echiochiloideae in the recovered topologies, in order to strengthen the hypothesis of the concatenated five markers matrix, in which individual sampling by species was reduced, both ingroup and outgroup. Concatenated five markers matrix (hereafter 5-loci matrix) included 43 terminals, 12 Antiphytum taxa (five taxa with only one individual), one Echiochilon taxa (one individual), and Ogastemma pusillum (one individual). As outgroup we used concatened GenBank sequences from Lithospermum californicum A.Gray and Echium vulgare L. Although sequences of these species from every marker came from different vouchers, we can rule out mixing of information, as these 70 species were recovered in the same phylogenetic position in the separate analysis and suffice the outgroup function. MP analyses were performed using the 5-loci matrix with nucleotides and microstructural characters with the described procedure. For Bayesian inference (BI) and maximum likelihood (ML) analyses we used only the nucleotide 5-loci matrix. Substitution models for each DNA region were selected based on the Akaike information criterion (AIC) implemented in jModelTest 2 (Darriba et al. 2012) using the following parameters: 203 substitution schemes, +F base frequencies, +I and +G rate variation, nCat=4 and SPR tree search. Every marker obtained a different model of molecular evolution: GTR+G for ITS; GTR for ndhF–rpl32; GTR+I for trnL–F; HKY for rps16, and GTR+G+I for matK. ML analysis was performed with RaXML (Stamatakis 2014) using raxmlGUI 2.0 (Edler et al. 2021) implementing the “ML+thorough bootstrap” option and 1000 ML bootstrap replicates for estimating branch supports. We considered nodes with weak (50–70%), moderate (71–80%) and strong support (81–100%) following to Ortíz-Rodríguez et al. (2018). BI analyses were conducted using MrBayes v.3.2.7 (Ronquist et al. 2012) with the following parameters: number of Markov Chain Monte Carlo generations=1,000,000, sampling frequency=1000, burn-in=100,000 generation. The burn-in was determined using Tracer 1.7.1 (Rambaut et al. 2018). We considered a clade to be well supported if posterior probabilities (PP) were ≥ 0.95 (according to Alfaro and Holder 2006), and not supported when PP < 0.85. Phylogenetic trees were visualized using FigTree 1.4.4 (Rambaut 2018). MP and BI consensus as well as the ML tree were graphically summarized, keeping only the supporting nodes by at least one inference method. Distribution Maps Maps were produced using QGIS Brighton (2.6.1). The distribution of the species was obtained from specimen labels or from data obtained directly in the fieldwork, using geographic coordinates when available. Specimens without coordinates were georeferenced using Google Earth 7.1.8.3036 (32 bit). 71 Results Morphology At the level of macromorphology, Antiphytum exhibits two well-defined groups: species bearing a non-marcescent basal rosette and species that lack it. While the first group is exclusively North American, species without basal rosettes are found both in North and South America. The rosette group includes A. floribundum, which is morphologically and geographically distant from the remaining three Mexican rosette-forming species A. caespitosum, A. paniculatum and A. parryi (from here, “eastern rosette species”). Cauline leaves are totally opposite in the South American species and in A. hintoniorum from North America; the remaining North American species have alternate phyllotaxy throughout the stem or branch, but sometimes are opposite at the base. Midvein is visible mainly on the adaxial side of the leaves, but secondary venation is besides only visible in two species, A. brevicalyx and A. floribundum. Glandular-capitate trichomes were solely found in A. berroi, A. cruciatum, and A. stoechadifolium from South America; and A. brevicalyx, A. geoffreyi, and A. peninsulare from North America. They are wider distributed on young branches, leaves and leafy organs such as bracts (pherophylls), flower-subtending bracts, and calyx in A. brevicalyx, A. geoffreyi, A. peninsulare, and A. stoechadifolium. This kind of trichomes were more difficult to observe in A. berroi (only on the calyx) and A. cruciatum (on the branchlets), probably because they are deciduous in the latter species, since not in all the specimens were possible to find. Ogastemma pusillum also has this kind of indumentum on the calyx and branchlets. The inflorescence in Antiphytum ends in a flower (monotelic) forming a terminal homocladic or heterocladic determinate thyrse (thyrsoid) with dichasial division. Except for A. charruasorum and A. hintoniorum, the basic unit of the inflorescence is the bracteate boragoid. Boragoids ranges from less of 10 mm long in A. parryi to 200 mm in A. heliotropioides. In A. charruasorum the inflorescence has only the terminal flower (without flower-subtending bracts) and no branch becomes a boragoid in every dichasial division subtending by two opposite pherophylls. In A. hintoniorum there is a reduced terminal dichasial cymoid to a one-flowered cymoid (only the terminal flower of each unique dichasial division). It could be interesting to highlight the arrangement of the boragoids in several species. For instance, in the blue-corolla species from South America there are ca. 3–4 paired boragoids at the previous nodes before the terminal dichasial 72 division of the last node, which bears the terminal flower and the last paired boragoids. Every paired boragoids arise from the leaf axils of a pair of opposite pherophylls. In the North American species, boragoids have alternate arrangement and arise from the leaf axil of one pherophyll. Antiphytum geoffreyi has the boragoids more proximal to the base of the plant, similar to Ogastemma pusillum, in contrast with the rather distal position common in the genus. The species with basal-rosettes have distal inflorescences bearing well-defined unbranching flowering stems, but in A. floribundum the thyrsoid is heterocladic, such as in the non-rosetted species A. brevicalyx. Both A. heliotropioides and A. peninsulare has foliose metaxyphylls (or intercalated sterile bracts) preceding the terminal flower (which apparently is concaulescent in A. heliotropioides) of every dichasial division; therefore, boragoids may appear to be isolated and axillary. Regarding corolla color, except for three exclusively South American species with blue corollas (A. berroi, A. cruciatum, and A. stoechadifolium), the genus exhibits mainly white corollas in all the North American species and in one of Uruguay (A. charruasorum). The presence of faucal appendages (or fornices) is usual, but not the rule, since A. floribundum and A. paniculatum lack of them. Fornices surfaces are often papillose, but hirsute trichomes are found in A. charruasorum. Calyx lobes are divided almost to the base, but except in A. brevicalyx, they always are equal or longer than the corolla tube. The corolla is hypocrateriform (rotate) when fornices are present, but A. paniculatum has also a rotate corolla, while rather it is infundibuliform-campanulate in A. floribundum. Corolla diameter ranges from 2–4 mm of diameter in A. parryi to 10–12 mm in A. peninsulare; other species with large corolla diameter are A. hintoniorum (8–9 mm) and A. charruasorum (7.5–8 mm). Flowers are sessile or subsessile (pedicel 0.2–2.7 mm in length), but in A. heliotropiodes are pedicelate (up to 12 mm in length). Pollen traits are common in Antiphytum and shared with Ogastemma: isopolar subprolate (in Antiphytum vs. oblate spheroidal in Ogastemma) tricolporate monads, ectoapertures with colpal membrane verrucate and endoapertures with converging transverse costae, meridional edges lacking or indistinct; large polar area, polar outline circular (Antiphytum and Ogastemma) to slightly concave (Antiphytum); tectate exin with perforations and columellar structure and microreticulate pattern (Fig. 2a). Although Ogastemma pusillum along with A. berroi and A. cruciatum have the largest polar area of all the sampling, the two South American blue corolla species have overall the monads with both the largest equatorial diameter and polar axis. The monads in the North American species varied in size from the smallest A. heliotropioides to the larger A. brevicalyx (Fig. 2a). In contrast, eremocarps discriminates very well among species of Antiphytum (although in the eastern rosette species group, eremocarps are very similar) and with respect to Ogastemma pusillum 73 (Fig. 2b). Diagnostic features are: 1) form and size; 2) number; 3) ornamentation of the pericarp; 4) modification of the ventral keel such as projection-like stipites or a kind of bump; 5) position and area of the cicatrix; 6) projection of the cicatrix tissue. Although the predominant eremocarp form in the genus is ovate (also in Ogastemma pusillum), rather it is clearly triangular in A. geoffreyi and A. peninsulare; the size ranges from the biggest eremocarps in A. charruasorum (up to 2.5 × 2.0 mm) to the more reduced in size such as in A. floribundum (up to 1.2 × 1.0 mm); A. cruciatum (up to 1.3 × 1.0 mm); A. parryi (1.35 × 1.0 mm); A. paniculatum (average 1.3–1.5× 1.0 mm, but the smallest measured eremocarp 1.17 × 1.0 mm) or A. caespitosum (1.5-1.7 × 1.2-1.4 mm). While from South America A. charruasorum has the biggest eremocarp, from North America A. peninsulare and A. hintoniorum have the biggest sizes with 2.0–2.3 × 1.2–1.6 mm and 1.8–2.0× 1.2–1.4 mm, respectively, close to those in the South American species A. stoechadifolium (1.8–2.0 × 1.4–1.6 mm). Usually there are four eremocarps per fruit, but reduction by abortion diminishes the number mainly in A. caespitosum (until two), A. brevicalyx (generally 2–3), to the extreme reduction to one (rarely two, but one almost abortive) in A. hintoniorum. Pericarp ornamentation varies in Antiphytum from tuberculate (or verrucate), papillate (or scabrate)- echinate, tuberculate-papillate (or verrucate-scabrate), to striate. Tubercules (or verrucas) could be aligned and form a striate pattern in some cases. Verrucate surfaces are characteristic to all the basal rosette-species and to the South American A. cruciatum; scabrate-echinate surfaces are found only in A. hintoniorum, while striate surfaces are privative of A. berroi. The remaining species of the genus, both from North and South American, have verrucate-scabrate ornamentation. Ogastemma pusillum has verrucas and a scale-like pattern in the surface. Three straight keels on the ventral side of the eremocarp could be present or not: two lateral (basal) and one ventral (apical), which is always present. This apical keel could be modified forming a projection-like stipites in all the blue corolla species from South America, or a kind of bump only in A. floribundum (Fig. 2b). While the cicatrix of the eremocarp varies in position and form in correspondence with the shape of the gynobase, the cicatrix area keeps correspondence with the presence of the projection of the ventral keel, since the smallest cicatrixes are found in the aforementioned species. Position of the cicatrix is necessary basal in the species with projection-like stipites, suprabasal in A. heliotropioides and A. hintoniorum; inframedial in A. charruasorum, A. geoffreyi and all the basal-rosette species; and medial in A. brevicalyx and A. peninsulare. Only in A. charruasorum there is a projection of the cicatrix tissue resembling a stipe, which could be an 74 elaiosome (Fig. 2b). In Ogastemma pusillum, the cicatrix extends from the middle to the infrabasal part of the eremocarp. Finally, gynobase columella ranges in Antiphytum from flat to triangular, whereas is almost narrowly triangular (length/width relation 5:2) in Ogastemma. Blue corolla species from South America have flat to slightly raised (1:6) gynobases, but also they are found in A. floribundum and A. hintoniorum from North America. Very shallowly triangular (1:3) are found in A. charruasorum, and A. stoechadifolium from South America and A. floribundum, which besides could has shallowly triangular gynobases (1:2). Broadly triangular gynobases (6:5) are found in A. caespitosum and A. heliotropioides; but also very broadly triangular or deltate (1:1 and 5:6), such as in A. paniculatum, A. parryi and A. peninsulare. Finally, A. geoffreyi has a triangular columella (2:1). Distribution Antiphytum is an American genus with species mainly with restricted distributions (Figs. 3,4). In some cases, known occurrences are from several adjacent localities with no more of 100 km of distance in total (e.g., A. hintoniorum and A. peninsulare); two or three disjunct localities in a determinate political state (e.g., A. brevicalyx, A. geoffreyi) or narrow region (i.e. no more of 100 km long (e.g., A. berroi¸ A. charruasorum); or only currently known of one reliable locality (e.g., A. paniculatum and A. stoechadifolium). Antiphytum caespitosum occurs mainly in Oaxaca state, Mexico and adjacent regions from the surrounding states; its distribution seems rather discrete, except for a southern locality disjunct to the main concentration of populations (Fig. 3). This species is sympatric in several localities with A. parryi, whose distribution range is the widest for the eastern species with basal-rosettes, and for many species of the genus. Antiphytum parryi distribution is only surpassed by the distribution ranges of A. floribundum and A. heliotropioides, which run from southern USA to central Mexico by the Sierra Madre Occidental or to southern Mexico by the Sierra Madre Oriental, respectively; both also in the Altiplano Mexicano. Except for these species, seven North American species of Antiphytum are endemic from Mexico (Fig. 3). In South America, nearly all the genus is either endemic or restricted to Uruguay, according to the updated geographical information (Fig. 4). Antiphytum cruciatum has the widest distribution, covering northern Uruguay and the entire longitudinal extension of Rio Grande do Sul State, in Brazil. Although A. berroi has historical records from the same Brazilian state, no current records have been made since then. 75 The genus exhibits in North America a marked edaphic affinity to calcareous and gypsum soils, being A. hintoniorum a gypsophile, A. peninsulare a gypsocline and, except A. floribundum, that never has been collected in gypsum, all are gypsovags (Meyer, 1986). Species gypsovags occurs out of gypsum only in calcareous soils, while A. peninsulare also grow in soils of volcanic origin, from which A. floribundum is restricted. The South American species also has preference for impoverished soils, and for A. charruasorum is known its preference to the calcareous ones. Phylogenetic Datasets Among the commonly used markers in Boraginaceae the spacer trnS–G could not be used in this study because of failure to amplifying it. Basic data set characteristics of the five loci used are provided in Table 2. Among the five amplified loci, matK and ndhF–rpl32 had the fewest available GenBank sequences for Echiochiloideae. Among the markers that successfully amplified, ITS was the easier to amplify, but most difficult to align. The intron rps16 was easy to amplify and relatively easy to align, but required a careful assessment of ambiguous alignment due to the relatively high number of SSRs, hotspots and complex gaps. The intergenic spacers trnL–F and ndhF–rpl32 behave quite different; trnL–F behaved similar to rps16, while ndhF–rpl32 was problematic to amplify requiring many trials, but it was one of the easier to align, including the inversion in one of the outgroups. The coding region of matK in general was easy to amplify for the South American species, but for the North American species it required an adjustment of the PCR program; as expected, it was easy to align. Because the terminals were combined only avoiding missing regions, the matrix with most terminals was ITS+trnL–F+rps16 with 113 terminals (102 from Echiochiloideae). In contrast, the combined matrix with the four chloroplast markers (matK+ndhF–rpl32 +trnL–F+rps16) had only 43 terminals (41 from Echiochiloideae); all of those terminals had sequences for ITS. Except for terminals of A. stoechadifolium, all Antiphytum species were available in each dataset, with good geographic representation for the more widely distributed species in North America. The combined matrix with all the regions had 594 DNA microstructural characters (Table 2; Appendix 1), of which only 29 were potentially parsimony informative, mainly from gaps, plus seven that were informative in the original matrices (Table 3). Tree topologies In all the analyses the earlier divergent clades are well resolved and supported while the tips of the trees are less resolved in greater or less degree depending on the marker. The combination of 76 ITS+ndhF–rpl32 has more terminals and retrieved the same topology obtained with the 5-loci dataset, although the support values may vary among them. Similarly, matK+ndhF–rpl32 also recovered the 5-loci topology with differences in some of the tip branches. As expected, the chloroplast markers contribute proportionally more to the resolution of the deeper branches, while ITS proportionally resolves better the grouping within a species (Fig. 5). All methods retrieve congruent trees with the consensus of the MP trees for the 5-loci trees (with more or less resolution) and therefore the consensus of these trees was used to map the support values obtained in the PP, MP-JK, ML-BS analyses (Fig. 6). The Echiochiloideae subfamily was retrieved as monophyletic in all analyses except for rps16. Both genera sampled with more than one species within Echiochiloideae, Antiphytum and Echiochilon, were recovered as monophyletic. The genus Echiochilon was the earliest divergent within the subfamily, although some of the sampled species were not recovered as monophyletic. Consistent with its phylogenetic position, E. persicum stands out by its branch lengths (number of changes), with 111 changes (in the 5-loci tree; Fig. 5). The phylogenetic position of Ogastemma is not consistent among the molecular markers, but all sampled individuals of this monotypic genus formed a clade in all cases. The genus is retrieved as sister to Antiphytum with the intergenic spacers ndhF–rpl32 and trnL–F, with the chloroplast combined matrix and with the 5-loci combined reduced matrix. With the other markers (ITS, matK, and rps16), as well as with ITS+trnL–F+rps16, Ogastemma appears nested in different positions within Antiphytum. In the 5-loci consensus, O. pusillum has a similar branch length (17 changes) as in any given species of Antiphytum (12.5 in average; Fig. 5). All loci were congruent supporting the monophyly of Antiphytum, including both the North and South American species. Three clades appear in at least one marker analysis –i.e. some locus or combined loci datasets retrieved at least one or none of these clades–: 1) a North American clade with all the basal-rosetted species (NA-rosette clade); 2) a North American clade without-rosette species (NA-clade), and 3) a clade of the three sampled South American species (SA-clade). None clade is retrieved with ITS, rps16 and the combination trnL–F+rps16; in contrast, the three clades appear at one time only with some combined datasets. In none of the consensus, the North American species form a monophyletic group. The resulting polytomy among these three clades can be caused by conflict or by lack of information, depending on the set of loci. These three clades appear in the 5-loci hypothesis (Fig. 6), no matter the method of reconstruction. However, the ML suggests a sister relationship between the SA-clade and the NA-rosette clade, but this relationship lacks statistical support. Whereas the NA-rosette clade and the SA-clade are supported in all the 5- 77 loci trees (PP, MP-JK, ML-BS), the NA-clade was supported only by PP in the Bayesian analysis and weakly supported by ML-BS in the ML tree (Fig. 6). This clade has weak or lacks statistical support (ML-BS or MP-JK, respectively) in the relationships among species (Fig. 6). In the 5-loci analysis (Fig. 6), all species in Antiphytum are supported as monophyletic, except for A. parryi that is paraphyletic with respect to A. paniculatum. Nevertheless, this is not the case for all the individual chloroplast marker analyses or their partial concatenations, where some species are not resolved (a polytomy includes more than one species) or they are retrieved as paraphyletic. In most chloroplast analyses A. berroi and A. cruciatum are unresolved, as well as A. caespitosum, A. paniculatum, and A. parryi. Two examples of paraphyletic species can be found in the ndhF–rpl32 analysis, with A. peninsulare paraphyletic to A. geoffreyi and A. brevicalyx paraphyletic to A. heliotropioides. The most problematic marker in this respect is rps16, that only retrieves two species as monophyletic. ITS provides an important contribution for supporting the monophyly of the species. For the 5-loci trees (Fig. 6), in the NA-rosette clade, A. floribundum is the earlier divergent species, followed by A. caespitosum, and A. parryi+A. paniculatum. In the NA-clade A. hintoniorum (gypsum edaphic endemic) is sister to two clades with the remaining species, one including A. heliotropioides and A. brevicalyx, the other including A. peninsulare and A. geoffreyi. In the SA- clade the white-flowered species, A. charruasorum, is retrieved as sister to the subclade with the two blue-corolla species, A. berroi and A. cruciatum. In general, relationships among Antiphytum species retrieved from chloroplast loci are not incongruent with those from the nuclear locus (from ITS), despite the lack of resolution (polytomies) for several subclades in the individual plastid loci trees. The main discrepancy depicted by both sources of evidence is the relationship between A. peninsulare and A. brevicalyx, which are sister in the ITS analysis, but appear in different clades in all the chloroplast analyses: A. peninsulare is sister to A. geoffreyi, while A. brevicalyx to A. heliotropioides. When ITS is combined with trnL–F+rps16 (the dataset with more terminals) the ITS phylogenetic signal is recovered while in the 5-loci analysis the chloroplast signal is recovered. In the 5-loci consensus there is a clear asymmetry in the different species branch lengths (Fig. 5): A. paniculatum (2), A. parryi (3), A. caespitosum (5), A. berroi (6) vs A. floribundum (16), A. geoffreyi (18), A. hintoniorum (24), A. charruasorum (29). The latest two are the earliest divergent species within their clades. The species with the broadest geographic distribution, A. heliotropioides, has the average number of changes (12), but notwithstanding provided mostly by chloroplast loci; within it, the branches are mostly supported by ITS. 78 Microstructural characters Microstructural characters were identified in all loci. For the 5-loci matrix, matK only had one informative SSR while ITS had 14, 13 of them gaps (Table 3; Fig. 7). Most of the microstructural characters are concentrated at the Echiochiloideae branch, mainly one- site or multiple-sites gaps in all regions but in matK (Fig. 7; Table 3). A few others are distributed along the 5-loci consensus without a clear pattern (Fig. 7). Two multiple-sites gaps in the ndhF– rpl32 locus characterize Echiochilon (Table 3). In the same region, a SSR (GAAAGA) in the aligned position 90–95 characterizes to Ogastemma, but the 5-loci dataset keept only one terminal. The sister relationship between Antiphytum and Ogastemma is characterized by two microstructural characters in the trnL–F: a synapomorphic 16-site gap (aligned positions 550–566) and one diagnostic but not synapomorphic SSR (CCTTG in the aligned positions 181–185), i.e., the insertion of the SSR occurred once within Echiochiloideae, but it is also present in the outgroup. A 10-sites gap in the same region (aligned positions 1115–1124) is diagnostic but not synapomorphic of Antiphytum (Table 3). Within Antiphytum no main clade is corroborated