REVISIÓN TAXONÓMICA Y CÓDIGOS DE BARRAS DE DNA PARA ZAMIA L. EN MEGAMÉXICO

TESIS QUE PRESENTA EL M. C. EDISON FERNANDO NICOLALDEMOREJÓN PARA OBTENER EL GRADO DE DOCTOR EN CIENCIAS

Xalapa, Veracruz, México 2009

Aprobación final del documento final de tesis de grado: REVISIÓN TAXONÓMICA Y CÓDIGOS DE BARRAS DE DNA PARA ZAMIA L. EN MEGAMÉXICO Nombre

Firma

Director: Dr. Andrew P. Vovides

________________________

Comité Tutorial: Dr. Dennis Wm. Stevenson

________________________

Dra. Victoria Sosa Ortega

________________________

Dr. Jorge González Astorga

________________________

Jurado Dr. Francisco Vergara Silva

________________________

2

RECONOCIMIENTOS •

A la Red Latinoamericana de Botánica (RLB) por la beca otorgada para estudios de doctorado (RLB-06-D2).

•

Al Instituto de Ecología, A. C., por el apoyo económico para desarrollar una estancia académica en el Instituto de Biología de la Universidad Nacional Autónoma de México.

•

A la Comisión Nacional para el Conocimiento y Uso de la Biodiversidad (CONABIO), que a través de proyecto GE004, financió parte de este trabajo de tesis.

•

A los siguientes investigadores por su apoyo: Andrew P. Vovides, Dennis Wm. Stevenson, Jorge González Astorga, Francisco Vergara Silva, Victoria Sosa, Alejandro Espinosa de los Monteros, Octavio Rojas Soto, Mario Vázquez Torres, Fernando Chiang, Gonzalo Castillo, Francisco Ornelas, Sergio Avendaño, Walter Palacios, Calaway H. Dodson, Carlos Cerón, John Janovec y David Neill.

•

A toda mi familia: mi madre Olga Marina, mis hermanos: Patricia, Silvia, Diego y Alexandra.

•

A Julia Hernández Villa técnico del Laboratorio de Genética de Poblaciones del Instituto de Ecología, A. C.

•

A mis amigos: David Martínez, Alejandro Abundis, Nadia Rivera, Pablo Carrillo, Dánae Cabrera, Lalo Ruiz, Deneb García, Xavier Barrientos, Carlos Durán, Fernando Vaz de Mello, Jaime Pacheco, Julián Bueno, Claudia Hornung.

•

A todo del personal del Jardín Botánico Francisco Javier Clavijero: Carlos Iglesias, Orlik Gómez, Víctor Luna, Philip J. Brewster, Julián Pérez, Daniel Hernández, Javier Hernández, Joel López y Genaro Justo.

3

DEDICATORIA

A Zamia Fernanda, Elvia Rubí y Olga Marina

4

DECLARACIÓN

Excepto cuando es explícitamente indicado en el texto, el trabajo de investigación contenido en esta tesis fue efectuado por el M. C. Edison Fernando Nicolalde Morejón como estudiante de la carrera de Doctorado en Ciencias entre septiembre del 2006 y septiembre del 2009, bajo la supervisión del Dr. Andrew P. Vovides.

Las investigaciones reportadas en esta tesis no han sido utilizadas anteriormente para obtener otros grados académicos, ni serán utilizadas para tales fines en el futuro.

Candidato: ________________________________________ M. C. Edison Fernando Nicolalde Morejón Director de tesis: ________________________________________ Dr. Andrew P. Vovides

5

ÍNDICE Resumen…………………………….…………………………………………...……...9 CAPITULO I—Introducción El género Zamia L. en Megaméxico: de la taxonomía alfa a los códigos de barras genéticos………………………………………………….…………………………....10 Resumen………………………………………………………………………...….12 Abstract……………………………………………………………………….....…12 Introducción………………………………………………………………….…..…14 Materiales y Métodos…………………………………………………………...….15 Resultados y Discusión……………………………………………………………..16 Morfología y grupos afines……………………………………………………..16 Megaméxico, diversidad y endemismo………………………...........................19 Intensidad de colectas botánicas………………..………………………………20 Taxonomía alfa y código de barras moleculares en Zamia…………………….21 Perspectivas del género Zamia en Megaméxico………………………………..24 Agradecimientos…………………………………………………………………....26 Literatura citada…………………………………………………………………….26 Figura 1……………………………………………………………………………..33 Tabla 1……………………………………………………………………………...34 Tabla 2………………………………………………………………………….…..35 Tabla 3……………………………………………………………………………...36 Tabla 4……………………………………………………………………………...38 CAPITULO II—Taxonomic revision of Zamia in Mega-Mexico……………….….39 Abstract………………………………………………………………………….….41 Resumen…………………………………………………………………………....41 Introduction…………………………………………………………………….......43 Materials and Methods...…….…………………………………………………......44 Results…………………………………………………………………………...…45 Habitat…………………………………………………………………………...…45 Morphology……………………………………………………………………...…46 Habit………………………………………………………………………...….46 6

Trichoms………………………………………………………………………..47 Reproductive structure…………………………………………………….……47 Chromosome numbers………………………………………………………….48 Distribution and endemism…………………………………………………………50 Taxonomic treatment……………………………………………………………….51 Species dubium…………………………………………………………………96 Acknowledgments………………………………………………………………….98 Literature cited……………………………………………………………..……….98 Figura 1……………………………………………………………………...….…105 Figura 2……………………………………………………………………...….…106 Figura 3……………………………………………………………………...….…107 Figura 4………………………………………………………………...……….…108 Figura 5……………………………………………………………...………….…109 Figura 6………………………………………………………………...………….110 Figura 7……………………………………………………………..….………….111 Figura 8……………………………………………………………...…………….112 Figura 9…………………………………………………………..….…………….113 Figura 10…………………………………………………………………………..114 Figura 11…………………………………………………………………………..115 Figura 12……………………………………………………………………..……116 Figura 13………………………………………………………………..…………117 CAPITULO III— DNA barcoding in the Mexican cycads: a character attribute organization system (CAOS) approach…………………………………….………118 Abstract……………………………………………………………………………120 Introduction………………………………………………………………….……121 Materials and Methods……………………………………………....................…126 Sampling of biological materials……………………………………...………126 Leaf genomic DNA extraction and PCR amplification (including DNA sequencing).…………………………………….……………..………………127 Sequence analysis.……………………………………………..……...………128 Character-based analysis/identification of ‘DNA diagnostics’ and determination of DNA barcodes in Mexican cycad species……………………..………...…128 7

Results and discussion…….………………………………………………………129 A DNA barcode for land plants: difficult roads toward a consensus................129 DNA barcoding in the cycads redux I: surprising results under the assumptions of a character-based method.............................................................................130 DNA barcoding in the cycads redux II: is the ‘Seberg-Petersen limit’ real?...136 Conclusions: gene quantity versus universality in plant DNA barcoding, and ‘the wisdom of the local’...................................................................................138 Acknowledgements……………………………………………………….………139 References……………………………………………………………...…………140 Tables………………………………………………………………….……..……148 Figures captions……………………………………………………….………..…149 Tabla 1………………………………………………………………….…………150 Tabla 2………………………………………………………………….…………151 Tabla 3……………………………………………………………………….……153 Tabla 4………………………………………………………………….…………154 Figure 1……………………………………………………………….…..……….155 Figure 2……………………………………………………………….…..……….156 CAPITULO IV—Discusión y conclusión general……………...………….…….…157 1. Implicaciones taxonómicas………………………………..................................158 2. Códigos de barras moleculares y la taxonomía del genero Zamia.………..…...159 3. Literatura citada…………………………………………………...…….…...…161

8

RESUMEN Los objetivos de esta tesis son 1) determinar y describir las especies del género Zamia L. en Megaméxico, área geográfica que abarca los países de México, Guatemala, Belice, Honduras, El Salvador y el Norte de Nicaragua, y 2) determinar la combinación optima de loci necesario para identificar molecularmente las especies del género, bajo la aproximación de los Códigos de Barras Moleculares. La tesis está dividida en cuatro capítulos que brevemente se describen a continuación. En el primer capítulo, el cual funge como introductorio, se presenta una síntesis de la variación morfológica y la formación de grupos afines, luego se describe la distribución y las áreas de mayor concentración de riqueza que el género tiene en Megaméxico; para finalmente presentar las perspectivas sistemáticas del género en el contexto de los códigos de barras moleculares. El segundo capítulo, corresponde a la revisión taxonómica del género Zamia, revisión que incluye una clave dicotómica, descripciones botánicas para cada una de las especies, material de herbario revisado, etimología y principales atributos morfológicos que representan caracteres diagnósticos para estas especies. También se indican nuevas designaciones de tipos nomenclaturales. En el tercer capítulo, es un estudio que aborda el análisis de sitios diagnósticos bajo la perspectiva de los códigos de barras moleculares para las especies del género Zamia de la revisión taxonómica, también se incluye en este análisis a todas la especies de Cycadas Mexicanas y al menos a un representante de los géneros no Mexicanas del orden Cycadales. En el cuarto capítulo se discuten los resultados más relevantes de esta tesis, su importancia y contribución al conocimento de la biología del género Zamia en Megaméxico.

9

CAPITULO I—Introducción Perspectivas sistemáticas de Zamia (Zamiaceae) en Megaméxico: de la taxonomía alfa a los códigos de barras genéticos

(Sometido a la Revista Mexicana de Biodiversidad)

10

Perspectivas sistemáticas de Zamia (Zamiaceae) en Megaméxico: de la taxonomía alfa a los códigos de barras genéticos

Fernando Nicolalde-Morejón1, 2*, Jorge González-Astorga1, Francisco Vergara-Silva3 y Andrew P. Vovides4

1

Laboratorio de Genética de Poblaciones, Departamento de Biología Evolutiva.

Instituto de Ecología, A. C., km 2.5 Antigua Carretera a Coatepec No. 351, Xalapa 91070, Veracruz, México 2

Instituto de Investigaciones Biológicas, Universidad Veracruzana, Av. Luis Castelazo

Ayala s/n, Col. Industrial Ánimas, Xalapa 91190, Veracruz, México 3

Laboratorio de Sistemática Molecular (Jardín Botánico), Instituto de Biología,

Universidad Nacional Autónoma de México, 3er Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México, D. F. México 4

Laboratorio de Biología Evolutiva de Cycadales, Departamento de Biología Evolutiva.

Instituto de Ecología, A. C., km 2.5 Antigua Carretera a Coatepec No. 351, Xalapa 91070, Veracruz, México

*Autor correspondiente: [email protected]

11

Resumen. El género Zamia en Megaméxico cuenta con 22 especies descritas y una entidad bajo el estatus de species dubium (Z. verschaffeltii). En las últimas décadas, el género Zamia ha recibido atención en tratamientos florísticos regionales y, de manera sobresaliente, en una monografía especializada. Además, algunas especies del género han sido objeto de varios estudios en citogenética, ecología y genética de poblaciones. El objetivo de este trabajo es presentar información actualizada sobre las especies de Zamia que se distribuyen en Megaméxico, con base en una revisión de ejemplares de herbario y trabajo de campo. Adicionalmente se hace énfasis en los complejos de especies que aún requieren investigación para esclarecer sus límites taxonómicos. La discusión se plantea bajo la perspectiva de la necesidad de investigación a nivel poblacional con datos moleculares, que se enmarca en la aproximación conocida como ‘códigos de barras de DNA’ (“DNA barcoding”). Se concluye que la creación de una base de datos moleculares que funcione como una ‘biblioteca de referencia de códigos de barras’ para todas las especies de Zamia en Megaméxico será de utilidad en varios aspectos de la investigación básica y aplicada de las cycadas neotropicales.

Palabras clave: Diversidad, DNA Barcoding, Megaméxico, Taxonomía, Zamia

Abstract. The genus Zamia in Megamexico consists of 22 valid species and one species dubium (Z. verschaffeltii) and over the last decades Zamia has undergone regional floristic treatments, one specialized monograph and some species have been object to a number of studies in cytotaxonomy, cytogenetics, ecology and population genetics. The objective of this study is to present updated information on the genus in Megamexico based herbarium voucher revisions and field studies. Emphasis is made on species complexes that still require further studies to elucidate taxonomic limits and

12

the need for population level studies using molecular data to underpin a DNA barcoding approach is discussed. We conclude that the creation of a molecular database that can function as a ‘DNA-barcode reference library’ covering Zamia species in Megamexico will be of utility to varying aspects of basic and applied research on neotropical cycads.

Key words: Diversity, DNA Barcoding, Megamexico, Taxonomy, Zamia.

13

Introducción El orden Cycadales consta de tres familias: Cycadaceae, Stangeriaceae y Zamiaceae (Stevenson, 1992). Zamiaceae incluye, con mucha diferencia, el mayor número de géneros descritos para el orden, y en el Neotrópico se distribuyen cinco géneros: Chigua D. W. Stev. y Microcycas (Miq.) A. DC. endémicos de Colombia y Cuba, respectivamente; Ceratozamia Brongn. y Dioon Lindl. endémicos de México y Zamia L. con una distribución neotropical. La primera monografía de Zamia incluyó diez especies (Miquel, 1842). Posteriormente, se registraron 23 especies (Miquel, 1851; 1861) de las cuales 14 se consideran sinónimos hoy en día (Hill et al., 2007); más tarde, Schuster (1932) reportó 26 especies. Trabajos taxonómicos regionales (Sabato, 1990; Stevenson, 1987; 1991a, b; 1993; 2001a, b; 2004; Norstog y Nicholls, 1997, Nicolalde-Morejón et al., 2008), así como la tipificación de nombres válidos (Stevenson y Sabato, 1986), han detectado errores nomenclaturales y taxonómicos en la revisión propuesta por Schuster (1932), debido principalmente al insuficiente trabajo de campo y a la escasa revisión de las colecciones botánicas. A la fecha, de acuerdo con las estimaciones taxonómicas más recientes, Zamia consta de 61 especies válidas (Hill et al., 2007; Taylor et al., 2009), distribuidas desde Georgia y Florida en Estados Unidos hasta Bolivia y el suroeste de Brasil (Balduzzi et al., 1982; Sabato, 1990; Norstog y Nicholls, 1997; Stevenson, 2001a). Dentro de este último grupo de especies, 22 son endémicas a Megaméxico. Zamia es uno de los géneros del orden Cycadales más difíciles de caracterizar (Norstog y Nicholls, 1997), debido a que sus patrones de variación morfológica, ecológica, citológica (Marchant, 1968; Vovides, 1983; Moretti y Sabato, 1984; Moretti, 1990; Moretti et al., 1991; Vovides y Olivares 1996; Norstog y Nicholls, 1997; Nicolalde-Morejón et al., 2009a), y genética (González-Astorga et al., 2006;

14

Limón, 2009) son muy complejos. Además, debido a los ciclos de vida largos y al tamaño y volumen de las estructuras reproductivas de las especies del género, escasos especímenes botánicos incluyendo estas estructuras son depositados en museos y herbarios. De hecho, los caracteres diagnósticos para identificar las especies de Zamia están basados principalmente en atributos foliares, los cuales son muy variables. Esta circunstancia ha traído confusión en la delimitación y reconocimiento de especies del género, lo cual se refleja directamente en un considerable número de sinonimias (58, para ser exactos; Hill et al., 2007). El presente artículo es parte inicial de un proyecto monográfico de largo alcance del género Zamia, el cual además de incorporar los aspectos tradicionales de una monografía incluirá evidencia molecular con la aproximación conocida como ‘DNA barcoding’ o ‘código de barras genético’ (Hebert et al., 2003). En línea con lo anterior, este artículo tiene los siguientes objetivos: (a) determinar cuántas y cuáles especies de Zamia se distribuyen en Megaméxico (sensu Rzedowski, 1991), (b) presentar un panorama histórico-taxonómico del género haciendo énfasis en los complejos de especies que se requieren estudiar a mayor profundidad o detalle, y finalmente (c) discutir la utilidad de los códigos de barras para identificar, e incluso delimitar especies en el grupo.

Materiales y Métodos Para la realización de este trabajo se revisó el material botánico disponible en los siguientes herbarios: B, BM, CIB, CHIP, CICY, ECOSUR, ENCB, F, FCME, FLAS, FTG, HEM, IBUG, IEB, K, LE, MEXU, MO, NY, SERO, U, UADY, UAMIZ, US, W, WIS, XAL, XALU y ZEA. La información se complementó con la consulta de

15

las descripciones originales así como de los tratamientos taxonómicos tanto regionales como de todo el género. Con el objeto de estudiar los endemismos con referencia geográfica natural y usando criterios históricos y florísticos, Rzedowski (1991) delimitó áreas para su definición y posterior análisis, áreas a las cuales llamó, precisamente, ‘Megaméxico’. Siguiendo el trabajo de dicho autor, nosotros hacemos uso del concepto ‘Megaméxico 2’, región que abarca a México, Guatemala, Honduras, Belice, El Salvador, así como el norte de Nicaragua, para la descripción y análisis de las especies del género Zamia que resultan endémicas para esta región.

Resultados y Discusión Morfología y grupos afines Zamia incluye plantas dioicas al igual que los demás géneros del orden Cycadales. Asimismo, se trata de un género entomófilo y de larga vida, con distribución restringida a los trópicos. Zamia presenta además una amplia diversidad de formas morfológicas y ecológicas: por ejemplo, mientras que la endémica colombiana Z. wallisii A. Brau presenta folíolos fuertemente acanalados de hasta 15 cm de ancho, con nervaduras conspicuas, Z. spartea A. DC. –endémica a Oaxaca, México– presenta folíolos de hasta 1 cm de ancho, totalmente lisos. El género también incluye especies con tallos arbóreos, como es el caso de Z. obliqua A. Braun (Colombia y Panamá), que puede llegar a medir hasta 5 m de alto (Stevenson, 2004). En contraste, también presenta especies como Z. paucijuga Wieland –otro endemismo de México– y Z. amazonum D. W. Stev. –proveniente de la Amazonía de Colombia, Ecuador, Perú y Brasil– que posee tallos subterráneos. Además, Z. pseudoparasitica Yates in Seem., especie endémica de Panamá, representa el único caso de especie epifita para el orden

16

Cycadales (Stevenson, 1993). En términos ecológicos, especies de Zamia se encuentran en hábitats muy contrastantes, como es el caso de Z. gentryi Dodson, endémica de Ecuador (Nicolalde-Morejón, 2007) que se distribuye en bosques húmedos premontanos (sensu Holdridge, 1978), con lluvias de hasta 6500 mm en promedio al año. Por su parte, Z. encephalartoides D. W. Stev. –endémica de Colombia– crece en hábitats secos (Stevenson, 2004), mientras que Z. roezlii Linden, distribuida en el Chocó Biogeográfico (Ecuador y Colombia), se conoce a partir de poblaciones creciendo en manglares. Altitudinalmente, Zamia se distribuye desde el nivel del mar y sobre dunas costeras (i. e. Z. furfuracea L. f., México) hasta los 2700 metros, como ocurre con Z. montana A. Braun, especie endémica a bosques húmedos premontanos en Antioquia, Colombia (Stevenson, 2001a). Tomando en cuenta la morfología –especialmente, hábito, hojas y folíolos– de múltiples especímenes dentro del género Zamia (F. Nicolalde-Morejón, observaciones no publicadas; Vovides et al., 2007), nosotros consideramos que en Megaméxico existen varios complejos de especies para el género, los cuales podrían ser de difícil identificación. Por ejemplo, en el noreste de México (Tamaulipas, San Luís Potosí, Querétaro, Hidalgo y Norte de Veracruz) ocurren dos especies en apariencia similares: Z. fischeri Miq. y Z. vazquezii D. W. Stev., Sabato, A. Moretti y De Luca. Este par de entidades, aquí denominado “complejo Zamia fischeri”, incluye plantas generalmente pequeñas con folíolos papiráceos, dentaciones sobre los bordes de los folíolos y tallos subterráneos; otros caracteres diagnósticos son principalmente atributos asociados a estructuras reproductivas femeninas (Nicolalde-Morejón et al., 2009a). Por otro lado, en el sureste de México (Veracruz, Oaxaca, Tabasco y Chiapas) se encuentra el “complejo Zamia katzeriana” que incluye a Z. katzeriana (Regel) Retting, Z. cremnophila Vovides, Schutzman & Dehgan, Z. lacandona Schutzman & Vovides y Z.

17

purpurea Vovides, J. D. Rees & Vázq. Torres. Estas especies se caracterizan por presentar folíolos anchos (3-11 cm), coriáceos y tallos subterráneos (Nicolalde-Morejón et al., 2008). Aunque Z. purpurea es incluida en este grupo, sus folíolos acanalados con nervaduras conspicuas –atributo único entre las especies en Megaméxico– la asemejan más al complejo basado en Z. skinneri Warsz. ex A. Dietrich, que se distribuye a lo largo del Chocó Biogeográfico. El complejo taxonómico que se construye alrededor de Zamia loddigesii (i.e. Z. loddigesii Miq., Z. paucijuga Wieland y Z. polymorpha D. W. Stev., A. Moretti y Vázq. Torres) incluye las especies con la más amplia distribución en Megaméxico (ver Figura 1), los números cromosómicos más altos del género (Moretti, 1990) y, desde el punto de vista morfológico, los taxa que presentan más dificultad en ser caracterizados. Entre los atributos fenotípicos del complejo están los folíolos coriáceos, linearlanceolados a oblanceolados y los tallos subterráneos. Sin embargo, bajo la misma perspectiva morfológico-vegetativa, consideramos que Z. spartea (endémica al Istmo de Tehuantepec-México) y Z. prasina W. Bull (endémica de Belice) deberían ser incluidas en este complejo, para posteriores estudios taxonómicos y evolutivos. Aunque no necesariamente existen problemas para su identificación, entre especies del grupo compuesto por Z. tuerckheimii Donn. Sm. (Guatemala), Z. soconuscensis Schutzman, Vovides y Dehgan (Chiapas, México) y Z. inermis Vovides, J. D. Rees & Vázq. Torres (Veracruz, México) se comparten caracteres como los tallos epigeos y los folíolos con los bordes enteros. Sin embargo, debido a la presencia de folíolos linear-lanceolados, coriáceos y bordes enteros, los ejemplares de herbario de Z. inermis (usualmente carentes de estructuras reproductivas) podrían ser confundidos con ejemplares de Z. encephalartoides D. W. Stevenson. Además es necesario considerar a Z. onanreyesii C. Nelson & G. Sandoval, especie que también presenta tallos aéreos de hasta 2 m de alto,

18

pero que a diferencia de las anteriores posee folíolos subcoriaceos y márgenes serrulados sobre el tercio distal. Las especies Zamia oreillyi C. Nelson, Z. sandovallii C. Nelson y Z. standleyi Schutzman, todas de reciente descripción, se caracterizan por tener folíolos lisos, subcoriaceos, bordes serrulados a dentados y tallos subterráneos. Aparte, dichas especies se distribuyen en una misma región (Honduras y Guatemala); como su identificación y delimitación a través de caracteres morfológicos tanto vegetativos como reproductivos esté bien definido, actualmente no existen problemas taxonómicos asociados a este grupo. Finalmente, tanto Z. variegata Warsz. como Z. furfuracea son especies fáciles de caracterizar e identificar. Z. variegata tiene folíolos variegados por el haz, atributo único dentro del género Zamia; en tanto, Z. furfuracea presenta tallos aéreos generalmente bifurcados, folíolos coriáceos, abovados a oblanceolados con indumento amarillento. Esta última especie crece de manera particular sobre dunas costeras en la región centro-sur del estado de Veracruz, México.