by microstructural characters, but some subclades and species are. One-site gaps in ITS supported both blue-corolla subclade and eastern rosette species subclade. On the other hand, the sister relationship between A. geoffreyi and A. peninsulare is characterizes by an SSR (CATTAA) in the ndhF–rpl32 region (aligned positions 343–348). From the same region, an SSR (AG in the aligned positions 688–690) is diagnostic for the eastern rosette species subclade, but it could be synapomorphic for the entire NA-rosette clade. Although several one-site gaps contribute to support species, all but one individual in A. floribundum have a notoriously big gap (263 sites) in the ndhF–rpl32 region (aligned positions 500– 762). Likewise, at the interior of the species, gaps provide evidence of divergence among individuals or populations. A prominent case is the 55-sites gap in the trnL–F region (aligned positions 977–1031) share by two individuals of A. brevicalyx. SSRs are less frequent than gaps and in general were not detected within species, except for one SSR (AAAT) in the trnL–F region (aligned positions 1573–1577) characteristic of a subclade within A. hintoniorum; this SSR is nevertheless not unique to the subclade occurring also outside Echiochiloideae. This species is by the way the only one corroborated by a microstructural character in the matK region, one SSR (CATTAA, aligned positions 343–348). In general, SSRs are 79 less prone to homoplasy than gaps, but every rule has an exception: in the 5-loci matrix only one SSR (TTCA in the rps16 aligned positions 157-160) is highly homoplastic (Fig. 7). Discussion Morphology of Antiphytum At first glance, morphological variation in Antiphytum might appear to be considerable, but it is rather a continuum or, better said, a set of modifications around a common morphological pattern. For instance, the inflorescence morphology varies around the number of pherophylls in each node, or the number of dichasial divisions, or the presence of foliose metaxyphylls, or the development or not of the boragoid in the dichasial division, but the inflorescence is in all the cases a terminal homocladic or heterocladic determinate thyrse. It would be bad enough that “the inflorescences from members of the same genus have to be associated with different inflorescence types” (Weberling 1989: 222). Although some features of Antiphytum such as corolla color, projection-like stipites, or presence a basal-rosette seem like a gap among the variation distribution, these features are associated to clearly delimited ecological o geographic groups in the genus, and hence, they could be explained in this context. Moreover, in contrast with the other genera of the subfamily, the variation is low, with the exception of the monotypic Ogastemma. In Echiochilon extreme variation is observed (Lönn 1999): from perennial to, even rare, annual plants with alternate or opposite phyllotaxy; glandular or eglandular trichomes; corollas actinomorphics to zygomorphics, with white, yellow, pink, blue, reddish, bluish or yellowish color; and projection-like stipites in the eremocarp of E. paciflorum (Stocks ex Wight) Långström & M.W.Chase. This genus is however monophyletic (Långström and Oxelman 2003; our results), although some species deserve more work. Within Antiphytum, we found that eremocarps are very useful for discriminating among species (Fig. 2b), but may be insufficient for discriminating some species of the eastern rosette species group, which could be related with recent divergence times. A curious pattern was found in the gynobase columella form, which southwards in the continent tends to be flat, but rather triangular northwards. Pollen morphology is not informative to separate Ogastemma from Antiphytum, but other morphological features related with the flower, such as presence of faucal appendages on the corolla throat, the corolla surpassing the calyx, and the height at which the stamens are inserted, for instance, clearly differentiates the genera. 80 At a morphological level, we therefore agree with Johnston (1923) that corolla color and phyllotaxy were scarcely generic differences to validate the segregation of Antiphytum. The monophyly of the genus as recovered in this study supports this agreement. Phylogeny The subfamily Echiochiloideae is monophyletic (Fig. 6), but some internal relationships and the corresponding taxonomic decisions are not free of conflict. Even though several ITS sequences for Echiochilon were taken from Långstrom and Oxelman (2003) through GenBank, the addition of other samples inside and outside the genus resulted in the possibility that E. johnstonii, E. longiflorum, and E. persicum are not monophyletic. Otherwise, our results are comparable with those of Långstrom and Oxelman (2003). This study raises awareness that the monotypic genus Ogastemma (O. pusillum) could be transferred to Antiphytum as not all the analyses support their sister relationship and contrary some place it well nested inside Antiphytum. Furthermore, the sister relationship would not prevent this taxonomic decision to be taken: the branch length of its species is like that in the many species of Antiphytum; fruit and inflorescence features have been compared to A. geoffreyi (Mendoza-Díaz et al. 2018); pollen morphology is almost identical to that in Antiphytum (see above). The fact that previous studies (Längstrom and Oxelman 2003; Weigend et al. 2013; Cohen 2014; Chacón et al. 2016) also recovered Ogastemma as sister to Antiphytum does not pose a strong argument because all those studies included very limited number of species of Antiphytum. Despite all this apparently overwhelming evidence, we consider that the 5-loci analyses are very robust and that the sister relation of Ogastemma and Antiphytum reflects the actual cladogenesis within the subfamily. Although the taxonomic concept of Antiphytum could be expanded to include the sister genus, there are several morphological differences among them. Within Echiochiloideae only Antiphytum has faucal appendages (see above); according to Lönn (1999) O. pusillum is annual, whereas all the species in Antiphytum are perennial; the otherwise diagnostic corolla surpassing the calyx, position of the stamen insertion within the corolla, and eremocarp cicatrix relative size and position in Antiphytum (see above) would have to become variable. At this time, we see no reason to destroy the clear diagnosis of Antiphytum and to disturb the nomenclatural stability of both genera: transferring the paleotropical Ogastemma to the neotropical Antiphytum would imply changing the current concepts that have been accepted in several regional floras. 81 Our study has the most comprehensive sampling in Antiphytum including 12 of the 13 recognized species as well as several representative individuals per species covering most of its distribution range and morphological variation. It corroborate the recognition of the three species recently described (Mendoza-Díaz et al. 2018, 2020a,b). More important, this is the first time that the South American taxa are included in a phylogeny. The effect of broadening the taxon sampling can be observed by comparing the results of Mendoza-Díaz et al. (2020a) with a sampling focused on the recognition of a new Mexican species using ITS+ndhF–rpl32. In that study, only the NA-rosette clade was retrieved, while the same loci combination retrieves three clades in our study. In the same way, the increasing of molecular markers with a subset of taxon sampling implemented with the 5- loci analyses, also retrieved those three clades. Notwithstanding, the reduced taxon sampling did not compromise the distribution range or morphological diversity representation for the genus. It is worth saying that it would have been ideal to have a more comprehensive loci coverage per sample, but even the fresh silica dried material posed amplification failures. The convergent results obtained when taxon representations or marker representation are privileged give us confidence that the backbone of the phylogenetic hypothesis, for Echiochiloideae in general and for Antiphytum in particular, has achieved stability. Nevertheless, it remains pending to further test the presence of the NA-clade and to solve its internal conflict depending on the marker; to confidently solve the relationships among the three major clades; to include A. stochaedifolium in phylogenetic analyses; and to further test the potential paraphyly of A. parryi. The potential paraphyly of A. parryi raises the possibility of merging it with A. paniculatum. Nevertheless, as currently circumscribed, both can be differentiated by the flower size and the absence or presence of faucal appendages (see above). Although both species are vegetatively very similar, as is the case in the entire subclade, these two floral features are strongly associated to the potential pollinators and thus can represent important traits driving divergence (Stebbins 1970; Bolnick and Fitzpatrick 2007); even though both species have not reached reciprocal monophyly at the molecular level. The subclade, which also includes A. caespitosum is supported by chloroplast and ITS changes, but the species inside are only supported by ITS (Fig. 5). This is an indication that the species within this subclade have diverged recently (Baldwin et al. 1995; Drouin et al. 2008; Poczai and Hyvönen, 2010; Leliaert et al. 2014). The geographic distribution of A. parryi overlaps with that of the other species, but A. paniculatum is only known from a single precise locality within the Tehuacan valley, where it is sympatric with A. parryi. It is possible that a genetic reproductive barrier has not yet be fully formed between both species, as A. paniculatum is only supported by two ITS synapomorphies while A. caespitosum is supported by five ITS changes, including three synapomorphies. 82 In contrast to the limited phylogenetic information supporting species within the A. caespitosum-A. paniculatum-A. parryi subclade, also in North America, A. hintoniorum outstands for the long branch supporting it (24 changes) and the several changes (4 and 5) distinguishing its individuals (Fig. 5). These long branches can be explained by being a relative ancient lineage, by an accelerated evolution due to edaphic factors, by other species in the lineage becoming extinct, or by a combination of all or some of them (as discussed by Phillipe et al. 2011; Swenson 2011; Bromham et al. 2015; but also see Sauquet 2013). Because the molecular support for A. hintoniorum is coupled with morphological uniqueness and given that it is the only species in the genus restritect to gypsum outcrops, which are naturally exposed as fragmented habitats and are believed to accelerate mutations (Moore and Jansen 2007; Aguirre-Liguori et al. 2014; Moore et al. 2014), and because the species is geographically restricted it would not be striking that all mentioned factors influenced this branch length pattern (see below for further discussion). With also a notoriously long branch (29 changes), but growing in South America, the recently described A. charruasorum (Mendoza-Díaz et al. 2020b) broke the paradigms for the genus because it has more in common with the NA species than with its sister group, unique with blue flowers. In addition to the flower color, there are many changes in the phenotypic traits among these species and the North American ones (see above). The long branch of A. charruasorum suggests that this species is more ancient than those with blue flowers (Swenson 2011), corroborating the polarity implied by the character optimization. Among the South American species, the phylogenetic position of A. stoechadifolium remains to be tested. Despite the relevance of this species for its historical role in the description of the genus, it is in general poorly known, with only two records from one reliable locality in Uruguay. Nevertheless, the morphological features in common with A. berroi (Mendoza-Díaz et al. 2022) suggest that they are probably sister. Microstructural characters Few sequence-based studies analyze the evolution of microstructural characters, (e.g., Graham et al. 2000; Kelchner 2000; Borsch et al. 2003, 2005; Drábková et al. 2004; Wanke et al. 2007; Borsch and Quandt 2009; Tian et al. 2019; Song et al. 2022) using them to support and diagnose clades in the phylogenetic hypothesis (e.g., Torres-Montúfar et al. 2017; Torres-Montúfar et al. 2020). For Boraginaceae this is very seldom (Mendoza-Díaz et al. 2020a). Despite the dramatic reduction in the number of potentially parsimony informative microstructural characters (from 594 to 29+7) in the 5-loci analyses, all markers contributed with at least some information and provided phylogenetic signal at several hierarchical levels, from the subfamily to 83 the individuals (Fig. 7). Contrary to the general pattern in the phylogenetic signal coming from the nts (transitions/tranversions), where the chloroplast signal is proportionally more concentrated towards the deepest branches and ITS towards the tips (Fig. 5), both the nuclear and chloroplast microstructural characters are more or less evenly distributed along the cladogram (Fig. 7). While short gaps coming from chloroplast are more concentrated in the deepest branches of the cladogram, chloroplast SSRs are more concentrated in the shallow branches. ITS in contrast only has one SSR that is autapomorphic for A. geoffreyi. Most gaps in both ITS and chloroplast are of one or two sites, but chloroplast also has longer gaps, all synapomorphic except the A. floribundum autapomorphic 263 positions deletion in ndhF–rpl32 (positions 500–762; character 21 in Table 3; Fig. 7), which is not present only in one of the sampled individuals of the species. In this molecular region there is an SSR that characterizes Ogastemma (GAAAGA; character 15 in Table 3, Fig. 7). Neither of the three main clades within Antiphytum is supported by microstructural characters. Within these clades, almost all SSRs are concentrated in the NA-clade, but only one of them is a synapomorphy supporting the conflicting relationship between A. peninsulare and A. geoffreyi. The broadening of this study sampling revealed that this SSR actually is a synapomorphy of the latter both species and not an autapomorphy of A. peninsulare, as was previously retrieved lacking A. geoffreyi samples (Mendoza-Díaz et al. 2020a). The other three SSRs are autapomorphic, and one more is even present at the level of individuals in A. hintoniorum. Worth mentioning is the matK SSR (CTAGAA) in A. hintoniorum (character 25 in Table 3, Fig. 7) because, as in any coding region, indels in multiples of three result in a differential number of amino acids (Metzgar et al. 2000), in this case adding two. This is the only matK microstructural change for the 5-loci matrix and coincides with the species inhabiting the most specialized habitat (it is a gypsophile) and that has one of the longest branches among Antiphytum. The type of character and hence its quality adds evidence to the already discussed scenarios (see above) that this species may represent an ancient lineage within Antiphytum, or it could reflect the distant gypsum outcrops where the species distributes. The single SSR outside the NA-clade is diagnostic of the eastern rosette species subclade, but this short insertion in ndhF–rpl32 (AG; character 22 in Table 3; Fig. 7) could have appeared earlier than the 263 positions deletion in the same marker in A. floribundum, thus being a synapomorphy of the entire NA-rosette clade. This clade hence is prone to microstructural mutations in this region of the marker. We have not found information on microstructural characters in Boraginaceae apart from Mendoza- Díaz et al. (2020a). Therefore, this study also provides the first insights into molecular evolution in the selected markers and sampling of Boraginaceae (Appendix 1). We found that the matK region 84 appeared in Boraginales as a functional gene by its trimeric structural mutations (Wanke et al. 2007). Also, within Echiochiloideae no inversions were found in any of the marker matrices, but outside the subfamily. Describing microstructural characters is important to generate a more complete understanding of molecular evolution in Antiphytum, but also in Boraginaceae. Marker selection In this study, we used some commonly used markers for assessing phylogenetic relationships at generic and higher levels in Boraginales. For example, to assess a phylogeny of Tiquilia Pers., Moore and Jansen (2006) used nuclear ITS; rps16 intron and matK and ndhF genes; in turn, Cohen and Davis (2012) a chloroplast set of five markers, including matK, for Lithospermum L.; Nazaire et al. (2014) used ITS and a set of 12 regions of chloroplast, including matK, ndhF–rpl32, trnL and trnL–trnF in their study on Mertensia Roth. For Heliotropium L., Luebert et al. (2011) used rps16, psbA–trnH, trnL–F, trnS–G, and ITS, almost the same set that Chacón et al. (2016) used in a broader sampling of Boraginaceae. In general, the combination of ITS with one of two chloroplast intergenic spacers, trnL–F or trnS–G, are commonly employed in phylogenetic analyses of genera of Boraginaceae (Cohen and Davis, 2012). Pourghorban et al. (2020) recently published a phylogenetic study of subtribe Cynoglossinae with ITS+rps16+trnL–F, which was the combined matrix with most terminals in our analyses. As in our case, they obtained a relatively poor topological resolution in the backbone at subtribe level. As it was previously shown (Mendoza-Díaz et al. 2020a) the combination of ITS+ndhF–rpl32 did not introduce significant conflict and provided support at different hierarchical levels, as the chloroplast genes evolved more slowly than the nuclear genes (Drouin et al. 2008). In our study, this combination retrieved the three described clades with internal resolution, but lacked support (MP- JK) for the Antiphytum entire clade, the three main clades and some internal relationships. Adding rps16+trnL–F corroborated the topology from and generally improved the support values. In this case, the sum of the five markers allowed “secondary signals” that maximized congruence among different sources of data in the phylogenetic inference (Nixon and Carpenter 1996). This includes the combination of chloroplast and nuclear markers because there is no evidence to suspect reticulation among species and, hence, the phylogenetic signal from both types of markers is congruent (Bull et al. 1993; Johnson and Soltis 1998). Despite the fact that rps16 is commonly used for phylogenetic analysis in Boraginaceae (e.g., Chacón et al. 2016; Chacón et al. 2019; Otero et al. 2019; Pourghorban et al. 2020) our results agree with those of Shaw et al. (2005 and references therein), who indicated that in general rps16 is not 85 variable enough to resolve infrageneric relationships. In our analyses, rps16 showed the lowest resolution even at subfamily level, not yet recovering Boraginaceae subfamilies, although it did support some tribes. After all, this marker added, at the most, little contribution to branch support in the combined matrix and probably could have been avoidable. In contrast, ndhF–rpl32 was a very useful marker with a good variation level, in agreement with the findings of Song et al. (2022), although is seldom used. To get a fully resolved phylogeny of Antiphytum, additional molecular information, including markers with an intermediate rate of evolution, is needed. Overall, we have achieved significant progress in our understanding of the group and the generation of much larger data sets, such as complete plastome data sets and/or hyb-seq data sets, can be guided by our results. Biogeography The retrieved monophyly of Antiphytum has not only implications for its systematics, but also for biogeography. First, our results confirm Antiphytum as another example of an American genus with amphitropical disjunct distribution (AAD), in this case trans-NA eastern-SA AAD (Simpson et al. 2017), since it occurred from southern USA to southern Mexico in North America and southern Brazil and mainly northern Uruguay in South America (Figs. 3,4). Species distribution agrees with the ecological zones characteristics of the amphitropical regions, SM, SBSh, SBWh in NA, and SCf in SA, but extends into southern Mexico in the TM zone (abbreviations and classification according to Davis and Holmgren 2001). Although Boraginaceae (sensu Chacón et al. 2016) has the second highest number of AAD representatives among all plant families (19 examples according to Simpson et al. 2017, without count Antiphytum), the south American distribution of the genus Antiphytum is unusual in that most Boraginaceae taxa occur mainly in western South America (Argentina, Chile, Colombia, Peru, and Venezuela), while Antiphytum is the only native Boraginaceae in Uruguay (according to records on Flora del Conosur, Zuloaga et al. 2018 and personal observation). In southern Brazil, Antiphytum occurs along with Moritzia DC. ex Meisn. and Thaumatocaryon Baill., unique genera from Boraginae tribe, subfamily Boraginoideae, and exclusively South American (Weigend et al. 2010). Second, with our data, we can propose an interpretation of the directionality of colonization of America from north to south. We argue that our evidence suggests a feasibly more recent blue- corolla subclade, in which occurred a corolla-pigment gain. Besides, two of the three possible scenarios for resolving the polytomy in the backbone of Antiphytum suggest that North American species do not share an immediate common ancestor. These hypotheses entail a single, more likely 86 origin in North America and subsequent dispersal into South America, rather than the need for two independent North America colonization events. In addition with our results, the extinct taxa from late Tertiary assignable to Echiochiloideae in North America (not published, but the fossils were erroneously described as Cryptantha by Thomasson 1979) along with the node age of the crown Antiphytum estimate in 9.4 Ma (Chacón et al. 2017), support a posterior colonization of Uruguay and Brazil from North America, very likely by long-distance dispersal (LDD), such as many examples in Boraginaceae (Chacón et al. 2017 and references therein; Guilliams et al. 2017; Mabry and Simpson 2018; Otero et al. 2019). This interpretation of the direction in the colonization may be associated with the biggest size of eremocarp in A. charruasorum, which may be disadvantageous for dispersal, but more advantageous at early stages of colonization after a dispersal event (Schenk and Saunders 2017; Otero et al. 2019). Therefore, we discuss further based on this hypothesis. Since Antiphytum is a small genus of suffruticose plants with all the South American species retrieved in one clade, no more than one LDD event is necessary to explain the colonization of South America. As in many cases, arises the question on how these plants reached this region, especially for Antiphytum, whose South American taxa are distributed outside of the general Pacific pattern of the others Boraginaceae genera, except for the endemics Moritzia and Thaumatocaryon (Weigend et al. 2010). Diaspore morphology usually is associated with some dispersal syndrome (van der Pijl 1982), but some LDD could be caused by both standard and nonstandard mechanisms of dispersal (i.e. not linked with the vector associated to the diaspore morphology) (Higgins et al. 2003). Schenk and Saunders (2017) argued, though, that the distance magnitude (7000–13,000 km) in AAD species implies that diaspores are more likely to disperse through their adaptive structures than in a nonstandard dispersal. Based on their results, they also argued that despite the effectiveness of endozoochory at shorter distances, epizoochory is a more likely dispersal mode, and that anemochory should not be discounted a priori as a dispersal syndrome in AAD plants (Schenk and Saunders 2017). For long distances, adhesive diaspores (epizoochory) could travel further than diaspores carried internally (endozoochory) or dispersed by the wind (anemochory), since it depend over the detection and removal by the vector (Sorensen 1986). Endozoochory has been postulated for Moritzia (Weigend et al. 2016; Chacón et al. 2017) and Cryptantha fruits (Mabry and Simpson 2018), with the latter example surmises the fruits (including the hispid calyces) being transported on the feathers of migratory birds flying south, perhaps in a single uninterrupted flight, as the best hypothesis (Mabry and Simpson 2018). Epizoochorous traits also were found in Omphalodeae and related with the LDD events in the tribe (Otero et al. 2019). 