Megaméxico, diversidad y endemismo Considerando las 22 especies de Zamia conocidas actualmente para Megaméxico (Nicolalde-Morejón et al., 2009a; ver Tabla 3), México cuenta con doce especies endémicas –la mayor diversidad para la región– seguido de Honduras, con tres, Guatemala con dos, y Belice con una especie (Tabla 1). En Megaméxico encontramos además dos centros de mayor diversidad (Figura 1). El primero está ubicado en el sureste de México (sur de Veracruz, Tabasco, sureste de Oaxaca y norte de Chiapas), donde se distribuyen siete especies, todas ellas simpátricas (i. e., Z. cremnophila, Z. katzeriana, Z. lacandona, Z. loddigesii, Z. purpurea, Z. polymorpha y Z. spartea), mientras que el segundo centro se ubica entre Guatemala (Alta Verapaz e Izabal) y

19

Honduras (Atlántida, Cortés y Santa Bárbara), en el cual ocurren siete especies (i. e. Z. monticola, Z. onanreyesii, Z. oreillyi, Z. sandovallii, Z. standleyi, Z. tuerckheimii y Z. variegata). Estos sitios tienen un clima cálido húmedo donde la temperatura promedio anual es de 25°C y la precipitación son de 3 000 a 4 000 mm por año (Toledo, 1982). Ambos sitios comparten tipos similares de vegetación, particularmente Bosque Tropical Perennifolio (sensu Rzedowski, 1978). Estos datos de acumulación de biodiversidad para Zamia son contrastantes con áreas como la planicie costera del Pacífico, donde se registran únicamente tres especies (i.e. Z. herrerae, Z. paucijuga y Z. soconuscensis) a lo largo de aproximadamente 2000 km distribuidos entre México, Guatemala y El Salvador. Esta amplia zona de distribución está asociada a diferentes tipos de vegetación, como Bosque Tropical Perennifolio, Bosque Tropical Subcaducifolio, Bosque de Coníferas y de Quercus (sensu Rzedowski, 1978); algo similar pasa en la península de Yucatán (México), el Petén en Guatemala y parte de Belice, donde ocurre únicamente Z. polymorpha (Figura 1).

Intensidad de colectas botánicas En términos de las colectas botánicas realizadas hasta la fecha, los dos sitios de alta concentración de riqueza de especies están muy poco representados. Por ejemplo, para Zamia cremnophila –correspondiente a la primera región– sólo hay cuatro colectas, caso que se repite en Z. tuerckheimii (segunda región). En contraste, las especies más colectadas son Z. polymorpha (con 116 colectas), Z. paucijuga (con 114) y Z. loddigesii (con 80), todos ellas taxa de amplia distribución. A su vez, debe notarse que sólo una pequeña parte de la distribución de estas especies converge con las áreas de alta riqueza, como sucede con Z. loddigesii y Z. polymorpha (ver Figura 1). Es

20

posible que este patrón se deba a dos razones: (i) al hecho de que las especies menos colectadas tienen rangos de distribución muy restringido, y en muchos casos sólo se conocen desde la localidad tipo; y (ii) a que las especies de amplia distribución, como Z. loddigesii, Z. paucijuga y Z. polymorpha, ocurren en áreas de fácil acceso y que además históricamente han sido colectadas ampliamente. Haciendo uso de los reportes actuales de colectas botánicas para este grupo (ver Tabla 2), consideramos que son necesarias más exploraciones botánicas, principalmente a regiones como Montes Azules (en Chiapas, México), la región de Alta Verapaz e Izabal en Guatemala, la región sureste de Honduras y el noreste de Nicaragua.

Taxonomía alfa y códigos de barras moleculares en Zamia Como mencionamos arriba, Megaméxico es una región biogeográfica con altos niveles de diversidad y endemismo para Zamiaceae, superada únicamente por Australia (Hill et al., 2007; Vovides et al., 2008a, b; Nicolalde-Morejón et al., 2009b, ver Tabla 4). En esta familia de cycadas, y en particular en el género Zamia, los avances en citogenética, ecología y genética de poblaciones (para una revisión actualizada sobre el tema, ver Vovides et al., 2007) están en vías de ser complementados por el uso de caracteres moleculares, semejantes a los que ya han sido usados para avanzar enormemente en la inferencia filogenética entre los géneros de cycadas a nivel mundial (Treutlein y Wink 2002; Hill et al., 2003; Bogler y FranciscoOrtega 2004; Rai et al., 2004; Chaw et al., 2005; Zgurski et al., 2008). De hecho, matrices de DNA para análisis filogenéticos ya han comenzado a ser empleadas en las cycadas neotropicales (Ceratozamia Brongn.; González y Vovides, 2002; Dioon Lindl.; Bogler y Francisco-Ortega 2004, González et al., 2008; Zamia, Caputo et al., 2004).

21

Entre las líneas de investigación actuales que abordan el estudio de la diversidad biológica mediante el uso de datos genómicos, mediante el uso de herramientas bioinformáticas, se encuentra la aproximación conocida como ‘DNA barcoding’ o ‘códigos de barras genéticos’, formalmente inaugurada con la propuesta hecha por Hebert et al. (2003a) usando parte de la secuencia que codifica para la enzima citocromo oxidasa 1 (CO1) para identificar especies de manera automatizada, confiable y rápida (para otros ejemplos de uso de los códigos de barras en el ámbito zoológico, ver Hebert et al., 2003b; Barrett et al., 2005; Hajibabaei et al., 2006; Stoeckle y Hebert, 2008). Para nuestros propósitos, y a manera de síntesis de una intensa labor académica en un espacio de apenas cinco años, podemos afirmar que los códigos de barras moleculares se presentan como una herramienta con grandes posibilidades de complementar el trabajo taxonómico tradicional. Esto es especialmente cierto en lo que atañe a la identificación de especímenes ya asignados a un binominal latino válido –es decir, a una hipótesis taxonómica robusta basada en análisis previos de caracteres morfológicos. Además de su papel en el plano de la identificación, el DNA barcoding podría ser un auxiliar en el descubrimiento de nuevas especies, aunque este tema no está exento de controversia (cf. Tautz et al., 2003; Seberg et al., 2003; DeSalle, 2006, 2007). La aplicación de los códigos de barras de DNA en plantas, particularmente provenientes de regiones neotropicales, está atrayendo mucho interés, pues la diversidad biológica en estas regiones del planeta está compuesta en gran medida por especies vegetales. Sin embargo, dada la complejidad involucrada en la selección de las regiones que podrían funcional mejor como fuentes de los códigos de barras en plantas, la oficialización de dicha selección apenas tuvo lugar en el mes de agosto de 2009 (Hollingsworth et al., 2009). En este contexto, las especies del género Zamia con distribución en Megaméxico mantienen su interés, debido a que constituyen un grupo de

22

especies relativamente pequeño (ver Tabla 4), con una clara relevancia evolutiva para el resto de las plantas con semilla, y en virtud de que incluye especies bajo alguna categoría de protección. En principio, cabría esperar que la identificación de especies del género Zamia en Megaméxico bajo la aproximación de los códigos de barras moleculares sea exitosa; existe el antecedente de un estudio piloto de códigos de barras genéticos en las cycadas del mundo que así lo sugieren (Sass et al., 2007). Dicho trabajo sería de gran utilidad para la corroboración y/o reforzamiento de las hipótesis individuales correspondientes a las 22 especies para esta región, propuesta en el último tratamiento taxonómico (Nicolalde-Morejón et al., 2009a). Las identificaciones moleculares alcanzadas con DNA barcoding en Zamia tendrían que ser, por supuesto, consideradas de manera paralela con aquellos aspectos que la taxonomía tradicional aun no ha logrado solventar, en especial al momento de identificar taxa que tengan una amplia variación morfológica entre y dentro de especies y, que a su vez presentan una amplia distribución geográfica (i. e. Z. paucijuga, Z. loddigesii y Z. polymorpha, ver Figura 1). Resumiendo, la posibilidad de contar con atributos diagnósticos moleculares, los cuales funcionarían de manera análoga a los caracteres morfológicos necesarios para la descripción de las especies, contribuirá a establecer los límites entre especies y/o al descubrimiento de ‘especies crípticas’. En última instancia, consideramos que el acercamiento a la identificación y delimitación de especies usando DNA como una nueva fuente de datos encuentra su marco conceptual de referencia óptimo de los conceptos de ‘taxonomía integrativa’ y ‘círculo taxonómico’ propuestos por DeSalle et al. (2005). Es importante enfatizar también que un objetivo ulterior de la construcción de una biblioteca molecular de referencia hecha de secuencias de nucleótidos para los

23

loci con mayor variabilidad –los códigos de barras de DNA, en sentido estricto- para el género Zamia y los otros géneros de cycadas en Megaméxico es su aprovechamiento por usuarios externos, en contextos de biología de la conservación. Esta utilización de la base de referencia de DNA barcodes se ejemplificaría bien, por ejemplo, en situaciones de recuperación de información de ejemplares de decomiso por instancias nacionales o internacionales (como las que se indican de acuerdo con la Convención sobre el Comercio Internacional de Fauna y Flora Silvestres en Peligro de Extinción, o CITES, por sus siglas en inglés), a partir de saqueos ilegales.

Perspectivas del género Zamia en Megaméxico Considerando la historia nomenclatural y la complejidad morfológica de las especies aquí analizadas, el “complejo Zamia loddigesii” es claramente sobresaliente. Consideramos que se trata de un ensamblaje taxonómico que aún requiere de investigación detallada para aclarar su taxonomía y nomenclatura, como se detalla a continuación. Zamia loddigesii representa sin duda la especie más compleja desde el punto de vista nomenclatural, con un total de 12 nombres afines publicados y que actualmente representan sinonimias (Nicolalde-Morejón et al., 2009a). A pesar de ser una especie ampliamente colectada a lo largo del Golfo de México desde el siglo XIX, hasta el año 2008 se desconocía de la existencia de un ejemplar de herbario que pudiera fungir como tipo nomenclatural, por lo que Stevenson y Sabato (1986) habían lectotipificado el protólogo. Sin embargo, investigaciones recientes muestran un espécimen [i.e. Van Houtte 3374 (U)] que predata a la descripción de la especie, y que a su vez concuerda con los atributos morfológicos descritos por Miquel (1843). Por esta razón, dicho

24

ejemplar ha sido considerado para su designación como lectotipo de esta especie (Nicolalde-Morejón et al., 2009a, ver Tabla 3). En contraste, aunque taxonómicamente Zamia paucijuga y Z. polymorpha no han experimentado cambio alguno, son junto a Z. loddigesii las especies catalogadas como de mayor dificultad para su identificación taxonómica. Esto se debe básicamente a que estas tres especies muestran patrones de variación morfológicos muy similares, lo cual complica su identificación desde ejemplares de herbario hasta colecciones vivas. Esta situación se agudiza en ausencia de estructuras femeninas, razón por la que convencionalmente su determinación siempre ha estado asociada a procedencias geográficas. En este contexto, consideramos que este complejo de especies requiere más investigación a nivel poblacional, donde se evalué a detalle la variación morfológica y nucleotídica que pudieran presentar dichas poblaciones y con ello clarificar su identificación tanto morfológica como molecular. Es también predecible que los estudios de filogeografía con datos moleculares (Avise, 2000) abran nuevas perspectivas acerca de los patrones y procesos que, a nivel poblacional, pudieran explicar la distribución espacial de la diversidad genética y fenotípica actual no sólo de los complejos, sino de las especies en general de Zamia en Megaméxico. En realidad, dichos estudios ya están coexistiendo con la aproximación de códigos de barras, para algunos taxa animales (e. g. Linares et al. 2009). Dicha convergencia analítica podría darse también en taxa vegetales como las cycadas de México. En cualquier caso, consideramos que el tipo de investigaciones basadas directamente en información molecular permitirán seguir estudiando la variación biológica en general, entre y dentro de especies, para resolver interrogantes biogeográficas, taxonómicas, sistemáticas y de biología evolutiva en las cycadas de Megaméxico.

25

Agradecimientos Este trabajo fue parcialmente financiado por el proyecto CONACYTSEMARNAT-2002. El primer autor expresa su agradecimiento a la Red Latinoamericana de Botánica por la Beca de Doctorado RLB-06-D2, y en especial al Dr. Javier Simonetti. Asimismo, agradecemos al Dr. Dennis Wm. Stevenson por facilitarnos imágenes de varios tipos nomenclaturales, a la Dra. Victoria Sosa (INECOL, Xalapa) por sus comentarios y edición al texto. A Miguel Ángel Pérez Farrera y Carlos Iglesias por su ayuda con el trabajo de campo. Por último, reconocemos a los curadores y personal de los herbarios mencionados por proporcionarnos las colecciones botánicas para el desarrollo de este estudio.

Literatura citada Avise, J. C. 2000. Phylogeography: The History and Formation of Species. Harvard University Press. Cambridge. Estados Unidos. 464 pp. Balduzzi, A. P., P. De Luca y S. Sabato. 1982. A phytogeographical approach to the New World Cycads. Delpinoa, n.s., 23-24. Barrett, R. D. H. y P. D. N. Hebert. 2005. Identifying spiders through DNA barcodes. Canadian Journal of Zoology 83: 481-491. Bogler, D. J. y J. Francisco-Ortega. 2004. Molecular systematic studies in cycads: evidence from trnL intron and ITS2 rDNA sequences. Botanical Review 70: 260– 273. Caputo, P., S. Cozzolino, P. De Luca, A. Moretti y D. W. Stevenson. 2004. Molecular phylogeny of Zamia (Zamiaceae). En Walters, T. y R. Osborne (eds.), Cycad

26

Classification: Concepts and Recommendations. Pp. 149-157. CABI Publishing, Wallingford, Inglaterra. Chaw, S. M., T. W. Walters, C. C. Chang, S. H. Hu y S. H. Chen. 2005. A phylogeny of cycadas (Cycadales) inferred from chloroplast matK gene, trnK intron, and nuclear rDNA ITS region. Molecular Phylogenetics and Evolution 37: 214-234. DeSalle, R., M. G. Egan y M. Siddall. 2005. The unholy trinity: taxonomy, species delimitation and DNA barcoding. Philosophical Transactions of the Royal Society B 360: 1905-1916. DeSalle, R. 2006. Species discovery versus species identification in DNA barcoding efforts: response to Rubinoff. Conservation Biology 20: 1545-1547. DeSalle, R. 2007. Phenetic and DNA taxonomy; a comment on Waugh. Bioessays 29: 1289-1290. González, D. y A. P. Vovides. 2002. Low intralineage divergence in Ceratozamia (Zamiaceae) detected with nuclear ribosomal DNA ITS and chloroplast DNA trnL-F non coding region. Systematic Botany 27: 654-661. González, D., A. P. Vovides y C. Bárcenas. 2008. Phylogenetic relationships of the neotropical genus Dioon (Cycadales, Zamiaceae) based on nuclear and chloroplast DNA sequence data. Systematic Botany 33: 229-236. González-Astorga, J., A. P. Vovides, P. Octavio-Aguilar, D. Aguirre-Fey, F. NicolaldeMorejón y C. Iglesias. 2006. Genetic diversity and structure of the cycad Zamia loddigesii Miq. (Zamiaceae): implications for evolution and conservation. Botanical Journal of the Linnean Society 152: 533-544. Hajibabaei, M., D. H. Janzen, J. M. Burns, W. Hallwachs y P. D. N. Hebert. 2006. DNA barcodes distinguish species of tropical Lepidoptera. Proceedings of the National Academy of Sciences USA 103: 968-971.

27

Hebert, P. D. N., A. Cywinska, S. L. Ball y J. R. deWaard. 2003a. Biological identifications through DNA barcodes. Proceedings of the Royal Society of London B 270: 313-321. Hebert, P. D. N., S. Ratnasingham y J. R. deWaard. 2003b. Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society of London B (Supplement) 270: S96-S99. Hill K. D., M. W. Chase, D. W. Stevenson, H. G. Hills y B. Schutzman. 2003. The families and genera of cycads: a molecular phylogenetic analysis of Cycadophyta based on nuclear and plastid DNA sequences. International Journal of Plant Sciences 164: 933-948. Hill, K. D., D. W. Stevenson y R. Osborne. 2007. The world list of cycads/La lista mundial de Cícadas. Memoirs of the New York Botanical Garden 97: 454-483. Holdridge, L. R. 1987. Ecología basada en zonas de vida. Instituto Interamericano de Cooperación para la Agricultura. San José, Costa Rica. 216 p. CBOL Plant Working Plant. Hollingsworth, P. M., L. L. Forrest, J. L. Spouge, M. Hajibabaei, S. Ratnasingham, M. van der Bank, M. W. Chase, R. S. Cowan, D. L. Erickson, A. J. Fazekas, S. W. Graham, K. E. James, Ki-Joong Kim, W. J. Kress, H. Schneider, J. van AlphenStahl, S. C. H. Barrett, C. van den Berg, D. Bogarin, K. S. Burgess, K. M. Cameron, M. Carine, J. Chacón, A. Clark, J. J. Clarkson, F. Conrad, D. S. Devey, C. S. Ford, T. A. J. Hedderson, M. L. Hollingsworth, B. C. Husband, L. J. Kelly, P. R. Kesanakurti, J. S. Kim, Y. D. Kim, R. Lahaye, H. L. Lee, D. G. Long, S. Madriñán, O. Maurin, I. Meusnier, S. G. Newmaster, ChongWook Park, D. M. Percy, G. Petersen, J. E. Richardson, G. A. Salazar, V. Savolainen, O. Seberg, M. J. Wilkinson, D. K. Yi, y D. P. Little. 2009. A DNA

28

barcode for land plants. Proceedings of the National Academy of Sciences USA 106: 12794-12797. Limón, F. L. 2009. Genética de Poblaciones de Zamia furfuracea L. f. (Zamiaceae); una cícada endémica al estado de Veracruz, México. Tesis de licenciatura, Universidad Veracruzana, México. Linares, M. C., I. D. Soto-Calderón, D. C. Lees y N. M. Anthony .2009. High mitochondrial diversity in geographically widespread butterflies of Madagascar: A test of the DNA barcoding approach. Molecular Phylogenetics and Evolution 50: 485-495. Marchant, C. J. 1968. Chromosome patterns and nuclear phenomena in the cycad families Stangeriaceae and Zamiaceae. Chromosoma 24: 100-134. Miller, S. E. 2007. DNA barcoding and the renaissance of taxonomy. Proceedings of the National Academy of Sciences USA 104: 4775-4776. Miquel, F. A. W. 1842. Monographia Cycadearum. Trajecti ad Rhenum/Utrecht pp. 3248, t. 4-5. ______. 1851. Cycadeae quaedam americanae, Partim Novae. Nederl. Inst. Verh. Erste Kl. (Series 3) 4: 181-189. Amsterdam. ______. 1861. Prodromus systematics cycadearum. C. v. d. Post Jr. Utrecht, The Netherlands. Moretti, A. y S. Sabato. 1984. Karyotype evolution by centromeric fission in Zamia (Cycadales). Plant Systematics and Evolution 146: 215-223. ______. 1990. Karyotypic data on North and Central American Zamiaceae (Cycadales) and their phylogenetic implications. American Journal of Botany 77: 1016-1029.

29

______, P. Caputo, L. Gaudio, & D.W. Stevenson. 1991. Intraspecific chromosome variation in Zamia (Zamiaceae, Cycadales). Caryologia 44: 1-10. Nicolalde-Morejón, F. 2007. Taxonomía, Distribución y Estado de Conservación de Zamia en Ecuador. Memoirs of The New York Botanical Garden 97: 45-63. Nicolalde-Morejón, F., A. P. Vovides, D. W. Stevenson y V. Sosa. 2008. The identity of Zamia katzeriana and Zamia verschaffeltii (Zamiaceae). Brittonia 60: 38-48. ______, A. P. Vovides, D. W. Stevenson. 2009a. Taxonomic revision of Zamia in Mega-Mexico. Brittonia. DOI 10.1007/s12228-009-9077-9. ______, F. Vergara-Silva, J. González-Astorga, A. P. Vovides y A. Espinosa de los Monteros. 2009b. Reciprocal illumination of morphological characters upon a molecular hypothesis supports the proposal of a new species of cycad from Mexico. Systematics and Biodiversity 7: 73-79. Norstog, K. y T. J. Nicholls. 1997. The Biology of the Cycads. Cornell University Press, Ithaca, Nueva York. Rai, H. S., H. E. O'Brien, P. A. Reeves, R. G. Olmstead y S. W. Graham. 2003. Inference of higher-order relationships in the cycads from a large chloroplast data set. Molecular Phylogenetics and Evolution 29: 350-359. Rzedowski, J. 1978. Vegetación de México. Limusa, S. A. México, D.F. ———, 1991. El endemismo en la flora fanerogámica mexicana: una apreciación analítica preliminar. Acta Botanica Mexicana 15: 47-64. Sabato, S. 1990. West Indian and South American cycads. Memoirs of the New York Botanical Garden 57: 173-185. Schuster, J. 1932. Cycadaceae. Das Pflanzenreich 99:1-168. Wilhelm Engelmann, Leipzig.