87 Lönn (1999) did not describe the diaspore in Ogastemma, but Chacón et al. (2017) considered that it presents two dispersal modes: epizoochory and anemochory. Lönn (1999) postulated the latter possibility of dispersal because sand storms should also be considered. Nevertheless, it is not clear how Chacón et al. (2017) classified the diaspora of Echiochiloideae members, since for both Antiphytum and Echiochilon sampled species are nutlets, but for Ogastemma is “fruit”, while there is also the “entire fruit” classification in other species (Chacón et al. 2017 in Supplementary Material). Calyx lobes in Echiochiloideae are divided almost to the base, and although the mature nutlets in many cases are kept within the acrescent calyces, they are easily visible and removable from. In Antiphytum, except for A. hintoniorum with only one eremocarp firmly appressed into the calyx, the calyx lobes extend horizontally and exhibit the eremocarps. Since features in Ogastemma pusillum fruit and individual eremocarps are pretty similar to those in Antiphytum, we suggest that the nutlet is mainly the diaspore in both cases, but it is possible that the entire fruit and/or a piece of plant , for example, a fruiting stem, could be the diaspore in epizoochory. Therefore, epizoocory may be applicable also for Antiphytum (and not only endozoocory, as postulated Chacón et al. 2017), but not anemocory, due that Antiphytum does not distribute in deserts where sand storms occur. Schenk and Saunders (2017) in their study used the term achory (without a dispersal syndrome) for some species that have fruits and seed that lack obvious structures associated with dispersal; they found that this was a common syndrome in AAD species (Schenk and Saunders 2017). These authors included in this category diaspores that were too large to be considered dust diaspores (>1 mm) and lacked structures to enhance dispersal, such as Antiphytum nutlets. If nutlet is the diaspore in Antiphytum, then achory implies that we know that the nutlet somehow dispersed, but we cannot hypothesize how (Schenk and Saunders 2017). Epizoochory by chance or endozoochory could be explanations of a concept that might be better characterized as polychorous (dispersal through multiple syndromes), according to Schenk and Saunders (2017). In any case, an in-depth study is needed to understand how these diaspores dispersed to successfully generate the AAD pattern. About the vectors, for the “long-distance dispersal” hypothesis concept, Guilliams et al. (2017) included both direct dispersal and stepwise dispersal and two nonmutually exclusive subhypothesis: dispersal by birds and dispersal by large mammals during the Great American Biotic Interchange (GABI). Biro and Whittal (2015) discussed over the plausibility of GABI as an alternative hypothesis to avian dispersal to explain amphitropic plant disjunctions for lineages whose dispersal events date coincide with the timing of the GABI migration (2.6-0.125 Ma) and without obvious characteristics that would facilitate bird dispersals. In general, american amphitropical disjunct studies often conclude that birds are the most likely animal vector, but due the long distance 88 between the North and South American localities, the challenge for the successful dispersal is related to the birds making such a travel without defecating (in endozoochory) or preening (in epizoochory) (Schenk and Saunders 2017 and references therein). Mammals, on the other hand, have dispersal limitations typical of terrestrial vertebrates (Guilliams et al. 2017), given their propensity to have small, intracontinental ranges and relatively show travel times (Schenk and Saunders 2017 and references therein). In the study of Guilliams et al. (2017), the GABI hypothesis was rejected for Amsinckiinae due to the asynchronous pattern inferred and the broad range of dates among the AAD events in the subtribe. Conversely, the estimated node age for crown Antiphytum (9.4 Ma by Chacón et al. 2017) does not preclude the possibility of dispersal affected by terrestrial vertebrates during the GABI, which joints with avian dispersal as scenarios to be tested in further studies. Patterns of diversification in Antiphytum The three clades retrieved in Antiphytum each have almost the same number of species. The NA- clade is the most diverse with five species and the NA-rosette and SA-clades has four species each (see above). Following the “taxonomic category” of Simpson et al. (2017), diversification in Antiphytum after the LDD could be assigned to a clade-clade category (diversification results in a clade that is sister to another clade). Simpson et al. (2017) found a correlation between this category and oldest divergence times, since it is expected that more time would be required for diversification of a single taxon into multiple species within a clade. Conversely, we argue that this pattern is not fulfilled in Antiphytum, due to our phylogenetic findings along with the expected recent divergence dates of the clades that comprise the genus. There are several interesting geographic, ecological and morphological aspects associated with each clade of Antiphytum that may be explaining its species richness. However, linking these patterns to process could be risky until further methods don't be applied to test those (Ng and Smith 2014). Despite this fact, we consider relevant and necessary to expound some hypothesis that could explain these patterns. NA-clade has species with all the edaphic preferences known in Antiphytum and includes the unique gypsophile, A. hintoniorum, a narrow edaphic endemic (Ferreira and Boldrini 2011) retrieved as sister of the remaining species, and the gypsocline A. peninsulare. This is the unique clade in Antiphytum in which its entire species have some degree of fidelity to gypsum (Meyer 1986). Distribution of this clade is equally broad, as in Mexico the species occur in the arid zones of five 89 biogeographical regions, that slightly varies according to the different definitions (Morrone 2019). These are the provinces of Baja Californian, the Chihuahuan Desert, the Sierra Madre del Sur, the Sierra Madre Oriental, the Transmexican Volcanic Belt and the Balsas Basin (Morrone 2019, Fig. 3). Verrucate-scabrate pericarp ornamentation of the eremocarp prevails in almost all the species of the clade (scabrate-echinate is only present in the gypsophile), that besides possesses the most different presentations of the inflorescence. Three species have glandular-capitate trichomes on young branches, leaves and leafy organs such as bracts (pherophylls), flower-subtending bracts, and calyx. Most of the NA-rosette clade species are gypsovags, but A. floribundum, which is the sister taxon of the of eastern rosette species subclade, is the only NA gypsophobe of the genus. The main morphological feature of this clade is the presence of a basal-rosette, which originates one time in Antiphytum. The clade species distributes in arid zones of six biogeographical regions in Mexico, which coincides with the aforementioned ones, except for the Baja Californian province, but additionally along both the Sierra Madre Occidental and the Transmexican Volcanic Belt provinces (Morrone 2019). This clade possesses the only two species without fornices in the corolla throat associated to two corolla shapes and, according to the branching pattern of the phylogeny, to two independently origins: the absent of fornices in A. floribundum is associated with a campanulate corolla, but in A. paniculatum the corolla keeps the infundibuliform rotate shape common to all the species, which suggests a different and perhaps more recent process to that occurred in A. floribundum, maybe for being an incipient species (see above). The inflorescences are thyrsoids heterocladics in A. floribundum, but rather homocladics in the eastern rosette species subclade. Ornamentation of the pericarp is homogeneus in this clade (verrucate). Western distribution of A. floribundum correlates with the presence of basaltic parental rock, product of the uplift of the Sierra Madre Occidental and the Transmexican Volcanic Belt, and the volcanic derived soils as islands in the Chihuahuan Desert (Aide and Auken 1985), because this species does not occur outside soils derived from this substrate. None species has glandular-capitate trichomes. None of the SA-clade species has affinity to gypsum soils, but the unique white-flowered species is distributed over calcareous soil, such as the majority of species of Antiphytum. The SA-subclade of blue corollas distributes over soils which could be associated to basaltic parental rock. Then, this is the unique clade in Antiphytum with none fidelity degree to gypsum. The inflorescence in the blue corolla subclade is homogeneous, but in the SA clade overall is characterized by the presence of two pherophylls subtending each node of the inflorescence. Three pericarp ornamentations exist in 90 this clade: verrucate, verrucate-scabrate, and striate. Glandular-capitate trichomes are found only in the subclade of blue corollas. As diversification factors, we could surmise the following: 1) in the NA-clade could be invoked as diversity causal the edaphic component plus tectonic events that leaded to the formation of the regions where the species distribute; 2) in the NA-rosette clade could be the rosette basal along with the edaphic and geographic factors; 3) in the SA-clade, we could argue for the dispersal event plus the gain of the corolla pigment. In order to discuss some of these scenarios, diversification in North America, the edaphic factor and the morphological traits that could lead diversification in the genus are discussed below. Diversification in North America Many results relate the diversification of many xeric plants from North America with the Neogene uplift and the subsequent formation and differentiation of the various deserts in the Miocene (e. g. Moore and Jansen 2007; Loera et al. 2012; De Nova et al. 2018; Sosa et al. 2018). These regions, among super-arid, arid and semi-arid, occupy in Mexico more than fifty percent of territory (Sousa et al. 2018). Wilson and Pitts (2010) have already suggested that “evolutionary events [in North America] dating anywhere between 15 and 2 Ma could be considered associated with mountain building or desert formation activity” (p. 434). We know that Antiphytum probably originates in America during the Late Miocene (the node age of the crown Antiphytum is 9.4 Ma, according to Chacón et al. 2017), and that the subfamily Echiochiloideae overall has features to deal with arid environments: the species of Echiochilon and Ogastemma are shrublets or perennial herbs with more or less obvious adaptations to dry conditions (Lönn 1999) and according to this study Antiphytum inhabiting arid or at least dry regions both in North and South America. Hence, we could surmise niche conservatism to dry conditions in Antiphytum that along with the expansion of available habitat in North America may drive the diversification of the genus in this area (Loera et al. 2012; Gándara et al. 2014). For instance, Chacón et al. (2019) suggested the aridity as promoting habitat openness, which in turn acts as precursor to soil specialization; thus the increasing availability of semiarid open habitats in the Mid- Miocene may have likely allowed the diversification of Lithospermeae through its adaptations to arid habitats (Chacón et al. 2019). Sosa et al. (2018) also argued that arid regions formed in the Miocene offered suitable areas for colonization by arid-adapted plants that later diversified there. 91 Distribution of Antiphytum in North America (Fig. 3) coincides with most of the complex geological history of Mexico: the mountain systems known as Sierra Madre Oriental and Sierra Madre Occidental that run from north to south; the Trans-Mexican Volcanic Belt that rather runs west to east; the Sierra Madre del Sur mountain system; the Chihuahuan Desert, the Balsas Basin; and the Baja California Peninsula, which in the past was connected to the Mexican mainland, with an generally accepted split around 6 Ma (Wilson and Pitts 2010). In this regard, Sosa et al. (2018) found that the most important areas of endemism and phylogenetic endemism for the Mexican flora are associated with zones of topographical complexity, such as the main mountain systems and desert areas. Also, De Nova et al. (2018) found that vicariance by tectonic events during the Neogene drove the rapid diversification of Fouqueria. According to Antiphytum distribution in the aforementioned regions, it is more feasible to suppose that the Mexican Plateau was the ancestral area for Antiphytum and that events in more than one step, such as vicariance by tectonic events, isolating by abiotic or biotic factors, and dispersal, played a substantial role in the diversification of Antiphytum in North America. For instance, rifting of the Baja California Peninsula from Mexican mainland that produced speciation with a subsequent narrowly distribution (Dolby et al. 2015) is the highly likely case for A. peninsulare origin, which has its sister species (A. brevicalyx) mainland. Likewise, dispersal could have occurred from the Mexican Plateau to the Sierra Madre Oriental and southwards, to Sierra Madre del Sur, including the Balsas Basin, in both the two NA clades. This latter pattern of dispersal, from Mexican Plateau to Sierra Madre del Sur, was found in the dry land Milla clade (Gándara et al. 2014). The diversification of a number of plant lineages from arid lands in North America has been correlated with events from different geological periods (Gándara et al. 2014); Sosa et al. (2018) found that the majority of clade divergences in vascular plants of Mexico occurred during the Miocene (5–20 Ma). The complex geological and climatic history of Mexico that has given rise to the country's present appearance is a mosaic of distinct geological provinces with different ages and origins (Mastretta-Yanes et al. 2015; Sosa et al. 2018). Therefore, it is not striking that Antiphytum found favorable conditions to diversify into the nine species known so far in North America once the genus originated in America, either by vicariance and/or by dispersal in the Mexican territory. Testing is needed to tackle this question. 92 Edaphic factors Along with the traits to dealing with arid habitats, plants of the Echiochiloideae subfamily have to cope with the soils of these environments, since they have been reported to grow over limestone, gypsum, granite, or volcanic ground (Lönn 1999; personal observation). Antiphytum shows a marked edaphic preference for a particular kind of impoverished or challenging soils. Most species of this genus have affinity to calcareous soils, but six species are facultative in gypsum (A. brevicalyx, A. caespitosum, A. geoffreyi, A. heliotropiodes, A. paniculatum, and A. parryi); one is an edaphic endemic to gypsum (A. hintoniorum) and four species only seems to grow in soils with basaltic parental material (A. floribundum and SA-subclade of blue corollas); A. peninsulare (a gypsocline) distributes both in gypsum and basaltic parental material, but not in calcareous soil; and A. charruasorum is only known from this kind of soil. The Antiphytum clades with gypsum affinities species distribute only in Mexico, where gypsum outcrops are most highly concentrated in the Chihuahuan Desert region, but outcrops also occur in Baja California, Campeche, Chiapas, Guerrero, Oaxaca, Puebla, and Tamaulipas (Ochoterena et al. 2020). Insofar is known, within Echiochiloideae lineage only A. hintoniorum is gypsophile, and examples of gygsovagues also are Echiochilon longiflorum and E. persicum, but Ogastemma lacks of gypsovagues (Lönn 1999). Preference to limestone is high in the subfamily; nevertheless, gypsum can also be misidentified when occurring with calcium carbonate or more soluble salts (Casby- Horton et al. 2015) and therefore records in gypsum may be underestimated. Calcareous soils and gypseous soils are found naturally in arid and semiarid regions, where the annual low rainfall is insufficient for the leaching of the calcium carbonate or calcium sulfate (gypsum) accumulated in the soil (Parsons 1976; Casby-Horton et al. 2015; Weil and Brady 2017; Jha and Sivapullaiah 2017). This kind of soils has high pH level, low organic matter and unavailability or reduced micronutrient availability (Weil and Brady 2017), but rather physical constraints to plant growth, including surface crusts and root-restrictive layers in the subsoil (Casby-Horton et al. 2015) explain the limited flora: seedlings are unable to penetrate the hard crust typical of these soils in semiarid and arid climates (Douglas and Manos 2007; Moore et al. 2014; Casby-Horton et al. 2015; Escudero et al. 2015, and references therein). However, at chemical level, when carbonate concentrations may be found at or near the soil surface, serious micronutrient and phosphorus deficiencies can be induced in plants that are not adapted to calcareous conditions (Weil and Brady 2017). 93 Lineages specialized on gypsum are unlikely to have arisen from lineages with no latent or expressed gypsum tolerance (Douglas and Manos 2007). Some taxonomic lineages may have ‘preadaptations’ to coping with restrictions in gypsum soils (Escudero et al. 2015), as we propose that is the case for Echiochiloideae. Although multiple origins of gypsophyly are typical within plant lineages that appear to be ancestrally tolerant of gypsum (Moore et al. 2014), this is not the case in this subfamily. Despite this, the clade containing Antiphytum hintoniorum (NA-clade) possesses one gypsocline and numerous other taxa that are gypsovags, meeting the pattern reviewed by Moore et al. (2014). To explain the origin of the gypsophyle lineage of Antiphytum, two possible scenarios could be referred. One is that gypsum promoted the differentiation and origin of A. hintoniorum with or without specific adaptive responses to these environments (Escudero et al. 2015); the other one is that the lineage originated in other stressful environments from which it colonized gypsum soils (Escudero et al. 2015). We ignore the leaf chemistry of A. hintoniorum to know if there are or not specialized chemical adaptations to cope with the chemical restrictions of gypsum soils and hence, if A. hintoniorum follows a refuge model, since it is narrowly distributed (Palacio et al. 2007; Escudero et al. 2015). Then, it remains pending to test if its restriction to gypsum soils is related with special adaptation to gypsum soils or with a low competitive ability linked to its nature as stress-tolerant plant (Palacio et al. 2007). Antiphytum hintoniorum is precisely the most morphological divergent species in the genus, and its branch has many changes (Fig. 5). Accumulation of molecular changes can be expected by time or by environmental pressures in a relative short time span (Phillipe et al. 2011; Bromham et al. 2015). Coupled with the latter, this species shows reduction of vegetative and fertile structures (the inflorescence is a terminal dichasial cymoid reduced to one flower and each fruit bears only one eremocarp) that could be a response to the impoverished soil or a characteristic of endemism. For instance, reduction in flowers and seeds as a common trait among most of endemic species of the French Mediterranean region was reported by Lavergne et al. (2004). That A. hintoniorum is the only gypsophyle species in Antiphytum contrasts with the gypsophile flora of the Chihuahuan Desert region reviewed by Moore et al. (2014). This lack of diversity could be explained by the fact that gypsophily may be a relatively recent phenomenon in Antiphytum, of less than 9 Ma (conversely to that hypothesized for gypsophile species in general; see Moore and Jansen 2007; Aguirre-Liguori et al. 2014 and references therein), or an evolutionary dead end (Ng and Smith 2014), leading to the extinction of other species in the lineage, if there ever were any. The few and separate localities where A. hintoniorum occurs could be the result of posterior habitat fragmentation from a wider distribution due to climatic reasons or have been independently 94 colonized by long-distance dispersal, a process that several authors relate to evolutionarily old taxa due to the time they need to colonize isolated gypsum outcrops by chance (Moore and Jansen 2007; Escudero et al. 2015). The position of A. hintoniorum being the earliest divergent species of the NA-clade perhaps corresponds with being relative ancient, although long branches do not necessarily imply long times (Sauquet 2013). It could also be the unique survivor species of its lineage. Morphological changes with respect to the other taxa of the genus, and at molecular level among the individuals of different gypsum outcrops, shows how gypsum soil restriction has promoted morphological change and has drove genetic differentiation in A. hintoniorum (Bromham et al. 2015) by the island-like pattern of the gypsum deposits that are found in North American deserts (Moore and Jansen 2007; Aguirre-Liguori et al. 2014; Moore et al. 2014). Edaphic factors could also be involved in the origin of A. floribundum. We could suppose a parapatric speciation explanation for this species, since the soils of the Chihuahuan Desert are predominantly calcium carbonate, whereas volcanic derived soils distributed more locally throughout the desert as islands (Aide and Auken 1985). Calcareous soils are primarily preferred by Antiphytum, thus individuals of the ancestral Antiphytum rosetted lineage populations colonizing volcanic outcrops on the Mexican Plateau may have acquired traits related to this different soil as a new niche. Colonization of Antiphytum floribundum from the Mexican Plateau to the Sierra Madre Occidental and the Trans-Mexican Volcanic Belt, where volcanic soils are common, could be posterior and reached by dispersal. Lastly, we would like to point out a couple of correlation patterns involving soils of volcanic origin. It is interesting that only A. floribundum and SA-subclade of blue corollas species have a modification of the ventral keel of the eremocarp, either as a stipe (SA-subclade) or as a hump (A. floribundum). Moreover, the loss of the fornices with the consequent modification of the corolla occurred in A. floribundum, while the gain of pigment in the corolla occurred in the SA species with this kind of parental material in the soils. This raises the question if the soil of basaltic origin has been a factor driven or influenced these modifications in the eremocarps and corolla. Basal rosette It is far known that arid regions of North America have a wide range of rosette plants, from annuals (Mulroy and Rundel 1977) to evergreen (Kemp and Gardetto 1982; Martorell and Ezcurra 2002). Martorell and Ezcurra (2002) postulated that succulent evergreen rosette plants in Mexico not only are good water storing plants, but also that the succulent rosette morphologies could be an adaptive 95 response to environmental conditions prevailing on the mid-slopes because of their high capacity to harvest fog and low-intensity rains. In the case of Antiphytum species inhabit arid regions, leaves arranged in rosettes are basal, non- succulent and in contact with the soil. In fact, we were able to observe in some localities individuals from the subclade of eastern rosette species growth as prostrates in often openness areas. The fact that four species of Antiphytum (A. caespitosum, A. floribundum, A. paniculatum, A. parryi) bearing such conspicuous morphological feature, were retrieved in a clade, arises the question over whether the apparition of this trait leads to the diversification in the genus. In general, it has been found that grazing increases the abundance of rosette plants (Díaz et al. 2007), maybe by protecting the apical meristems. In another way, rosette growth-form was found to have a positive correlation with aridity (Nunes et al. 2017). Further, Cramer and Morrow (2007) found that prostrate leaves allow a small proportion of the total photosynthetic CO2 to be derived from soil CO2 with consequently reduced H2O loss. They concluded that probably other prostrate- leaved plants (e.g., geophytes, rosette-form herbs) also derive advantage from CO2 fluxing from the soil (Cramer and Morrow 2007). Maybe this trait has some advantages under drought (Nunes et al. 2017) and to conquer bare new habitats. Corolla color Flower color is evolutionarily labile, with examples of evolutionary divergence between closely related species from diverse genera and families (Wessinger and Rausher 2012). In Antiphytum, flower color transition occurred one time within the SA-clade by the gain of anthocyanin pigmentation from white; therefore, the blue color of the corollas in the subclade (A. berroi, A. cruciatum) is a derived trait, posterior to the long-distance dispersal event and, although differing in flower color does not by itself implicate a change at speciation, as other characters could have caused the initial divergence with flower color evolving anagenetically (Smith and Goldberg 2015), probably product of cladogenesis. In contrast, white flowers (lacking floral anthocyanins) is the condition for all the North American species of Antiphytum and the Uruguayan A. charruasorum, the sister taxa to the SA-subclade with blue corollas. Anthocyanin pigments are a flavonoid subclass that are a major contributor to the diversity of flower colors found across angiosperms (Ng et al. 2018) and are synthetized by the anthocyanin biosynthetic pathway (ABP), which is conserved across angiosperms and is well-characterized genetically (Wessinger and Rausher 2012). The ABP and its regulation are conserved through large 96 spans of evolutionary time, regardless of whether pigment is produced in flowers (Smith and Goldberg 2015), because nearly all higher plants produce anthocyanin in vegetative tissue (Sobel and Streisfeld 2013), with exception of the Caryophyllales (anthocyanins are entirely replaced by betalain pigments [Winkel 2006]). Anthocyanin pigments are also expressed in sepals, stems, leaves and even roots where they can be crucial to a plant's abiotic and biotic stress response (Dick et al. 2011), since they are involved in a variety of physiological functions (Winkel-Shirley 2001): in addition to their role in pigmentation in flowers, fruits, seeds, and leaves, which is useful for attracting pollinators and fruit dispersal agents (Rausher 2006), flavonoids have key roles for the defense and survival of the plants, and protection against UV radiation and drought (Winkel-Shirley 2001; Rausher 2006). According to Smith and Goldberg (2015), this range of functions may explain the deep conservation of the pathway across flowering plants. While almost all the functions of flavonoids are ancient (Rausher 2006), since the “early” steps in the pathway are found in the bryophytes (Winkel-Shirley 2001), the role of anthocyanin pigments as visual signals in angiosperms was acquired late in the evolutionary diversification of flavonoids (Rausher 2006). Although gains of novel traits are expected to occur less frequently than losses because there are typically more ways to break a pathway than to restore it (Sobel and Streisfeld 2013), this is not the case for the ABP, and therefore, gains of floral anthocyanin pigmentation may occur more readily than gains in other phenotypic traits (Sobel and Streisfeld 2013). A pattern common to all floral color transitions is that a mutation at a single locus is sufficient to cause a shift in flower color (Ellis and Field 2016), hence, transitions between pigmented and unpigmented flowers are frequent and reversible (e.g., Smith and Goldberg 2015; Ellis and Field 2016; Roguz et al. 2020). According to Ng et al. (2018), this macroevolutionary lability appears to arise from regulatory and coding mutations in rapidly evolving MYB transcription factors, which control anthocyanin expression and whose changes at these loci underlie polymorphisms within species and differences between closely related taxa, such as in this case between A. charruasorum and the SA-subclade of blue corollas. The gaining floral anthocyanin in this Antiphytum lineage likely did not require the restoration of a non-functioning pathway or the re-evolution of the entire pathway de novo, but instead a simple shift in regulation of the ABP genes into this expression domain through activation of this existing pathway in petals (Smith and Goldberg, 2015). Although flower color divergence may occur through genetic neutral drift or linkage to pleiotropic traits experiencing selection, in many cases such change is likely adaptive (Wessinger and Rausher 2012). Therefore, this morphological novelty in Antiphytum raises the question of whether the change was adaptative (or not), and if so, whether it was to biotic or abiotic factors, i.e. new 97 pollinators, herbivores or nectar robbers or dispersal agents, or by the environment and the geography, respectively. It is worthy not underrate the role of non-pollinator agents of selection in color traits mediated by flavonoids, because due to the numerous non-pollinator functions, correlations in anthocyanin expression in different plant tissues may lead to indirect selection on flower color or even predispose plants to a particular evolutionary trajectory (Strauss and Whittall 2006). Thus, in some systems, flower color may not be the primary target of natural selection, nor pollinators the primary agent (Vaidya et al. 2018). Since we don’t know the pollinators of Antiphytum in both disjunct regions, we are not able to compare and conclude changes in pollinators’ preferences. On the other side, although some studies have shown that in species with flower color polymorphism (FCP) –i.e. individuals produce white or pigmented mediated by anthocyanins flowers- white phenotypes have disadvantages relating to pollinator visits, sensibility to drought and water stress, greater inbreeding depression under stressful conditions, foliar herbivory, or reduction in the fitness (reviewed by Narbona et al. 2018), North American species of Antiphytum are distributed in arid environments and exposed to stressful drought and water conditions, but they are successful and diverse. Therefore, stress-related functions of anthocyanine are active in the vegetative tissues of these species, and probably the genus Antiphytum can independently regulate pigment expression in petals and vegetative tissues avoiding the deleterious effects of pigment loss in vegetative organs (Dick et al. 2011). At least the arid component in the environments that inhabits Antiphytum is shared among the species in North and South America. Other abiotic factors have regulatory effects on anthocyanin expression (Stimpson et al. 2021). In species with FCP, for example, changes to floral color can occur either through regulation of the amount of anthocyanin produced, through modulation to the pH of the vacuole, or through complexation with metal ions accumulated from soils (Stimpson et al. 2021). Anthocyanins are stored in the vacuole and vacuolar pH plays an important function in coloring anthocyanin pigments (Grotewold 2006). In petunia flowers and flowers of Ipomoea, blue flowers are produced when the vacuolar pH increases (Grotewold 2006), but in Hydrangea macrophylla (Thunb.) Ser. (Hydrangaceae), more acidic vacuoles by accumulation of aluminum due to acidic soils (at lower pH, aluminum bioavailability increases) produce a color change of the floral sepals to blue by the formation of an stable blue complex from the otherwise red anthocyanin (Stimpson et al. 2021). Then, edaphic conditions like soil pH and mineral nutrient concentration can also induce variation in flower color and flower color variation could be maintained as a phenotypically plastic trait driven by environmentally variable conditions (Vaidya et al. 2018). Nevertheless, it is mostly 98 unknown how soil properties influence flower color in nature (Vaidya et al. 2018). For Antiphytum with blue corollas, it is known that A. cruciatum grows in some arid localities of Rio Grande do Sul with marked acidic soils (Setubal 2010). Whether the edaphic factor triggered color change in the genus requires that the ancestral population in which the gain of pigmentation mutant appeared was already settled on acid soils and the floral ABP was functional. In contrast, a new floral color different from the ancestral white flowered lineage maybe not only leaded to the formation of a new lineage (SA-subclade of blue- corollas), but also was able to promote dispersal to a new region, altering patterns of gene flow and therefore, affecting the diversification in the genus (Ng & Smith 2014), since the species with blue- corolla are spatially disjunct with respect to A. charruasorum, which distributes only over calcareous soils in Uruguay, far away from the localities where are distributed the blue-corolla species. At this point it could be a circular argument, because we don't know what factor actually triggered the expression of floral anthocyanins in South America: if the mutant in the ancestral population appeared first and afterwards the colonization of environments with acid soils ocurred, or the mutant appeared over acid soils that promoted the blue color and sympatric speciation. We can suspect due to the above said, that the ABP likely is decoupled and therefore expressed in the vegetative tissues in white flowered lineage, as well as in the ancestral lineage of blue-corolla flowers and therefore, prone to become active in the petals. Nevertheless, even if we ignore all these factors, we can conclude that the pigment gain allowed the formation of the SA-subclade of blue-corollas, whose known diversity currently are represented by three species, and therefore it is a trait that affected the diversification in Antiphytum. Whether the origin of this novel trait was adaptive or not, and to what extent was adaptive, remains an open question worthy of further biological investigation. Conclusions The results of this work improve our understanding of the circumscription and phylogenetic relationships of Antiphytum, although some interspecific relationships are not resolved. Antiphytum, like many Boraginaceae genera, was long an elusive genus, but it is nowadays clear that it belongs to Echiochiloideae subfamily and that is a sister clade of Ogastemma. The genus is the second richest in the subfamily with 13 species (compared to 15 in Echiochilon) disjunct in North and South America and retrieved in three clades: NA-with basal-rosette plants; NA-without basal- rosette plants and SA-without basal rosette plants; the relationships among them remain to be determined. México is the most diverse country for the genus, since it hosts nine species, while 99 Uruguay has four species. USA and Brazil are the northernmost distribution for two species each in its corresponding region, respectively. The direction of the colonization of America is proposed from north to south and the disjunction is explained by a single event, probably long-distance dispersal, as in almost all the Boraginaceae groups with American amphitropical disjunct distribution. One event of corolla color (white to blue) and cicatrix position (basal in SA) change occurred likely after the dispersal event. Edaphic preferences, patterns of distribution and morphological traits associated to each clade of the genus open several questions about the process behind the diversification in Antiphytum. We suggest that tectonic events together with specialization to harsh soils were the main drivers of diversification in NA, while blue corolla probably promoted the colonization of new habitats in SA likely by the recruitment of novel pollinator or dispersal agents. The discoverer of A. charruasorum was fundamental to place relevant morphological, geographical and phylogenetic patterns, but overall the prior species recognition through taxonomic revision was the most important step before the phylogenetic reconstruction. We realized the dramatic change in the phylogenetic hypothesis as a function of sampling, even with the same markers. Therefore, we encourage continuing field exploration, mainly in South America, whose species were long behind a veil, in order to improve the field knowledge and sampling, to discover more populations or to include lacking species, such as A. stoechadifolium. Moreover, our results with ITS nuclear marker reveal some problems within Echiochilon, hence the whole subfamily Echiochiloideae requires additional study. Future studies should test other markers or methods in addition to a more comprehensive sampling to get a better phylogenetic resolution. Since the success in PCR amplification of all the markers and the informative degree of these are not guaranteed, next-generation sequencing methods might be a better option. Nevertheless, despite the lacking of resolution on the backbone of the phylogeny of Antiphytum, this work has the most comprehensive sampling of the genus, with almost all the known species from North and South America. With this study, we contribute with initial insights about its phylogeny, diversification and biogeographic patterns, and posit several questions to be dealt in-depth in the future. Identification key to the species of Antiphytum 1a. Suffruticose plants with white corolla flowers arranged or not in scorpioid cymes; eremocarp cicatrix at suprabasal, inframedial or medial position (known from North and South America)…..……………………………………………...…………………………………..………2 2a. Cauline leaves opposite throughout the plant, basal-rosette absent; flowers not arranged in scorpioid cymes (known from North and South America)……………………...…………….3 100 3a. Inflorescences with only one dichasial branching, therefore, branches with only one terminal flower (one-flowered cymoid); faucal appendages papillose; one whitish eremocarp per fruit generally; eremocarp cicatrix tissue not forming a projection; cicatrix at suprabasal position (known from North America)……………………….….....A. hintoniorum L.C.Higgins & B.L.Turner 3b. Inflorescences with more than one dichasial branching, therefore, branches with more than a terminal flower; faucal appendages hirsute-pubescent; four brown eremocarp per fruit generally; eremocarp cicatrix tissue forming a projection; cicatrix at inframedial position (known from South America)………………………………….A. charruasorum N.Mend. & Mar.Díaz 2b. Cauline leaves alternate throughout the plant, but sometimes opposite at the base, basal- rosette present or absent; flowers arranged in scorpioid cymes (known from North America)..…………………………………………………………………………...….……...4 4a. Plants with a basal-rosette; eremocarp cicatrix at inframedial position……………………………………………………………………………....5 5a. Flowers with five faucal appendages on the throat opposite to the corolla lobes……………………………………..………………………….6 6a. Corolla limb 5–8 mm diameter; scorpioid cymes up to 40 mm in length…...………………………..……A. caespitosum I.M.Johnst. 6b. Corolla limb (1–)2–4 mm diameter; scorpioid cymes up to 10 mm in length...........................................………A. parryi S.Watson 5b. Flowers without five faucal appendages on the throat opposite to the corolla lobes…………..………………………………………………….....7 7a. Corolla hypocrateriform (rotate); homocladic thyrsoids; scorpioid cymes up to 20 mm in length; eremocarp apical keel straight and flat ……………...……….A. paniculatum I.M.Johnst. 7b. Corolla infundibuliform-campanulate; heterocladic thyrsoids; scorpioid cymes up to 40 mm in length; eremocarp apical keel forming a bump…………………...A. floribundum (Torr.) A.Gray 101 4b. Plants without a basal-rosette; eremocarp cicatrix at suprabasal, inframedial or medial position ………………………………………………………...…………....8 8a. Calyx lobes shorter than the corolla tube at anthesis; cicatrix at medial position……………...…..A. brevicalyx N.Mend., Flores Olv. & H.Ochot. 8b. Calyx lobes equal or longer than the corolla tube at anthesis; cicatrix at suprabasal, inframedial or medial position……………………...………….9 9a. Scorpioid cymes proximal to the base of the plant; inflorescences without foliose metaxyphylls; gynobase columella triangular (2:1); cicatrix at medial position........A geoffreyi N.Mend. & Flores Olv. 9b. Scorpioid cymes distal to the base of the plant; inflorescences with foliose metaxyphylls; gynobase columella broadly triangular (6:5) to very broadly triangular or deltate (1:1 and 5:6); cicatrix at suprabasal or medial position…………………………………..…10 10a. Flowers subsessile, pedicels 1.2 mm in length; corolla limb 10–12 mm diameter; gynobase columella very broadly triangular or deltate (1:1 and 5:6); cicatrix at medial position……..................A. peninsulare (Rose) I.M.Johnst. 10b. Flowers pedicellate, pedicels up to 14 mm in length; corolla limb 4–6 mm diameter; gynobase columella broadly triangular (6:5); cicatrix at suprabasal position…………………..…...….A. heliotropioides A.DC. 1b. Suffruticose plants with blue corolla flowers arranged in scorpioid cymes; eremocarp cicatrix at basal position (known from South America)……………..………………………………...……….11 11a. Leaves 3–4(–6) mm wide, oblanceolate; eremocarps 1.3–1.4 × 1.0–1.1 mm……………………………………….…………….………….A. cruciatum (Cham.) DC. 11b. Leaves less than 3 mm wide, linear-lanceolate or linear-oblong to linear-oblanceolate; eremocarps 1.6–2.0 × 1.3–1.6 mm………………………………...…………………………12 102 12a. Pericarp ornamentation more or less striate, leaves up to 3 cm long ………………………………...…….A. berroi (Arechav.) N.Mend. & Flores Olv. 12b. Pericarp ornamentation tuberculate-granulate; leaves ca. 5 cm long…………………...…………………………..A. stoechadifolium (Cham.) DC. References Aguirre-Liguori, J. A., E. Scheinvar, and L. E. Eguiarte. 2014. Gypsum soil restriction drives genetic differentiation in Fouquieria shrevei (Fouquieriaceae). Amer. J. Bot. 101(4): 730– 736. https://doi.org/10.3732/ajb.1400031 Aide, M. and O. W. Van Auken. 1985. Chihuahuan desert vegetation of limestone and basal slopes in West Texas. Southwest. Nat. 30(4): 533–542. https://doi.org/10.2307/3671047 Alfaro, M. E. and M. T. Holder. 2006. The posterior and the prior in Bayesian phylogenetics. Annual Rev. Ecol. Evol. Syst. 37:19–42. https://doi.org/10.1146/annurev.ecolsys.37.091305.110021 Baldwin, G. B., M. J. Sanderson, J. M. Porter, M. F. Wojciechowski, C. S. Campbell, and M. J. Donoghue. 1995. The ITS region of nuclear ribosomal DNA: A valuable source of evidence on Angiosperm phylogeny. Ann. Missouri Bot. Gard. 82(2): 247–277. https://doi.org/10.2307/2399880 Bentham, G. and J. D. Hooker. 1876. Boragineae. In: Genera Plantarum: ad exemplaria imprimis in Herberiis Kewensibus servata definita. Vol. 2. Part 2. London: Reeve and Co. and Williams and Norgate. Pp. 836–865. Biro, P. and J. B. Whittal. 2015. The age of amphitropic plant disjuncts coincides with the Great American Biotic Interchange. California Native Plant Society 2015 Conservation Conference, San Jose, California, USA. Available at https://www.researchgate.net/publication/267288563_The_age_of_amphitropic_plant_disjun cts_coincides_with_the_Great_American_Biotic_Interchange Bolnick, D. I. And B. M. Fitzpatrick. 2007. Sympatric speciation: Models and empirical evidence. Annu. Rev. Ecol. Evol. Syst. 38: 459–487. https://doi.org/10.1146/annurev.ecolsys.38.091206.095804 Borsch, T. and D. Quandt. 2009. Mutational dynamics and phylogenetic utility of noncoding chloroplast DNA: Plant Syst. Evol. 282: 169–199. https://doi.org/10.1007/s00606-009-0210-8 Borsch, T., W. Hilu, D. Quandt, V. Wilde, C. Neinhuis, and W. Barthlott. 2003. Non-coding plastid trnT–trnF sequences reveal a well resolved phylogeny of basal angiosperms. J. Evol. Biol. 16: 558–576. https://doi.org/10.1046/j.1420-9101.2003.00577.x 103 Borsch, T., C. Löhne, K. Müller, K. W. Hilu, S. Wanke, A. Worberg, W. Barthlott, C. Neinhuis, and D. Quandt. 2005. Towards understanding basal angiosperm diversification: Recent insights using rapidly evolving genomic regions. Nova Acta Leopoldina NF 92(342): 85–110. Brand, A. 1926. Decas specierum novarum sexta. Repert. Spec. Nov. Regni Veg. 22:100–105. Brand, A. 1929. Verweisung des Gattungsnamens Antiphytum in die Synonymie. Repert. Spec. Nov. Regni Veg. 27: 145–149. Brand, A. 1931. Borraginaceae-Borraginoideae-Cryptantheae. Pflanzenr. 27: 1–236 Bromhan, L., X. Hua, R. Lanfear, and P. F. Cowman. 2015. Exploring the relationships between mutation rates, life history, genome size, environment, and species richness in flowering plants. Amer. Naturalist. 185(4): 507–524. https://doi.org/10.1086/680052 Bull, J. J., J. P. Huelsenbeck, C. W. Cunningham, D. L. Swofford, and P. J. Waddell. 1993. Partitioning and combining data in phylogenetic analysis. Sys. Biol. 42: 384–397. https://doi.org//10.1093/sysbio/42.3.384 Candolle, A. P. de and Candolle, A. L. P. P. de. 1846. Borragineae. In: Candolle, A. P. (Ed.) Prodromus systematis naturalis regni vegetablis, sive, Enumeratio contracta ordinum generum specierumque plantarum huc usque cognitarium, juxta methodi naturalis, normas digesta, Pars Decima. Paris: Victoris Masson. Pp. 117–178. Casby-Horton, S., J. Herrero and N. A. Rolong. 2015. Gypsum soils—their morphology, classification, function, and landscapes. Advances Agron. 130: 231–290. https://doi.org/10.1016/bs.agron.2014.10.002 Chacón, J., F. Luebert, H. H. Hilger, S. Ovcinnikova, F. Selvi, L. Cecchi, C. M. Guilliams, K. Hasenstab-Lehman, K. Sutorý, M. G. Simpson, and M. Weigend. 2016. The borage family (Boraginaceae s.str.): A revised infrafamilial classification based on new phylogenetic evidence, with emphasis on the placement of some enigmatic genera. Taxon 65: 523–546. https://doi.org/10.12705/653.6 Chacón, J., F. Luebert, and M. Weigend. 2017. Biogeographic events are not correlated with diaspore dispersal modes in Boraginaceae. Front. Ecol. Evol. 5: 26. https://doi.org/10.3389/fevo.2017.00026 Chacón, J., F. Luebert, F. Selvi, L. Cecchi, and M. Weigend. 2019. Phylogeny and historical biogeography of Lithospermeae (Boraginaceae): Disentangling the possible causes of Miocene diversifications. Molec. Phylogen. Evol. 141: 106626 https://doi.org/10.1016/j.ympev.2019.106626 104 Cohen J. I. and J. I. Davis. 2012. Molecular phylogenetics, molecular evolution, and patterns of clade support in Lithospermum (Boraginaceae) and related taxa. Syst. Bot. 37(2): 490–506. https://doi.org/10.1600/03634412X635539 Cohen, J. I. 2014. A phylogenetic analysis of morphological and molecular characters of Boraginaceae: Evolutionary relationships, taxonomy, and patterns of character evolution. Cladistics 30(2): 139–169. https://doi.org/10.1111/cla.12036 Cramer, M. D. and C. K. Morrow. 2007. Does the prostrate-leaved geophyte Brunsvigia orientalis utilize soil-derived CO2 for photosynthesis? Ann. Bot. 99(5): 835–844. https://doi.org/10.1093/aob/mcm019 Darriba, D., G. L. Taboada, R. Doallo, and D. Posada. 2012. jModelTest 2: More models, new heuristics and parallel computing. Nature Meth. 9: 772. https://doi.org/10.1038/nmeth.2109 Davis, R. and P. Holmgren. 2001. FRA 2000: Global ecological zoning for the global forest resources assessment 2000, final report. Forest Resources Assessment Programme, Working Paper, vol. 56. Rome: Food and Agriculture Organization. De Nova, J. A., L. L. Sánchez-Reyes, L. E. Eguiarte, and S. Magallón. 