30

Seberg, O., J. Ch. Humphries, S. Knapp, D. W. Stevenson, G. Petersen, N. Scharff y N. M. Andersen. 2003. Shortcuts in systematics? A commentary on DNA-based taxonomy. 2003. Trends in Ecology and Evolution 18(2): 63-65. Stevenson D. W. 1987. The West Indian Zamias. Fairchild Tropical Garden Bulletin.42: 23-27. ———. 1991a. Flora of the Guianas. Serie. A. Fascicule 9: 7-11. ———. 1991b. The Zamiaceae in the Southeastern United States. Journal of the Arnold Arboretum, Supplemental Series 1: 367-383. ———. 1992. A formal classification of the extant cycads. Brittonia 44: 220-223 ———. 1993. The Zamiaceae in Panama with comments on phytogeography and species relationships. Brittonia 45: 1-16. ———. 2001a. Orden Cycadales. Flora de Colombia. Monografía No. 21. Editorial Unibiblos. Bogotá D.C. ———. 2001b. Zamiaceae. Flora Nicaragua 1: 6-7. ———. 2004. Zamiaceae of Bolivia, Ecuador, and Perú. En Walters, T. y R. Osborne (eds.), Cycad Classification: Concepts and Recommendations. Pp. 173-194. CABI Publishing, Wallingford, Inglaterra. ——— y S. Sabato. 1986. Typification of names in Zamia L. and Aulacophyllum Regel. (Zamiaceae). Taxon 35: 134-144. Stoeckle, M. Y. y P. D. N. Hebert. 2008. Barcode of life. Scientific American 299: 8288. Taylor, A. S., J. L. Haynes y G. Holzman. 2009. Taxonomical, nomenclatural and biogeographical relevations in the Zamia skinneri complex of Central America (Cycadales: Zamiaceae). Botanical Journal of the Linnean Society 158: 399-429.

31

Toledo, V. M. 1982. Pleistoceno changes of vegetation in tropical Mexico. Pp. 93-111. En: Prance, G. T. (ed). Biological Diversification in the Tropics. Columbia University Press, New York. Treutlein, J. y M. Wink. 2002. Molecular phylogeny of cycads inferred from rbcL sequences. Naturwissenschaften 89:221-225. Tautz, D., P. Arctander, A. Minelli, R. H. Thomas y A. P. Vogler. 2003. A plea for DNA taxonomy. Trends in Ecology and Evolution 18: 70-74. Vovides, A. P. 1983. Systematics studies on the Mexican Zamiaceae I. Chromosome numbers and karyotypes. American Journal of Botany 70: 1002-1006. ______ y M. Olivares. 1996. Karyotype polymorphism in the cycad Zamia loddigesii (Zamiaceae) of the Yucatan peninsula, Mexico. Botanical Journal of the Linnean Society 120: 77-83. ______, J. González-Astorga, M. A. Pérez-Farrera, D. González, C. Bárcenas y C. Iglesias. 2007. The Cycads of Mexico: 25 years of research and conservation. Memoirs of the New York Botanical Garden 97: 611-641. ______, S. Avendaño, M. A. Pérez-Farrera y J. González-Astorga. 2008a. A new species of Ceratozamia (Cycadales, Zamiaceae) from Veracruz, Mexico. Novon 18: 109-114. ______, M. A. Pérez-Farrera, J. González-Astorga y C. Iglesias. 2008b. A new species of Ceratozamia (Zamiaceae) from Oaxaca, Mexico with comments on habitat and relationships. Botanical Journal of the Linnean Society 157: 169-175. Zgurski, J. M., H. S. Rai, Q. M. Fai, D. J. Bogler, J. Francisco-Ortega J y S. W. Graham. 2008. How well do we understand the overall backbone of cycad phylogeny? New insights from a large, multigene plastid data set. Molecular Phylogenetics and Evolution 47: 1232-1237.

32

Figura 1. Distribución de: A) Zamia cremnophila, Z. katzeriana, Z. lacandona, Z. purpurea, Z. spartea; B) Z. monticola, Z. onareyesii, Z. oreillyi, Z. sandovallii, Z. standleyi, Z. tuerckheimii, Z. variegata; C) Z. paucijuga; D) Z. loddigesii; E) Z. polymorpha; F) Z. herrerae; G) Z. soconuscensis; H) Área inexplorada para Zamia.

33

Tabla 1. Especies, distribución y tipos de vegetación de Zamia en Megaméxico No Especie País Estado/Departamento Tipo de Vegetación 1 Z. cremnophila México Tabasco BTP 2 Z. fischeri México Hidalgo, Querétaro, San Luís BTsC, Potosí BMM, BQBC 3 Z. furfuracea México Veracruz BTP 4 Z. herrerae El Salvador, ES: Sonsonate, M: Chiapas BTP, BTsC Guatemala, México 5 Z. inermis México Veracruz BTC 6 Z. katzeriana México Chiapas, Tabasco, Veracruz BTP 7 Z. lacandona México Chiapas BTP 8 Z. loddigesii México Chiapas, Hidalgo, Oaxaca, BTP Puebla, Tabasco, Tamaulipas, Veracruz 9 Z. monticola Guatemala Alta Verapaz BTP 10 Z. onanreyesii Honduras Cortés BTP 11 Z. oreillyi Honduras Atlántida BTP 12 Z. paucijuga México Colima, Guerrero, Jalisco, BTsC, Michoacán, Nayarit, Oaxaca BTC, BQ 13 Z. polymorpha Belice, B: Distrito Belice, Distrito Cayo; BTP, BTsC Guatemala, G: Peten; M: Campeche, México Chiapas, Quintana Roo, Tabasco, Yucatán 14 Z. prasina Belice Distrito Toledo BTP 15 Z. purpurea México Oaxaca, Veracruz BTP 16 Z. sandovallii Honduras Atlántida BTP 17 Z. soconuscensis México Chiapas BTsC 18 Z. spartea México Oaxaca BTP, BTC 19 Z. standleyi Guatemala, G: Izabal; H: Atlántida, Santa BTP Honduras Bárbara 20 Z. tuerckheimii Guatemala Alta Verapaz BTP 21 Z. variegata Guatemala, G: Alta Verapaz, Izabal; M: BTP México Chiapas 22 Z. vazquezii México Veracruz BTP Simbología aquí utilizada para los tipos de vegetación, según Rzedowski (1978). BTP Bosque Tropical Perennifolio; BTsC Bosque Tropical Subcaducifolio; BMM Bosque Mesófilo de Montaña, BQ Bosque de Quercus; BC Bosque de Coníferas.

34

Tabla 2. Especies, distribución y cantidad de colectas de Zamia en Megaméxico No Especie País Estado/Departamento No. de colectas botánicas 1 Z. cremnophila México Tabasco 4 2 Z. fischeri México Hidalgo, Querétaro, San Luís 20 Potosí 3 Z. furfuracea México Veracruz 36 4 Z. herrerae El Salvador, ES: Sonsonate, M: Chiapas 14 Guatemala, México 5 Z. inermis México Veracruz 12 6 Z. katzeriana México Chiapas, Tabasco, Veracruz 25 7 Z. lacandona México Chiapas 22 8 Z. loddigesii México Chiapas, Hidalgo, Oaxaca, 80 Puebla, Tabasco, Tamaulipas, Veracruz 9 Z. monticola Guatemala Alta Verapaz 2 10 Z. onanreyesii Honduras Cortés 3 11 Z. oreillyi Honduras Atlántida 2 12 Z. paucijuga México Colima, Guerrero, Jalisco, 114 Michoacán, Nayarit, Oaxaca 13 Z. polymorpha Belice, B: Distrito Belice, Distrito Cayo; 116 Guatemala, G: Peten; M: Campeche, México Chiapas, Quintana Roo, Tabasco, Yucatán 14 Z. prasina Belice Distrito Toledo 3 15 Z. purpurea México Oaxaca, Veracruz 15 16 Z. sandovallii Honduras Atlántida 4 17 Z. soconuscensis México Chiapas 7 18 Z. spartea México Oaxaca 24 19 Z. standleyi Guatemala, G: Izabal; H: Atlántida, Santa 8 Honduras Bárbara 20 Z. tuerckheimii Guatemala Alta Verapaz 5 21 Z. variegata Guatemala, G: Alta Verapaz, Izabal; M: 16 México Chiapas 22 Z. vazquezii México Veracruz 4

35

Tabla 3. Especies, publicación y tipos nomenclaturales de Zamia en Megaméxico Especie Año Publicación Z. cremnophila Vovides, Schutzman & 1988 Bot. Gaz. 149: 351 Dehgan Z. fischeri Miq. 1845 in Lem., Hort. Vanhoutt. 1: 20 Z. furfuracea L. f. 1789 in Aiton, Hortus Kew. 3: 477 Z. herrerae Calderón & Standl.

1924

Z. inermis Vovides, J.D. Rees & Vázq. Torres Z. katzeriana (Regel) Retting Z. lacandona Schutzman & Vovides Z. loddigesii Miq.

1983 1896 1998 1843

Z. monticola Chamb. Z. onanreyesii C. Nelson & G. Sandoval Z. oreillyi C. Nelson Z. paucijuga Wieland

1926 2008 2005 1916

Z. polymorpha D.W. Stev., A. Moretti & Vázq. Torres Z. prasina W. Bull Z. purpurea Vovides, J.D. Rees & Vázq. Torres Z. sandovallii C. Nelson Z. soconuscensis Schutzman, Vovides & Dehgan Z. spartea A. DC. Z. standleyi Schutzman

199596 1881 1983 2005 1988 1868 1989

Tipo H-MEXU

N-U L-pl. 210 in Herm. Parad. Bat., 1698 Proc. Wash. Acad. Sci. 14(4): 93, H-US fig. 1 Flora de Veracruz 26: 22-24, fig. H-XAL 3 Acta Horti Petrop. 4(4):298 L-LE Novon 8(4): 441-446, figs. 1-3 H-FLAS Tijdschr. Natuurl. Gesch. L-U Physiol. 10: 72-73 Bot. Gaz. 81: 219-223, figs. 1-3 H-MO Ceiba 49(1): 135-136, figs. 1-6 H-TEFH Ceiba 46(1-2): 56, figs. 1-3 H-TEFH American Fossil Cycads 2: 212, LE—fig. 86 fig. 86 Delpinoa n.s. 37-38: 3-8 (issued H-NY 1998), fig. 1 Retail List: 20 H-K Flora de Veracruz 26: 28-31, fig. H—XAL 5 Ceiba 46(1-2): 55, figs. 1-10 H-TEFH Bot. Gaz. 149(3): 347-351, figs. H—F 1-3 Prodr. 16(2): 539 H—G-DC Syst. Bot. 14(2): 214-219, figs. 1- H—FLAS

36

Isotipos CSAT, FCM, MO, UAMIZ

F XAL NY TEFH

BRH, FTG, MO, NY, U

TEFH CR, MEXU, MICH ENA, FTG

Z. tuerckheimii Donn. Sm.

1903

Z. variegata Warsz. Z. vazquezii D.W. Stev., Sabato, A. Moretti & De Luca

1845 199596

2 Bot. Gaz. (Crawfordsville) 35(1): H—US 8, pl. 1 Allg. Gartenzeitung 32: 252-253 N—NY Delpinoa n.s. 37-38: 9-17 (issued N—NY 1998), fig. 3

37

K U, XAL CIB, FTG, MO, NY, U

Tabla 4. Diversidad y endemismos de Cycadas en Megaméxico Géneros de Especies descritas Especies endémicas a Cycadas Megaméxico 2 Ceratozamia

25

25

Dioon

14

14

Zamia

61

22

Total

100

61

38

CAPITULO II Taxonomic revision of Zamia in Mega-Mexico (Brittonia: Aceptado DOI 10.1007/s12228-009-9077-9)

39

Taxonomic revision of Zamia in Mega-Mexico

Fernando Nicolalde-Morejón1, Andrew P. Vovides1 and Dennis W. Stevenson2*

1

Departamento de Biología Evolutiva. Instituto de Ecología, A. C. km 2.5 Antigua

Carretera a Coatepec No. 351, Xalapa 91070, Veracruz, Mexico. 2

The New York Botanical Garden, Bronx, New York, 10458-5120, USA.

*Corresponding author: [email protected]

40

ABSTRACT. The genus Zamia is revised for Mega-Mexico, with 22 species recognized and described and one species dubium. The study presents a taxonomic clarification for the genus in Mesoamerica, a contribution that provides the foundation for a future monograph for Zamia in the Neotropics. The largest proportion of species richness and endemism for the genus is concentrated in southeastern Mexico, among the states of Chiapas, Oaxaca, Tabasco and Veracruz, an area that is considered highly diverse in floristic terms. Distribution maps and a key to species are also provided, as well as complete descriptions of the specimens examined, including information on nomenclatural types, habitats, synonymies, and etymologies. Zamia spartea is illustrated for the first time and chromosome numbers for Z. herrerae are reported and illustrated. Finally, scanning electron micrographs of leaflet trichome character states are presented, along with a discussion of their systematic implications within the group. KEYWORDS: Endemism, Mexico, gymnosperms, cycads, floristic richness, Zamia.

RESUMEN. El género Zamia es revisado para Mega-México, con 22 especies descritas y una species dubia. El estudio está orientado al esclarecimiento taxonómico del género en Mega-México, una contribución que siente los fundamentos para una futura monografía para Zamia en el Neotrópico. La mayor riqueza y endemismo para el género se concentra en el sureste de México, entre los estados de Chiapas, Oaxaca, Tabasco y Veracruz, área de alta biodiversidad florística. Mapas de distribución y una clave para las especies son presentadas, como también descripciones completas, tipos, hábitat, sinónimos, etimología y especimenes examinadas. Z. spartea es ilustrada por primera vez y números cromosómicos para Z. herrerae son presentados. Finalmente, se

41

presentan fotografías de tricomas al microscopio electrónico y una discusión de sus implicaciones en la sistemática del grupo.

42

Introduction According to Stevenson (1992), Zamiaceae comprises eight genera distributed in tropical and subtropical Africa, Australia, Greater Antilles, North, Central and South America. Five genera with 94 species are known from the Neotropics (Hill et al., 2007). The genera Ceratozamia Brongn. (21 spp.) and Dioon Lindl. (13 spp.) are both endemic to Mexico and a neighboring biogeographic region of Central America that is floristically similar to the southern part of Mexico (Mega-Mexico) while Microcycas A. DC. (1 sp) and Chigua D.W. Stev. (2 spp.) are endemic to Cuba and Colombia, respectively. The type genus for the family, Zamia L. (59 spp.), is distributed throughout tropical and sub-tropical America as well as the Caribbean, with exception to the Lesser Antilles. Zamia is the widest distributed genus of the order Cycadales in the Neotropics. Its northern range starts in Georgia and Florida (United States of America), reaching Bolivia and the Mato Grosso of Brazil in South America (Balduzzi et al., 1982; Sabato, 1990; Norstog & Nicholls, 1997; Stevenson, 2001a, b). A remarkable morphological and cytological variation has been documented (Vovides, 1983; Moretti & Sabato, 1984; Moretti, 1990a, b; Stevenson et al., 1995-96a; Vovides & Olivares, 1996; Norstog & Nicholls, 1997), and also high levels of genetic variation (González-Astorga et al., 2006). As a consequence of this complexity, Zamia´s taxonomy is controversial; although the genus comprises 75 species, their circumscription and limits have yet to be determined (see Hill et al., 2007). The most recent exhaustive taxonomic treatment for Zamia was published by Schuster (1932) and later work by Sabato (1990) and Stevenson (1987; 1991a, b; 1993; 2001a, b; 2004) underlined nomenclatural and taxonomic anomalies in Zamia, principally owing to insufficient fieldwork and a scarcity of good quality botanical

43

collections. The taxonomic history for the genus in Mexico began with the publication of Zamia fufuracea L. f. in 1789 from the central-south coastal region of Veracruz, which also represents the first cycad species described from the American continent. During the 19th century, six species were subsequently described for Mexico; Z. fischeri Miq., Z. katzeriana (Regel) Rettig, Z. lawsoniana Dyer, Z. loddigesii Miq., Z. spartea A. DC. and Z. verschaffeltii Miq., whereas the remaining known species were documented and characterized during the 20th century, with a marked tendency for taxonomic activity during the last thirty years (see: Vovides et al., 1983; Schutzman et al., 1988; Stevenson et al., 1995-96a,b; Schutzman et al., 1998; Vovides, 1999). The present taxonomic revision includes endemic species of the cycad genus Zamia that occur in ‘Mega-Mexico 2’, a term coined by Rzedowski (1991) that associates the Central American territories of Guatemala, Belize, Honduras and northern Nicaragua to to the Mexican states of Nayarit on the Pacific to southern Tamaulipas on the Gulf of Mexico. using biotic (mainly floristic) criteria. Rzedowski’s concept will be referred to as simply ‘Mega-Mexico’ hereafter, given that the slight differences in the boundaries of ‘Mega-Mexico 1’ with respect to ‘Mega-Mexico 2’ do not affect the biogeographic aspects of our description of the species. This study is intened to provide a taxonomic clarification of the genus in Mesoamerica and to provide the basis for a future monograph of Zamia.

Materials and Methods The present taxonomic revision is based on more than 450 specimens from the following herbaria: B, BM, CIB, CHIP, CICY, ECOSUR, ENCB, F, FCME, FLAS, FTG, HEM, IBUG, IEB, K, LE, MEXU, MO, NY, SERO, U, UADY, UAMIZ, US, W,

44

WIS, XAL, XALU, ZEA. Unfortunately, we were not able to obtain vouchers on loan from NAP; therefore, material from this herbarium is not cited. Chromosome counts for Zamia herrerae Calderón & Standley were performed on five individuals held at the Jardín Botánico Fco. Javier Clavijero, Instituto de Ecología, A.C (JBC). These plants has been previously collected at the Acacoyagua and Tonalá regions, located in the state of Chiapas and represent the species range in southern Mexico. Plants from its full biogeographic range, which would include El Salvador and Guatemala, were not available for this study. A modified root-tip squash method was used for examining somatic metaphase cells described by Vovides (1983) with a 12 to 15 hour ice-water (0°C) follow-up soak after the 0.2% colchicine pretreatment at ambient temperature (Schutzman et al., 1988). Counts were made from the best 10-15 metaphase cells and karyotype noted according to the classification of Schlarbaum and Tsuchiya (1984). Photomicrographs were produced using a Zeiss Fomi III photomicroscope fitted with planapochromatic objectives and Kodak Plus-X pan ASA 125 film. Scanning electron micrographs (SEM) were taken on young leaflet material from living plants cultivated in the JBC. Samples were placed on sample stubs with double sided adhesive tape and then introduced into a dessicator for 24 hours. All samples were sputter coated with gold-palladium at 1.5 kv at 5 mA for 8 minutes with a Jeol Fine Coat JFC 1100 sputter coater. Observations were made with a Jeol JSM-5600LV SEM. In all cases, the types have been examined by one or more of the authors.

Results HABITAT

45

Of the 22 species included in this revision, 18 occur in specific habitats, accounting for the restricted distribution of the majority of the taxa. The species with the widest distribution are associated with two or more vegetation types, namely (a) Zamia paucijuga Wieland, found in pine-oak, oak and tropical dry forests; (b) Z. polymorpha D.W. Stev., A. Moretti & Vázq. Torres, located in evergreen tropical rainforest, subdeciduous tropical forest and their secondary succession stages; (c) Z. herrerae, which occurs in evergreen tropical rainforest, sub-deciduous tropical forest and tropical dry forest and their secondary succession stages and finally, to a lesser extent, (d) Z. loddigesii, present in evergreen tropical rainforest, but more commonly in subdeciduous tropical forest and its secondary succession stages.

MORPHOLOGY HABIT. — All adult cycad stems are pachycaulous and may be columnar and arborescent or subterranean and tuber-like. The genera Dioon, Microcycas, Ceratozamia and Lepidozamia Regel are usually columnar arborescent in habit, while the subterranean forms are characteristic of Bowenia Hook. ex Hook. f., Chigua and Stangeria T. Moore. Cycas L., Encephalartos Lehm., Macrozamia Miq. and Zamia have both stem morphologies, either subterranean tuber-like or columnar arborescent (Stevenson, 1980). In this context, the species of Zamia in Mesoamerica represent both growth forms, with the subterranean tuber-like habitat predominant. Only four species, Z. inermis, Z. onanreyesii, Z. soconuscensis and Z. tuerckheimii, have arborescent stems reaching up to 100 cm or more in height. Some species branch dichotomously with age (namely, Z. fischeri Miq., Z. furfuracea, Z. inermis Vovides, Rees & Vázq. Torres, Z. soconuscensis Schutzman, Vovides & Dehgan, Z. loddigesii and Z. herrerae), with the

46

coastal dune species Z. furfuracea being the most notable, with branches reaching up to 80 cm long in adult plants. TRICHOMES.— Trichomes of cycads leaves are bi-celled, consisting of a small basal cell and a longer free apical portion (Stevenson, 1981). All the trichomes analyzed here show the same bifurcate pattern, with one arm proportionally longer than the other (Fig 1a). Each trichome presents a rounded basal cell and a more extensive cylindrical bifurcate free portion, which in most cases observed had evidence of collapse and twisting (Fig. 1b-f). The pubescence was significantly denser in emerging than in older adult leaves with the latter becoming completely glabrous. An exception to this condition is found in Zamia furfuracea, which maintains a great part of its original indument on the abaxial surface of each leaflet. According to Stevenson (1981), four types of trichomes occur in Zamia; of these, the transparent ramified and the colored ramified are the types found in the present work. In contrast to what was found in the aforementioned study by Stevenson, no trend or correlation was found between aerial stems (i.e. Z. soconuscensis or Z. inermis) and bifurcate trichomes with equal length branches. Trichomes with unequal sized prevail among taxa with both aerial and subterranean stems. In this context, we consider that a more extensive and detailed sampling of the genus Zamia across the Neotropics would be necessary to corroborate any correlation between trichome morphology and stem habit.

REPRODUCTIVE STRUCTURES — Although the characters employed for the identification of species of Zamia have been obtained from leaf morphology (see Miquel, 1861, 1869; Regel, 1857, 1876; De Candolle, 1868; Schuster, 1932; Eckenwalder 1980; Vovides et al., 1983; Newell,1986; Schutzman & Vovides, 1998, Schutzman et al., 1988;

47

Stevenson, 1993, 2001a, b, 2004, Nicolalde-Morejón et al., 2008), the evaluation of reproductive characters, especially those corresponding to ovulate strobili, is essential to discriminate among closely related taxa. Outstanding attributes that should be considered in this case are (i) the form and shape of the cone apex (Stevenson, 1987) ; (ii) the peduncle position with respect to the vertical axis of the cone, when mature (Schutzman et al., 1988); and (iii) the overall color of the cone (see descriptions and Figs. 2 & 3). In contrast to the ovulate strobili, pollen strobili show scarce variation at the species level, and their utility to discriminate among species that might possess high degrees of genealogical affinity is relatively low. For the description of the pollen reproductive axes, the terminology introduced by Mundry and Stützel (2003) has been followed.

CHROMOSOME NUMBERS —Zamia shows the highest chromosome numbers and karyotype variation throughout the Order Cycadales, with 2n counts of 16, 17, 18, 22, 23, 24, 25, 26, 27 and 28 (see Marchant, 1968; Norstog, 1980, 1981; Vovides, 1983; Moretti & Sabato, 1984; Moretti, 1990a, b; Vovides & Olivares, 1996). In contrast, all Ceratozamia species studied so far have stable diploid chromosome numbers (2n = 16) and karyotypes, as do all the Dioon species analyzed to date with stable diploid chromosome numbers (2n = 18) and karyotypes (Marchant, 1968; Vovides, 1983, 1985; Moretti, 1990a, b). Chromosome counts are presented and illustrated (Fig. 4) here for the first time for Zamia herrerae. This species has, but to a lesser extent, cytotype polymorphisms similar to those found for Z. paucijuga (Moretti and Sabato, 1984) and Z. polymorpha (Vovides and Olivares, 1996; Stevenson et al., 1995-96b). Zamia herrerae has 2n = 23, 24 from two populations along its Mexican range in Chiapas (Fig. 4). Both the m

48

(median region of the chromosomes) and T (telocentric) chromosomes vary in number (6-11 T, 4-6 m) and are large (6-11 µm for T and 8-12 µm for m) and their arms can be longer than half the spindle axis, which can cause mitotic instability during telophase (Schubert, 2007). Karyotype differences are probably due to centric fissions occurring on some of the larger m chromosomes, giving rise to telocentrics with part of the centromere still present. In this context, there are two general hypotheses to explain karyotypic evolution in Zamia: first, Norstog’s hypothesis (Norstog 1980, 1981) relating karyotype simplification and symmetry with progressive fusion of telocentric chromosomes, which predicts low diploid number in taxa with a high number of metacentric chromosomes, and secondly that of more recent research on Zamia, which postulates centric fission rather than fusion producing a progressively higher diploid number and asymmetric karyotypes with a high number of telocentric chromosomes (Moretti and Sabato, 1984; Vovides & Olivares, 1996; Caputo et al., 2004). For a more in-depth discussion of mechanisms of chromosome evolution in seed plants, see Jones (1998). The highly asymmetric karyotypes and somatic chromosome numbers in both Z. paucijuga and Z. polymorpha, which also appear to be the pattern for Z. herrerae, seem to be correlated with the highest morphological variation and widest geographic distribution of the genus in Mesoamerica. Zamia herrerae has a range of about 1,000 km, spanning El Salvador, Guatemala and Chiapas (Mexico); therefore, we suggest investigating the distribution of chromosome character states in this species throughout its range. This karyotype asymmetry contrasts with their congeners of a more restricted distribution, which often have less morphological variation and a tendency towards constant chromosome number and karyotype, e. g. Z. cremnophila Vovides, Schutzman & Dehgan, Z. fischeri, Z. inermis, Z. katzeriana, Z. purpurea and Z. soconuscensis (all

49

2n = 16), as well as Z. furfuracea, Z. spartea and Z. vazquezii (all 2n = 18). Vovides and Olivares (1996) and Jones (1998) comment that atypical chromosome number increase attributed to fission is probably a result of stressful influences.

DISTRIBUTION AND ENDEMISM Seventy-five percent of the species in this revision are endemic to the type locality and nearby areas. They are limited to two or three close populations with low population densities. These attributes are consistent with Rabinowitz’s (1981) evaluation criteria for species rarity, which mainly considers information related to geographic range, habitat specificity and local population size. In congruence with these criteria, the endemic Zamia species of Mesoamerica are considered rare, threatened or endangered and are listed under the IUCN Red List (2005, Hill et al., 2007). With 21 endemic cycad species in three genera, Mexico has the highest cycad diversity and number of endemics of the region. Six species of Zamia are known in Guatemala, two of which are endemic (Z. monticola Chamberlain and Z. tuerckheimii J. Donnell Smith); in Honduras, three species are known of which three recently described species are endemic (Z. oreillyi C. Nelson , Z. sandovalii C. Nelson, and Z. onanreyesii C. Nelson & G. Sandoval). Two further species are known for Belize, of which one is endemic (Z. prasina W. Bull) and finally, El Salvador is represented by one broadly distributed species, Z. herrerae, with a range that runs along the Pacific seaboard through Guatemala and the Sierra Madre de Chiapas in Mexico. Other species of the genus in Mexico with broad distributions are (a) Z. paucijuga, distributed along the Pacific seaboard of Mexico ranging from Nayarit (northwest Mexico) to Oaxaca (southwest); (b) Z. polymorpha, distributed widely throughout the Yucatán penninsula in Mexico, Belize and the Petén region of Guatemala; (c) Z. loddigesii, ranging along

50

the Gulf of Mexico seaboard from Tamaulipas (northeast Mexico) to Tabasco (southeast); and finally (d) Z. variegata Warsz., distributed between Guatemala and southern Mexico. The southern and southeastern regions of Mexico, comprising the states of Veracruz, Oaxaca, Chiapas and Tabasco, are the most diverse area of Mexico for the genus Zamia, with seven endemic species (Z. cremnophila, Z. lacandona, Z. loddigesii, Z. katzeriana, Z. purpurea, Z. polymorpha and Z. spartea). The Gulf of Mexico region has three micro-endemics Z. furfuracea, Z. inermis and Z. vazquezii, whereas two micro-endemic species, Z. fischeri and Z. soconuscensis, are known respectively from the Sierra Madre Oriental and the Sierra Madre de Chiapas. Finally, in spite of the status of Z. paucijuga as a Mexican endemic, its distribution is extremely wide within Mega-Mexico, covering a range of about 1,000 km between the states of Oaxaca and Nayarit. The northern limit of Z. paucijuga in the latter state represents the northernmost distribution for the genus along the Pacific seaboard of the Neotropics.

Taxonomic Treatment Zamia L., Sp. Pl. ed. 2: 1659. 1763. nom. cons. Type species: Zamia pumila L. Palma-Felix Adanson, Fam. Pl. 2: 21, 587. 1763 Aulacophyllum Regel, Gartenflora 25: 140. 1876 Stems hypogeous and epigeous, erect to decumbent, sometimes dichotomously branched in mature plants. Cataphylls chartaceous to membranaceous, stipulate, persistent or deciduous, base triangular, apex long acuminate to aristate, tomentose, generally reddish-brown to yellowish. Ptyxis erect to inflexed. Leaves stipulate,

51

ascending to descending to spreading, reddish-brown or green when emerging; petiole sometimes blackish in young leaves, terete or subterete, without prickles or heavily to lightly armed with straight or bifurcate prickles; rachis subterete generally with few prickles along the proximal third or without prickles, with up to 60 pairs of leaflets. Leaflets articulate, sessile, papyraceous to coriaceous, linear, linear-lanceolate, lanceolate, ovate, obovate obpyriform to elliptic, opposite to subopposite, falcate or non-falcate, imbricate to non-imbricate, generally acute at apex and symmetric, attenuate at base, margins entire to dentate along upper third, subrevolute, articulations green, yellowish or dark brown in young leaflets. Pollen strobili usually 1-2(4), with sterile tip, erect, cylindrical to conical, light brown to purple, tomentulose, apex acute to apiculate, generally with densely tomentose peduncles; pollen sporangiophores cuneiform, distal face truncate hexagonal, 0.3-0.55 cm long, fertile abaxial surface 2 lobed with 2-14 bisporangiate synangia per lobe, sporangia dehiscent by longitudinal slit. Ovulate strobili usually solitary, erect to decumbent, cylindrical to ellipsoid, purple to yellowish, generally tomentulose, apex acute to apiculate; peduncle densely tomentose; ovulate sporangiophores cuneiform-peltate to scutiform, distal end truncatehexagonal when not scutiform. Seeds ovoid, sarcotesta white to pink when immature, red at maturity, sclerotesta smooth but sometimes with several furrows running longitudinally from micropylar end.

Key to the species of Zamia in Mega-Mexico

1. Leaflets chartaceous to papyraceous 2. Leaflet margin dentate 3. Leaflets elliptic, adaxial surface with yellow to cream variegation

52

Z. variegata 3. Leaflets long-lanceolate, without variegation 4. Leaflets imbricate, peduncle of pollen strobili decumbent, up to 16 cm long

Z. oreillyi

4. Leaflets not imbricate, peduncle of pollen strobili erect, up to 8 cm long Z. herrerae 2. Leaflet margin serrulate to entire, 5. Leaflet margin entire, chartaceous 6. Leaflets oblong-lanceolate, glossy, 4-6 cm wide; ovulate strobili iridescent blue-green at maturity

Z. tuerckheimii

6. Leaflets linear-lanceolate, not glossy, 0.6-1.5 cm wide, ovulate strobili darkbrown at maturity

Z. soconuscensis

5. Leaflet margin serrulate, papyraceous 7. Leaflets sub-falcate basally, long-acuminate, strongly apically curved

Z. monticola

7. Leaflets straight, acuminate, not curved at apex 8. Leaflets elliptic to lanceolate, ovulate strobili cylindrical to ovoid, darkgreen and glabrous when mature, from San Luís Potosí and Querétaro, Mexico

Z. fischeri

8. Leaflets ovate to obpyriform, ovulate strobili ovoid-cylindrical, gray to brown tomentulose when mature. Endemic to Veracruz, Mexico Z. vazquezii 1. Leaflets coriaceous 9. Leaflet margin entire to serrulate 10. Leaflet margin entire, petiole unarmed

53

Z. inermis

10. Leaflet margin serrulate, petiole prickly 11. Leaflets long acuminate apically; stems arborescent

Z. onanreyesii

11. Leaflets rounded to acute apically; stems subterranean 12. Leaflets obovate to oblanceolate, keeled adaxially, apex rounded Z. furfuracea 12. Leaflets linear to oblanceolate, flat adaxially, apex acute 13. Leaflets linear, 0.4-0.6 cm wide

Z. spartea

13. Leaflets lanceolate, ≥0.7 cm wide 14. Leaflets falcate

Z. sandovallii

14. Leaflets not falcate 15. Leaflets lanceolate-oblanceolate, ovulate strobili dark-brown tomentulose

Z. polymorpha

15. Leaflets linear-lanceolate, ovulate strobili brown to yellowish 16. Ovulate strobili ellipsoid to cylindrical, apex acute to apiculate, yellowish-brown. From the Pacific seaboard of Mexico Z. paucijuga 16. Ovulate strobili ellipsoid to conical, apex acute, yellowish. From the Gulf of Mexico seaboard

Z. loddigesii

9. Leaflet margin distinctly dentate 17. Leaflets channeled adaxially between veins, appearing plicate Z. purpurea 17. Leaflets smooth, not channeled adaxially between veins, not appearing plicate 18. Leaflets linear-lanceolate, imbricate, petiole strongly armed with straight or bifurcate prickles up to 6 mm long

54

Z. cremnophila 18. Leaflets lanceolate to oblanceolate, not imbricate, petiole armed with small straight prickles, generally between 2-4 mm 19. Leaflets with brilliantly shining cuticle on adaxial surface; ovulate strobili decumbent when mature Z. katzeriana 19. Leaflets without brilliantly shining cuticle on adaxial surface; ovulate strobili erect when mature 20. Distal leaflets sub-falcate, petiole with bulbous base, blackish in young leaves

Z. lacandona

20. All leaflets straight, petiole with no bulbous base, greenish in young leaves 21. Leaflets oblong to lanceolate, bright grass-green, without conspicuously denticulate along margins; ovulate strobili green glabrous when mature, apex acuminate

Z. prasina

21. Leaflets long-lanceolate, conspicuous dentate, up to 4 mm long; ovulate strobili brown tomentulose when mature, apex long-apiculate Z. standleyi

Zamia cremnophila Vovides, Schutzman & Dehgan. Bot. Gaz. 149(3): 351. 1988. Type: Mexico. Tabasco: 18 Aug 1981, M. A. Magaña & S. Zamudio 343 (holotype: MEXU; isotypes: CSAT, FCME, MO, UAMIZ).

55

Stem hypogeous, generally unbranched, 5-34 cm long, 4-11.3 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 5.4 cm long, 2.9 cm at base, reddish-brown tomentose. Ptyxis inflexed. Leaves 2-3(4), 57.5-179 x 40.2-76.1 cm, descending, reddish-brown when emerging; petiole 10.2-95.3 cm long, blackish in young leaves, subterete, heavily armed with straight or bifurcate prickles up to 6 mm long; rachis subterete, up to 84 cm long, with few prickles along the proximal third. Leaflets 5-28 pairs, sessile, coriaceous, lanceolate, opposite to subopposite, imbricate, apex acute, base attenuate, margins dentate along distal third, subrevolute; articulations dark brown in young leaflets, 0.8-1.2 cm; the median leaflets 22.7-38 x 3.1-4.4 cm wide. Pollen strobili usually 1-2, erect, cylindrical, up to 8 cm long and 1.4 cm in diam, light brown, apex acute; peduncle densely light brown tomentose, 4.5 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face truncate-hexagonal, 0.35 cm long, fertile abaxial surface with 3 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, ellipsoid, 13.2 cm long, 4.9 cm in diam, brown to reddish, tomentulose, apex acuminate; peduncle densely brown tomentose, 2.9 cm long, 1.6 cm in diam; distal face hexagonal-truncate with a horizontal longitudinal depression, 1-1.2 cm high, 1.1-1.4 cm wide. Seeds ovoid, sarcotesta white when immature, red at maturity, 2-2.8 cm long, 1.4-2 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 16 (Schutzman et al., 1988). Diversity and genetic structure.— The average of alleles per locus is A = 1.98, the percentage of polymorphic loci is P = 94.3, the expected heterozygosity is HE = 0.347 and the genetic differentiation between the two populations currently under study is Fst = 0.093 (González-Astorga et al. unpubl. data). Distribution and habitat.— Endemic to Tabasco-Mexico (Fig. 5), on karstic rocks and cliffs of the Sierra El Madrigal, between 50-150 m. The vegetation type where this

56

species grows is evergreen tropical rain forest or bosque tropical perennifolio of Rzedowski (1978). Etymology.— The specific epithet is derived from the Greek word for cliff-friend (κρεµνóσ = cremnos = cliff, and φιλοσ = filos = friend/lover) (Schutzman et al., 1988), because of its unusual habitat. Distinguishing features.— The species is characterized by its exclusive habit on the rocky walls of limestone cliffs, in addition to descendent leaves. Petioles are densely armed with prickles that are sometimes branched, and also lanceolate, imbricate leaflets that are visibly dentate along the distal third. Additional specimens examined.— MEXICO. TABASCO: Teapa, F. NicolaldeMorejón et al. 1497 (XAL), M. A. Pérez-Farrera 293 (HEM), M. A. Pérez-Farrera 900 (HEM, MEXU).

Zamia fischeri Miq., in Lem., Hort. Vanhoutt. 1: 20. 1845. Type: ex Horto Petropolitano in H. Houtte. vecta, Miquel s.n. (neotype: designated by Stevenson & Sabato, 1986a: U). (Fig. 7).

Stem subterranean, dichotomously branching in older plants, up to 30 cm long, 4-8 cm in diam. Cataphylls membranaceous, persistent, base triangular, apex aristate, 4.5 cm long, 1.2 cm at base, yellowish tomentose. Ptyxis inflexed. Leaves 1-5(8), 15-45 x 820 cm, ascending to spreading, dark-brown when emerging; petiole 9-14 cm long, blackish in young leaves, unarmed, terete; rachis subterete, up to 31 cm long, unarmed. Leaflets 20-35 pairs, sessile, papyraceous, elliptic to oblanceolate, alternate to subopposite, apex acute symmetric, base cuneate, margins serrulate along distal third, subrevolute; articulations light-brown in young leaflets, 0.3-0.4 cm wide, median

57

leaflets 5-9 x 1.5-4.5 cm. Pollen strobili usually 1-3, erect, conical, 4-6 cm long, 1.5-2.2 cm in diam, gray tomentose, apex acute; peduncle yellowish tomentose, 2.8 cm long, 0.9-1.1 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.3 cm long, fertile abaxial surface with 3-4 bisporangiate synangia per lobe. Ovulate strobili usually 1-2, erect, cylindrical to ovoid, up to 9 cm long, 4.5 cm in diam, brown to reddish tomentulose when young, dark-green and glabrous when mature, apex acute; peduncle brown tomentose, 3 cm long, 1.1 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate when immature, scutiform when mature 0.8-1.2 cm high, 1.3-1.9 cm wide. Seed ovoid, 1.6 cm long, 1.3 cm in diam, sarcotesta pink when young and orange at maturity. Chromosome number.— 2n = 16 (Marchant, 1968; Moretti et al., 1991; Stevenson et al., 1995-96a). Distribution and habitat.—Zamia fischeri is endemic to Mexico (Tamaulipas, San Luis Potosí, Querétaro and Hidalgo) (Fig. 5), between 140-900 m elevation. The vegetation type of its habitat is pine-oak forest, tropical deciduous forest and mountain tropical forest (sensu Rzedowski, 1978). Etymology.— The specific epithet is in honor of Friedrich Fischer, a German cycad horticulturist of the 19th century. Distinguishing features.— Zamia fischeri are small fern-like plants (up to 100 cm tall), with papyraceous but serrulate leaflets and unarmed petiole and rachis. Ovulate strobili greenish upon maturity. Additional specimens examined.— MEXICO. QUERÉTARO: Jalopan, López 438 (XAL), Servín 1471 (XAL), Vovides 330 (XAL). SAN LUIS POTOSÍ: O. M. Clark 6839 (MO), J. Rees 1686 (XAL), Stevenson et al. 566 (MEXU, NY), Vovides 753 (XAL); El Naranjo, F. Nicolalde-Morejón & J. González Astorga 1614 (XAL), 1615 (XAL), 1616

58

(XAL), 1617 (XAL), 1618 (MEXU, XAL), 1619 (XAL); Ciudad Valles, F. NicolaldeMorejón & J. González-Astorga 1620 (XAL), 1621 (XAL), 1622 (XAL), 1623 (MEXU, XAL), 1624 (MEXU, XAL), 1625 (MEXU, XAL), 1626 (MEXU, XAL). HIDALGO: Pisaflores, O. Alcántara-Ayala & R. Mayorga-Saucedo 3325 (FCME).

Zamia furfuracea L. fil. in Aiton, Hortus Kew. 3: 477. 1789. Type: Palma americana crassis rigidisque foliis, pl. 210, in Herm. Paradisus Batavus. 1698 (lectotype: designated by Stevenson & Sabato, 1986a). Zamia furfuracea var. trewii A. DC., Prodr. 16(2): 541. 1868. Type: Palmifolia fructu clavato polypireno[polyspermo]. C. J. Trew, PI. Select. Tab. 26. 1752 (holotype: G; typotype: designated by Stevenson & Sabato, 1986a: BM). Zamia murieata var. obtusifolia Miquel, Tijdschr. nat. Gesch. Physiol. 10(1): 71-72. 1843. Type: Tab. VII, fig. a in Linnaea 19(4): 1847. (neotype: by Stevenson & Sabato, 1986a). Zamia latifolia Loddiges ex Miquel, Tijdschr. wis-en natuurk. Wet. 2(4): 298. 1849. Basionym: Z. muricata var. obtusifolia Miq.

Stem hypogeous, becoming epigeous with age, often dichotomously branched, up to 60 cm long, 20 cm in diam. Cataphylls coriaceous, persistent, base triangular, apex long aristate, up to 10 x 3-4.5 cm at base, yellowish tomentose. Ptyxis inflexed. Leaves 3 to many, 45-190 x 10-30 cm, diffuse, brown-yellowish when emerging; petiole 17-50 cm long, brown-yellowish when young, subterete, armed with small prickles up to 3 mm long; rachis terete, up to 120 cm long, with few prickles along the proximal third. Leaflets 8-18 pairs, sessile, coriaceous, obovate to oblanceolate, opposite to subopposite, imbricate, keeled, apex rounded to sub-acute, base attenuate, margins

59

slightly serrulate along the ½ distal portion, subrevolute; articulations yellow in young and juvenile leaflets, 0.4-0.7 cm wide; median leaflets 14-20 x 4-7.5 cm. Pollen strobili usually 2-4, erect, cylindrical, up to 17 cm long, 1.8 cm in diam, yellowish to brown, apex acute; peduncle densely light-brown tomentose, 12 cm long, 1.1 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.45-0.55 cm long, fertile abaxial surface with 8-9 bisporangiate synangia per lobe. Ovulate strobili usually 1 per crown, erect, cylindrical, up to 25 cm long, up to 10.5 cm in diam, yellowishgreen tomentulose when immature, light brown upon maturity, apex apiculate; peduncle yellowish tomentose, up to 8 cm long, up to 1.3 cm in diam; megasporangiophores peltate, distal end hexagonal-truncate, 0.9 cm high, 1.3 cm wide. Seeds ovoid, sarcotesta yellowish-green when immature turning red at maturity, 1.6 cm long, 1.2 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 18 (Moretti, 1990a ,b). Diversity and genetic structure.— The average of alleles per locus is A = 2.05, the percentage of polymorphic loci is P = 90.7, the expected heterozygosity is HE = 0.356 and the genetic differentiation between the two populations currently under study is Fst = 0.161 (González-Astorga et al. unpubl. data). Distribution and habitat.— Endemic to Mexico in central-south coastal Veracruz, along a coastal stretch of approximately150 km in stable dunes and basalt cliffs (Fig. 5). Etymology.— The specific epithet alludes to the persistent brown-yellowish trichomes of the leaves throughout the developmental stages of the plant. Distinguishing features.— Leaves strongly keeled with highly imbricate and coriaceous obovate to oblanceolate leaflets with brown-yellowish indumentum persisting with age.