2018. Recent radiation and dispersal of an ancient lineage: The case of Fouquieria (Fouquiericeae, Ericales) in North American deserts. Molec. Phylogen. Evol. 126: 92–104. https://doi.org/10.1016/j.ympev.2018.03.026 Díaz, S., S. Lavorel, S. McIntyre, V. Falczuk, F. Casanoves, D. G. Milchunas, C. Skarpe, G. Rusch, M. Sternberg, I. Noy-Meir, J. Landsberg, W. Zhang, H. Clark, and B. D. Campbell. 2007. Plant trait responses to grazing-a global synthesis. Global Change Biol. 13(2):313–341. https://doi.org/10.1111/j.1365-2486.2006.01288.x Dick, C. A., J. Buenrostro, T. Butler, M. L. Carlson, D. J. Kliebenstein, and J. B. Whittall. 2011. Arctic mustard flower color polymorphism controlled by petal-specific downregulation at the threshold of the anthocyanin biosynthetic pathway. PLoS One 6(4): e18230. https://doi.org/10.1371/journal.pone.0018230 Dolby, G. A., S. E. K. Benett, A. Lira-Noriega, B. T. Wilder, and A. Munguía-Vega. 2015. Assessing the geological and climatic forcing of biodiversity and evolution surrounding the Gulf of California. J. Southwest 57(2–3): 391–456. Doyle, J. J. and J. L. Doyle. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19: 11–15. Douglas, N. A. and P. S. Manos. 2007. Molecular phylogeny of Nyctaginaceae: taxonomy, biogeography, and characters associated with a radiation of xerophytic genera in North America. Amer. J. Bot. 94(5): 856–872. https://doi.org/10.3732/ajb.94.5.856 105 Drábková, L., J. Kirschner, C. Vlček, V. Pačes. 2004. TrnL–trnF intergenic spacer and trnL intron define major clades within Luzula and Juncus (Juncanceae): Importance of structural mutations. J. Mol. Evol. 59: 1–10. https://doi.org/10.1007/s00239-004-2598-7 Drouin, G., H. Daoud, and J. Xia. 2008. Relative rates of synonymous substitutions in the mitochondrial, chloroplast and nuclear genomes of seed plants. Molec. Phylogen. Evol. 49: 827–831. https://doi.org/10.1016/j.ympev.2008.09.009 Edler, D., J. Klein, A. Antonelli, and D. Silvestro. 2021. raxmlGUI 2.0: A graphical interface and toolkit for phylogenetic analyses using RAxML. Methods Ecol. Evol. 12: 373– 77. https://doi.org/10.1111/2041-210X.13512 Ellis, T. J. and D. L. Field. 2016. Repeated gains in yellow and anthocyanin pigmentation in flower colour transitions in the Antirrhineae. Ann. Bot. 117: 1133–1140. https://doi.org/10.1093/aob/mcw043 Erdtman, G. 1960. The acetolysis method. Svensk Bot. Tidskr. 54: 561–564. Escudero, A., S. Palacio, F. T. Maestre, and A. L. Luzuriaga. 2015. Plant life on gypsum: a review of its multiple facets. Biol. Rev. 90: 1–18. https://doi.org/10.1111/brv.12092 Farris, J. S., V. A. Albert, M. Källersjö, D. Lipscomb, and A. G. Kluge. 1996. Parsimony jackknifing outperforms neighbor-joining. Cladistics 12: 99–124. https://doi.org/10.1006/clad.1996.0008 Ferreira, P. M. A., and I. I. Boldrini. 2011. Potential reflection of distinct ecological units in plant endemism categories. Conservation Biol. 25(4): 672–679. https://doi.org/10.1111/j.1523- 1739.2011.01675.x Fresenius, J. B. G. W. 1857. Familia Borragineae. Vol. 8 Part 1, Fasc. 19. In: von Martius, C. F. P., A. W. Eichler, and I. Urban (Eds.) Flora brasiliensis. Col. 63–64. Frohlich, M. W., M. Thulin and M. W. Chase. 2020. Ninety-three new combinations in Euploca for species of Heliotropium section Orthostachys (Boraginaceae sensu APG). Phytotaxa 434(1): 013–021. https://doi.org/10.11646/phytotaxa.434.1.2 Gándara, E., C. D. Specht, and V. Sosa. 2014. Origin and diversification of the Milla Clade (Brodiaeoideae, Asparagaceae): A Neotropical group of six geophytic genera. Molec. Phylogen. Evol.75: 118–125. https://doi.org/10.1016/j.ympev.2014.02.014 Goloboff, P. A. 1999. Analyzing large data sets in reasonable times: Solutions for composite optima. Cladistics 15: 415–428. https://doi.org/10.1006/clad.1999.0122 Goloboff, P. A., S. Farris, and K. Nixon. 2000. TNT (tree analysis using new technology) (BETA) ver. 1.1. Tucumán, Argentina: Published by the authors. 106 Graham, S. W., P. A. Reeves, A. C. E. Burns, and R. G. Olmstead. 2000. Microstructural changes in noncoding chloroplast DNA: Interpretation, evolution and utility of indels and inversions in basal angiosperm phylogenetic inference. Int. J. Plant Sci. 161: S83–S96. https://doi.org/10.1086/317583 Grotewold, E. 2006. The genetics and biochemistry of floral pigments. Annual Rev. Pl. Biol. 57(1): 761–780. https://doi.org/10.1146/annurev.arplant.57.032905.105248 Guilliams, C. M., K. E. Hasenstab-Lehman, M. E. Mabry, and M. G. Simpson. 2017. Memoirs of a frequent flier: Phylogenomics reveals 18 long-distance dispersals between North America and South America in the popcorn flowers (Amsinckiinae). Amer. J. Bot. 104(11): 1717– 1728. https://doi.org/10.3732/ajb.1700369 Higgins, L. and B. L. Turner. 1983. Antiphytum hintoniorum (Boraginaceae), a Bizarre New Gypsophile from North-Mexico. Southwest Nat. 28(4): 457–458. Higgins, S. I., R. Nathan, and M. L. Cain. 2003. Are long-distance dispersal events in plants usually caused by nonstandard means of dispersal? Ecology 84(8): 1945–1956. https://doi.org/10.1890/01-0616 Hilger, H. H. 2014. Ontogeny, morphology, and systematic significance of glochidiate and winged fruits of Cynoglosseae and Eritrichieae (Boraginaceae). Pl. Diversity Evol. 131: 167–214. https://doi.org/10.1127/1869-6155/2014/0131-0080 Holstein, N. and M. Weigend. 2016. Lectotypifications and nomenclatural notes in the Boraginaceae s.str. for the Flora of Ecuador. Phytotaxa 278(1): 69–73. http://dx.doi.org/10.11646/phytotaxa.278.1.9 Jha, A. K., and P. V. Sivapullaiah. 2017. Unpredictable Behaviour of Gypseous/Gypsiferous Soil: An Overview. Indian Geotech. J. 47(4): 503–520. https://doi.org/10.1007/s40098-017-0239- 5 Johnson, L. A. and D. E. Soltis. 1998. Assessing congruence: Empirical examples from molecular data. In: Soltis, D. E., P. S. Soltis, and J. J. Doyle (Eds.) Molecular systematics of plants II. DNA sequencing. Boston: Kluwer Academic. Pp. 297–348. Johnston, I. M. 1923. Studies in the Boraginaceae. Contr. Gray Herb. 68: 43–80. Johnston, I. M.1924. Studies in the Boraginaceae III. 1. The Old Genera of the Boraginoideae. Contr. Gray Herb. 73:42–73. Johnston, I. M. 1927. Studies in the Boraginaceae VI: A revision of the South American Boraginoideae. Contr. Gray Herb. 78: 3–118. Johnston, I. M. 1957. Studies in the Boraginaceae XXIX: Echiochilon and related genera. J. Arnold Arbor.38: 255–294. 107 Johnston, I. M. 1966. Antiphytum. In: Lundell, C. L. (Ed.) Flora of Texas. Vol.1. Texas: Research Foundation. Pp. 196–200. Kelchner, S. A. 2000. The evolution of non-coding chloroplast DNA and its application in plant systematics. Ann. Missouri Bot. Garden 87: 482–498. https://doi.org/10.2307/2666142 Kemp, P. R. and P. E. Gardetto. 1982. Photosynthetic pathway types of evergreen rosette plants (Liliaceae) of the Chihuahuan Desert. Oecologia 55:149–156. https://doi.org/10.1007/BF00384480 Lavergne, S., J. D. Thompson, E. Garnier, and M. Debussche. 2004. The biology and ecology of narrow endemic and widespread plants: A comparative study of trait variation in 20 congeneric pairs. Oikos 107: 505–218. https://doi.org/10.1111/j.0030-1299.2004.13423.x Långström, E. and M. W. Chase. 2002. Tribes of Boraginoideae (Boraginaceae) and placement of Antiphytum, Echiochilon, Ogastemma and Sericostoma: A phylogenetic analysis based on atpB plastid DNA sequence data. Pl. Syst. Evol. 234: 137–153. https://doi.org/10.1007/s00606-002-0195-z Långström, E. and B. Oxelman. 2003. Phylogeny of Echiochilon (Echiochileae, Boraginaceae) based on ITS sequences and morphology. Taxon 52(4): 725–735. https://doi.org/10.2307/4135545 Leliaert, F., H. Verbruggen, P. Vanormelingen, F. Steen, J. M. López-Bautista, G. C. Zuccarello, and O. de Clerck. 2014. DNA-based species delimitation in algae. Eur. J. Phycol. 49(2): 179–196. http://doi.org/10.1080/09670262.2014.904524 Loera, I., V. Sosa, and S. M. Ickert-Bond. 2012. Diversification in North American arid lands: Niche conservatism, divergence and expansion of habitat explain speciation in the genus Ephedra. Molec. Phylogen. Evol. 65(2): 437–450. https://doi.org/10.1016/j.ympev.2012.06.025 Lönn, E. 1999. Revision of the Boraginaceae genera Echiochilon, Ogastemma and Sericostoma. Bot. J. Linn. Soc. 130: 185–259. https://doi.org/10.1111/j.1095-8339.1999.tb00521.x Luebert, F., G. Brokamp, J. Wen, M. Weigend, and H. H. Hilger. 2011. Phylogenetic relationships and morphological diversity in Neotropical Heliotropium (Heliotropiaceae). Taxon 60(3): 663–680. https://doi.org/10.1002/tax.603004 Luebert, F., T. L. Couvreur, M. Gottschling, H. H. Hilger, J. S. Miller and M. Weigend. 2016. Historical biogeography of Boraginales: West Gondwanan vicariance followed by long distance dispersal? J. Biogeogr. 44(1): 158–169. https://doi.org/10.1111/jbi.12841 108 Mabry, M. E. and M. G. Simpson. 2018. Evaluating the Monophyly and Biogeography of Cryptantha (Boraginaceae). Syst. Bot. 41(1): 53–76. https://doi.org/10.1600/036364418X696978 Macbride, J. F. 1916. Notes on Certain Borraginaceae. Contr. Gray Herb. 48: 39–58. Martens, M. and H. Galeotti. 1844. Enumeratio synoptica plantarum phanerogamicarum ab Henrico Galeotti in regionibus Mexicanus collectarum. Bull. Acad. Roy. Sci. Bruxelles 11(2): 319– 340. Martorell, C. and E. Ezcurra. 2002. Rosette scrub occurrence and fog availability in arid mountains of Mexico. J. Veg. Sci. 13: 651–662. https://doi.org/10.1111/j.1654-1103.2002.tb02093.x Mastretta-Yanes, A., A. Moreno-Letelier, D. Piñero, T. H. Jorgensen, and B. C. Emerson. 2015. Biodiversity in the Mexican highlands and the interaction of geology, geography and climate within the Trans-Mexican Volcanic Belt. J. Biogeogr. 42: 1586–1600. https://doi.org/10.1111/jbi.12546 Meisner, C. F. 1840. Plantarum vascularium genera. Leipzig: Libraria Weidmannia. Mendoza-Díaz, N., H. Flores-Olvera, M. G. Simpson, and M. J. Moore. 2018. A new and unusual endemic species from the Chihuahuan Desert, Mexico: Antiphytum geoffreyi (Boraginaceae, Echiochiloideae). Phytotaxa 367(3): 275–283. https://doi.org/10.11646/phytotaxa.367.3.7 Mendoza-Díaz, N., H. Ochoterena, M. J. Moore, and H. Flores-Olvera. 2020a. Molecular and morphological evidence reveals a new species of Antiphytum (Echiochiloideae, Boraginaceae) from Guerrero, Mexico. Syst. Bot. 45(1): 190–199. https://doi.org/10.1600/036364420X15801369352469 Mendoza-Díaz, N., M. Díaz, P. Brussa, F. Muñoz, J. M. Bonifacino, and H. Flores-Olvera. 2020b. White flowers in South America: A new species of Antiphytum (Echiochiloideae, Boraginaceae). Syst. Bot. 45(4): 913–920. https://doi.org/10.1600/036364420X16033963649291 Mendoza-Díaz, N., J. M. Bonifacino, M. Díaz, and H. Flores-Olvera. 2022. A Taxonomic Reevaluation of the South American Myosotis Species Described by José Arechavaleta. Novon 30: 80–91. https://doi.org/10.3417/2022743 Metzgar, D., J. Bytof, and C. Wills. 2000. Selection against frameshift mutations limits microsatellite expansion in coding DNA. Genome Res. 10(1): 72–80. https://doi.org/10.1101/gr.10.1.72 Meyer, S. E. 1986. The ecology of gypsophile endemism in the eastern Mojave Desert. Ecology 64: 1303–1313. https://doi.org/10.2307/1938686 109 Mociño, J. M. and M. Sessé. 2010. La Real Expedición Botánica a Nueva España. Vol. 5. Siglo XXI Editores y Universidad Nacional Autónoma de México. Pp. 360. Moore, M. J. and R. K. Jansen. 2006. Molecular evidence for the age, origin, and evolutionary history of the American desert plant genus Tiquilia (Boraginaceae). Molec. Phylogen. Evol. 39: 668–687. https://doi.org/10.1016/j.ympev.2006.01.020 Moore, M. J. and R. K. Jansen. 2007. Origins and biogeography of gypsophily in the Chihuahuan Desert plant group Tiquilia subg. Eddya (Boraginaceae). Syst. Bot. 32(2): 392–414. https://doi.org/10.1600/036364407781179680 Moore, M. J., J. F. Mota, N. A. Douglas, H. Flores-Olvera, and H. Ochoterena. 2014. The ecology, assembly and evolution of gypsophile floras. In: Rajakaruna, N., R. S. Boyd, and T. B. Harris (Eds.) Plant ecology and evolution in harsh environments. New York: Nova Science, Hauppauge. Pp. 97–128. Morrone, J. J. 2019. Regionalización biogeográfica y evolución biótica de México: Encrucijada de la biodiversidad del Nuevo Mundo. Revista Mex. Biodivers. 90: e902980. https://doi.org/10.22201/ib.20078706e.2019.90.2980 Mousa, M. O. and S. Sh. Shahatha. 2021. Taxonomic study for the new record Ogastemma pusillum (Boraginaceae) in Iraq. Iraqi Journal of Agricultural Sciences 52(3): 724–735. Mulroy, T. W. and P. W. Rundel. 1977. Annual Plants: Adaptations to desert environments. BioScience 27(2): 109–114. https://doi.org/10.2307/1297607 Müller, J., K. Müller, C. Neinhuis, and D. Quandt. 2010. PhyDE-Phylogenetic Data Editor v. 0.9971. Available from: http://www.phyde.de/ Narbona, E., H. Wanh, P. L. Ortiz, M. Arista and E. Imbert. 2018. Flower color polymorphism in the Mediterranean Basin: Occurrence, maintenance and implications for speciation. Pl. Biol. 20: 8–20. https://doi.org/10.1111/plb.12575 Nazaire, M. and L. Hufford. 2012. A broad phylogenetic analysis of Boraginaceae: Implications for the relationships of Mertensia. Syst. Bot. 37(3):758–783. https://doi.org/10.1600/036364412X648715 Nazaire, M., X. Wang, and L. Hufford. 2014. Geographic origins and patterns of radiation of Mertensia (Boraginaceae). Amer. J. Bot.101(1): 104–118. https://doi.org/10.3732/ajb.1300320 Nixon, K. C., and J. M. Carpenter. 1996. On simultaneous analysis. Cladistics 12: 221–241. https://doi.org/10.1006/clad.1996.0016 Nixon, K. C. 1999. The Parsimony Ratchet, a new method for rapid parsimony analysis. Cladistics 15: 407–414. https://doi.org/10.1006/clad.1999.0121 110 Ng, J. and S. D. Smith. 2014. How traits shape trees: New approaches for detecting character state- dependent lineage diversification. J. Evol. Biol. 27: 2035–2045. https://doi.org/10.1111/jeb.12460 Ng, J., L. B. Freitas, and S. D. Smith. 2018. Stepwise evolution of floral pigmentation predicted by biochemical pathway structure. Evolution 72: 2792–2802. https://doi.org/10.1111/evo.13589 Nunes, A., M. Köbel, P. Pinho, P. Matos, F. de Bello, O. Correia, and C. Branquinho. 2017. Which plant traits respond to aridity? A criptical step to assess functional diversity in Mediterranean drylands. Agric. Forest Meteorol. 239: 176–184. https://doi.org/10.1016/j.agrformet.2017.03.007 Ochoterena, H. 2009. Homology in coding and non-coding DNA sequences: A parsimony perspective. Pl. Syst. Evol. 282: 151–168. https://doi.org/10.1007/s00606-008-0095-y Ochoterena, H., H. Flores-Olvera, C. Gómez- Hinostrosa, and M. J. Moore. 2020. Gypsum and plant sciences: A marvel of Cuatro Ciénegas and the Chihuahuan Desert. In: Mandujano, M. C., I. Pisanty, and L. E. Eguiarte (Eds.) Plant diversity and ecology in the Chihuahuan Desert, Cuatro Ciénegas Basin: An endangered hyperdiverse oasis. Switzerland: Springer Nature. Pp. 129–165. https://doi.org/10.1007/978-3-030-44963-6 Ortíz-Rodríguez, A. E., J. F. Ornelas, and E. Ruiz-Sanchez. 2018. A jungle tale: Molecular phylogenya and divergence time estimates of the Desmopsis-Stenanona clade (Annonaceae) in Mesoamerica. Molec. Phylogen. Evol. 122: 80–94. Otero, A., P. Jiménez-Mejías, V. Valcárcel, and P. Vargas. 2019. Worldwide long-distance dispersal favored by epizoochorous traits in the biogeographic history of Omphalodeae (Boraginaceae). J. Syst. Evol. 57(6): 579–593. https://doi.org/10.1111/jse.12504 Oxelman, B., M. Liden, and D. Berglund. 1997. Chloroplast rps16 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Plant Syst. Evol. 206: 393–410. https://doi.org/10.1007/BF00987959 Palacio, S., A. Escudero, G. Monserrat-Martí, M. Maestro, R. Milla, and M. J. Albert. 2007. Plants living on gypsum: Beyond the specialist model. Ann. Bot. 99: 333–343. https://doi.org/10.1093/aob/mcl263 Parsons, R. F. 1976. Gypsophily in plants-A review. Amer. Midl. Naturalist. 96(1): 1–20. https://doi.org/10.2307/2424564 Phillippe, H., H. Brinkmann, D. V. Lavrov, D. T. J. Littlewood, M. Manuel, G. Wörheide, and D. Baurain. 2011. Resolving difficult phylogenetic questions: why more sequences are not enough? PLoS Biol. 9(3): e1000602. https://doi.org/10.1371/journal.pbio.1000602 111 Poczai, P. and J. Hyvönen. 2010. Nuclear ribosomal spacer regions in plant phylogenetics: problems and prospects. Mol. Biol. Rep. 37: 1897–1912. https://doi.org/10.1007/s11033-009- 9630-3 Pourghorban, Z., Salmaki, Y., and M. Weigend. 2020. Phylogenetic relationships within the subtribe Cynoglossinae (Cynoglossoideae: Boraginaceae): New insights from nuclear and plastid DNA sequence data. Pl. Syst. Evol. 306: 45. https://doi.org/10.1007/s00606-020- 01671-x Rambaut, A. 2018. FigTree v. 1.4.4. Available from: http://tree.bio.ed.ac.uk/software/figtree/ Rambaut, A., A. J. Drummond, D. Xie, G. Baele, and M. A Suchard. 2018. Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7. Syst. Biol. 67(5):901–904. https://doi.org/10.1093/sysbio/syy032 Rausher, M. D. 2006. The evolution of flavonoids and their genes. In: Grotewold, E. (Ed.) The Science of Flavonoids. New York: Springer. Pp: 175–211. Roguz, K., M. K. Gallagher, E. Senden, Y. Bar-Lev, M. Lebel, R Heliczet, and Y. Sapir. 2020. All the colors of the rainbow: Diversification of flower color and intraspecific color variation in the genus Iris. Frontiers Pl. Sci. 11: 569811. https://doi.org/10.3389/fpls.2020.569811 Ronquist, F., M. Teslenko, P. van der Mark, D. L. Ayres, A. Darling, S. Höhna, B. Larget, L. Liu, M. A. Suchard, and J. P. Huelsenbeck. 2012. MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol. 61(3): 539–542. https://doi.org/10.1093/sysbio/sys029 Rose, J. N. 1890. Borraginaceae. Contr. U. S. Natl. Herb. 1: 85. Sang, T., D. J. Crawford, and T. F. Stuessy. 1997. Chloroplast DNA phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae). Am. J. Bot. 84 (9), 1120–1136. https://doi.org/10.2307/2446155 Sauquet, H. 2013. A practical guide to molecular dating. C. R. Palevol. 12: 355–367. https://doi.org/10.1016/j.crpv.2013.07.003 Schenk, J. J. and K. Saunders. 2017. Inferring long-distance dispersal modes in American amphitropically disjunct species through adaptive dispersal structures. Am. J. Bot. 104(1): 1756–1764. https://doi.org/10.3732/ajb.1700178 Sessé M. and J. M. Mociño. 1887. Plantae Novae Hispaniae. Mexici: Apud Ignatium Escalante. Bajos de San Agustín. Número 1. Pp. 21. Setubal, R. B. 2010. Vegetação campestre subtropical de um morro granítico no Sul do Brasil, Morro São Pedro, Porto Alegre, RS. Tesis de Maestría. Universidade Federal do Rio Grande do Sul. Pp.148. 112 Shaw, J., E. B. Lickey, J. T. Beck, S. B. Farmer, W. Liu, J. Miller, K. C. Siripun, C. T. Winder, E. E. Schilling, and R. L. Small. 2005. The tortoise and the hare II: Relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Am. J. Bot. 92(1):142–66. https://doi.org/10.3732/ajb.92.1.142 Simmons, M. P. and H. Ochoterena. 2000. Gaps as characters in sequence-based phylogenetic analyses. Syst. Biol. 49: 369–381. Simpson, M. G., L. A. Johnson, T. Villaverde and C. M. Guilliams. 2017. American amphitropical disjuncts: Perspectives from vascular plant analyses and prospects for future research. Amer. J. Bot. 104(11): 1600–1650. https://doi.org/10.3732/ajb.1700308 Smith, S. D. and E. E. Goldberg. 2015. Tempo and mode of flower color evolution. Amer. J. Bot. 102(7): 1014–1025. https://doi.org/10.3732/ajb.1500163 Sobel, J. M. and M. A. Streisfeld. 2013. Flower color as a model system for studies of plant evo- devo. Frontiers Pl. Sci. 4: 321. https://doi.org/10.3389/fpls.2013.00321 Song, Y., W. Zhao, J. Xu, M. Li, and Y. Zhang. 2022. Chloroplast genome evolution and species identification of Styrax (Styracaceae). Biomed Res. Int.: 5364094. https://doi.org/10.1155/2022/5364094 Sorensen, A. E. 1986. Seed dispersal by adhesion. Annual Rev. Ecol. Syst. 17: 443–463. https://doi.org/10.1146/annurev.es.17.110186.002303 Sosa, V., J. A. de Nova, and M. Vásquez-Cruz. 2018. Evolutionary history of the flora of Mexico: Dry forests cradles and museums of endemism. J. Syst. Evol. 56(5): 523–536. https://doi.org/10.1111/jse.12416 Stamatakis, A. 2014. RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30(9): 1312–1313. https://doi.org/10.1093/bioinformatics/btu033 Stebbins, G. L. 1970. Adaptative radiation of reproductive characteristic in Angiosperms, I: Pollination mechanisms. Annual Rev. Ecol. Syst. 1: 307–326. https://doi.org/10.1146/annurev.es.01.110170.001515 Stimpson, M. L., R. D. B. Whalley, L. McLean, N. J. Sadgrove, G. F. Padilla-Gonzalez, B. Van Wyk, J. Clay, and J. J. Bruhl. 2021. Colour of floral styles in the Banksia spinulosa Sm complex (Proteaceae) relates to the anthocyanin and flavonol profile, not soil pH. Phytochemistry 192: 112931. https://doi.org/10.1016/j.phytochem.2021.112931 Strauss S. Y. and J. B. Whittall. 2006. Non-pollinator agents of selection on floral traits. In: Harder, L. D. and S. C. H. Barrett (Eds.) Ecology and evolution of flowers. Oxford: Oxford University Press. Pp. 120–138. 113 Swenson, N. G. 2011. The role of evolutionary processes in producing biodiversity patterns, and the interrelationships between taxonomic, functional and phylogenetic biodiversity. Amer. J. Bot.98(3): 472–480. https://doi.org/10.3732/ajb.1000289 Taberlet, P., L. Gielly, G. Pautou, and J. Bouvet. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Mol. Biol. 17: 1105–1109. https://doi.org/10.1007/BF00037152 Thomasson, J. R. 1979. Late Cenozoic grasses and other Angiosperms from Kansas, Nebraska, and Colorado: Biostratigraphy and relationships to living Taxa. Kansas Geol. Surv. Bull. 218: 1–68. Tian, X., J. Ye, and Y. Song. 2019. Plastome sequences help to improve the systematic position of trinerved Lindera species in the family Lauraceae. PeerJ 7: e7662. https://doi.org/10.7717/peerj.7662 Torres-Montúfar, A., T. Borsch, S. Fuentes, T. Clase, B. Peguero, and H. Ochoterena. 2017. The new Hispaniolan genus Tainus (Rubiaceae) constitutes an isolated lineage in the Caribbean biodiversity hotspot. Wildenowia 47: 259–270. https://doi.org/10.3372/wi.47.47309 Torres-Montúfar, A., T. Borsch, S. Fuentes, J. Gutierrez, and H. Ochoterena. 2020. It is a not disaster: molecular and morphologically based phylogenetic analysis of Rondeletia complex (Cinchonoideae, Rubiaceae). Pl. Syst. Evol. 306: 26. https://doi.org/10.1007/s00606-020- 01630-6 Torrey, J. 1859. Boraginaceae. In: Emory, W. H. (Ed.) Report on the United States and Mexican boundary survey made under the direction of the Secretary of the Interior Vol. 2, Part. 1. Washington: Cornelius Wendell Printer. Pp. 135–142. Vaidya, P., A. McDurmon, E. Mattoon, M. Keefe, L. Carley, C. Lee, R. Bingham, and J. T. Anderson. 2018. Ecological causes and consequences of flower color polymorphism in a self-pollinating plant (Boechera stricta). New Phytol. 218: 380–392. https://doi.org/10.1111/nph.14998 Van der Pijl, L. 1982. Principles of dispersal in higher plants, Third edition. Heidelberg: Springer- Verlag. Pp. 218. https://doi.org/10.1007/978-3-642-87925-8 Wanke, S., M. A. Jaramillo, T. Borsch, M. Smain, D. Quandt, and C. Neinhuis. 2007. Evolution of Piperales—matK gene and trnK intron sequence data reveal lineage specific resolution contrast. Molec. Phylogen. Evol. 42: 477–497. https://doi.org/10.1016/j.ympev.2006.07.007 Watson, S. 1883. Contributions to American Botany. Proc. Amer. Acad. Arts. 18: 96–196. https://doi.org/10.2307/25138689 Weberling, F. 1989. Morphology of flowers and inflorescences. Cambridge: Cambridge University Press. Pp. 348. 114 Weigend, M., M. Gottschling, F. Servi and H. H. Hilger. 2010. Fossil and extant western hemisphere Boraginae, and the polyphyly of “Trigonitideae” Riedl (Boraginaceae: Boraginoideae). Syst. Bot. 35(2): 409–419. https://doi.org/10.1600/036364410791638423 Weigend, M., F. Luebert, F. Selvi, G. Brokamp and H. H. Hilger. 2013. Multiple origins for Hound´s tongues (Cynoglossum L.) and Navel seeds (Omphalodes Mill.). The phylogeny of the borage family (Boraginaceae s. str.). Molec. Phylogen. Evol. 68: 604-618. https://doi.org/10.1016/j.ympev.2013.04.009 Weigend, M., F. Luebert, M. Gottschling, T. L. P. Couvreur, H. H. Hilger and J. S. Miller. 