60

Additional specimens examined.— MEXICO. VERACRUZ: Ibarra-Mariquez 316 (MEXU, MO), Ibarra-Mariquez 1952, 1959 (MEXU); Alvarado, Rees 1650, 1651, 1652 (XAL), M. Vázquez-Torres et al. 4871 (CIB); Catemaco, Calzada 1475 (MEXU), Calzada 2451 (XAL), Cedillo 2610 (MEXU, XAL), González-Quintero 1520 (ENCB), Ibarra 316 (MEXU), Lot 1277-14 (F, XAL), Menendez 115 (MEXU, MO, XAL), F. Nicolalde-Morejón et al. 1484 (XAL), 1485 (XAL), 1486 (XAL), J. Rees 1651 (IBUG, MEXU, XAL), J. Rees 1652 (XAL), Schatz & Nee 207 (XAL), Vovides 567 (MEXU, XAL); Mecayapan, Calzada et al. 11325 (XAL), Castillo-Campos 12732 (XAL); Lerdo De Tejada, Vovides 828 (XAL), 829 (XAL), 830 (XAL), 831 (XAL), 832 (XAL), 833 (XAL), 839 (XAL); San Andrés Tuxtla, Castillo-Campos et al. 13881 (XAL), Chazaro B. 512 (XAL), Hammel & Merello 15499 (MO), Hernández-M 1216A (F, MEXU), Lorence 4978 (MEXU), Sousa 3099 (F, MEXU, MO), Vovides & Iglesias 1148 (XAL).

Zamia herrerae Calderón & Standl., Proc. Wash. Acad. Sci. 14(4): 93. 1924. Type: Salvador. Vicinity of Sonsonate: 17 Jul 1923, S. Calderón 1682 (holotype: US).

Stem hypogeous, bifurcate in adult plants, 6-26 x 4.5-13.5 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 4.6 x 1.2 cm wide at base, yellowish tomentose. Ptyxis inflexed. Leaves 2-4, erect, green to light-brown when emerging, 61-96 x 24.5-29.5 cm; petiole 19-34 cm long, brownish in young leaves, terete, armed with prickles up to 3 mm long; rachis subterete, up to 63 cm long, with few prickles along the proximal third. Leaflets 15-32 pairs, sessile, papyraceous, lanceolate, alternate to subopposite, apex acute, base symmetric attenuate; margins dentate along distal 2/3, subrevolute; articulations dark brown when young, 0.3-0.5 cm wide; median leaflets 22-38 x 3.1-4.4 cm. Pollen strobili usually 2-3, erect, cylindrical

61

to conical, 4.3-7.5 cm long, 1.3-2.1 cm in diam, light brown tomentulose, apex mucronate; peduncle densely light-brown tomentose, 6.2-7.8 cm long, 1.1-1.4 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal, 0.35 cm long, fertile abaxial surface with 3-4 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, cylindrical to ovoid, 7.1-11.6 cm long, 4.1-4.9 cm in diam, brown, tomentulose, apex acute; peduncle densely brown tomentose, 3.5-4.2 cm long, 1.1-1.3 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 1.3-1.6 cm high, 2.2-2.8 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 1.6-1.9 cm long, 1.2-1.5 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 23, 24 (Fig. 4). Distribution and habitat.— The species range is El Salvador, Guatemala and southern Mexico in Chiapas, between 100-600 m elevations (Fig.6). It generally grows in tropical deciduous forest (sensu Rzedowski, 1978) on deep clay soils. Also, this cycad may be found in secondary growth forests and pastures. Etymology.— The specific epithet honors Hector Herrera, a scientist from El Salvador. Distinguishing features.— Zamia herrerae is distinguished by its long-lanceolate papyraceous leaflets with dentate margins along the distal 2/3. Additional specimens examined.— MEXICO. CHIAPAS: Escuintla, Matuda 16368 (MEXU), 16871 (MEXU), 17332 (MEXU), 18332 (MEXU), M. A. PérezFarrera 143 (HEM), Schutzman 526 (XAL), 527 (XAL), 528 (XAL); Tonalá, Farrera 2489 (CHIP), F. Nicolalde-Morejón & J. González-Astorga 1579 (XAL), 1580 (XAL), 1581 (XAL), M. A. Pérez-Farrera 744 (CIB). GUATEMALA. P. C. Standley 67306 (F). NICARAGUA. MANAGUA: Cultivated, A. Grijalva 3658 (MO).

62

Zamia inermis Vovides, J.D. Rees & Vázq. Torres, Flora de Veracruz 26: 22. 1983. Type: Mexico. Veracruz: 6 Jun 1981, Vovides 666 (holotype: XAL; isotype: F). (Fig. 8).

Stem epigeal, erect, dichotomously branching in mature plants, 15-43 cm long, 8.626.4 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 2.12.6 x 4.1-5.6 cm wide at base, yellowish tomentose. Ptyxis inflexed to erect. Leaves 1035, erect, light to yellowish-green when emerging, 30-95 x 43.5-60 cm; petiole 18-41 cm long, greenish in young leaves, subterete, without prickles; rachis subterete, 15-19 cm long, unarmed. Leaflets 27-32 pairs, sessile, coriaceous, linear-lanceolate, opposite to subopposite, apex acute, base attenuate; margins entire, subrevolute; articulations 0.4-0.6 cm wide; median leaflets 20-30.5 x 0.9-1.2 cm. Pollen strobili cylindrical, usually 1-2 per crown, erect, up to 9.1 cm long, up to 2.8 cm in diam, beige-yellowish, apex acute; peduncle densely light-yellowish tomentose, up to 4.5 cm long, 1.1 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.35 cm long, fertile abaxial surface with 5-6 bisporangiate synangia per lobe. Ovulate strobili usually 1-2 per crown, erect, cylindrical, 13-23 cm long, 8-9.8 cm in diam, light-brown to beige tomentulose, apex apiculate; peduncle brown tomentose, 6-8 cm long, 1.2-1.4 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 0.7-0.9 cm high, 1.1-1.3 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 1.7-2.5 cm long, 1.4-2.1 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 16 (Vovides, 1983).

63

Distribution and habitat.— Endemic to Mexico in a small mountain range in central Veracruz (Fig. 5) at 150-300 elevation on basaltic soils. The vegetation type of the habitat is tropical deciduous forest (sensu Rzedowski, 1978). Etymology.— The specific epithet alludes to the absence of prickles on the petiole and rachis and the entire leaflet margins. Distinguishing features.— Zamia inermis differs from its congeners in Mexico by a total absence of prickles along the petiole and rachis, as well as having totally entire leaflets and light-yellowish to clear-beige tomentulum on the pollen and ovulate strobili. Additional specimens examined.— MEXICO. VERACRUZ: Actopan, Acosta & Acosta 234 (XAL), F. Nicolalde-Morejón & Vovides 1415 (XAL), 1416 (XAL), 1417 (XAL), Schutzman 570 (XAL), 571 (XAL), 572 (XAL), 575 (XAL), 576 (XAL), 577 (XAL), J. Rees et al. 681 (XAL).

Zamia katzeriana (Regel) Rettig, Gartenflora 45: 148. 1896. Type: ex Horto Katzer., Regel s.n. (lectotype: designated by Stevenson & Sabato 1986a : LE). Ceratozamia katzeriana Regel, Acta Horti Petrop. 4(4): 298. 1876. Type: ex Horto Katzer., Regel s.n. (lectotype: designated by Stevenson & Sabato, 1986b: LE). Zamia splendens Schutzman, Phytologia 55(5): 299. 1984. Type: Cultivated in Fairchild Tropical Garden, Miami, accession no. FTG 76-1046, 11 Apr 1984, J. Watson s.n. (holotype: NY; isotypes: FLAS, FTG, MEXU).

Stem hypogeous, unbranched, up to 25 cm long, up to 7 cm in diam. Cataphylls chartaceous, semidecidious, base triangular, apex aristate, 5.3 x 1.4 cm at base, yellowish tomentose. Ptyxis inflexed. Leaves 1-2 (3), 49-220 x 35-58 cm, ascending to

64

descending, bright reddish-pink with lustrous cuticle when emerging, dark green when mature; petiole 20-130 cm long, terete, armed with few simple prickles; rachis subterete, up to 84 cm long, unarmed. Leaflets 3-7 pairs sessile, coriaceous, oblonglanceolate, opposite to subopposite, adaxial surface with brilliantly shining cuticle throughout life of leaflet, apex acute, base attenuate; margins dentate along distal third, subrevolute; articulations brown in young leaflets turning green with age 0.6-1.6 cm wide; median leaflets 18-35 x 3.5-12 cm. Pollen strobili 1-5, conical, light-brown tomentulose, erect upon emergence becoming prostrate to decumbent upon maturity, up to 3.9 cm long, 1.1 cm in diam, apex acute; peduncle light-brown tomentose, up to 5.8 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal scutiform, 0.35 cm long, fertile abaxial surface with 5-6 bisporangiate synangia per lobe. Ovulate strobili usually solitary, decumbent to erect, elliptic, 8-12 cm long, 4.5-6 cm in diam, brown to yellowish tomentose, apex aristate; peduncle brown tomentose, 2.1-4.3 cm long, 1.1-1.3 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate to scutiform, 1-1.2 cm high, 1.1-1.3 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 1.1-1.4 cm long, 1.6-1.8 cm in diam, sclerotesta smooth.

Chromosome number.— 2n = 16 (Schutzman, 1984; Moretti, 1990a). Diversity and genetic structure.— The average of alleles per locus is A = 1.95, the percentage of polymorphic loci is P = 84.4, the expected heterozygosity is HE = 0.280 and the genetic differentiation between the two populations currently under study is Fst = 0.194 (González-Astorga et al. unpubl. data).

65

Distribution and habitat.— Endemic to Mexico and known from the states of Chiapas, Tabasco and Veracruz (Fig. 5) at 200-700 m in evergreen tropical forest (sensu Rzedowski, 1978). Etymology.— Specific epithet in honor of Katzer, inspector of the gardens in Paullowsk (Stevenson and Sabato, 1986a). Distinguishing features.— Zamia katzeriana is easily distinguished from its congeners by having leaves with highly lustrous or shiny cuticles. Emerging leaves are a bright reddish-pink. Both pollen and ovulate strobili are borne on long peduncles that become descendent to prostrate. Additional specimens examined.— MEXICO. VERACRUZ: Las Choapas, Martínez & Martínez-M. 825 (HEM), F. Nicolalde-Morejón et al.1436 (XAL), 1437 (XAL). Tabasco, Teapa, Hdez-Najarro 622 (CHIP), M. A. Magaña 1905 (MEXU), M. A. Pérez-Farrera s.n. (XAL), M. A. Pérez-Farrera 899 (HEM, MEXU), Walters s.n. (FTG accession 12-2, XAL). CHIAPAS: San Fernando, Vovides et al. 1266 (XAL), Palacios 383 (CHIP), F. Nicolalde-Morejón & Pérez-Farrera 1420 (XAL), M. A. Pérez-Farrera s.n. (XAL), Walters s.n. (FTG accession 23-2, XAL); Ocozocoautla, Gómez-Pompa 705 (FCME, MEXU), F. Nicolalde-Morejón et al. 1453 (XAL), 1454 (XAL), 1455 (XAL), 1456 (XAL), 1457 (XAL), 1458 (XAL), 1459 (XAL), 1460 (XAL), M. A. Pérez-Farrera 29 (CHIP, CIB, MEXU); Tila, Vovides et al. 1340 (XAL), 1343 (XAL), 1341 (XAL).

Zamia lacandona Schutzman & Vovides. Novon 8(4): 441. 1998. Type: Mexico. Chiapas: Selva Lacandona, July 1984, Schutzman 517 (holotype: FLAS; isotype: XAL).

66

Stem hypogeous, unbranching, 6-14 cm long, 5-9 cm in diam. Cataphylls coriaceous, persistent, base triangular, apex long-aristate, 8.3 x 3.6 cm at base, reddishbrown tomentose. Ptyxis inflexed, reddish-brown. Leaves usually solitary, up to 3 under cultivation, 46-171 x 32-75 cm, ascending, reddish-brown when emerging; petiole 1495.3 cm long with bulbous base, blackish in young leaves, subterete, proximal section strongly channeled, armed with prickles up to 5 mm long; rachis subterete, 31-76 cm long, with few prickles along proximal third. Leaflets 4-17 pairs, sessile, coriaceous, lanceolate, opposite to subopposite, subfalcate, apex acute, base attenuate, margins dentate along distal third, subrevolute; articulations dark-brown when young, 0.4-1.1 cm wide; median leaflets 15.6-37 x 2.9-6 cm. Pollen strobili usually 2-3, erect, conical, 5.4-6.6 cm long, 1.5-1.7 cm in diam, light-brown, apex acute; peduncle light-brown tomentose, 5.8-6.9 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal truncate, 0.35 cm long, fertile abaxial surface with 4-5 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, ellipsoid, 13.2 cm long, 4.9 cm in diam, dark-brown, tomentulose, apex acute to slightly apiculate; peduncle densely brown tomentose, 6.5 cm long, 1.2 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 1-1.2 cm high, 1-1.5 cm wide. Seeds irregularly ovoid, sarcotesta pink when immature, red at maturity, 2-2.4 cm long, 1.3-1.9 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 16, 17, 18 (Schutzman & Vovides, 1998). Diversity and genetic structure.— The average of alleles per locus is A = 1.78, the percentage of polymorphic loci is P = 67.9, the expected heterozygosity is HE = 0.191 and the genetic differentiation between the two populations currently under study is Fst = 0.108 (González-Astorga et al. unpubl. data).

67

Distribution and habitat.— Endemic to Chiapas (Fig. 5) in the vicinity of the Lacandon forest at 80-200 m elevation, in evergreen tropical forest (sensu Rzedowski, 1978). It is also found in secondary succession stages of the tropical forest. Etymology.— The epithet is derived from the name of the 1.8 million hectare Selva Lacandona (Lacandona Jungle) in southeastern Chiapas, which itself bears the name of the Lacandona Maya Indians who inhabit the forest. Distinguishing features.— This species is distinguished by having a solitary large leaf (up to three leaves may be maintained on plants under cultivation) with a stout erect petiole and a strongly bulbous base. The leaf is reddish-brown at emergence and petiole dark purplish-brown turning dark-brown with age. Ovulate cone generally solitary, with acute apex. Additional specimens examined.— MEXICO. CHIAPAS: Palenque, Schutzman 510 (XAL), 511 (XAL), 512 (XAL), 513 (XAL), 514 (XAL), 515 (XAL), 516 (XAL), 517 (XAL), 518 (XAL), 519 (XAL), 520 (XAL), F. Nicolalde-Morejón & N. Martínez 1418 (XAL), M. A. Pérez-Farrera 890 (HEM, MEXU), Walters s.n. (FTG accession 14-2, XAL), M. Vázquez-Torres et al. 3925 (CIB); San Jerónimo Tulija, Chavelas et al. ES=315 (ENCB, MEXU), Schutzman 521 (XAL), 522 (XAL), 524 (XAL), 523 (XAL), 525 (XAL).

Zamia loddigesii Miq., Tijdschr. Natuurl. Gesch. Physiol. 10: 72. 1843. Type: cultivated by Van Houtte 3374 (lectotype, here designated: U) Zamia galeotti De Vriese, in Hoven & De Vries. Tijdschr. Natuurl. Gesch. Physiol. 12: 24. 1845. Type: Mexico, Veracruz. 5 July 1983. D. W. Stevenson 538 (neotype, here designated: NY, isoneotype: XAL). Zamia leiboldii Miq. Linnaea 19: 425. 1847. Type: E. Mexico in Hortum

68

Loehrianum Lipsiae attulit Liebold, 1845, Miquel s.n. (holotype: U). Zamia loddigesii var. angustifolia Regel, Bull. Soc. Nat. Moscou 30(1): 190. 1857. Type: ex horto Petropolitano, 1856, Regel s.n. (holotype: LE) Zamia loddigesii var. obtusifolia Regel, Bull. Soc. Nat. Moscou 30(1): 190. 1857. Type : t. 186, figs 27-28 in Gartenflora 6: 1857. (lectotype, designated by Stevenson & Sabato, 1986a: LE). Zamia mexicana Miquel, Prodr. syst. Cycad. 13. 1861. Type: Eriozamia mexicana H. Belg., 1847, Miquel s.n. (holotype:U). Zamia loddigesii var. leiboldii (Miquel) A. DC., Prodr. 16(2): 541. 1868. Basionym: Zamia leiboldii Miquel. Zamia leiboldii var. angustifolia Regel Trudy Imp. S.-Petersburgsk. Bot. Sada 4(4): 307. 1876. Type: Mexico. Oaxaca: 15 July 1983. D. W. Stevenson 559 (neotype, here designated: NY; isoneotype, XAL). Zamia leiboldii var. latifolia Regel, Trudy Imp. S. Petersburgsk. Bot. Sada 4(4): 307. 1876. Type: ex Horto Petropolitano, 1875, Regel s.n. (holotype: LE). Zamia lawsoniana Dyer in Hemsley, Biol. Centr.-Amer., Bot. 3(16): 195. 1884. Type: Mexico. Oaxaca: Fielding 209 (holotype: OX; isotype: K). Zamia cycadifolia Dyer in Hemsley, BioI. Cent.-Amer., Bot. 3(16): 195. 1884, non Jacquin 1809. Type: México. Bourgeau s.n. (holotype : K; isotype : C). nomen illegit. Zamia sylvatica Chamberlain, Bot. Gaz. 81: 223. 1926. Type: Mexico, Oaxaca, Tuxtepec, Sep 1910, C. J. Chamberlain s.n. (lectotype, designated by Stevenson & Sabato, 1986a: NY; isolectotype, F-3 sheets).

69

Zamia loddigesii var. cycadifolia Schuster, Pflanzenreich 99: 148. 1932. Basionym: Zamia cycadifolia Dyer in Hemsley, Biol. Cent.-Amer., Bot. 3(16): 195. 1884, non Jacquin 1809. Zamia loddigesii var. angustifolia (Regel) J. Schust., in Engl., Pflanzenr. 4(1): 148. 1932. Type: México. Veracruz: savanne bei Mundo nuevo, Karwinski 1028b (lectotype, designated by Stevenson & Sabato, 1986a: LE; isolectotype, LE). Zamia loddigesii var. longifolia J. Schust., in Engl., Pflanzenr. 4(1): 147. 1932. Type: Mexico: Veracruz, Colipa, Karwinski 1029 (lectotype, designated by Stevenson & Sabato, 1986a : LE ; isolectotype: LE).

Stem hypogeous, branching dichotomously with age, 10-45 cm long, 8-15 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 8.4 x 3.7 cm at base, yellowish tomentose. Ptyxis inflexed. Leaves 2-3 (4) ascending to spreading, 4596 x 30-41 cm, light-green when emerging, green to dark-green when mature; petiole 15-25 cm long, green in young leaves, subterete, armed with prickles up to 4 mm long; rachis subterete, up to 57 cm long, with few prickles on the proximal third. Leaflets 1223 pairs, sessile, coriaceous, linear-lanceolate, opposite to subopposite, apex acute, base attenuate, margins serrulate along distal third, subrevolute; articulations 0.4-0.7 cm wide; median leaflets 16-26 x 1.8-3.1 cm. Pollen strobili 1-2 per crown, up to 6 (7) when multiple crowned, erect, cylindrical, 8-14 cm long, 1.8-3.5 cm in diam, light brown tomentulose, apex acute; peduncle light-brown tomentose, 6 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.3 cm long, fertile abaxial surface with 6-8 bisporangiate synangia per lobe. Ovulate strobili usually 1-2 per crown, erect, ellipsoid to conical, up to 16 cm long, up to 6 cm in diam, beige-

70

tomentulose, apex acute; peduncle brown-tomentose, up to 6 cm long, 1.6 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 0.7-1 cm high, 1.9-2.6 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 1.4-1.8 cm long, 0.8-1 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 18 (Norstog, 1980; Moretti, 1990a, b). Diversity and genetic structure.— The average of alleles per locus is A = 1.8, the percentage of polymorphic loci is P = 66.6, the expected heterozygosity is HE = 0.266 and the genetic differentiation between the two populations currently under study is Fst = 0.179 (González-Astorga et al. unpubl. data). Distribution and habitat.— Endemic to Mexico and distributed widely within the states of Tamaulipas, Hidalgo, Veracruz, Tabasco and parts of Oaxaca with a single known locality in Chiapas (Fig. 6). The vegetation types are evergreen tropical forest, tropical deciduous and sub-deciduous forests (sensu Rzedowski, 1978), as well as a variety of secondary succession and disturbed habitats such as pastures and cornfields, as well as road-side vegetation. Etymology.— The specific epithet honors Conrad Loddiges (1738-1826), a German horticulturist who lived in London and cultivated American cycads. Nomenclatural notes.— The description of Zamia loddigesii was lectotypified by Stevenson & Sabato, (1986). However, it is no longer permissible to designate a description to serve as a type. Moreover, since then a specimen, Van Houtte 3374 at U, was found that was sent to Miquel by Van Houtte. This specimen predates the publication of Z. loddigesii Miq. and matches the description. Thus, we are designating it the lectotype. Distinguishing features.— Zamia loddigesii in contrast to Z. paucijuga, has been widely collected along the Gulf of Mexico seaboard and on the Yucatan peninsula since

71

the 19th century, largely by British and Russian botanists, and recently by Mexican and North American researchers. The high morphological variation presented by this species has resulted in the publication of 10 affine names (see Hill et al., 2007) that at times, has been a basis for the separation of natural populations, i.e. that of the Yucatan peninsula (formerly Z. loddigesii sensu Vovides & Olivares, 1996) that is now a separate entity Z. polymorpha (see Stevenson et al., 1995-96b), a decision based on both vegetative and reproductive characters that differ from Z. loddigesii mainly on pollen and ovulate strobilus shape and indument color. Zamia loddigesii is similar to Z. polymorpha in leaf morphology; however, there are differences in the reproductive structures. Pollen cones of Z. loddigesii are beige in color with a blunt apex whereas those of Z. polymorpha are dark reddish-brown or maroon with an acute apex. Ovulate cones of Z. loddigesii are cylindrical and beige in color and those of Z. polymorpha are ovoid and dark maroon in color. Additional specimens examined.— MEXICO. CHIAPAS: Ocozocoautla, M. A. Pérez-Farrera 81 (CHIP, HEM). HIDALGO: Atlapexco, San Juan 15 (XAL), 16 (XAL), 17 (XAL), 18 (XAL), 19 (XAL), 20 (XAL), 21 (XAL). PUEBLA: Sarukán et al. 4632 (FCME, MEXU). Oaxaca, Tuxtepec, J. Chamberlain s.n. (MO). TAMAULIPAS: R. L. Dressler 1858 (MO), Mayfiel et al. 791 (MEXU); Aldama, F. Nicolalde-Morejón & J. González-Astorga 1585 (XAL), 1586 (XAL), 1587 (XAL), 1588 (XAL). VERACRUZ: Chavelas et al. ES-4231 (MEXU), Dorantes et al. 964 (MEXU, MO), Dorantes et al. 1112 (MEXU), Lot 733 (MEXU), Medrano et al. 2725 (MEXU), Nevling & GómezPompa 140 (MEXU), Santos 353 (XAL, XALU); Acayucan, Vovides et al. 1376 (XAL), 1377 (XAL); Actopan, J. I. Calzada et al. 6369 (MEXU, XAL), Lot 1027 (XAL), A. Vovides 754 (XAL), 755 (XAL), 817 (XAL), 818 (XAL), 819 (XAL), 820 (XAL), 821 (XAL), 822 (XAL), 823 (XAL), 824 (XAL); Alto Lucero, J. Rees 1627

72

(XAL), 1629 (XAL), 1630 (XAL), 1631 (XAL), 1632 (XAL), 1637 (XAL), Vovides 846 (XAL); Atoyac, Acevedo & Castillo-Campos 240 (XAL); Cotaxtla, González 82 (MEXU); Chicontepec, J. Rees 1615 (MEXU, XAL); Choapas, Vovides et al. 1373, 1374, 1735 (XAL); Coatepec, J. Rees & Vovides 1670 (XAL); Coatzacoalcos, Castillo & Acosta 16220 (XAL); Colipa, J. Rees 1634, 1635 (XAL); Cosautlan, Vovides 35 (XAL); Emiliano Zapata, J. Rees 1763 (XAL), Stevenson et al. 538 (MEXU, NY), M. Vázquez-Torres 8071 (CIB); Huejultla, Stresser 291 (MEXU); Hueyapan de Ocampo, Gómez-Pompa 4424 (XAL), Vovides et al. 1378 (XAL), 1379 (XAL), 1380 (XAL); Jalcomulco, Castillo & Zamora C. 7542 (XAL), Castillo & Gómez-Pompa 2588 (XAL), Castillo-Campos 2727 (XAL); Mecayapan, Castillo et al. 13681 (XAL), 13792 (XAL), 13843 (XAL), 13861 (XAL), 13865 (XAL), 13866 (XAL), A. Calatayud & J. MartínezGándara 124 (CIB); Moloacan, J. Rees 1656 (XAL); Puente Nacional, Castillo & Medina 4261 (XAL); Papantla, R. Cuevas et al. 4652 (ZEA); Soteapan, Leonati 42 (MEXU); Soteapan, A. Calatayud & J. Benítez R 285 (CIB), M. A. Santos R. 352 (CIB); Tampico Alto, Ortega & Ortega O. 2437 (XAL); Tezonapa, Robles 370 (XAL); Totutla, J. Rees 1661 (XAL); Yecuatla, J. Rees 1633 (XAL).