2014. From capsules to nutlets –phylogenetic relationships in the Boraginales. Cladistics 30: 508- 518.https://doi.org/10.1111/cla.12061 Weigend, M., F. Selvi, D. C. Thomas, and H. H. Hilger. 2016. Boraginaceae. In: Kadereit, J. W. and V. Bittrich (Eds) The families and genera of vascular plants Vol. 14. Flowering plants: Eudicots; Aquifoliales, Boraginales, Bruniales, Dipsacales, Escalloniales, Garryales, Paracryphiales, Solanales (except Convolvulaceae), Icacinaceae, Metteniusaceae, Vahliaceae. Cham: Springer International Publishing. Pp. 41–102. Weil, R. R. and N. C. Brady. 2017. The Nature and Properties of Soils, Fifteenth Edition. Boston: Pearson. Pp. 1104. Wessinger, C. A. and M. D. Rausher. 2012. Lessons from flower colour evolution on targets of selection. J. Exp. Bot. 63(16): 5741–5749. https://doi.org/10.1093/jxb/ers267 White, T. J., T. Bruns, S. Lee, and J. Taylor. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M. A., D. H. Gelfand, J. J. Sninsky, and T. J. White (Eds) PCR Protocols: A guide to methods and applications. San Diego: Academic Press. Pp. 315–322. Wilson, J. S. and J. P. Pitts. 2010. Illuminating the lack of consensus among descriptions of earth history data in the North American deserts: A resource for biologists. Prog. Phys. Geogr. 34(4): 419–441. https://doi.org/10.1177/0309133310363991 Winkel, B. S. J. 2006. The biosynthesis of flavonoids. In: Grotewold, E. (Ed.) The Science of Flavonoids. New York: Springer. Pp: 71–95. Winkel-Shirley, B. 2001. Flavonoid biosynthesis. A colorful model for genetics, biochemistry, cell biology, and biotechnology. Pl. Physiol. 126: 485–493. https://doi.org/10.1104/pp.126.2.485 Zuloaga, F. O., O. Morrone, and M. J Belgrano. 2018. Catálogo de las plantas vasculares del Cono Sur. Instituto de Botánica Darwinion. San Isidro. Buenos Aires, Argentina. 115 Table 1. Antiphytum species achieved through taxonomic revision of the genus Species Publication date Nomenclatural synonyms Taxonomic synonyms Geographic distribution A. berroi (Arechav.) N.Mend. & Flores Olv. comb. nov. 2022 Myosotis berroi Arechav. (basionym, 1911) – SA A. brevicalyx N.Mend., Flores Olv. & H.Ochot. 2020 – – NA A. caespitosum I.M.Johnst. 1923 Amblynotopsis caespitosa (I.M.Johnst.) Brand (1929) Antiphytum nudicalces I.M.Johnst. (1923) Amblynotopsis nudicalces (I.M.Johnst.) Brand (1929) NA A. charruasorum N.Mend. & Mar.Díaz 2020 – – SA A. cruciatum (Cham.) DC. 1846 Anchusa cruciata Cham. (basionym, 1829) Chamissoniophila cruciata (Cham.) Brand (nom. superfl., 1929) – SA A. floribundum (Torr.) A.Gray 1874 Eritrichium floribundum Torr. (basionym, 1859) Krynitzkia floribunda (Torr.) A. Gray (1885) Amblynotopsis floribunda (Torr.) J.F.Macbr. (1916) Amsinckia mexicana M.Martens and Galeotti (1844) Amphibologyne mexicana (M.Martens and Galeotti) Brand (nom. superfl., 1931) Amblynotopsis durangensis J.F.Macbr. (1916) NA A geoffreyi N.Mend. & Flores Olv. 2018 – – NA A. heliotropioides A.DC. 1846 Eritrichium heliotropioides (A.DC.) Torr. (1859) Krynitzkia heliotropioides (A.DC.) A.Gray (1885) Amblynotopsis ehrenbergii Brand (1926) NA 116 Cryptantha heliotropioides (A.DC.) Loes. (1913) Amblynotopsis heliotropioides (A.DC.) J.F.Macbr. (1916) A. hintoniorum L.C.Higgins & B.L.Turner 1983 - – NA A. paniculatum I.M.Johnst. 1923 Lithospermum linifolium M.Martens and Galeotti (replaced synonym, 1844) Amblynotopsis paniculata (I.M.Johnst.) Brand (1929) Amblynotopsis linifolia (M.Martens and Galeotti) Brand (nov. comb., 1931) – NA A. parryi S.Watson 1883 Krynitzkia parryi (Watson) A.Gray (1885) Amblynotopsis parryi (Watson) J.F.Macbr. (1916) Amblynotopsis humilis Brand (1926) NA A. peninsulare (Rose) I.M.Johnst. 1923 Krynitzkia peninsularis Rose (basionym, 1890) Amblynotopsis peninsularis (Rose) J.F.Macbr. (1916) – NA A. stoechadifolium (Cham.) DC. 1846 Anchusa stoechadifolia Cham. (basionym, 1829) Chamissoniophila stoechadifolia (Cham.) Brand (nom. superfl., 1929) – SA 117 Table 2. Individual and combined datasets characteristics. nts= nucleotides; mch=non-gap microstructural characters ITS ndhF–rpl32 trnL–F rps16 matK ITS+trnL– F+rps16 Nuclear-plastid combined (5-loci dataset) Number of sequences (Echiochiloideae/Outgroup) 253 (207/46) 134 (95/39) 322 (123/199) 286 (115/171) 193 (65/128) 113 (102/11) 43 (41/02) Sequence length (nts) variation (minimum–) most common (–maximum) (267–)551— 663 (305–)494— 698(–749) (534–)848— 909(–922) (560–)720— 862(–897) (584–)725— 1017 NA NA Aligned sequence length (with hotspots/without hotspots) 721/712 849/836 1609/1463 1350/893 1053 NA/2213 NA/3907 Number of mch 13 19 90 50 9 153 181 Number of coded gaps 73 36 220 69 15 362 413 Potentially parsimony informative characters (nts/mch/gaps) 330/11/58 94/04/20 342/59/124 192/29/43 336/06/07 343/15/49 263/06/23 118 Table 3. Description of the parsimony informative microstructural characters in the 5-loci reduced matrix (43 terminals). The interpretation of the informativeness of the event is based on both, the consensus of the reduced matrix and the optimization of each character onto the full corresponding matrices Number Locus Microstructural character Aligned positions Derived state Supported group Interpretation 1 ITS Gap 139–141 – – Sensitive to actran or deltran optimization 2 ITS Gap 237 Deletion Antiphytum brevicalyx Autapomorphic but not diagnostic 3 ITS Gap 241 Deletion Echiochiloideae Synapomorphic and diagnostic 4 ITS Gap 245 Deletion A. hintoniorum Autapomorphic within Antiphytum, but also present in one species outside the genus 5 ITS Gap 512–513 Insertion Echiochiloideae Ambiguous mapping of the insertion: either synapomorphic for the subfamily or for the second branch within it 6 ITS Gap 513 Insertion Blue-corolla subclade Diagnostic, but not synapomorphic within Antiphytum 7 ITS Gap 518–519 Insertion – Homoplastic for A. geoffreyi and the OG 8 ITS Gap 518 Deletion Eastern rosette species subclade Diagnostic, but not synapomorphic 9 ITS Gap 519 Insertion Eastern rosette species subclade Diagnostic, but not synapomorphic 10 ITS SSR (TG) 522–523 Insertion A. geoffreyi Autapomorphic and diagnostic 11 ITS Gap 610 Insertion A. floribundum Autapomorphic and diagnostic 12 ITS Gap 617 Insertion Echiochiloideae Diagnostic but also present in Cynoglossoideae 13 ITS Gap 651 Insertion A. brevicalyx Autapomorphic and diagnostic 14 ITS Gap 674 Deletion Echiochiloideae Synapomorphic and diagnostic 15 ndhF– rpl32 SSR ( GAAAGA) 90–95 Insertion Ogastemma Autapomorphic and diagnostic 16 ndhF– rpl32 Gap 308 Insertion A. heliotropioides Autapomorphic and diagnostic 17 ndhF– rpl32 Gap 334–338 Deletion Echiochilon Synapomorphic and diagnostic 18 ndhF– rpl32 SSR (CATTAA) 343–348 Insertion A. geoffreyi +A. peninsulare Synapomorphic and diagnostic 19 ndhF– rpl32 Gap 412 Deletion Echiochiloideae Synapomorphic and diagnostic 20 ndhF– rpl32 Gap 472–691 Deletion Echiochilon Synapomorphic and diagnostic 21 ndhF– rpl32 Gap 500–762 Deletion A. floribundum Autapomorphic but not diagnostic because it is absent in one indivudual 119 22 ndhF– rpl32 SSR (AG) 688–690 Insertion Eastern rosette species subclade Diagnostic, but maybe synapomorphic for all the NA-rosette clade 23 ndhF– rpl32 Gap 835–345 Deletion A. brevicalyx Potentially autapomorphic and diagnostic, but with ambiguous mapping because it is close to the end of the reading 24 ndhF– rpl32 Gap 835–837 Insertion A. floribundum Potentially autapomorphic and diagnostic, but with ambiguous mapping because it is close to the end of the reading 25 matK SSR (CTAGAA) 161–166 Insertion A. hintoniorum Diagnostic, but also present outside Echiochiloideae 26 rps16 SSR (TTCA) with mutations 157–160 Deletion – Highly homoplastic 27 rps16 Gap 361–365 Deletion Echiochiloideae Synapomorphic and diagnostic 28 trnL– F SSR (CCTTG) 181–185 Insertion Antiphytum+Ogastemma Diagnostic, but also present outside Echiochiloideae 29 trnL– F Gap 349 Insertion Echiochiloideae Diagnostic, but also present outside Echiochiloideae 30 trnL– F Gap 550–566 Deletion Antiphytum+Ogastemma Synapomorphic and diagnostic 31 trnL– F Gap 977– 1031 Deletion A. brevicalyx Shared by two individuals of the species 32 trnL– F Gap 1115– 1124 Deletion Antiphytum Diagnostic, but also present outside Echiochiloideae 33 trnL– F SSR with mutation (GGAA/TGAA) 1209– 1216 Deletion Echiochiloideae Synapomorphic and diagnostic in the reduced matrix, plesiomorphic in the complete trnL–F matrix 34 trnL– F Gap 1420– 1425 Deletion Echiochiloideae Synapomorphic and diagnostic in the reduced matrix, plesiomorphic in the complete trnL–F matrix 35 trnL– F Gap 1460– 1465 Deletion Echiochilon Synapomorphic and diagnostic 36 trnL– F SSR (AAAT) with mutations 1573– 1577 Insertion Subclade of A. hintoniorum Shared by several individuals of the species and also present out of Echiochiloideae 120 Fig. 1 Distribution of subfamily Echiochiloideae: Antiphytum is American (black dots); Ogastemma (black diamonds) and Echiochilon (empty gray squares) are found in Africa and the Arabian Peninsula. The boxes in America show the regions where Antiphytum is found 121 Fig. 2 Pollen and eremocarps of Antiphytum in comparison to Ogastemma pusillum. The species are divided by genus and, within of Antiphytum, they are sorting according to geographical distribution and size. a Pollen in equatorial or oblique (above) and polar (below) view: i O. pusillum; ii A. berroi; iii A. cruciatum, iv A. brevicalyx, v A. caespitosum; vi A. floribundum; vii A. hintoniorum; viii A. peninsulare; ix A. heliotropioides. Scale bar equals to 10 μm. b Eremocarps in ventral view. The basal-rosette bearing species are highlighted with a bright green box: i O. pusillum; ii A. charruasorum; iii A. stoechadifolium; iv A. berroi; v A. cruciatum; vi A. peninsulare; vii A. hintoniorum; viii A. brevicalyx; ix A. geoffreyi; x A. heliotropioides, xi A. floribundum; xii A. caespitosum; xiii A. parryi; xiv A. paniculatum. Scale bar equals to 0.5 mm 122 Fig. 3 Distribution of the nine North American species of Antiphytum: A. brevicalyx (bright green diamonds); A. caespitosum (green pentagons); A. floribundum (wine-colored dots); A. geoffreyi (blue diamonds); A. heliotropioides (bright orange dots); A. hintoniorum (bright pink dots); A. paniculatum (red-lined empty circle); A. parryi (purple dots); A. peninsulare (bright red dots). Biogeographical regions in Mexico (according to Morrone 2019): a Baja Californian; b Sierra Madre Occidental; c Chihuahuan Desert; d Sierra Madre Oriental; e Transmexican Volcanic Belt; f Balsas Basin; g Sierra Madre del Sur provinces 123 Fig. 4 Distribution of the four South American species of Antiphytum: A. berroi (turquoise diamonds); A. charruasorum (pink triangles); A. cruciatum (wine-colored dots); A. stoechadifolium (green square). Distribution of A. berroi in Brazil is not shown (see Mendoza-Díaz et al. 2022). Departments of Uruguay: a Paysandú; b Río Negro; c Tacuarembó; d Rivera; e Cerro Largo; f Treinta y Tres; g Maldonado 124 Fig. 5 Strict consensus tree of the MP trees from the 5-loci matrix showing branch lengths (number of changes) and the contribution of either the chloroplast markers (on the left of the forward slash) or the nuclear ITS region (on the right). The label is red when most of the changes are from chloroplast markers; blue when due to ITS. The thicker branches are relevant for its number of changes and their length was corroborated against all MP trees. The explanation for other figure notations can be found in Fig. 6 125 Fig. 6 Consensus tree based on the MP analysis of the 5-loci dataset. In shaded green is highlighted the Echiochiloideae subfamily. The boxes show the three clades retrieved in Antiphytum: SA-clade (solid line box); NA-clade of species bearing basal-rosette (broken line box), and NA-clade of species lacking basal-rosette (dotted line box). The SA-subclade of blue corolla species is shown with a blue square, and in front, the pigment gain representation. In front of NA clade with basal- rosette images from this feature are shown. The NA clade lacking basal-rosette has species with some degree of fidelity to gypsum (soil shown in the figure) and the unique gypsophile species of the whole subfamily, A. hintoniorum (shown in the figure). Support values are shown on each branch (PP/ML-BS above; MP-JK below); MP-JK values below 64 are not shown 126 Fig. 7 Informative microstructural characters in any given matrix mapped onto the consensus of the MP trees of 5-loci dataset. Red hashmarks represent SSRs and black hashmarks gaps; solid hashmarks when the character state shows no homoplasy. Numbers above the hashmark correspond to the character number of the 43-terminal matrix after deletion of the non-potentially informative characters. Letters below hashmarks indicate if the character state corresponding to a deletion (D) or an insertion (I). Marker codes are as follows: n32=ndhF–rpl32; LF=trnL–F; r16=rps16. Green asterisks in front of terminals indicate that the apparently autapomorphic microstructural character(s) is synapomorphic in the full corresponding matrix. Table 3 describes the microstructural characters that group the main clades in Echiochiloideae. The explanation for other figure notations can be found in Fig. 6 127 Appendix 1. Lists of microstructural characters A. List of microstructural characters and their aligned positions found in the ITS region coded as absent (0) or present (1) Gaps 1. 4 2. 6 3. 24 4. 35–38 5. 35 6. 50–52 7. 51–52 8. 52–53 9. 52 10. 59 11. 62 12. 65 13. 68–85 14. 69 15. 81–82 16. 81 17. 84–87 18. 84–85 19. 93 20. 103–104 21. 104 22. 107 23. 125 24. 139–141 25. 140–141 26. 143 27. 148–151 28. 150 29. 161–163 30. 162–163 31. 171 32. 188 33. 201 34. 219 35. 237 36. 239 37. 241 38. 245 39. 251 40. 253 41. 264–266 42. 276–381 43. 448 44. 452–453 45. 453–456 46. 453 47. 479 48. 507–519 49. 507 50. 512–513 51. 513 52. 518–519 53. 518 54. 519 55. 537 56. 566–569 57. 584 58. 610 59. 617 60. 618 61. 627 62. 651 63. 654–672 64. 656 65. 660 66. 662–663 67. 663 68. 673–674 69. 673 70. 674 71. 677 72. 683 73. 694 Inversions 74. GAGTTAAAAAA to TTTTTTAACTC 36–43; 47–49 75. AA to TT 636–637 SSRs 76. AAC 44–46 77. GC 72–73 78. TTA 114–116 79. GGC 130–132 80. AT 133–134 81. GT 148–149 82. AT 199–200 83. CA 459–460 84. TGTG double with some mutations in the first (CGTG) and inversion of the second 499–506 85. TG 522–523 86. TT 638–639 B. List of microstructural characters and their aligned positions found in the ndhF–rpl32 region coded as absent (0) or present (1) unless otherwise stated Gaps 1. 18–19 2. 58–61 3. 114–125 4. 167–173 5. 174–179 6. 181 7. 249–254 8. 308 9. 316–320 10. 326 11. 334–338 12. 406–414 128 13. 406–407 14. 412 15. 414 16. 419–420 17. 431–716 18. 436–677 19. 436–668 20. 449–458 21. 451–458 22. 463–466 23. 472–691 24. 482–490 25. 500–762 26. 502–505 27. 571–578 28. 594–600 29. 613–615 30. 634–637 31. 662–666 32. 686–689 33. 688–749 34. 724 35. 835–845 36. 835–837 Inversions 37. PolyA to polyT 669–672 SSRs 38. CTTTTTTAAT with change of G for C 36–46 39. GAAAGA 90–95 40. AAAAAAGA 116–123 41. AATATT 154–159 42. TCA 242–244 43. AAGATAACA with change of T for G 298–307 44. CATTAA 343–348 45. CACTT 377–381 46. CACTTTATAAA 388–398 47. ATT 488–490 48. TTA 494–499 without SSR=0; simple SSR=1; double SSR=2 49. AGGTTTA with extra G 512–518 50. ATCGATT with shorter T strings 538–544 51. ACACGGC 552–558 52. TTT 649–651 53. AG 688–690 54. TTTTCATCC 707–715 55. ATT 756–758 C. List of microstructural characters and their aligned positions found in the matK region coded as absent (0) or present (1) unless otherwise stated Gaps 1. 7–12 2. 13–27 3. 16–215 4. 52–75 5. 150–152 6. 157–171 7. 157–159 8. 382–384 9. 394–399 10. 484–906 11. 694–696 12. 706–711 13. 706–708 14. 802–825 15. 820–825 Inversions 16. AAA to TTT 598–600 SSRs 17. TCA 421–423 18. TTC 439–450 without SSR=0; simple SSR=1; double SSR=2; quadruple SSR=3. 19. ATTTCT 16–21 20. AATGAT with intermediate triplet 31–49 21. CTAGAA 160–165 22. GCTAAA 601–606 23. AAA 721–723 24. ATAAACCAA 802–810 129 D. List of microstructural characters and their aligned positions found in the trnL–F region coded as absent (0) or present (1) unless otherwise stated Gaps 1. 7 2. 24 3. 31 4. 35–37 5. 36 6. 56 7. 64–71 8. 70–72 9. 74–76 10. 75–76 11. 76 12. 121 13. 175–191 14. 175–188 15. 180–192 16. 197–418 17. 201–210 18. 217–231 19. 217–219 20. 233–259 21. 236–349 22. 236–259 23. 243 24. 253–259 25. 256–259 26. 263–270 27. 265–270 28. 262–264 29. 269–303 30. 269–272 31. 309–346 32. 309–326 33. 323–325 34. 323–324 35. 348–349 36. 349 37. 351 38. 361–500 39. 361–418 40. 361–414 41. 361–406 42. 374–375 43. 410–420 44. 414–443 45. 424–443 46. 424–434 47. 435–443 48. 436–437 49. 439–443 50. 439 51. 437 52. 442–443 53. 445–448 54. 445 55. 446–490 56. 446–448 57. 448–490 58. 447–448 59. 449–492 60. 449–490 61. 451–492 62. 451–512 63. 451–500 64. 451–491 65. 451–493 66. 454–456 67. 489–490 68. 494–512 69. 509–640 70. 512 71. 550–761 72. 550–566 73. 550–554 74. 550 75. 554 76. 557 77. 559 78. 562–573 79. 562–570 80. 583–611 81. 596–611 82. 615–618 83. 634–642 84. 639–640 85. 676–684 86. 688–696 87. 692–696 88. 731–747 89. 745 90. 771–807 91. 778 92. 787 93. 795–796 94. 796 95. 833–837 96. 833 97. 835–839 98. 837–859 99. 838–856 100. 839 101. 859 102. 860–862 103. 861–867 104. 866–869 105. 872–884 106. 872–874 107. 876–898 108. 901–924 109. 901–902 110. 901 111. 928–952 112. 929–956 113. 929–952 114. 933–1059 115. 952 116. 960–962 117. 967–969 118. 971–1006 119. 977–1031 120. 1020–1077 121. 1021–1071 122. 1028–1060 123. 1028–1031 124. 1034 125. 1038 126. 1041–1061 127. 1041–1052 128. 1082–1221 129. 1087–1101 130. 1087–1089 131. 1089 132. 1089–1100 133. 1089–1092 134. 1091–1094 135. 1091–1092 136. 1093–1094 137. 1094 138. 1097–1223 139. 1097–1123 140. 1097–1111 141. 1097–1101 142. 1101 143. 1105–1124 144. 1115–1126 145. 1115–1124 146. 1123 147. 1126–1131 148. 1127–1158 149. 1134–1138 150. 1140–1158 151. 1141–1158 152. 1147–1161 153. 1149–1191 154. 1161–1163 155. 1163–1164 156. 1178–1180 130 157. 1190–1191 158. 1194–1206 159. 1194–1203 160. 1195–1199 161. 1202–1301 162. 1204–1365 163. 1207–1223 164. 1209–1301 165. 1210–1216 166. 1211–1365 167. 1211–1217 168. 1219–1223 169. 1220–1223 170. 1223–1465 171. 1223 172. 1223–1243 173. 1223–1241 174. 1220–1221 175. 1236–1311 176. 1241–1257 177. 1242–1311 178. 1239–1241 179. 1242–1244 180. 1250–1263 181. 1250–1261 182. 1261 183. 1299–1301 184. 1308–1314 185. 1309–1314 186. 1338–1347 187. 1337–1352 188. 1338–1365 189. 1338–1350 190. 1372–1393 191. 1373–1404 192. 1373–1395 193. 1380–1391 194. 1401–1417 195. 1398–1404 196. 1400–1404 197. 1409–1425 198. 1420–1448 199. 1420–1425 200. 1425–1448 201. 1428–1429 202. 1432 203. 1435–1460 204. 1435–1459 205. 1435–1455 206. 1460–1465 207. 1461–1464 208. 1464 209. 1466–1507 210. 1466–1469 211. 1468 212. 1475–1515 213. 1498–1500 214. 1550 215. 1554 216. 1555–1556 217. 1573–1597 218. 1578–1603 Inversions 219. TTTGAATGA to TCSTTCAAA 1236–1244 SSRs 220. AAGG 112–118 without SSR=0; simple SSR=1; double SSR with deletion of one A=2 221. SRR ATAGG 125–129 222. CCTTG 181–185 223. TC 213–214 224. TCATAAA 220–226 225. TAAA 227–230 226. TG 236–237 227. GTG 238–240 228. GAAA 244–247 229. TGAAA 248–252 230. CA 269–270 231. AT 276–279 without SSR=0; simple SSR=1; double SSR=2 232. ATTTGT 280–285 233. GAAAGTATATACATTTTT 286–303 234. TGTATTGA 309–316 235. TATTGA 317–322 236. ACTTTATCTTACTTT triple with flipflop to TC in the second repetition 332–346 237. ATCAAA 355–360 238. TGA 410–412 239. TAAT 424–427 240. AAT double 428–433 241. CTAATATATT 451–463 242. CAAAATATATT 464–474 243. TTAATATATT with TT instead of AA 475–484 244. TATT 485–488 245. CTGTT 496–500 246. TTAA 563–566 247. ATAGAA583–588 248. AAGAA 591–600 without motif=0; motif without SSR=1; motif and SSR=2 249. GAAGAAAGAA 601–610 250. TAA 629–631 131 251. GATCAAA 634–640 252. AATCA 644–648 253. TTTT 676–679 254. TTA 693–695 255. TC 699–700 256. GTC 720–722 257. CAA 751–753 258. TTG 843–851 without SSR=0; simple SSR=1; double SSR=2; triple SSR=3 259. TTGA 853–856 260. TA triple 879–884 without motif= 0; motif without SSR=1; motif and triple SSR=2 261. TATCT 888–894: with repetition of the last CT 262. ATCT 895–898 263. TTATCTATTT 904–915: with changes of the first nt 264. TATCTATTT 916–924 265. GT 929–930 266. TTA 933–935 267. TAGT 938–941 268. TTTAGTTAGT 942–951 269. CTT 973–975 270. TTCCAAATTCCTCATCTTT 977–995 271. CTCATCTTT 996–1004 272. ATCATT 1011–1016 273. ACTCT 1023–1027 274. ATT 1035–1037 275. GTT 1041–1043 276. GTTTT 1046–1050 277. ACAA 1056–1059 278. TAGAAATG 1064–1071 279. GAT 1075–1077 280. CTGG with mutations in some terminals 1082–1085 281. CTTT 1097–1100 282. CATC 1105–1108 283. ATC 1109–1111 284. TCATCTTA 1115–1122 285. TCTT 1130–1137 without motif=0; motif without SSR=1; motif and SSR=2 286. TGAGAAACGT 1149–1158 287. ATGATATAC with mutation C to T 1168–1176 288. TACAAA 1184–1189 289. GGAA/TGAA with mutations 1209–1216 without motif=0; motif without SSR=1; motif and SSR=2 290. TTATTTAC 1250–1257 291. TTC 1309–1311 292. TTAT 1355–1358 293. TCTTAT with insertion of one T 1359–1365 294. TTG 1368–1370 295. TGAA 1373–1376 296. GAA 1377–1379 297. ACAGTA 1412–1417 298. GATAAAAGT with flip flop 1435–1443 299. TAAAA 1444–1448 300. GTT 1452–1454 301. AAAGTTGGTAATCTTCTTTAATA 1475–1497 302. TTTAATA 1501–1507 303. CTTT 1512–1515 304. TTTAATACTTTTAATTGACATA 1527–1548 305. AAAAT 1573–1577 306. ATAAAT 1578–1584 132 307. AAT 1586–1591 without SSR= 0; simple SSR=1; double SSR=2 308. AAG 1595–1597 E. List of microstructural characters and their aligned positions found in the rps16 region coded as absent (0) or present (1) unless otherwise stated Gaps 1. 6 2. 30–50 3. 37–52 4. 43–54 5. 70–73 6. 79–84 7. 123–126 8. 138–141 9. 138 10. 164 11. 188–191 12. 239–274 13. 242–247 14. 244–248 15. 287–295 16. 302–312 17. 304–312 18. 350–362 19. 350 20. 353–359 21. 354–365 22. 354–360 23. 361–365 24. 364–365 25. 400–421 26. 418–419 27. 425–426 28. 433–435 29. 434–437 30. 441–461 31. 455–471 32. 521–552 33. 549 34. 558–564 35. 559 36. 565–578 37. 565–567 38. 572–584 39. 632–635 40. 657–668 41. 659–722 42. 660–672 43. 669–678 44. 670–678 45. 670–676 46. 675–677 47. 694–700 48. 698–710 49. 707–722 50. 707–711 51. 714–731 52. 716–728 53. 725–726 54. 731–732 55. 738–740 56. 792 57. 793–796 58. 820–832 59. 822 60. 824–826 61. 827–830 62. 850–860 63. 874 64. 1202–1278 65. 1212–1270 66. 1216–1225 67. 1282–1284 68. 1309 69. 