Zamia monticola Chamb., Bot. Gaz. 81: 219. 1926. Type: cultivated from a single seed collecting opposite the crater of Naolinco, near Xalapa, Veracruz, Oct 1925, C. J. Chamberlain s.n. (holotype: MO; isotype: NY).

Stem epigeal, up to 30 cm tall, 18-20 cm in diam. Cataphylls base triangular, apex linear-lanceolate, 3-6 x 1-2 cm wide at base. Leaves 5-20, 100-200 cm long, erect to slightly curved; petiole 50-75 cm long, terete, armed with stout prickles in lower half; rachis terete, up to 100 cm long, with few prickles along the lower half. Leaflets 30-40

73

pairs, sessile, chartaceous to papyraceous, linear-lanceolate, opposite to subopposite, subfalcate near the base, apex long-acuminate and often strongly curved, base attenuate, margins serrulate only near the base, subrevolute; articulations 0.4-0.7 cm wide; the median leaflets 25-30 x 4-6 cm. Pollen strobili usually 2-6, erect, cylindrical to oblong, 12-20 cm long and 2-4 cm in diam, cream to light brown, apex acute; peduncle light brown tomentose, 10-20 cm long; pollen sporangiophores cuneiform, distal face hexagonal, 0.4 cm long, fertile abaxial surface 2-lobed with 10-16 bisporangiate synangia per lobe. Ovulate strobili unknown. Chromosome number.— Unknown. Etymology.— The specific epithet alludes to its mountainous habitat type, originally thought to be near Naolinco near the city of Xalapa, Veracruz in Mexico. It is now known that this species is not known from Mexico and is endemic to Guatemala. Distribution and habitat.— Endemic to Guatemala on rocky outcrops in primary and secondary evergreen tropical rainforest (Fig. 6). Distinguishing features.— Characterized by its consistently chartaceous to papyraceous long acuminate leaflets that are strongly curved near the apex with light serrulations. Additional specimens examined.— GUATEMALA. ALTA VERAPAZ: H. Förther 2621/592 (NY).

Zamia onanreyesii C. Nelson & G. Sandoval. Ceiba 49(1): 135. 2008. Type: Honduras. Cortés, 7 Jan 2008, O. Reyes 406 (holotype: TEFH). Zamia bussellii Schutzman, R. S. Adams, J. L. Haynes & Whitelock. The Cycad Newsletter 31(2/3), 22. 2008. TYPE: Honduras. Cortés, June 2003, Whittington 2003/01 (holotype: FLAS).

74

Stem up to 2 m tall, up to 16 cm in diam. Ptyxis inflexed. Leaves 3-15(44), 60-180 x 15-50 cm, erect to slightly curved, tomentulose when emerging; petiole 15-40 cm long, terete, sparsely to moderately armed with prickles; rachis terete, up to 40-120 cm long, with few prickles along proximal third. Leaflets up to 30 pairs, sessile, subcoriaceous, oblong-lanceolate, opposite to subopposite, long acuminate apically, base attenuate, margins serrulate along distal third, subrevolute; articulations yellowish; median leaflets to 36 x 4 cm. Pollen strobili usually 1-3, erect, cylindrical, up to 27.5 cm long, up to 4 cm in diam, light brown to tan, apex acute; peduncle brown to tan, tomentose, up to 8.5 cm long, up to 1.6 cm in diam; pollen sporangiophore cuneiform, distal face hexagonal truncate, fertile abaxial surface with up to 9 bisporangiate sori per lobe. Ovulate strobili usually solitary, erect, cylindrical, up to 43 cm long, 12 cm in diam, brown to greenish, tomentulose, apex conical; peduncle densely brown tomentose, up to 5 cm long, up to 2.5 cm in diam, distal face hexagonal-truncate, 2.9 cm high, 4.7 cm wide. Seeds ovoid, sarcotesta red at maturity, up to 3 cm long, up to 2 cm in diam, sclerotesta smooth. Chromosome number.— Unknown. Distribution and habitat.— The species range is Honduras (Fig. 6), between 0-1300 m elevations in evergreen tropical forest. Etymology.— The specific epithet is in honor of Onán Reyes, a Honduran biologist. Distinguishing features.— Stems arborescent up to 2 m tall, leaflets coriaceous, long acuminate apically, with margin serrulate. Nomenclatural note. — Both Z. onanreyesii and Z. busselllii were published in 2008. The dates of issue are however different. The date of issue for Z. onanreyesii is 6 September 2008 and the date of issue for Z. bussellii is 16 October 2008. Thus, under

75

Article 29.1 of the Interantional Code of Botanical Nomenclature (McNeill et al., 2007), Z. onanreyesii has priority and is used here. Additional specimens examined.— HONDURAS. Departamento Cortés: J. Haynes et al. 044A, 044B (TEFH).

Zamia oreillyi C. Nelson, Ceiba 46(1-2): 56. 2005. Type: Honduras. Atlántida: 8 Apr 2006, G. Sandoval et al. 1157 (holotype: TEFH). (Fig. 9).

Stem hypogeous, non-branching, up to 25 cm long, up to 7.5 cm in diam. Leaves usually 1(2), up to 78.2 cm long, 31-35 cm wide, ascending to descending; petiole up to 47 cm long, subterete, armed with small prickles; rachis subterete, up to 35-40 cm long, with few prickles along the proximal third. Leaflets 29-31 pairs, sessile, papyraceous to sub-coriaceous, linear-lanceolate, opposite to subopposite, imbricate, apex acuminate, base attenuate, margins dentate to rarely entire along distal third, up to 0.3 cm, subrevolute; articulations brown in young leaflets, 0.3-0.5 cm wide; the median leaflets up to 16 cm long, 1 cm wide. Pollen strobili usually solitary, decumbent, cylindrical, up to 2.5 cm long and 1 cm in diam, light brown, apex acuminate; peduncle light brown tomentose, up to 15.9 cm long, 0.2 cm in diam; pollen sporangiophores cuneiform, distal face truncate-hexagonal, 0.4 cm long, fertile abaxial surface, with 2-3 bisporangiate synangia per lobe. Ovulate strobili unknown. Chromosome number.— Unknown. Etymology.— The specific epithet is in honor of Carlos Manuel O´Reilly, a Honduran biologist. Distribution and habitat.— Endemic to Honduras, between 0-200 m in evergreen tropical rainforest (Fig. 6).

76

Distinguishing features.— Characterized by strongly imbricate linear-lanceolate to oblong leaflets with dentate margins along distal third; Pollen strobili usually solitary, decumbent, with fertile section up to 2.9 cm long and decumbent peduncle up to 15.9 cm long. Additional specimens examined.— HONDURAS. ATLÁNTIDA. Balick 1711 (NY, TEFH).

Zamia paucijuga Wieland, American Fossil Cycads 2: 212 . 1916. Type: Fig. 86 in American Fossil Cycads 2: 212 .1916.. (lectotype, designated by Stevenson & Sabato, 1986a).

Stem hypogeous, branching dichotomously with age, 15-27 cm long, 8-13 cm in diam. Cataphylls coriaceous, persistent, base triangular, apex aristate, 4.5 x 3.4 cm at base, brown tomentose. Ptyxis inflexed. Leaves 2-3, ascending to descending 41-95 x 29-36 cm wide, brownish when emerging; petiole 10.2-32 cm long, green in young leaves, subterete, armed with prickles up to 4 mm long; rachis subterete, up to 56 cm long, with few prickles along the proximal third. Leaflets 5-28 pairs sessile, coriaceous, lanceolate, opposite to subopposite, apex acute, base attenuate, margins serrulate to slightly dentate along distal third, subrevolute; articulations brownish in young leaflets, 0.4-0.6 cm wide; the median leaflets 14-19 x 2.3-3.4 cm. Pollen strobili usually 1-2, erect, cylindrical, 6.3-11 cm long, 2.1-2.6 cm in diam, light brown tomentulose, apex acute; peduncle light-brown tomentose, 6.3 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.3 cm long, fertile abaxial surface with 6-8 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, ellipsoid to cylindrical, 8.1 cm long, 5.2 cm in diam, brown-yellowish, tomentulose,

77

apex apiculate; peduncle densely brown tomentose, 3.9 cm long, 1.3 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 0.6-0.8 cm high, 1.1-1.5 cm wide. Seeds ovoid, sarcotesta pink when immature, orange at maturity, 2-2.8 cm long, 1.5-1.7 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 23, 34, 25, 26, 27, 28 (Moretti & Sabato, 1984). Distribution and habitat.— Endemic to Mexico. Known from the states of Nayarit, Jalisco, Colima, Michoacan, Guerrero and Oaxaca (Fig. 6). Found in varied habitats from dry open woodlands to understory evergreen tropical forest. Etymology.— The specific epithet alludes to few leaflet pairs per leaf. Distinguishing features.— Zamia paucijuga is highly variable morphologically and in chromosome number and karyotype (Moretti, 1990a, b). This may represent more than one entity under the concept of Z. paucijuga sensu stricto Wieland (1916). This species shares a number of characteristics with Z. loddigesii, being small plants with underground contractile stems and having coriaceous leaflets with marginal teeth. Nevertheless, it differs from Z. loddigesii in having highly coriaceous leaflets with longer marginal teeth, and seeds with an orange sarcotesta at maturity as contrasted to the red sarcotesta of Z. paucijuga. Zamia paucijuga occurs along the Pacific seaboard of Mexico whereas Z. loddigesii is known generally from the Gulf of Mexico seaboard. Additional specimens examined.— MEXICO. COLIMA: McVaugh 15768 (FCME, MEXU). GUERRERO: Acapulco, N. Noriega-Acosta 463, 546 (FCME), W. Thomas & J. L. Contreras 3744 (FCME); Chilpancingo, Kruse 902 (FCME, MEXU), R. M. Fonseca 1210 (FCME), F. Nicolalde-Morejón et al. 1566, 1567, 1568, 1569 (XAL), José Azueta, Vovides et al. 1426 (XAL); Petatlán, Vovides et al. 1427, 1428, 1429, 1430, 1431, 1432, 1433, 1434 (XAL); Unión de Isidro Montes De Oca, Vovides et al. 1416, 1417, 1418, 1420, 1421 (XAL); La Unión, G. Lozano-Valdez 331 (FCME), J.

78

Jiménez 331 (FCME). JALISCO: Gómez-Pompa 4876 (MEXU); El Arenal, Castillo et al. 9822 (XAL); Cabo Corrientes, Castillo et al. 10147, 11733, 10280, 10466 (XAL), J. Ceja et al 1437 (UAMIZ), J. Ceja et al. 1470 (UAMIZ), F. Nicolalde-Morejón et al. 1524, 1525 (XAL); Cihuatlán, J. Borocio R. s.n. (ZEA); Cuautitlán, Cochrane et al. 10886 (IBUG, WIS, ZEA), R. Cuevas et al 7025 (ZEA), L. Guzmán & J. Santana M. 745, 947 (ZEA), F. Nicolalde-Morejón et al. 1528, 1529 (XAL), Pérez de la Rosa 1039 (FCME, IBUG, MEXU), 1040 (FCME, IBUG, XAL), 1041 (CIB), 1518 (IBUG), Ramírez 425 (IBUG), M. Rosales & L. Cruz 75 (ZEA), J. Santana M. et al. 5296 (ZEA); La Huerta, Cuevas et al. 4861 (IBUG, ZEA); San Sebastián, F. NicolaldeMorejón et al. 1422 (XAL), 1423 (XAL), 1424 (XAL), 1425 (XAL), 1426 (XAL), 1427 (XAL), Pérez de la Rosa 1084 (IBUG, FCME, MEXU) 1097, 1098 (IBUG); Tuito, A. Flores et al. 614 (UAMIZ), F. Nicolalde-Morejón et al. 1429 (XAL), Pérez de la Rosa 1438, 1439 (IBUG); Vallarta, Pérez de la Rosa 1413, 1415 (IBUG); Villa Purificación, Pérez de la Rosa 1885. (IBUG). NAYARIT: Gentry & Gilly 10496 (MEXU), McVaugh 19211 (FCME, MEXU), Vovides et al. 1487 (XAL), 1488 (XAL), 1489 (XAL), 1490 (XAL), 1491 (XAL), 1493 (XAL); Compostela, F. NicolaldeMorejón et al. 1521 (XAL), 1522 (XAL), 1523 (XAL); Tepic, R. Dressler 1026 (MO), H. S. Gentry & C. I. Gilly 10498 (FCME). Oaxaca: Miranda 4205 (MEXU), F. Nicolalde-Morejón et al. 1465 (XAL), 1466 (XAL), 1467 (XAL), 1468 (XAL), 1469 (XAL), 1470 (XAL), 1471 (XAL), 1472 (XAL), 1473 (XAL); Pochutla, Schutzman 543 (XAL), 544 (XAL), 545 (XAL), 546 (XAL), 547 (XAL), 548 (XAL), 550 (XAL), 551 (XAL), 552 (XAL), 553 (XAL), 554 (XAL), 555 (XAL), 556 (XAL), 557 (XAL), 558 (XAL), 560 (XAL), 561 (XAL), 562 (XAL), 563 (XAL), 565 (XAL), A. Nava-Zafra & J. Pascual 35 (SERO, FCME); Puerto Escondido, J. Rees 1603 (MO, XAL), Walters sn (FTG accession 7-14, XAL); San Gabriel Mixtepec, F.

79

Nicolalde-Morejón et al. 1474 (XAL); San Pedro Pochulta, J. Lomelí et al. 2967 (MEXU).

Zamia polymorpha D.W. Stev., A. Moretti & Vázq. Torres. Delpinoa n.s. 37-38: 4. 1995-96 (issued 1998). Type: Belize. Cayo: 22 Jan 1989, D. W. Stevenson et al. 1119 (holotype: NY; isotypes: BRH, FTG, MO, NY, U). (Fig. 10).

Stem hypogeous, branching dichotomously with age, up to 32 cm long, up to 14 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 3-6.4 x 1.32.6 cm at base, brown tomentose. Ptyxis inflexed. Leaves 2-3(4), ascending, 30-105 x 29-45 cm, brown when emerging, green when mature; petiole 10.2-95.3 cm long, greenish in young leaves, subterete, armed with prickles up to 4 mm long; rachis subterete, up to 67 cm long, with few prickles along the proximal third. Leaflets 3-12 pairs, sessile, coriaceous, lanceolate to oblanceolate, opposite to subopposite, apex acute, base attenuate, margins serrulate along upper third, subrevolute; articulations brown when young, green when mature, 0.4-0.8 cm wide; the median leaflets 17-35 x 23.5 cm. Pollen strobili usually 1-2, erect, conical, 6.5-7.3 cm long, 1.1-1.4 cm in diam, light to dark-brown tomentulose, apex acute; peduncle light brown tomentose, 6.8 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.4 cm long, fertile abaxial surface with 4-5 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, cylindrical to ovoid, 8.7-16.3 cm long, 5-8.3 cm in diam, dark-brown tomentulose, apex acute; peduncle brown tomentose, 4-7.5 cm long, 1.1 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 0.8 cm high, 1.82.1 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 1.4-2.1 cm long, 0.5-0.9 cm in diam, sclerotesta smooth.

80

Chromosome number.— 2n = 17, 22, 23, 24, 25, 26, 27, 28 (Stevenson et al., 199596; Vovides & Olivares, 1996). Distribution and habitat.— This species is known from Belize, Guatemala and Mexico from 0-200 m elevation. In Mexico, its range includes the states of Quintana Roo, Yucatán, Campeche, Tabasco and Chiapas (Fig. 6). Etymology.— The specific epithet alludes to the extreme variation in leaf and leaflet morphology presented by this species (Stevenson et al., 1995-96b). Distinguishing features.— Zamia polymorpha shares many morphological attributes with Z. loddigesii of the Gulf of Mexico drainage. However, there are clear differences in the pollen and ovulate reproductive structures. In Z. loddigesii, the pollen strobili are beige in color with an acute apex and the ovulate strobili are beige and cylindrical whereas in Z. polymorpha the pollen strobili are maroon with an acuminate apex and the ovulate strobili are dark-maroon and ovoid. Additional specimens examined.— MEXICO. CAMPECHE: Hernández et al. ES184 (MEXU), Schutzman 502 (XAL), 503 (XAL), 504 (XAL), 505 (XAL), 506 (XAL), 507 (XAL), 508 (XAL), 509 (XAL); Benito Juárez, Vovides et al. 1312 (XAL), 1313 (XAL), 1314 (XAL), 1315 (XAL), 1316 (XAL), 1317 (XAL); Calakmul, Madrid et al. 736 (MEXU), Martínez 30420-A (MEXU); Ciudad del Carmen, Flores et al. 9586 (XAL); Champoton, Chan 3719 (CICY, XAL), Vovides 853 (XAL), 854 (XAL), 855 (XAL), 1326 (XAL), 1328 (XAL), 1329 (XAL), 1330 (XAL), 1331 (XAL), 1332 (XAL), 1333 (XAL), 1334 (XAL), 1335 (XAL), 1336 (XAL), 1337 (XAL), 1338 (XAL), 1339 (XAL), 1527 (XAL); Hopelchen, Ortega & Ucán 1562 (UADY, XAL), Ucán et al. 7293 (UADY, XAL), 7307 (UADY, XAL), 7398 (UADY, XAL). CHIAPAS: Palenque, Aguilar & Aguilar 1355 (MEXU), F. Nicolalde-Morejón & N. Martínez 1419 (XAL), Schutzman 508 (XAL), Walters s.n. (FTG accession 13-2, XAL); Ocosingo,

81

Walters s.n. (FTG accession 17-2, XAL). QUINTANA ROO: Cabrera et al. 2574 (MEXU), Davidse et al. 20075 (MEXU, MO), Téllez 1415 (MEXU), Trejo 225 (CICY, MEXU); Adolfo Huerta, Álvarez et al. 9495 (MEXU); Chetumal, Vovides 852 (XAL), Flores & Burgos 9635 (XAL), 9643 (XAL), Othon P. Blanco, Vovides et al. 1318 (XAL), 1319 (XAL), 1320 (XAL), 1321 (XAL), 1322 (XAL), 1323 (XAL), 1324 (XAL), 1325 (XAL). TABASCO: Balancan, Matuda 3117 (MEXU), Méndez 214 (XAL), Novelo 169 (MEXU, XAL), Puig 788 (MEXU). Macuspana, Vovides et al. 1344, 1345 (XAL). Yucatán, G. F. Gaumer 2430 (MO), Lundell & Gentle 827 (MEXU), May 743 (CICY, MEXU); Tekon, Enríquez 94 (MEXU); Tzucacab, Vovides et al. 1303 (XAL), 1306 (XAL), 1307 (XAL), 1308 (XAL), Flores & Burgos 9642 (XAL); Valladolid, Vovides 856, 857 (XAL), Vovides et al. 867 (XAL), 868 (XAL), 869 (XAL) 871 (XAL), 872 (XAL), 873 (XAL), 874 (XAL), 875 (XAL), 881 (XAL), 877 (XAL), 870 (MEXU, XAL), 876 (MEXU, XAL), 877 (MEXU, XAL), 878 (MEXU, XAL), 880 (MEXU, XAL); Yaxcaba, Vovides et al. 1309 (XAL), 1310 (XAL), Vovides 1311 (XAL). GUATEMALA. PETEN: W. E. Harmon & J. A. Fuentes 5735 (MO). BELIZE. BELIZE DISTRICT: Estrada 234 (CICY), D. L. Spellman 1548 (MO), C. Whitefoord 2603 (MO). CAYO DISTRICT: D. W. Stevenson et al. 1121 (FTG, MO, NY, U), 1122 (FTG, MO, NY, U), M. J. Balick et al. 1803 (MO, NY), M. J. Balick et al. 2058 (NY), T. B. Croat 23732 (MO), D. R. Hotel & L. Thomas 1130 (NY), J. S. Huston s.n. (MO), R. W. Long 3238 (MO), J. A. Ratter 5195 (MO), D. L. Spellman 1974 (MO), D. L. Spellman & W. W. Newey 1962 (K), J. R. Wiley 333 (MO). Orange Walk, G. Davidse & A. E. Brant 32768 (MO). Stann Creek, R. L. Walter 1099 (MO). Sin datos (K).

82

Zamia prasina W. Bull, Retail List: 20. 1881. Type: cultivated from Br. Honduras, W. Bull s.n. (holotype: K).

Stem hypogeous to epigeous, rarely branching. Cataphylls chartaceous, persistent, base triangular, apex aristate, 3-4.5 x 1.2-2.7 cm at base, brown tomentose. Ptyxis inflexed. Leaves 2-4(6), 57-100 x 25-35 cm, ascending to spreading, brown when emerging; petiole 12-30 cm long, brown-greenish young leaves, subterete, sparsely to densely armed with prickles up to 4 mm long; rachis subterete, up to 70 cm long, with few prickles along the proximal third. Leaflets 12-18 pairs, sessile, coriaceous, oblong to oblanceolate, opposite to subopposite, apex acute to acuminate, base cuneate, margins serrate to denticulate in the upper two third, subrevolute; articulations brownyellowish in young leaflets, 0.3-0.6 cm wide; the median leaflets 15-20 x 4-6 cm. Pollen strobili usually 1-2, erect, cylindrical to ovoid, up to 6-10 cm long and 2-4 cm in diam, light brown, apex acute; peduncle densely light brown tomentose, 2-4 cm long, 1.5 cm in diam. Ovulate strobili usually solitary, erect, cylindrical, 10-15 cm long, 5-7 cm in diam, green, glabrous when mature, apex acute; peduncle brown-greenish tomentose, 3.5 cm long, 1.5 cm in diam; megasporangiophores distal face hexagonal-truncate, 1-1.4 cm high, 1.1-1.8 cm wide. Seeds ovoid, sarcotesta light red when immature, red at maturity, 1.5-2 cm long, 0.5-0.8 cm in diam, sclerotesta smooth. Chromosome number.— Not known. Distribution and habitat.— Endemic to Belize (Fig. 6) on rocky outcrops between 100-200 m elevation in evergreen tropical rainforest. Etymology.— The specific epithet alludes to the bright grass-green leaflets.

83

Distinguishing features.— Plants with subterranean or epigeal stems, distinctly serrulate leaflet margins and bright-green leaflets; ovulate strobili ovate, green and glabrous when mature. Additional specimens examined.— BELIZE. TOLEDO: G. Davidse & A. E. Brant 32179 (MO), 32232 (MO).

Zamia purpurea Vovides, J.D. Rees & Vázq.Torres. Flora de Veracruz 26: 28. 1983. Type: Mexico. Veracruz: 30 January 1982, Vovides 734 (holotype: XAL). (Fig. 11).

Stem hypogeous, dichotomously branching with age, up to 30 cm long, 4-6 cm in diam. Cataphylls membranaceous, deciduous, base triangular, apex acuminate, 5.x 1.8 cm at base, reddish-brown tomentose. Ptyxis inflexed. Leaves 1-6, ascending to spreading, 34-90 x 32-38 cm, reddish-brown when emerging, turning green to darkgreen at maturity; petiole 16-29 cm long, blackish in young leaves, subterete, armed with simple prickles up to 4 mm long; rachis subterete, 17-45 cm long, with few prickles along the proximal third. Leaflets 3-4(6) pairs, sessile, coriaceous, elliptic to lanceolate, opposite to subopposite, veins highly prominent on adaxial surface, though relatively inconspicuous veins have been reported on some individuals, apex acute, base attenuate, margins dentate along upper third, subrevolute; articulations brown in young leaflets, 0.4-0.8 cm wide; the median leaflets 6-27 x 2-8.1 cm. Pollen strobili usually 12, erect, conical, 2-4.3 cm long, 0.5-1.1 cm in diam, light-brown tomentulose, apex acute; peduncle light-brown tomentose, 2-3.5 cm long, 0.7-0.9 cm in diam; pollen sporangiophores cuneiform peltate, distal hexagonal face and truncate to scutiform, 0.25 cm long, fertile abaxial surface with 2 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, conical, 6-9 cm long, 3-4.5 cm in diam, purplish-brown

84

tomentulose when immature turning dark-purple glabrescent when mature, apex acute; peduncle densely dark-brown tomentose, 2.9 cm long, 1.6 cm in diam, distal face hexagonal-scutiform, 0.9-1.2 cm high, 1.6-2.1 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 0.8-1.1 cm long, 0.6-0.8 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 16 (Vovides, 1983). Diversity and genetic structure.— The average of alleles per locus is A = 2.10, the percentage of polymorphic loci is P = 100, the expected heterozygosity is HE = 0.481 and the genetic differentiation between the two populations currently under study is Fst = 0.037 (González-Astorga et al. unpubl. data). Distribution and habitat.— Zamia purpurea is endemic to Mexico from the Río Uxpanapa drainage system within the states of Veracruz and Oaxaca (Fig. 5) between 50-200 m in evergreen tropical forest (sensu Rzedowski, 1978) classification. Etymology.— The specific epithet alludes to the dark purple color of the ovulate strobili. Distinguishing features.— Characterized by the prominent leaflet veins giving the coriaceous leaflets a channeled appearance in contrast to the rest of its congeners in Mesoamerica, and the dark purple color of the ripe ovulate strobili. Additional specimens examined.—MEXICO. Oaxaca: Santa María Chimalapa, S. H. Salas M. 982 (SERO), Sánchez et al. 40 (B, MEXU); Santa María Lachixio, Cerón et al. 266 (XAL); San Juan Guichicovi, F. Nicolalde-Morejón & J. Torres 1503 (XAL), 1404 (XAL), M. Vázquez-Torres et al. 1470 (MO, XAL), Walters s.n. (FTG accession 10-1, XAL). Veracruz: M. Vázquez-Torres 4038 (CIB); Hidalgotitlán, Calzada 8374 (XAL), M. Vázquez-Torres et al. 224 (MO); Jesús Carranza, A. González-

85

Christen s.n. (CIB), F. Nicolalde-Morejón & J. Torres 1502 (XAL), M. Vázquez-Torres et al. 2373 (CIB), M. Vázquez-Torres et al. V-2532 (CHAPA, CIB, XAL).

Zamia sandovallii C. Nelson, Ceiba 46(1-2): 55.. 2005. Type: Honduras. Atlántida: Jan 2006, G. Sandoval et al. 1154 (holotype: TEFH).

Stem hypogeous, non-branching, 15.7 cm long, 10.1 cm in diam. Cataphylls persistent, base triangular. Leaves 1-3, ascending to descending up to 210 x 0.55 cm; petiole 68 cm long, subterete, armed with small prickles; rachis subterete, up to 140 cm long, unarmed. Leaflets 68 pairs sessile, sub-coriaceous, lanceolate, opposite to subopposite, falcate, apex acuminate, base attenuate, margins serrulate along distal third, subrevolute; the median leaflets 16-31 x 2-3 cm. Pollen strobili usually 2, erect, cylindrical, 11.2 cm long, 2.6 cm in diam, brown-reddish tomentulose, apex apiculate; peduncle light-brown tomentose, 16.7 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal truncate, 0.8 cm long, fertile abaxial surface with 2-3 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, ellipsoid to cylindrical, 12 cm long, 4.5 cm in diam, brown-yellowish, tomentulose, apex longacuminate; peduncle densely brown tomentose, 1.5 cm long; megasporangiophores peltate, distal face hexagonal-truncate, 2-2.5 cm high, 1.8-2 cm wide. Seeds ovoid, sarcotesta white when immature, 1.2-1.8 cm long, 0.5-0.7 cm in diam, sclerotesta smooth. Chromosome number.— Unknown. Distribution and habitat.— Endemic to Honduras (Fig. 6), between 200-350 m in evergreen tropical forest.

86

Etymology.— The specific epithet honors Germán Sandoval, biologist of the Universidad Nacional Autonoma de Honduras (Nelson, 2005). Distinguishing features.— This species is characterized by glabrous, subcoriaceous, lanceolate, falcate leaflets; cylindric, ovulate strobili with a strongly acuminate apex; and seeds with a white sclerotesta. Additional specimens examined.— HONDURAS. ATLÁNTIDA: J. Haynes et al. 37 (TEFH), G. Sandoval et al. 1155 (TEFH), 1156 (TEFH).

Zamia soconuscensis Schutzman, Vovides & Dehgan, Bot. Gaz. 149(3): 347. 1998. Type: Mexico. Chiapas: Feb 1939, Matuda 2659 (holotype: F; isotypes: CR, MEXU, MICH).

Stem epigeal, erect to decumbent in adult plants, branching dichotomously with age, 30-65 cm long, 10-31.5 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 7.1 x 2.6 cm at base, reddish-brown tomentose. Ptyxis inflexed to erect. Leaves 3-15 or more per apex, 120-190 x 45-62 cm, ascending, distal portion descending to spreading, brown when emerging turning green at maturity; petiole 38-72 cm long, green-yellowish in young leaves, terete, armed with prickles up to 5 mm long; rachis subterete, up to 84 cm long, with few prickles along the proximal third. Leaflets 41-52 pairs, sessile, coriaceous, linear-lanceolate, alternate to subopposite, subfalcate, apex acute, base attenuate; margins entire, subrevolute; articulations brown in young leaflets, 0.4-0.8 cm wide; the median leaflets 12-35 x 0.6-1.5 cm. Pollen strobili usually 1-3 per apex, erect, cylindrical to conical, 9-15 cm long, 1.2-2.4 cm in diam, darkbrown tomentulose, apex apiculate; peduncle light-brown tomentose, up 7.2 cm long, 1.2 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.45

87

cm long, fertile abaxial surface with 5 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, cylindrical, 12-15 cm long, 6.1-7.3 cm in diam, dark-brown to reddish, tomentulose, apex aristate; peduncle brown puberulent, 2.1 cm long, 1.2 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 0.7-0.9 cm high, 1.6-1.8 cm wide. Seeds ovoid to angular, sarcotesta white when immature turning salmon-pink when mature, sclerotesta light-beige, smooth with 6-8 light furrows running longitudinally and sometimes dichotomizing, 2.1-2.6 cm long, 1.4-1.9 cm diam. Chromosome number.— 2n = 16 (Schutzman et al., 1988). Distribution and habitat.— Endemic to Chiapas-Mexico, between 900-1,400 m in the Soconusco mountain range of southern Chiapas. It inhabits the understory herbaceous layer of the transition zone between evergreen tropical forest and cloud forest (Fig. 5). Etymology.— The specific epithet alludes to the Sierra del Soconusco mountain range also known as the Sierra Madre de Chiapas, being the region where this species is native (Schutzman et al., 1998). Distinguishing features.— Zamia soconuscensis is the only species of the genus in Mexico that approaches an arborescent habit with leaves that can reach almost two meters long that gracefully arch toward the terminal portion. The linear-lanceolate leaflets have totally entire margins; the only other Mexican congener with entire margins is Z. inermis. Ovulate strobili are short-pedunculate and solitary with a darkbrown velvety tomentulum. Additional specimens examined.— MEXICO. CHIAPAS: Matuda 2087 (MEXU), 2535 (MEXU), 2590 (MEXU), 2656 (MEXU); Acacoyagua, García 149 (CHIP, MEXU), M. A. Pérez-Farrera 141 (CIB, HEM, MEXU).

88

Zamia spartea A. DC., Prodr. 16 (2): 539. 1868. Type: Mexico. Oaxaca: prope Acayucam, Verapa, Chimalapi, 1832, Alaman s.n. (holotype: G-DC). (Fig. 12). Zamia loddigesii var. spartea (A. DC.) Schuster, Pflanzenr. 99: 148. 1932. Basionym: Zamia spartea A. DC.

Stem hypogeous, branching dichotomously with age, 5-40 cm long, 5-8 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 6 x 1.4 cm at base, yellowish tomentose. Ptyxis inflexed. Leaves 2-5(8) per crown, 35-60 x 38-52 cm, ascending to gracile, reddish-brown when emerging, turning green at maturity; petiole 12-21 cm long, green-yellowish in young leaves, subterete, heavily armed with straight to sometimes bifurcate prickles up to 4 mm long; rachis subterete, up to 42 cm long, with few prickles along the proximal third. Leaflets 15-27 pairs sessile, coriaceous, linear, alternate to subopposite, apex acute, base attenuate; margins serrulate along extreme distal portion, subrevolute; articulations light-orange when young, turning yellowish with age, 0.3-0.4 cm wide; the median leaflets 20-35 x 0.3-0.6 cm. Pollen strobili usually 2-3, erect, cylindrical, 6.5-8.5 cm long, 1.4-1.9 cm in diam, yellowishbeige tomentulose, apex acute; peduncle densely light-brown tomentose, 6-8 cm long, 0.9-1.1 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal-truncate, 0.3 cm long, fertile abaxial surface with 10-14 bisporangiate synangia per lobe. Ovulate strobili usually solitary, erect, cylindrical to oval-cylindrical, 9-12 cm long, 4.2-4.8 cm in diam, brown tomentulose, apex acute; peduncle densely brown tomentose, 3.4-3.8 cm long, 0.8-1.1 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate, 0.8-1.1 cm high, 1-1.3 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, 1.2-1.6 cm long, 0.7-0.9 cm in diam, sclerotesta smooth.

89

Chromosome number.— 2n = 18 (Vovides, 1983; Moretti, 1990a). Distribution and habitat.— Endemic to Oaxaca-Mexico. Known in southern areas of the Isthmus of Tehuantepec (Fig. 5), between 200-400 m elevation, associated with tropical deciduous forests (sensu Rzedowski, 1978). Etymology.— The epithet is from the broom genus Spartium (Fabaceae), in reference to the narrow and tapered leaflets of the cycad. Distinguishing features.— Small gracile plants reaching up to 80 cm tall with very narrow-linear leaflets with an almost entire margin with a few very small serrulations only on the extreme distal portion. Additional specimens examined.— MEXICO. OAXACA: Meave del Castillo & García 2388 (MEXU); Matías Romero, F. Nicolalde-Morejón & J. Torres 1505 (XAL), 1506 (XAL), Schutzman 529 (XAL), 530 (XAL), 531 (XAL), 532 (XAL), 533 (XAL), 534 (XAL), 535 (XAL), 536 (XAL), 537 (XAL), 538 (XAL), 539 (XAL), 540 (XAL), 541 (XAL), 542 (XAL), Vovides & Perales 600 (XAL), Walters s.n. (FTG accession 93, XAL); San Juan Guichicovi, N. Antonio-Barrera 83b (CIB), Santa María Chimalapa, A. Espejo et al 6485 (UAMIZ), R. García S. 341 (XAL, SERO), Torres 653 (XAL); San Miguel Chimalapa, M. Vázquez-Torres 4039 (CIB).

Zamia standleyi Schutzman, Syst. Bot. 14(2): 214. 1989. Type: Honduras. Atlantida: Lanatilla Valley near Tela, Aug 1984, B. Schutzman 449 (holotype: FLAS; isotypes: ENA, FTG).

Stem hypogeous and tuberous, 6-14 cm long, 5-9 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex long-aristate, to 12 cm x 1.5 cm at base, reddish-brown tomentose. Ptyxis inflexed. Leaves 1-5, 20-100 x to 55 cm, slightly to

90

slightly recurved, tomentulose when emerging; petiole 35-60 cm long, terete, sparsely to heavily armed with prickles; rachis terete, up to 70 cm long, with few prickles along proximal third. Leaflets 10-15 pairs, sessile, subcoriaceous-coriaceous, long-lanceolate, opposite to subopposite, recurved, apex acute, base attenuate, margins dentate along distal third, subrevolute; articulations dark brown when young, 0.3-0.5 cm wide; median leaflets 20-45 x 1-4 cm. Pollen strobili usually 1-3, decumbent, cylindrical, 6-10 cm long, 1-2 cm in diam, light-brown tomentulose, apex acute; peduncle light-brown tomentose, 2-4 cm long, 1.1 cm in diam; pollen sporangiophores cuneiform, distal face hexagonal truncate, fertile abaxial surface with 4 bisporangiate sori per lobe; Ovulate strobili usually solitary, erect, cylindrical to slightly ovoid, 8-12 cm long, 3-8 cm in diam, brown, tomentulose, apex long-apiculate; peduncle densely brown tomentose, 2.5-4 cm long, 1.3 cm in diam, distal face hexagonal-truncate, 0.7-1.2 cm high, 1.5-2.1 cm wide. Seeds ovoid, sarcotesta pink when immature, red at maturity, up to 3 cm long, up to 2 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 16 (Schutzman, 1989). Distribution and habitat.— The species range is Honduras and Guatemala (Fig. 6), between 0-200 m in evergreen tropical forest. Etymology.— The specific epithet honors Paul C. Standley a prominent botanist of the flora of Mexico and Central America (Schutzman, 1989). Distinguishing features.— Its falcate leaflets and conspicuous marginal teeth up to 4 mm long, and long-apiculate ovulate strobili as well as its cylindrical characterize this species. Additional specimens examined.— GUATEMALA. PROVINCIA IZABAL: H. Förther 10234/252 (MSB, W).

91

HONDURAS. DEPARTAMENTO ATLÁNTIDA: Tela, V. Severen 1450 (NA), Standley 53721 (F, US); Puerto Sierra, Wilson 537 (NY). DEPARTAMENTO SANTA BARBARA: San Pedro Sula, Thieme 144 (US). Departamento Yoro, Coyoles, T. G. Yuneker et al. 8186 (F, G, GH, MO, NY). DEPARTAMENTO CORTES: Montaña Santa Ana, Molina R. 3628 (F, GH).

Zamia tuerckheimii Donn. Sm., Bot. Gaz. (Crawfordsville) 35(1): 8. 1903. Type: Guatemala. Dept. Alta Verapaz: Cubilquitz, Jul 1900, von Tuerckheim 7786 (holotype: US; isotype: K).

Stem epigeal, rarely branched with age, up to 100 cm long, 10-12 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 8.3 x 2.7 cm at base, light brown-greenish tomentose. Ptyxis inflexed to erect. Leaves 8-15, 100-200 x 20-45 cm, ascending, distal portion descending to spreading, green when emerging; petiole 30-50 cm long, green in young leaves, terete, armed with minute prickles to unarmed; rachis subterete, up to 150 cm long, unarmed. Leaflets 8-15 pairs, sessile, papyraceous, oblong-lanceolate, glossy, iridescent blue-green, alternate to subopposite, apex abruptly acuminate, base attenuate; margins entire, subrevolute; articulations green, 0.4-0.6 cm wide; the median leaflets 14-18 x 4-6 cm x. Pollen strobili usually solitary, erect, cylindrical, 12-15 cm long, 2-4 cm in diam, gray-brown at maturity, apex acute; peduncle light-brown tomentose, up 6.5 cm long, 1.4 cm in diam. Ovulate strobili usually solitary, erect, cylindrical, 12-18 cm long, 4-8 cm in diam, brownish when young, iridescent blue-green at maturity, apex aristate; peduncle light brown puberulent, 3.4 cm long, 1.5 cm in diam; megasporangiophores peltate, distal face hexagonal-

92

truncate, 1.4-2 cm high, 3-3.6 cm wide. Seeds elongate-ovoid, sarcotesta light red when immature, red at maturity, smooth, 2.1-2.6 cm long, 1.3-1.6 cm diam. Chromosome number.— 2n = 16 (Moretti, 1990b). Distribution and habitat.— Endemic to Guatemala, between 250-1000 m in the Alta Verapaz department mountain range. It inhabits the understory herbaceous layer of tropical forest (Fig. 6). Etymology.— The specific epithet honors Hans von Tüerkheim, who collected this species for the first time in Guatemala. Distinguishing features.— The principal distinguishing features of this species are arborescent stems up to one meter tall, papyraceous leaflets with entire margins and ovulate cylindrical strobilus with abruptly acuminate apex and aristate apex, the cone becoming blue-green iridescent when mature. Additional specimens examined.— GUATEMALA. ALTA VERAPAZ: T. B. Croat 41647 (MO), H. Förther 11034 (MSB, W), J. A. Steyermark 44484 (MO); Rubeltem, H. Förther 10918 (MSB, W).

Zamia variegata Warsz., Allg. Gartenzeitung 32: 253. 1845. Type: Mexico. Chiapas, Lacandona, on border with Guatemala, 12 January 1987 D. W. Stevenson 685 (neotype, here designated: NY; isoneotypes: U, XAL). Zamia picta Dyer in Hemsley, Biol. Cent.-Amer., Bot. 3(16): 194. 1884. Basionym: Zamia muricta var. picta Miquel, Tijdschr. Wis-en natuurk. Wet.1(4): 198-199. 1848, non Von Houtte 1846. Type: ex Horto Amsterdam (Z. picta H. Belg.), Miquel s.n. (holotype, U).

93

Stem hypogeous, 9-16 cm long, 4-10.5 cm in diam. Cataphylls chartaceous, base triangular, apex aristate, 4.6-7.1 x 2.1-2.4 at base, yellowish tomentose. Ptyxis inflexed. Leaves 1-2(3), 40-291 x 24-44 cm, ascending; petiole 35-177 cm long, dark-green with characteristic yellow variegation; petiole subterete, heavily armed with prickles up to 5 mm long; rachis subterete, up to 105 cm long, with few prickles along the proximal third. Leaflets 3-10 pairs, sessile, papyraceous, elliptic, opposite to subopposite, darkgreen with yellow or cream variegation, apex acute, base attenuate; margins dentate along distal third, subrevolute; articulations brown in young leaflets turning green at maturity, 0.4-0.8 cm wide; the median leaflets 12-22 x 3.1-8.8 cm. Pollen strobili up to 6, erect, long-cylindrical, 7-11 cm long, 1.9-2.5 cm in diam, yellowish-beige, apex acute; peduncle densely light-brown tomentose, 3-6 cm long, 0.7-0.9 cm in diam. Ovulate strobili 1-2 ovoid to cylindrical, 12 cm long, 4.5 cm in diam, gray-greenish tomentulose; megasporangiophores peltate, distal face hexagonal-truncate, 0.8-1.3 cm high, 1.7-2.1 cm wide. Seeds ovoid, red at maturity, 1.1-1.5 cm long, 0.7-1 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 21, 22 (Moretti et al., 1991). Diversity and genetic structure.— The average of alleles per locus is A = 2.02, the percentage of polymorphic loci is P = 97.3, the expected heterozygosity is HE = 0.355 and the genetic differentiation between the two populations currently under study is Fst = 0.085 (González-Astorga et al. unpubl. data). Distribution and habitat.— Zamia variegata was described from plants collected in Guatemala by Warszewicz (Stevenson & Sabato, 1986a). In Mexico, this plant is known only from Chiapas, in lowland areas near the Montes Azules Biosphere Reserve (Fig. 5). Its habitat is evergreen tropical rainforest (sensu Rzedowski, 1978).