1347 SSRs 70. ATT 27–32 without motif=0; motif without SSR=1; motif with SSR=2 71. TGTTC 38–42 72. TCCAC 46–50 73. GGG 79–81 74. TGT 91–93 75. AAATA with C insertion 99–104 76. GTA 130–132 77. GTA 150–152 78. TTCA with mutations 157–160 79. AATCAAT with T deletion 199–204 80. AAATT with change of T to A 205–209 81. AAATAT with change of A to G 227–232 82. GATATAGAAA 255–264 83. ATAGAAA 265–271 84. CCAATT with insertion of A 306–312 85. TGAT 354–357 86. TGTA 392–395 87. TCGGAA double 406–417 88. ATAAA 448–452 89. ATAAAAG 455–461 90. GGA with mutation 469–471 91. CAAAA 482–486 92. TAAAA 560–564 133 93. AAA 572–574 94. CAAA with change of C to G 575–578 95. AAGGAT 589–594 96. CAGAA 602–606 97. TTTTTAATTGTCTCAA 639–654 98. ATAACT 660–665 99. AAT 670–672 134 DISCUSIÓN GENERAL I. Reevaluación taxonómica de las especies de Antiphytum El género Antiphytum nació con problemas taxonómicos que requirieron la lectotipificación del mismo por Johnston (1923), uniendo el nombre a las especies sudamericanas, que desde entonces eran de difícil acceso y poco conocidas. Un estudio del género debía, por fuerza, incluir la diversidad austral para poder plantear escenarios de distribución y diversificación. Empero, las especies sudamericanas representaban un problema logístico, pues no había disponibilidad de colectas para estudios filogenéticos moleculares. Los herbarios brasileños no permiten ni el envío ni el muestreo de ejemplares de sus colecciones biológicas, mientras que el herbario uruguayo MVFA no cuenta con recursos para envío de material. Por otra parte, además de los pocos ejemplares de herbarios internacionales (B, G, GH, K, MO, NY, P, US), trabajos como el de De Melo y de Stapf (2014) rechazaban la existencia de Antiphytum en Brasil, contradiciendo las colectas de herbario y trabajos recientes en que se ha documentado el género, como el de Setubal (2010) y las bases de datos de los herbarios brasileños. La disponibilidad de ejemplares para una adecuada revisión taxonómica afecta la circunscripción genérica y de especies. Por ejemplo, en el tratamiento del género, Johnston (1923) consideró A. stoechadifolium la misma especie que A. cruciatum; posteriormente y con reserva, el mismo autor reconoció a las dos especies sudamericanas, aunque el material que pudo consultar fue más bien pobre (Johnston 1927). Mendoza-Díaz (2015) concordó con el reconocimiento de estas dos especies principalmente por la ornamentación de las clusas y coincidió en lo exiguo del material para concluir la delimitación de especies en Sudamérica: la información sobre la distribución de los especímenes sudamericanos es escasa en concordancia con la antigüedad de los mismos, precisando si acaso el país o la región donde fueron colectados. Empero, las conclusiones taxonómicas preliminares de Mendoza-Díaz (2015) para estas especies son diferentes con respecto a lo obtenido en este trabajo. La posibilidad de muestrear en Sudamérica y consultar los herbarios de la región fue la piedra angular para los resultados de este proyecto, evidenciando que el trabajo de campo y de herbario siguen siendo pasos vitales en Sistemática. El estudio completo de material perteneciente a Antiphytum permitió delimitarlo dentro de Echiochiloideae en términos morfológicos y reconocer nuevas especies como parte del género. Resultado de ello fue la publicación de una nueva especie sudamericana, Antiphytum charruasorum, que coincide morfológicamente con las especies distribuidas en Norteamérica y que estableció importantes patrones morfológicos, biogeográficos y 135 filogenéticos para entender el género; la transferencia a Antiphytum y lectotipificación de una especie sudamericana, A. berroi, que no sólo aumentó la riqueza de especies en Sudamérica, sino que clarificó que esta especie, además de A. cruciatum, se distribuye en el sur de Brasil. Finalmente, de las dos especies incluidas en la publicación original de Antiphytum (Meisner 1840), A. stoechadifolium era la menos entendida, pero gracias a la revisión de los herbario MVFA e ICN, pudimos concluir que no se distribuye en el actual Brasil, sino sólo en Uruguay, en concreto al sur de este país. Las etiquetas de las colectas de Sellow (ejemplares tipo), que sólo especifican como localidad “Brasilia meridionalis”, aluden a un territorio en el que no se hacía distinción de la entonces inexistente República de Uruguay. Sin embargo, esta especie sigue siendo objeto de estudio, pues aparte de las colectas tipo del siglo XIX, dos ejemplares del MVFA (MVFA-25860, Bonifacino s.n.; MVFA-s.n, Marchesi 1420) son los únicos registros recientes (siglo XX) que aclaran la distribución de la especie en el Departamento de Maldonado, disyunta al resto de las especies uruguayas distribuidas en la zona norte del país. La preservación de los herborizados con alcohol-mercurio impidió la extracción de ADN para incluir a la especie en la reconstrucción filogenética molecular. No obstante, la morfología permite anticipar que se recuperará dentro del subclado sudamericano de especies con corolas azules e hipotetizar una relación hermana con A. berroi. En Norteamérica, donde pareciera que el número de especies estaba claro, material herborizado permitió el reconocimiento de dos nuevas especies que cambiaron el muestreo disponible del género, y por ende, la hipótesis filogenética de Mendoza-Díaz (2015), aun empleando los mismos marcadores de ese trabajo; ampliando, además, la distribución geográfica conocida del mismo. Por todo lo anterior, se considera que el más valioso resultado de este trabajo fue la actualización taxonómica del género (primera parte de la tesis) que deriva en 13 especies distribuidas de forma disyunta en América, y que fue el paso preliminar para la segunda parte de esta tesis. En términos morfológicos, Antiphytum se puede diferenciar principalmente por la combinación de los siguientes caracteres: sufrutices perennes como forma de crecimiento; corola con simetría radial y sobrepasando el cáliz; presencia de apéndices faucales; fruto no encerrado por el cáliz, sino el cáliz extendido y subtendiendo al fruto; forma de la ginobase de plana a triangular; cicatriz en la mitad proximal del eremocarpo, no rebasando la zona media de la cara ventral de éste. La circunscripción actualizada de Antiphytum incrementa los 19 ejemplos de distribución anfitropical en América (AAD por sus siglas en inglés) de Boraginaceae, la segunda familia con mayor número de AAD (Simpson et al. 2017b), sumándose a los patrones previamente estudiados en esta familia, pero añadiendo la particularidad geográfica y morfológica de Antiphytum. 136 Antiphytum presenta una distribución que, de acuerdo a Simpson et al. (2017b), puede considerarse como trans-NA oriental-SA anfitropical, porque tiene representantes en zonas que se encuentran en el Neotrópico de Norteamérica. Por otro lado, la distribución en Sudamérica es atípica para la familia Boraginaceae, que principalmente se distribuye en la franja occidental del subcontinente. Los únicos géneros en el oriente de Sudamérica, entre el sur de Brasil y norte de Uruguay, son Moritzia y Thaumatocaryon (subfamilia Boraginoideae, tribu Boragineae, sólo en Brasil) y Antiphytum (subfamilia Echiochiloideae, en ambos países), los tres endémicos del continente americano. México alberga la mayor riqueza de especies de Antiphytum, con A. heliotropioides y A. floribundum alcanzando el sur de EUA, y las demás especies norteamericanas distribuidas principalmente sobre las cadenas montañosas de las Sierras Madres, el Altiplano Mexicano y una en la Península de Baja California, teniendo a Oaxaca como su distribución más sureña en Norteamérica. Asociado a esta distribución, se encuentra la preferencia edáfica del género, pues a excepción de A. floribundum, que sólo crece sobre suelos volcánicos, las especies norteamericanas se encuentran en suelos calcáreos y son facultativas para suelos yesosos, siendo A. hintoniorum la única especie gipsófila del género y A. peninsulare una gipsoclina. Reapareciendo hasta Uruguay y el sur de Brasil –en el estado de Río Grande do Sul–, las especies sudamericanas de corola azul se distribuyen sobre suelos de origen volcánico, algunos de baja fertilidad, de acidez acentuada, con deficiencia de P, Ca y Mg y con granulometría de limos, arcillas y arenas (Setubal 2010), mientras que A. charruasorum, de corola blanca, se distribuye sobre suelos calcáreos como sus contrapartes norteamericanas a las que se parece más morfológicamente. II. Filogenia y patrones de diversificación de Antiphytum. El primer resultado que deriva de los análisis filogenéticos de este estudio es la monofilia de Antiphytum en concordancia con la delimitación morfológica del género. Las tres posibles topologías para resolver la politomia por conflicto de los cinco marcadores moleculares son ambiguas respecto a la interpretación de la direccionalidad de colonización, pues el grupo hermano, Ogastemma, es africano. Sin embargo, dos de ellas sugieren que las especies norteamericanas no comparten un ancestro común inmediate e implican un origen único y más probable en Norteamérica, en contraste a la necesidad de invocar dos colonizaciones independientes de esta región. Además, el clado de las especies sudamericanas de corola azul y de clusas estipitadas es derivado, y de acuerdo al número de cambios del árbol consenso de cinco marcadores, más reciente con respecto a A. charruasorum. En conjunción con lo anterior, la distribución geográfica de los otros géneros de Echiochiloideae, la mayor diversidad en Norteamérica, que incluye taxones 137 extintos asignables a la subfamilia (no publicado, descripción como Cryptantha por Thomasson 1979) y la edad del nodo Antiphytum (Chacón et al. 2017), apoyan una colonización de norte a sur del continente, que es la que se propone en este trabajo. Esto coincide con los patrones hallados en la familia y, especialmente por distribución geográfica, a los de Moritzia y Thaumatocaryon (Weigend et al. 2010). Estos últimos autores postularon una direccionalidad de colonización del linaje americano de Boragineae de norte a sur, apoyado en la actual presencia de Moritzia y Thaumatocaryon y la presencia de fósiles del grupo en Norteamérica. Además, la revisión de Simpson et al. (2017b) infirió que cerca de 53.8% de taxa desérticos de diferentes linajes (sin incluir helechos) con eventos AAD tuvieron direccionalidad de Norte a Sudamérica. El clado sudamericano presenta dos cambios morfológicos, ganancia de pigmento en las corolas y presencia de un estípite, los cuales se plantean posteriores al evento de colonización. El primero se propone como promotor de diversificación del linaje sudamericano, mientras que el segundo se correlaciona con el tipo de suelo de origen volcánico. Ninguna especie sudamericana presenta rosetas basales ni registros en yeso. En Norteamérica, se definen dos clados que morfológicamente se pueden separar por la presencia o ausencia de roseta basal. El clado sin rosetas presenta un mayor número de especies, todas con algún registro en suelos yesosos y con la especie gipsófila A. hintoniorum como hermana al resto de las especies de este clado. En contraposición, el clado de arrosetadas, aunque presenta especies gipsovagas, también incluye a A. floribundum, la única especie norteamericana que no se distribuye en yesos, sino sólo en suelos de origen volcánico. Algunas de las áreas de distribución de Echiochilon y Ogastemma (subfamilia Echiochiloideae) en África y la Península Arábiga presentan yeso. En el género Echiochilon también hay especies gipsovagas, por lo que la subfamilia puede considerarse pre-adaptada a zonas áridas de suelos pobres con yeso o calizas, y por ello, estas pre-adaptaciones pudieron permitir la colonización y posterior endemismo edáfico de A. hintoniorum. En México los afloramientos yesosos se concentran principalmente en la región del Desierto Chihuahuense, aunque también hay en Baja California, Campeche, Chiapas, Guerrero, Oaxaca, Puebla y Tamaulipas (Ochoterena et al. 2020). En contraste, afloramientos de yeso en Uruguay son menos comunes. De acuerdo al Ministerio de Industria y Energía de Uruguay (1987), el yeso se encuentra en los materiales Devónicos en la localidad de Blanquillo, en el Departamento de Durazno, y en materiales Cuaternarios. En ninguna de estas zonas hay algún registro conocido de Antiphytum. La explotación de yeso sólo ocurre en el Departamento de Río Negro, en la localidad de Bellaco, aproximadamente a 35 km al NO de la población de A. charruasorum en Río Negro. 138 La disyunción biogeográfica de Antiphytum es un patrón clado-clado, de acuerdo a la terminología de Simpson et al. (2017b) e implica una diversificación de especies de ambos linajes tras un evento de AAD, y por ende, tiempos mayores de divergencia. Esto no coincide con lo observado en el género, pues más bien se infiere una diversificación rápida y reciente de sus especies, en concordancia con los datos de Chacón et al. (2017), que fechan el nodo de Antiphytum en 9.4 Ma. Aunque en Norteamérica se presentan nueve especies del género, cada clado tiene aproximadamente el mismo número de especies (clado sudamericano 4; clado norteamericano arrosetado 4, clado norteamericano sin rosetas 5). La preferencia edáfica es más amplia en Norteamérica sólo por la presencia de especies gipsovagas y una gipsófila. El clado sin rosetas presenta individuos en tres tipos de suelo: calcáreo, yesoso y volcánico (A. peninsulare se distribuye en yeso y suelos volcánicos). El clado de arrosetadas se distribuye en suelos de origen volcánico (sólo A. floribundum), calcáreos y yesosos (como gipsovagas). La roseta basal se originó en un único evento dentro del género y se plantea su importancia como promotor de diversificación. En el clado de especies con rosetas se presenta la única especie no recuperada como monofilética, el grado de A. parryi dentro del subclado de especies arrosetadas del este de México. La especie, sin embargo, reúne otras características morfológicas y de distribución para considerarse un linaje independiente de A. paniculatum, del que es parafilética. Conjuntamente, la relación entre la distribución de las especies norteamericanas y los eventos geológicos que han configurado el relieve de México parecen sugerir eventos relacionados con la tectónica, preferencia edáfica y dispersiones posteriores en la especiación de Antiphytum en esta zona. El clado sudamericano sólo presenta distribución en suelos calcáreos y de origen volcánico. Ya que, tanto el subclado de especies con roseta del este de México, como el subclado sudamericano de corolas azules, no se resuelven en los análisis individuales y concatenados con marcadores de cloroplasto, se pueden especular como recientes. Por otra parte, aunque las especies del clado norteamericano sin rosetas se recuperaron en la mayoría de los análisis concatenados e individuales como monofiléticos, este clado no presentó soporte en los análisis de parsimonia, pero sí en el bayesiano y débil en el de máxima verosimilitud, lo que podría asociarse con una rápida diversificación del clado. Ante la falta de resolución entre los tres clados previamente descritos en la filogenia recuperada de Antiphytum con cinco marcadores moleculares, la morfología tampoco es concluyente. El subclado de corolas azules presenta tricomas glandulares capitados, también presentes en tres especies del clado norteamericano sin rosetas (ausentes en el resto de las especies); ambos sin roseta basal. La ginobase plana característica del clado de corolas azules y la modificación de la quilla ventral se 139 presenta en A. floribundum (especie arrosetada sin fornículos), con la que, además, comparte la preferencia a suelos volcánicos, por lo que estos cambios morfológicos en corola y eremocarpos podrían estar correlacionados al tipo de suelo. Con el clado arrosetado de Norteamérica, el clado sudamericano también comparte la ornamentación de tubérculos (o verrugas) en la mayoría de las especies. También, la poca variación molecular en los marcadores de cloroplasto en los subclados de especies arrosetadas del este de México y de corolas azules de Sudamérica en los marcadores de cloroplasto puede apoyar un escenario donde ambos clados sean hermanos. Aun sin resolución del esqueleto en la filogenia propuesta de Antiphytum, este estudio permite concluir un género natural con circunscripción taxonómica y geográfica actualizada, lo que descarta el color de la corola como característica de segregación taxonómica. Los resultados sugieren que la distribución disyunta ocurrió en un único evento de norte a sur del continente, con la ganancia de pigmento en la corola como posterior al evento de colonización y como un posible promotor de diversificación. Por ello, los resultados, aunque no resuelven la totalidad de preguntas que derivan de Antiphytum, plantean escenarios de futura investigación sobre bases sustentadas, aunque perfectibles, pues más muestreo en campo y trabajo de herbario siempre aportarán nuevas y valiosas evidencias. También, se proporciona una herramienta para la identificación de las especies del género. 140 CONCLUSIONES GENERALES -La Sistemática es una disciplina que se nutre de diferentes líneas de evidencia para hacer propuestas respecto a la circunscripción y explicación de la biodiversidad. Sus primeros pasos de descubrir, describir y nombrar la diversidad son claves para atender preguntas sobre diversificación y distribución de los taxa bajo estudio. -Bajo la aseveración anterior, la publicación de tres nuevas especies del género Antiphytum (A. brevicalyx, A. charruasorum y A. geoffreyi) es un resultado de gran envergadura, porque cubre con datos necesarios los espacios en blanco de nuestro conocimiento sobre el género, no sólo en cuanto a diversidad y distribución, sino a caracteres morfológicos y moleculares que se añaden al concepto genérico y de especies. Además, propicia el interés de la comunidad por explorar zonas potenciales de distribución y aumentar los registros y conocimiento del hábitat de las especies. Como ejemplo de ello, tras la publicación de la especie uruguaya A. charruasorum se sumó una localidad de distribución para la especie, gracias al ingeniero agrónomo uruguayo Andrés González, quien pudo asociar su hallazgo con la nueva especie (com. pers.). -La aclaración de las especies sudamericanas de Antiphytum por medio de la transferencia genérica de Myosotis berroi a Antiphytum y su delimitación geográfica y morfológica con respecto a A. stoechadifolium, que a su vez se ha delimitado con respecto a A. cruciatum, han sido fundamentales para conocer la diversidad del género en Sudamérica y desenmarañar la oscura situación taxonómica que se había heredado desde los tiempos de publicación del género. Prueba de ello han sido las constantes muestras de interés por el reconocimiento de A. stoechadifolium como especie independiente de A. cruciatum, bajo la cual se etiquetaba todo el material sudamericano (p. ej. Dr. José Iranildo Miranda de Melo, com. pers.). -México es el país en el que se encuentra la mayor riqueza y diversidad de Antiphytum, con dos nuevas especies publicadas como producto de este trabajo, además de las conclusiones taxonómicas que reducen algunos nombres a la sinonimia taxonómica, p. ej. Antiphytum nudicalces (≡Amblynotopsis nudicalces) =Antiphytum caespitosum; Amblynotopsis ehrenbergii = Antiphytum heliotropioides; Amblynotopsis humillis =Antiphytum parryi; Amphibologyne mexicana = Antiphytum floribundum, y Amblynotopsis durangensis = Antiphytum floribundum; o que se denotan como superfluos (Chamissoniophila cruciata, C. stoechadifolia). Cambios taxonómicos y nomenclaturales derivados de este estudio requieren la integración en una síntesis monográfica de Antiphytum. 141 -La circunscripción de Antiphytum resultado de este trabajo apoya a un género endémico del continente americano, monofilético y de distribución disyunta (trans-AAD) integrado por 13 especies conocidas, nueve en Norteamérica (NA) y cuatro en Sudamérica (SA). En Norteamérica se encuentran A. brevicalyx, A. caespitosum, A. floribundum, A. geoffreyi, A.heliotropiodes, A. hintoniorum, A. paniculatum, A. parryi y A. peninsulare. De Sudamérica se conocen A. berroi, A. charruasorum, A. cruciatum y A. stoechadifolium. Con excepción de A. parryi, que se recuperó como parafilética a A. paniculatum, todas las especies en el género se recuperaron como monofiléticas en el análisis combinado. Bajo el concepto general de especie, A. parryi cuenta con suficientes características morfológicas y geográficas para ser delimitada de A paniculatum. El color azul de las flores y la filotaxia totalmente opuesta no son características morfológicas que segreguen al género por distribución geográfica, dados los patrones geográficos y filogenéticos obtenidos. -Antiphytum se recuperó como integrante de la subfamilia Echiochiloideae, ratificando la relación hermana con Ogastemma obtenida en trabajos anteriores. Aunque algunos marcadores individuales (ITS, matK, rps16) recuperaron a Ogastemma al interior de un clado más grande con todos los terminales de Antiphytum, se concluye que este último es un género diferente por la evidencia filogenética de los marcadores concatenados y características diagnósticas propias del mismo. -El análisis de parsimonia con todos los marcadores de cloroplasto y los tres métodos de reconstrucción filogenética con los cinco marcadores moleculares recuperaron tres clados al interior de Antiphytum: clado NA con rosetas basales; clado NA sin rosetas basales; y clado SA (sin rosetas basales). El clado NA sin rosetas basales tiene la particularidad de estar integrado por especies que tienen algún grado de afinidad al yeso y contener a la única especie gipsofila del género (A. hintoniorum), la cual se recupera como la primeramente divergente del clado, con una rama con muchos cambios moleculares. El clado NA con rosetas presenta elementos gipsovagos y la única especie norteamericana gipsofoba (A. floribundum), hermana al resto. El clado sudamericano tiene especies sin presencia en yeso, pero con afinidad a suelos calcícolas (A. charruasorum, corolas blancas) como en la mayoría de gipsovagas de NA; o a suelos de origen volcánico (especies sudamericanas de corolas azules), como en las norteamericanas A. floribundum y A. peninsulare. -Las filogenias consenso de los cinco marcadores obtenidas por parsimonia y reconstrucción bayesiana no resuelven las relaciones entre los tres clados, ya que se recuperan dos posibles escenarios: 1) las especies de NA son monofiléticas y hermanas al clado SA, o 2) las especies NA no comparten un ancestro común inmediato y, por ende, el clado de SA es hermano al clado NA con rosetas o al clado NA sin rosetas, respectivamente. El árbol de máxima verosimilitud presenta 142 una relación hermana entre el clado SA y el clado NA de rosetas, sin soporte estadístico y con una rama muy corta. -Aun con la incertidumbre filogenética, el escenario del árbol de máxima verosimilitud se presenta factible por las características morfológicas en común entre las especies del clado arrosetado y las especies sudamericanas, la distribución hacia el sur en Norteamérica del primero y al hecho de que ambos clados posiblemente son linajes más recientes por la poca variación molecular presente en los marcadores de cloroplasto. -Morfológicamente es posible identificar a Antiphytum por ser sufrútices perennes con corolas con simetría radial sobrepasando el cáliz y con apéndices faucales; por poseer lóbulos del cáliz divididos casi hasta la base y acrescentes durante la fructificación, que se extienden y subyacen a los cuatro eremocarpos expuestos (o menos por aborto, según la especie). Las ginobases van de planas a triangulares, con las correspondientes cicatrices en la mitad proximal del eremocarpo, no rebasando la zona media de la cara ventral, ovadas o triangulares en posición basal (asociada a un estípite formado por la quilla ventral), suprabasal, inframedial o medial. Las especies son fácilmente reconocibles por los eremocarpos, a excepción del grupo de especies con rosetas basales integrado por A. caespitosum, A. paniculatum y A. parryi. La morfología del polen es común para las especies de Antiphytum muestreadas de Sudamérica y Norteamérica, y es muy similar en Ogastemma. -Características morfológicas como presencia o ausencia de roseta basal, presencia o ausencia de fornículos en la garganta de la corola, filotaxia opuesta o alterna, presencia o no de un estípite en el eremocarpo, ornamentación de los eremocarpos, o el color blanco o azul de las corolas, definen morfológicamente no sólo a las especies, sino a grupos al interior del género. -La filotaxia totalmente opuesta no es privativa de las especies sudamericanas, sino que está también presente en la especie norteamericana A. hintoniorum. Las ginobases planas también se encuentran tanto en tres especies sudamericanas (las de corola azul) como en dos norteamericanas. El carácter de rosetas basales tuvo un solo origen en el género Antiphytum y deberá investigarse su papel como promotor de diversificación. -El color blanco de las flores no es particular de las especies norteamericanas, ya que también se encuentra en un representante del linaje sudamericano. El color azul de las flores se propone como una adquisición posterior al evento de colonización de Sudamérica, al ser A. charruasurom, de corola blanca, el elemento tempranamente divergente en el clado recuperado con todas las especies sudamericanas. La ganancia de pigmento característico del subclado sudamericano se sugiere como 143 un carácter que afectó la diversificación en Antiphytum, probablemente por facilitar la especiación o la colonización de nuevos ambientes al incorporar nuevos polinizadores o agentes de dispersión. -El género Antiphytum presenta una marcada preferencia edáfica a suelos empobrecidos, principalmente a los calcáreos, pero casi la mitad de las especies, seis, son facultativas a yeso, una es gipsófila y una es una gipsoclina. Las especies con afinidad a yeso se distribuyen sólo en México. La preferencia a caliza es alta en la subfamilia, y ejemplos de gipsovagas también existen en el género Echiochilon. Suelos de origen volcánico son también colonizados por Antiphytum y algunas modificaciones morfológicas se presentan en las especies que habitan este tipo de suelos. Por ende, se plantea la preferencia edáfica como un factor involucrado en la diversificación del género. -Los antecedentes bibliográficos respecto a la biogeografía de la familia Boraginaceae y la probable no monofilia de las especies norteamericanas, con el clado sin rosetas probablemente más antiguo, permiten suponer que en Antiphytum también ocurrió una colonización de norte a sur del continente por un evento de dispersión a larga distancia. Probablemente el linaje sudamericano comparta un ancestro común con el linaje norteamericano de rosetas basales. -La mayor riqueza del género en México parece asociada a eventos geológicos más complejos del territorio mexicano, así como a la diversidad en el tipo de suelos restrictivos donde se establecieron las especies (yesosos, calcáreos y de origen volcánico). -La subfamilia Echiochiloideae en conjunto requiere de mayor trabajo, pues se hallaron conflictos al interior de las especies de Echiochilon con el marcador ITS y queda pendiente obtener una filogenia resuelta de Antiphytum. El escenario provisional biogeográfico sugiere una dispersión desde África hacia Norteamérica (donde se originó Antiphytum), a través de Europa y como último punto de colonización el subcontinente de América del Sur. Fósiles de Echiochiloideae en el centro de EUA respaldan esta direccionalidad y son prueba de la presencia durante el Neogeno de representantes en latitudes mayores del continente, antes de restringirse a la distribución actual de Antiphytum en la zona sur de EUA, en lo que se considera Megaméxico 1. -El presente estudio es un claro ejemplo de la necesidad del trabajo de campo y herbario para generar datos fiables previos al abordaje de preguntas sobre biogeografía y diversificación. En particular, la recolección de material sudamericano y la visita a los herbarios MVFA e ICN fue vital para completar la circunscripción del género Antiphytum y ubicar el momento del cambio de color de la corola en el género. 144 -La gran diferenciación de los eremocarpos a nivel del reconocimiento de las especies hace deseable conocer el desarrollo ontogenético de los mismos, para entender cómo se acomodan espacialmente con respecto a la ginobase y dan esa posición característica de la cicatriz. Asimismo, lo es entender el elongamiento de la quilla ventral para formar un estípite o una giba. -Es necesario el uso de marcadores moleculares con una tasa de sustitución más rápida, o el uso de una mayor cantidad de genes usando técnicas de secuenciación masiva, pues el género tiene elementos muy recientes, y no sólo la relación entre los clados no se resuelve, sino que la separación entre los linajes internos del grupo sin rosetas en NA no presenta una resolución concluyente. Por ello, tanto el género como la subfamilia siguen siendo candidatos a estudios a diferentes niveles: genéricos, específicos, poblacionales, o de individuos. -A pesar de las limitaciones impuestas por los resultados, este trabajo recopila la mayor información del género Antiphytum y aclara la situación de las especies del género en Sudamérica, proveyendo las primeras hipótesis de relación entre las especies y los primeros patrones de diversificación y biogeografía del mismo. 145 REFERENCIAS BIBLIOGRÁFICAS Al-Shehbaz, I. 1991. The genera of Boraginaceae in the southeastern United States. Journald of the Arnold Arboretum Suppl. Ser. 1:1–169. Amsinckiinae Working Group. 2022. Systematics of Amsinckiinae (Boraginaceae): The popcorn flowers. Website. https://plants.sdsu.edu/amsinckiinae Baillon, H. 1890. Boraginacées. En: Baillon, H. (Ed.). Histoire des Plantes. Paris: Librairie Hatchette and Cies. Pp. 343–402. Bentham, G. y J. D. Hooker. 1876. Boragineae. En: Genera Plantarum: ad exemplaria imprimis in Herberiis Kewensibus servata definita vol. 2, parte 2. London: Reeve and Co. and Williams and Norgate. Pp. 836–865. Brand, A. 1929. Verweisung des Gattungsnamens Antiphytum in die Synonymie. En: Fedde, F. (Ed.). Repertorium specierum novarum regni vegetabilis vol. 27. Berlin-Dahlem: Selbstverlag. Pp 145–149. Brand, A. 1931. Borraginaceae-Borraginoideae-Cryptantheae. En: Engler, A. (Ed.). Das Pflanzenreich Regni vegetabilis conspectus vol. 27. Leipzig: Wilhelm Engelmann. Pp. 1– 236. Brikiatis, L. 2014. The De Geer, Thulean and Beringia routes key concepts for understanding early Cenozoic biogeography. Journal of Biogeography 41: 1036–1054. Calvillo-Canadell, L. y R. S. Cevallos-Ferriz. 2005. Diverse assemblage of Eocene and Oligocene Leguminosae from México. International Journal of Plant Siences 166(4): 671–692. Cevallos-Ferriz, S. R. S. y J. L. Ramírez. 2004. Bosquejo de la evolución florística. En: García- Mendoza, A. J., M. J. Ordóñez y M. Briones-Salas (Eds.). Biodiversidad de Oaxaca. México, D.F.: Universidad Nacional Autónoma de México, Fondo Oaxaqueño para la Conservación de la Naturaleza & World Wildlife Fund. Pp. 87–104. Chacón, J., F. Luebert, H. H. Hilger, S. Ovchinnikova, F. Selvi, L. Cecchi, C. M. Guilliams, K. Hasenstab-Lehman, K. Sutorý, M. G. Simpson y M. Weigend. 2016. The borage family (Boraginaceae s.str.): A revised infrafamilial classification based on new phylogenetic evidence, with emphasis on the placement of some enigmatic genera. Taxon 65(3): 523– 546. Chacón, J., F. Luebert y M. Weigend. 2017. Biogeographic events are not correlated with diaspore dispersal modes in Boraginaceae. Frontiers in Ecology and Evolution 5: 26. Chamisso de Boncourt, L. C. A. 1829. De Plantis in expeditione speculatoria Romanzoffia observatia disserere pergunt Ad. Dd Chamisso et Died. de Schelechtendal. Boragineae Juss. Linnaea. Ein Journal für die Botanik in ihrem ganzen Umfange 4: 435–496. 146 Cohen, J. I. 2011. A phylogenetic analysis of morphological and molecular characters of Lithospermum L. (Boraginaceae) and related taxa: evolutionary relationships and character evolution. Cladistics 27(5): 559–580. Cohen J. I. 2014. A phylogenetic analysis of morphological and molecular characters of Boraginaceae: evolutionary relationships, taxonomy, and patterns of character evolution. Cladistics 30(2): 139–169. Cohen, J. I. 2015. Adelinia and Andersonglossum (Boraginaceae), two new genera from New World species of Cynoglossum. Systematic Botany 40(2): 611–619. Cohen, J. I. 2022. Phylogenomics, floral evolution, and biogeography of Lithospermum L. (Boraginaceae). Molecular Phylogenetics and Evolution 166: 107317. Cohen, J. I. y Davis J. I. 2009. Nomenclatural changes in Lithospermum (Boraginaceae) and related taxa following a reassessment of phylogenetic relationships. Brittonia 61(2): 101–111. Cohen, J. I. y J. I. Davis. 2012. Molecular phylogenetics, molecular evolution, and patterns of clade support in Lithospermum (Boraginaceae) and related taxa. Systematic Botany 37(2): 490– 506. Cronquist, A. 1981. An integrated system of classification of flowering plants. New York: Columbia University Press. Pp. 1262. De Candolle, A. P. y Candolle, A. L. P. P. de. 1846. Borragineae. En: Candolle, A. P. (Ed.) Prodromus systematis naturalis regni vegetablis, sive, Enumeratio contracta ordinum generum specierumque plantarum huc usque cognitarium, juxta methodi naturalis, normas digesta, Pars Decima. Paris: Victoris Masson. Pp. 117–178. De Melo, J. I. M. y M. N. de Stapf. 2014. Clave para los géneros de Boraginaceae sensu lato nativos del Brasil. Biotemas 27(4): 43–50. Erkens, R. H., J. W. Maas y T. L. P. Couvreur. 2009. From Africa via Europe to South America: migrational route of a species-rich genus of Netropical lowland rain forest trees (Guatteria, Annonaceae). Journal of Biogeography 36: 2338–2352. Fresenius, J. B. G. W. 1857. Familia Borragineae vol. 8, parte 1, pág. 64. En: von Martius, C. F. P., A. W. Eichler y I. Urban (Eds.). Flora brasiliensis (1840-1906). Frohlich, M. W., M. Thulin y M. W. Chase. 2020. Ninety-three new combinations in Euploca for species of Heliotropium section Orthostachys (Boraginaceae sensu APG). Phytotaxa 434(1): 013–021. 147 Givnish, T. J. y S. S. Renner. 2004. Tropical intercontinental disjunctons: Gondwana breakup, immigration from the Boreotropics, and transoceanic dispersal. International Journal of Plant Sciences 165 (4 Suppl.): S1–S6. Gottschling, M., F. Luebert, H. H. Hilger y J. S. Miller. 2014. Molecular delimitations in the Ehretiaceae (Boraginales). Molecular Phylogeny and Evolution 72: 1–6. Gray, A. 1885. Contributions to the Botany of North America: 1. A revision of some Borragineous genera. Proceedings of the American Academy of Arts and Sciences 20: 257–286 Grønlie, G. 1979. Tertiary paleogeography of the Norwegian-Greenland Sea. Norsk Polarinstitutt Skrifter 170: 49–61. Guilliams, C. M. 2015. Diversification, biogeography, and classification of Amsinckiinae (Boraginaceae), with an emphasis on the popcorn flowers (Plagiobothrys). Tesis de doctorado, Universidad de California, Berkeley. Guilliams, C. M., B. A. Veno, M. G. Simpson y R. B. Kelley. 2013. Pectocarya anisocarpa, a new species of Boraginaceae, and a revised key for the genus in western North America. Aliso: A Journal of Systematic and Evolutionary Botany 31(1): 1–13. Guilliams, C. M., K. E. Hasenstab-Lehman, M. E. Mabry y M. G. Simpson. 2017. Memoirs of a frequent flier: Phylogenomics reveals 18 long-distance dispersals between North America and South America in the popcorn flowers (Amsinckiinae). American Journal of Botany 104(11): 1717–1728. Gürke, R. L. A. M. 1897. Fam. Borraginaceae (Asperifoliaceae). En: Engler, A. y K. Prantl (Eds.). Die natürlichen Pflanzenfamilien nebst ihren Gattungen und wichtigeren Arten, insbesondere den Nutzpflanzen, unter Mitwirkung zahlreicher hervorragender Fachgelehrten begründet parte 4, sección a. Leipzig: Wilhelm Engelmann. Pp. 71–131. Hasenstab-Lehman, K. E. y M. G. Simpson. 2012. Cat´s eyes and popcorn flowers: phylogenetic systematics of the genus Cryptantha s.l. (Boraginaceae). Systematic Botany 37(3): 738– 757. Higgins, L. y B. L. Turner. 1983. Antiphytum hintoniorum (Boraginaceae), a bizarre new gypsophile from North-Mexico. The Southwestern Naturalist 28(4): 457–458. Hilger, H. H. y E. Zippel. 2001. “Studies in the Boraginaceae”. An index to the publications of Ivan M. Johnston dealing with the Borage Family. Haussknechtia 11: 4–10. Holstein, N. y M. Weigend. 2016. Lectotypifications and nomenclatural notes in the Boraginaceae s.str. for the Flora of Ecuador. Phytotaxa 278(1): 69–73. 148 Holstein, N., J. Chacón, A. Otero, P. Jiménez-Mejías y M. Weigend. 2016a. Towards a monophyletic Omphalodes—or an expansión of North American Mimophytum. Phyotaxa 288(2): 131–144. Holstein, N., J. Chacón, H. H. Hilger y M. Weigend. 2016b. No longer shipwrecked—Selkirkia (Boraginaceae) back on the mainland with generic rearrangements in South American “Omphalodes” based on molecular data. Phytotaxa 270(4): 231–251. Johnston, I. M. 1923. Studies in the Boraginaceae. Contributions from the Gray Herbarium of the Harvard University 68: 43–80. Johnston, I. M. 1924. Studies in the Boraginaceae II. 1. A synopsis of the American natural and immigrant Borages of the subfamily Boraginoideae. Contributions from the Gray Herbarium of the Harvard University 70: 3–55. Johnston, I. M. 1925. Concerning the range and identity of certain American species. Contributions from the Gray Herbarium of Harvard University 75: 40–49. Johnston, I. M. 1927. Studies in the Boraginaceae VI: A revision of the South American Boraginoideae. Contributions from the Gray Herbarium of Harvard University 78: 3–118. Johnston, I. M. 1948. Studies in the Boraginaceae XVI: Species chiefly from Mexico and Western United States. Journal of the Arnold Arboretum 29: 227–241. Johnston I. M. 1954. Further revaluations of the genera of the Lithospermeae. Journal of the Arnold Arboretum 35: 1–81. Johnston, I. M. 1957. Studies in the Boraginaceae XXIX: Echiochilon and related genera. Journal of the Arnold Arboretum 38: 255–294. Johnston I. M. 1966. Antiphytum. En: Lundell, C. L. (Ed.). Flora de Texas vol. 1. Texas: Research Foundation. Pp. 196–200. Jørgensen, P. M., M. H. Nee y S. G. Beck. 2014. Catálogo de las plantas vasculares de Bolivia. 127(1–2): i–viii, 1–1744. En: Jørgensen, P. M., M. H. Nee y S. G. Beck (Eds.) Catálogo de Plantas Vasculares de Bolivia. St. Louis: Missouri Botanical Garden Press. Kadereit, J. W. y B. G. Baldwin. 2012. Western Eurasian–western North American disjunct plant taxa: The dry-adapted ends of formerly widespread north temperate mesic lineages—and examples of long-distance dispersal. Taxon 61(1): 3–17. Långström, E. y M. W. Chase. 2002. Tribes of Boraginoideae (Boraginaceae) and placement of Antiphytum, Echiochilon, Ogastemma and Sericostoma: A phylogenetic analysis based on atpB plastid DNA sequence data. Plant Systematics and Evolution 234: 137–153. 149 Långström, E. y B. Oxelman. 2003. Phylogeny of Echiochilon (Echiochileae, Boraginaceae) based on ITS sequences and morphology. Taxon 52(4): 725–735. Lavin, M. y M. Luckow. 1993. Origins and relationships of tropical North America in the context of the Boreotropics Hypothesis. American Journal of Botany 80: 1–14. Lönn, E. 1999. Revision of the Boraginaceae genera Echiochilon, Ogastemma and Sericostoma. Botanical Journal of the Linnean Society 130: 185–259. Luebert, F., L. Cecchi, M. W. Frohlich, M. Gottschling, C. M. Guilliams, K. E. Hasenstab-Lehman, H. H. Hilger, J. S. Miller, M. Mittelbach, M. Nazaire, M. Nepi, D. Nocentini, D. Ober, R. G. Olmstead, F. Selvi, M. G. Simpson, K. Sutorý, B. Valdés, G. K. Walden y M. Weigend. 2016. Familial classification of the Boraginales. Taxon 65(3): 502–522. Luebert, F., T. L. Couvreur, M. Gottschling, H. H. Hilger, J. S. Miller y M. Weigend. 2017. Historical biogeography of Boraginales: West Gondwanan vicariance followed by long‐ distance dispersal? Journal of Biogeography 44(1): 158–169. Mabry, M. E. y M. G. Simpson. 2018. Evaluating the monophyly and biogeography of Cryptantha (Boraginaceae). Systematic Botany 41(1): 53–76. Macbride, J. F. 1916. Notes on Certain Borraginaceae. Contributions from the Gray Herbarium of Harvard University 48: 39–58. Martens, M. y H. Galeotti. 1844. Enumeratio synoptica plantarum phanerogamicarum ab Henrico Galeotti in regionibus Mexicanus collectarum. Bulletin de l'Academie Royale des Sciences et Belles-lettres de Bruxelles 11(2): 319–340. McVaugh, R. 2000. Botanical results of the Sessé & Mociño expedition (1787-1803). VII. A guide to relevant scientific names of plants. Hunt Institute for Botanical Documentation. Pittsburgh: Cargenia Mellon University. Pp. 626. McLoughlin, S. 2001. The breakup history of Gondwana and its impact on pre-Cenozoic floristic provincialism. Australan Journal of Botany 49: 271–300. Meisner, C. F. 1840. Plantarum vascularium genera: secundum ordines naturales digesta eorumque differentiae et affinitates tabulis diganosticis expositae. Parte uno: Tabulae diagnosticae. Pp. 280. Parte dos: Commentarius. Pp. 188. Mendoza-Díaz, N. 2015. Sistemática de Antiphytum DC. ex Meisn. (Boraginaceae). Tesis Maestría en Ciencias. Posgrado de Ciencias Biológicas UNAM. Ministerio de Industria y Energía de Uruguay. 1987. Memoria explicativa de la carta de materias primas minerales no metálicas. República Oriental del Uruguay. Mociño, J. M. y M. Sessé. 2010. La Real Expedición Botánica a Nueva España vol. 5. Siglo XXI Editores y Universidad Nacional Autónoma de México. Pp. 360. 150 Moore M. J., A. Tye y R. K. Jansen. 2006. Patterns of long-distance dispersal in Tiquilia subg. Tiquilia (Boraginaceae): Implications for the origins of amphitropical disjuncts and Galápagos Islands endemics. American Journal of Botany 93(8): 1163–1177 Nazaire, M. y L. Hufford. 2012. A broad phylogenetic analysis of Boraginaceae: Implications for the relationships of Mertensia. Systematics Botany 37(3): 758–783. Nazaire, M. y L. Hufford. 2014. Phylogenetic systematics of the genus Mertensia (Boraginaceae). Systematic Botany 39(1): 268–303. Nazaire M., X. Wang y L. Hufford. 2014. Geographic origins and patterns of radiation of Mertensia (Boraginaceae). American Journal of Botany 101(1): 104–118. Ochoterena, H., H. Flores-Olvera, C. Gómez-Hinostrosa y M. J. Moore. 2020. Gypsum and plant sciences: A marvel of Cuatro Ciénegas and the Chihuahuan Desert. En: Mandujano, M. C., I. Pisanty y L. E. Eguiarte (Eds.). Plant Diversity and Ecology in the Chihuahuan Desert, Cuatro Ciénegas Basin: An endangered hyperdiverse oasis. Switzerland: Springer Nature. Pp. 129–165. Otero, A., P. Jiménez-Mejía, V. Valcárcel y P. Vargas. 2014. Molecular phylogenetics and morphology support two new genera (Memoremea and Nihon) of Boraginaceae s.s. Phytotaxa 173: 241–277. Otero, A., P. Jiménez-Mejías, V. Valcárcel y P. Vargas. 2019. Worldwide long-distance dispersal favored by epizoochorous traits in the biogeographic history of Omphalodeae (Boraginaceae). Journal of Systematics and Evolution 57(6): 579–593. Pérez-García, E. A., J. A. Meave y S. R. S. Cevallos-Ferriz. 2012. Flora and vegetation of the seasonally dry tropics in Mexico: Origin and biogeographical implications. Acta Botánica Mexicana 100: 149–193. Pourghorban, Z., Y. Salmaki y M. Weigend. 2020. Phylogenetic relationships within the subtribe Cynoglossinae (Cynoglossoideae: Boraginaceae): New insights from nuclear and plastid DNA sequence data. Plant Systematics and Evolution 306: 45. Raven, P. H. 1963. Amphitropical relationships in the floras of North and South America. The Quarterly Review of Biology 38(2): 151–177. Raven, P. H. y D. I. Axelrod. 1974. Angiosperm biogeography and past continental movements. Annals of the Missouri Botanical Garden 61(3): 539–673. Rebman, J. P. y M. G. Simpson. 2022. Three new species of Cryptantha (Boraginaceae) from the Southern Channel Islands of California. Madroño 69(1): 111–134. Refulio-Rodríguez, N. y R. G. Olmstead. 2014. Phylogeny of Lamiidae. American Journal of Botany 101(2): 287–299. 151 Riedl, H. 1967. Boraginaceae. En: Rechinger, K. H. (Ed.). Flora Iranica Lfg. 48. Graz: Akademische Druck- und Verlagsanstalt. Pp. 1–1281. Riedl, H. 1968. Die neue Tribus Trigonotideae and das System der Boraginoideae. Österreichische Botanische Zeitschrift 115: 291–321. Rose, J. N. 1890. List of plants colleted by Dr. Edward Palmer in Lower California and western Mexico in 1890. Contributions from the United States National Herbarium 1(3): 85. Sanmartín, I., H. Enghoff y F. Ronquist. 2001. Patterns of animal dispersal, vicariance and diversification in the Holartic. Biological Journal of the Linnean Society 73: 345–390. Selvi, F., A. Coppi y L. Cecchi. 2011. High epizoochorous specialization and low DNA sequence divergence in Mediterranean Cynoglossum (Boraginaceae): Evidence from fruit traits and ITS region. Taxon 60: 969–985. Sessé, M. y J. M. Mociño. 1887. Plantae Novae Hispaniae. Mexici: Apud Ignatium Escalante. Bajos de San Agustín. Número 1. Pp. 21. Setubal, R. B. 2010. Vegetação campestre subtropical de um morro granítico no Sul do Brasil, Morro São Pedro, Porto Alegre, RS. Tesis de Maestría. Universidade Federal do Rio Grande do Sul. Pp.148. Simpson, M. G., C. M. Guilliams, K. E. Hasenstab-Lehman, M. E. Mabry y L. Ripma. 2017a. Phylogeny of the popcorn flowers: Use of genome skimming to evaluate monophyly and interrelationships in subtribe Amsinckiinae (Boraginaceae). Taxon 66(6): 1406–1420. Simpson, M. G., L. A. Johnson, T. Villaverde y C. M. Guilliams. 2017b. American amphitropical disjuncts: Perspectives from vascular plant analyses and prospects for future research. American Journal of Botany 104(11): 1600–1650. Simpson, M. G., M. E. Mabry y K. Hasenstab-Lehman. 2019. Transfer of four species of Cryptantha to the genus Johnstonella (Boraginaceae). Phytotaxa 425(5): 279–289. Simpson, M. G. y J. P. Rebman. 2021. Research in Boraginaceae: A new variety of Cryptantha maritime, Cryptantha pondii resurrected, and Johnstonella echinosepala transferred back to Cryptantha. Phytotaxa 509(2): 185–210. Smerdmark, J. E. y A. A. Anderberg. 2007. Boreotropical migration explains hybridization between geographically distant lineages in the pantropical clade Sideroxyleae (Sapotaceae). American Journal of Botany 94(9): 1491–1505. Sutorý, K. 2010. Oncaglossum, a new genus of the Boraginaceae, tribe Cynoglosseae, from Mexico. Novon 20: 463–469. Takhtajan, A. 1980. Outline of the classification of flowering plants (Magnoliophyta). Botanical Review 46: 225–359. 152 Takhtajan, A. 1997. Diversity and classification of flowering plants. New York: Columbia University Press. Pp. 643. Thomasson, J. R. 1979. Late Cenozoic grasses and other Angiosperms from Kansas, Nebraska, and Colorado: Biostratigraphy and relationships to living Taxa. Kansas Geological Survey Bulletin 218: 1–68. Tiffney, B. H. 1985. The Eocene North Atlantic land bridges: Its importance in Tertiary and modern phytogeography of the Northern Hemisphere. Journal of the Arnold Arboretum 66: 243– 273. Tiffney, B. H. y S. R. Manchester. 2001. The use of geological and paleontological evidence in evaluating plant phytogeographic hypotheses in the Northern Hemisphere Tertiary. International Journal of Plant Sciences 162 (6 Suppl.): S3–S17. Torrey, J. 1859. Boraginaceae. En: Emory, W. H. (Ed.). Report on the United States and Mexican boundary survey made under the direction of the secretary of the interior vol. 2, parte 1. Washington: Cornelius Wendell. Pp. 135–142. Watson, S. 1883. Contributions to American Botany. Proceedings of the American Academy of Arts and Sciences 18: 96–196. Weigend, M., M. Gottschling, F. Selvi y H. H. Hilger. 2009. Marbleseeds are gromwells- systematics and evolution of Lithospermum and allies (Boraginaceae tribe Lithospermeae) based on molecular and morphological data. Molecular Phylogenetics and Evolution 52(3): 755–768. Weigend, M., M. Gottschling, F. Servi y H. H. Hilger. 2010. Fossil and extant western hemisphere Boraginae, and the polyphyly of “Trigonitideae” Riedl (Boraginaceae: Boraginoideae). Systematic Botany 35(2): 409–419. Weigend, M., F. Luebert, F. Selvi, G. Brokamp y H. H. Hilger. 2013. Multiple origins for Hound´s tongues (Cynoglossum L.) and Navel seeds (Omphalodes Mill.). The phylogeny of the borage family (Boraginaceae s. str.). Molecular Phylogenetics and Evolution 68: 604–618. Weigend, M., F. Luebert, M. Gottschling, T. L. P. Couvreur, H. H. Hilger y J. S. Miller. 2014. From capsules to nutlets –phylogenetic relationships in the Boraginales. Cladistics 30: 508–518. Weigend, M., F. Selvi, D. C. Thomas y H. H. Hilger. 2016. Boraginaceae. En: Kadereit, J. W. y V. Bittrich (Eds.). The families and genera of vascular plants Vol. 14. Flowering plants: Eudicots; Aquifoliales, Boraginales, Bruniales, Dipsacales, Escalloniales, Garryales, Paracryphiales, Solanales (except Convolvulaceae), Icacinaceae, Metteniusaceae, Vahliaceae. Springer International. Pp. 41–102. 153 Wikström, N., V. Savolainen y M. W. Chase. 2001. Evolution of the angiosperms: Calibrating the familiy tree. Proceedings of the Royal Society of London B 268: 2110–2220. Wolfe, J. A. 1975. Some aspects of plants geography of the Northern Hemisphere during the Late Cretaceous and Tertiary. Annals of the Missouri Botanical Garden 62(2): 264–279. Zuloaga, F. O., O. Morrone y M. J. Belgrano. 2021 en curso. Catálogo de las plantas vasculares del Cono Sur. Instituto de Botánica Darwinion. San Isidro. Buenos Aires, Argentina