94

Etymology.— The specific epithet alludes to the variegated nature of the leaflets, an attribute unique to this species of Zamia. Distinguishing features.— Its yellow-cream variegated papyraceous leaflets easily distinguish Zamia variegata. The variegations are in the form of irregular yellow blotches that are most apparent on the adaxial side of the lamina. Additional specimens examined.— MEXICO. Chiapas: Ocosingo, CastilloCampos et al. 3848 (XAL), 3855 (XAL), 3885 (XAL), M. Vázquez-Torres et al. 3924 (CIB); Margaritas, F. Nicolalde-Morejón et al. 1443 (XAL), 1444 (XAL), 1445 (XAL), 1446 (XAL), 1447 (XAL), 1448 (XAL), 1449 (XAL), 1450 (XAL), 1451 (XAL), 1452 (XAL); Lacandona, Stevenson 692 (NY, XAL). GUATEMALA. Alta Verapaz: J. A. Steyermark 45048 (F, NY). Izabal: J. J. Castillo & D. R. Hodel 2138 (MO), M Véliz 6893 (BIGUA, MEXU).

Zamia vazquezii D.W. Stev., Sabato, A. Moretti & De Luca, Delpinoa n.s. 37-38: 14. 1995-1996 (issued 1998). Type: Mexico. Veracruz: 22 Jan 1989, M. Vázquez-Torres et al. 3990 (holotype: NY; isotypes: FTG, MO, NY, U, CIB). (Fig. 13).

Stem hypogeous, dichotomously branching with age, 35 cm long, 12 cm in diam. Cataphylls chartaceous, persistent, base triangular, apex aristate, 5.9 x 3.8 cm at base, brown tomentose. Ptyxis inflexed. Leaves 4-6 to many, 35-100 x 16-29 cm, ascending, brownish when emerging, turning green at maturity; petiole 21-45 cm long, terete, unarmed or rarely with prickles, rachis terete, up to 65 cm long. Leaflets 14-26 pairs, sessile, papyraceous, ovate to obpyriform, opposite to subopposite, apex acuminate, base cuneate; margins with numerous serrations along distal third, subrevolute, articulations brown in young leaflets, 0.5-0.7 cm wide; the median leaflets 7.1-14.6 x

95

2.8-4.1 cm. Pollen strobili usually 1-2 per apex, erect, ovoid to ovoid-cylindrical, up to 10.6 cm long and 2.6 cm in diam, light-brown tomentulose, apex acute; peduncle densely light-brown tomentose, 4.5 cm long, 1.2 cm in diam; pollen sporangiophore distal face hexagonal-truncate. Ovulate strobili usually solitary, erect, cylindrical to ovoid-cylindrical, up to 15 cm long, 7.3 cm in diam, gray to brown tomentulose, apex apiculate; peduncle densely brown tomentose, 5.3 cm long, 1.4 cm in diam; megasporangiophores peltate, distal face hexagonal-truncate to scutiform, 1.2 cm high, 2.9 cm wide. Seeds ovoid, sarcotesta pink when immature, orange-red to red at maturity, 1.6 cm long, 1.2 cm in diam, sclerotesta smooth. Chromosome number.— 2n = 18 (Stevenson et al., 1995-1996a). Distribution and habitat.— Endemic to Veracruz, Mexico (Fig. 5), found between 50-350 m in evergreen tropical forest sensu Rzedowski (1978) classification, on predominantly deep clayey soils. Etymology.— The specific epithet honors Mario Vázquez-Torres, a Mexican biologist and cycad specialist. Distinguishing features.— Zamia vazquezii has more leaves (up to six), longer with each up to 100 cm long, and wider leaflets than Z. fischeri. The two species are disjunct in distribution and have different chromosome numbers, Z. vazquezii 2n = 18 and Z. fischeri 2n = 16 (Stevenson et al., 1995-1996a). Additional specimens examined.— MEXICO. VERACRUZ: Papantla, M. VázquezTorres 4568 (CIB), Allen 1970-2141 (XAL); Tiguatlán, J. Rees 1617 (XAL).

Species dubium

Socorro is the locality mentioned by Schuster (1932) for Miquel’s (1870) source of

96

Zamia verschaffeltii, although no locality information is given on the type specimen, which is the only specimen at U where Miquel’s herbarium resides (Stafleu, 1966). The original description by Miquel gives the source of material as “Ex imperio Mexicano introduxit A. Verschaffelt, qui in Catalogis Z. fuscam latifoliam dixit” and makes no mention of a specific locality. However, following the information given by Schuster (1932), we have conducted exhaustive searches over the years at two possible localities with the name of El Socorro, one in Tabasco and the other at Ruta del Socorro in Veracruz, It is important to note that the whole area and surrounding regions have been converted into vast sugar cane plantations so that historically Zamia could have been present but would now be extirpated. We can confidently say that no species of Zamia were found in or near the two localities of the name Socorro, and no other locality of this name was located within the distribution range of this species complex in the study. Also, no individuals or populations of Zamia studied here conform to the original description of Z. verschaffeltii; because we were unable to find another record of Z. verschaffeltii since its publication in 1870, we believe that this species is probably extinct (Nicolalde-Morejón et al., 2008). It is interesting that Schuster actually described a form latifolia Schuster does not cite any specimens of this form, but only that it is cultivated in "Garten Verschaffelt." We have been unable to locate any specimens of this form that were seen by Schuster, and if one did exist at B it was destroyed. Therefore, all source material remains enigmatic.

Zamia verschaffeltii Miq., Verh. Kon. Ned. Akad. Wetensch., Afd. Natuurk. 2(4): 31. 1870. Type: Mexico. "Zamia fusca latifolia Versch.", Miquel s.n. (holotype, U). Zamia verschaffeltii forma latifolia Schuster, Pflanzenr. 99: 138. 1932. TYPE: description (lectotype, designated by Stevenson & Sabato, 1986a).

97

Acknowledgments

The first author thanks the Red Latinoamericana de Botánica for the award of a Ph. D. fellowship (RLB-06-D2; Systematics Program, Instituto de Ecología, A.C., Xalapa, Mexico) and the Mellon Foundation for a stipend for a six months residence at the New York Botanical Garden during 2002. This research was supported partially by CONACyT-SEMARNAT grant No. 2002-CO1-0183 to AV and NSF Grants BSR8607049 and EF-0629817 to DWS. The authors thank to Francisco Vergara Silva, Jorge González Astorga, and Victoria Sosa for their comments on a previous version of this manuscript, and Pablo Carrillo-Reyes, Eduardo Ruiz for their comments on the diagnostic key to species of Zamia, Carlos Iglesias for assistance and guidance in the field, and Edmundo Saavedra for the illustrations of Zamia spartea. We thank the Curators and staff of the herbaria mentioned for making their collections available for study, as well as the staff of the Jardín Botánico Fco. J. Clavjiero of the Instituto de Ecología, A.C. for access to the living specimens of the Mexican National Cycad Collection. Finally, the authors would like to tahnk Roy Osborene for his salient discussions and diligence in reviewing this manuscript throughout the preparation of this work.

Literature Cited

Balduzzi, A. P., P. De Luca & S. Sabato. 1982. A phytogeographical approach to the New World Cycads. Delpinoa, n.s., 23-24.

98

de Candolle, A. 1868. Cycadaceae. Prodromus systematis naturalis regni vegetabilis 16 (2): 522-548. Caputo, P., S. Cozzolino, P. De Luca, A. Moretti & D. Stevenson. 2004. Molecular phylogeny of Zamia. In: T. Walters & R. Osborne, eds. Cycad Classification: Concepts and Recommendations, Pp. 149-158. CABI Publishing, Oxford. Eckenwalder, J. 1980. Taxonomy of the West Indian cycads. Journal of the Arnold Arboretum. 61(4): 701-722. González-Astorga, J., A. P. Vovides, P. Octavio-Aguilar, D. Aguirre-Fey, F. Nicolalde-Morejón & C. Iglesias. 2006. Genetic diversity and structure of the cycad Zamia loddigesii Miq. (Zamiaceae): implications for evolution and conservation. Botanical Journal of the Linnean Society 152: 533-544. Hill, K. D., D. W. Stevenson & R. Osborne. 2007. The world list of cycads/La lista mundial de Cícadas. Memoirs of the New York Botanical Garden 97: 454-483. Hill, K. 1998-present.The cycad pages. http://plantnet.rbgsyd.nsw.gov.au/ PlantNet/cycad IUCN. 2005. IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge, UK. ii + 30 pp. Jones, K. 1998. Robertsonian fusion and centric fission in karyotype evolution of higher plants. The Botanical Review 64: 273-289. Marchant, C. J. 1968. Chromosome patterns and nuclear phenomena in the cycad families Stangeriaceae and Zamiaceae. Chromosoma 24: 100-134. McNeill, J., F.R. Barrie, H.M. Burdet, V. Demoulin, D.L. Hawksworth, K. Marhold, D.H. Nicolson, J. Prado, P.C. Silva, J.E. Skog, J.H. Wiersema, & N.J. Turland. 2007. International Code of Botanical Nomenclature (Vienna

99

Code),adopted by the Seventeenth International Botanical Congress. Vienna, Austria. July 2005. Gantner, Ruggell, Liechtenstein. Miquel, F. A. W. 1861. Prodromus systematics cycadearum. C. v. d. Post Jr. Utrecht, The Netherlands. . 1869. Neiuwe bijdragen tot der kennis der Cycadeen. Vijfde gedeelte. Verslagen en medeelingen van de afdeeling natuurkunde: koninkljkie akademie van wetenschappen. Amsterdam. ser. 2, 3(2): 196-206. . 1870. Neiuwe bijdragen tot der kennis der Cycadeen. Zesde gedeelte.Verslagen en medeelingen van de afdeeling natuurkunde: koninkljkie akademie van wetenschappen. Amsterdam.. 2(4): 31. 1870. Moretti, A. 1990a. Karyotypic data on North and Central American Zamiaceae (Cycadales) and their phylogenetic implications. American Journal of Botany 77: 1016-1029. ______. 1990b. Cytotaxonomy of cycads. Memoirs of the New York Botanical Garden 57: 114-122. ______, P. Caputo, L. Gaudio, & D.W. Stevenson. 1991. Intraspecific chromosome variation in Zamia (Zamiaceae, Cycadales). Caryologia 44: 1-10. ——— & Sabato, S. 1984. Karyotype evolution by centromeric fission in Zamia (Cycadales). Plant Systematics and Evolution 146: 215-223. Mundry, M. & T. Stützel. 2003. Morphogenesis of male sporangiophores of Zamia amblyphyllidia D. W. Stev. Plant Biology 5: 297-310 Nelson, C. H. 2005. Dos plantas nuevas del género Zamia (Zamiaceae, Gymnospermae) en Honduras. Ceiba 46: 55-58. Newell, S. J. 1986. Variations in leaflet morphology among three populations of Zamia L. in Puerto Rico. Taxon 35: 234-242.

100

Nicolalde-Morejón, F., A. P. Vovides, D. W. Stevenson & V. Sosa. 2008. The indentity of Zamia katzeriana and Zamia verschaffeltii (Zamiaceae). Brittonia 60: 38-48. Norstog, K. J. 1980. Chromosome number in Zamia (Cycadales). Caryologia 33: 419428. ______. 1981. Karyotypes of Zamia chigua (Cycadales). Caryologia 34: 255-260. ______. & T. J. Nicholls. 1997. The Biology of the Cycads. Cornell University Press, Ithaca. Rabinowitz, D. 1981. Seven forms of rarity. In: Synge H, (ed). Biological Aspect of Rare Plant Conservation. Pp. 205-217, John Wiley & Sons Ltd. Regel, E. 1857. Die Cycadeen des Botanischen Gartens in Petersburg. Gartenfora 6: 516. ———, 1876. Cycadearum generum specierumque revisio. Acta Horti Petropoltani 4(4): 273-320. Rzedowski, J. 1978. Vegetación de México. Limusa, S. A. México, D.F. ———, 1991. Diversidad y orígenes de la flora fanerogámica de México. Acta Botánica de México 14: 3-21. Sabato, S. 1990. West Indian and South American cycads. Memoirs of the New York Botanical Garden 57: 173-185. Schlarbaum, S. E. & T. Tsuchiya. 1984. The chromosomes of Cunninghamia konishii, C. lanceolata and Taiwania cryptomeroides (Taxodiaceae). Plant Systematics and Evolution 145: 169-181. Schubert, I. 2007. Chromosome evolution. Current Topics in Plant Biology 10: 109115.

101

Schuster, J. 1932. Cycadaceae. Das Pflanzenreich 99:1-168. Wilhelm Engelmann, Leipzig. Schutzman, B. 1984. A new species of Zamia L. (Zamiaceae, Cycadales) from Chiapas, Mexico. Phytologia 55: 299–304. ______. 1989. A new species of Zamia from Honduras. Systematic Botany 14: 214-219. ______, A. P. Vovides & B. Dehgan. 1988. Two new species of Zamia (Zamiaceae, Cycadales) from southern Mexico. Botanical Gazette. 149: 347360. ______ & A. P. Vovides. 1998. A new Zamia (Zamiaceae-Cycadales) from eastern Chiapas, Mexico. Novon 8: 441-446. Stevenson, D. W. 1980. Radial growth in the Cycadales. American Journal of Botany 67: 465-475. ______. 1981. Observations on ptyxis, phenology, and trichomes in the Cycadales and their systematics implications. American Journal of Botany 68: 1104-1114. ______. 1987. The West Indian Zamias. Fairchild Tropical Garden Bulletin.42: 23-27. ———. 1991a. Flora of the Guianas. Serie. A. Fascicule 9: 7-11. ———. 1991b. The Zamiaceae in the Southeastern United States. Journal of the Arnold Arboretum, Supplemental Series 1: 367-383. ———. 1992. A formal classification of the extant cycads. Brittonia. 44: 220-223 ———. 1993. The Zamiaceae in Panama with comments on phytogeography and species relationships. Brittonia 45: 1-16. ———. 2001a. Orden Cycadales. Flora de Colombia. Monografía No. 21. Editorial Unibiblos. Bogotá D.C.

102

———. 2001b. Zamiaceae. Flora Nicaragua 1: 6-7. ———. 2004. Zamiaceae of Bolivia, Ecuador, and Peru. In: T. Walters & R. Osborne (eds.), Cycad classification: Concepts and Recommendations. Pp. 173-194. CABI Publishing, Wallingford, England. ——— & S. Sabato. 1986a. Typification of names in Zamia L. and Aulacophyllum Regel. (Zamiaceae). Taxon 35: 134-144. ——— & ———.1986b. Typification of names in Ceratozamia Brongn., Dion Lindl., and Microcycas A.DC. (Zamiaceae). Taxon 35: 578-584. ———,. ———, A. Moretti & P. De Luca. 1995-1996a. What is Zamia fischeri? Delpinoa ns. 37-38: 9-17 (issued 1998). ______, A. Moretti & L. Gaudio. 1995-96b. A new species of Zamia (Zamiaceae) from Belize and the Yucatan Peninsula of Mexico. Delpinoa, sn. 37-38: 3-8 (issued 1998). Vovides, A. P. 1983. Systematic studies on the Mexican Zamiaceae I. Chromosome numbers and karyotypes. American Journal of Botany 70: 1002-1006. ______. 1985. Systematic studies on Mexican Zamiaceae II. Additional notes on Ceratozamia kuesteriana from Tamaulipas, Mexico. Brittonia 37: 226-361. ______. 1999. Zamiaceae. Flora del Bajío y de Regiones Adyacentes. Fascículo 71. ______, J. D. Rees & M. Vázquez-Torres. 1983. Zamiaceae. Flora de Veracruz, fasiculo 26. In: A. Gómez-Pompa, ed., Flora de Veracruz, Instituto Nacional de Investigaciones sobre Recursos Bióticos, Xalapa, Veracruz, México. ______ & M. Olivares. 1996. Karyotype polymorphism in the cycad Zamia loddigesii (Zamiaceae) of the Yucatan peninsula, Mexico. Botanical Journal of the Linnean Society 120: 77-83.

103

Wieland, G. R. 1916. American Fossil Cycads. Vol 2. Carnegie Institute, Washington, D.C.

104

FIG. 1. Trichomes. A-B. Zamia furfuracea. C-D. Z. katzeriana. E-F. Z. polymorpha.

105

FIG. 2. Illustrations of ovulate strobili. A. Zamia inermis. B. Z. vazquezii. C. Z. paucijuga. D. Z. lacandona. E. Z. spartea.

106

FIG. 3. Illustrations of ovulate strobili. A. Z. fischeri. B. Z. furfuracea. C. Z. cremnophila. D. Z. katzeriana. E. Z. polymorpha

107

FIG. 4. Chromosomes of Zamia herrerae. A. 2n = 23. B. 2n = 24.

108

FIG. 5. Distribution of Zamia cremnophila ( ), Z. furfuracea ( ), Z. inermis ( ), Z. katzeriana ( ), Z. lacandona (■), Z. monticola ( ), Z.onanreyesii ( ), Z. prasina ( ), Z. sandovallii (▲), Z. soconuscensis ( ), Z. vazquezii ( ).

109

FIG. 6. Distribution of Zamia fischeri ( ), Z. herrerae ( ), Z. loddigesii ( ), Z. oreillyi ( ), Z. paucijuga ( ), Z. polymorpha (■), Z. purpurea (▲), Z. spartea ( ), Z.standleyi ( ), Z. tuerckheimii ( ), Z. variegata ( ).

110

FIG. 7. Zamia fischeri A. Habit. B. Leaflets.

111

FIG. 8. Zamia inermis. A-B. Habit. C. Cataphyll. D. Pollen strobilus. E. Microsporophyll, abaxial and adaxial view. F-H. Ovulate strobilus. I. Ovulate sporangiophores. J. Seed.

112

FIG. 9. Zamia oreillyi. A. Habit. B. Cataphyll. C-E. Leaflet variability. C. Broad and imbricate with dentate margins. D. Narrow and imbricate with dentate margins. E. Narrow and slighly imbricate with entire margins. F. Ovulate strobilus. G. Pollen strobilus. H. microsporophyll, abaxial view.

113

FIG. 10. Zamia polymorpha. A. Habit. B. Leaflets. C. Ovulate strobilus. D. Pollen strobilus. E. Cataphyll.

114

FIG. 11. Zamia purpurea. A. Habit. B. Pollen strobilus. C. Microsporophyll, abaxial view. D. Ovulate strobilus. E. Ovulate sporangiophores. F. Seed.

115

Fig. 12. Zamia spartea. A. Habit. B. Leaflet. C. Pollen strobilus. D-E. Microsporophyll, abaxial and adaxial view. F-G. Petiole and cataphyll. H. Ovulate strobilus and peduncle. I. Ovulate sporangiophore. J. Ovulate sporangiophores, distal end truncate hexagonal. K. Seed with sarcotesta. L-M-N. Seeds.

116

Fig. 13. Zamia vazquezii. A. Habit. B. Leaflet. C. Ovulate strobilus. D. Pollen strobilus. E. Cataphyll.

117

CAPITULO III DNA barcoding in the Mexican cycads: a character attribute organization system (CAOS) approach (Sometido a Cladistics)

118

DNA barcoding in the Mexican cycads: a character attribute organization system (CAOS) approach

Running title: DNA barcoding in the Mexican cycads using CAOS

Fernando Nicolalde-Morejón1, 2*, Francisco Vergara-Silva3, Jorge González-Astorga1, Dennis W. Stevenson4, Andrew P. Vovides5 & Victoria Sosa6

1

Laboratorio de Genética de Poblaciones, Biología Evolutiva. Instituto de Ecología, A.

C., km 2.5 Antigua Carretera a Coatepec No. 351, Xalapa 91070, Veracruz, México 2

Instituto de Investigaciones Biológicas, Universidad Veracruzana, Av. Luis Castelazo

Ayala s/n, Col. Industrial Ánimas, Xalapa 91190, Veracruz, México 3

Laboratorio de Sistemática Molecular (Jardín Botánico), Instituto de Biología,

Universidad Nacional Autónoma de México, 3er Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México, D. F. México 4

The New York Botanical Garden, Bronx, New York, 10458-5120, USA

5

Laboratorio de Biología Evolutiva de Cycadales, Biología Evolutiva. Instituto de

Ecología, A. C., km 2.5 Antigua Carretera a Coatepec No. 351, Xalapa 91070, Veracruz, México 6

Laboratorio Molecular, Biología Evolutiva. Instituto de Ecología, A. C., km 2.5

Antigua Carretera a Coatepec No. 351, Xalapa 91070, Veracruz, México

*Corresponding author: [email protected]

119

Abstract A DNA barcoding study was conducted to determine the optimal combination of loci needed for successful species-level molecular identification in three extant cycad genera, Ceratozamia, Dioon and Zamia that occur in Mexico. Based on conclusions of a previous study in representative species of all genera in the Cycadales, we tested the DNA barcoding performance of seven chloroplast coding (matK, rpoB, rpoC1 and rbcL) and non-coding regions (atpF/H, psbK/I and trnH-psbA), plus sequences of the nuclear ITS. We analyzed data under the assumptions of the “Character Attributes Organization System” (CAOS), a character-based approach whereby species are identified through the presence or ‘DNA diagnostics’. In Ceratozamia, five chloroplast regions were needed to achieve >70% of unique species identification, whereas the twogene atpF/H+psbK/I and the four-gene combination atpF/H+psbK/I+rpoC1+ITS2 were needed to reach 79% and 75% of unique species identification in Dioon and Zamia, respectively. The combinations atpF/H+psbK/I and atpF/H+psbK/I+rpoC1+ITS2 include loci previously considered by the international DNA barcoding community. However, our results suggest that the optimal combination for DNA barcoding in cycads does not coincide with the ‘core barcode’ of chloroplast markers (matK +rbcL) recently proposed for universal use in the plant kingdom.

Keywords: Plant DNA barcoding, Ceratozamia, Dioon, Zamia, Mexican cycads, taxonspecific DNA barcode combinations, character attribute organization system (CAOS)

120

Holding approximately one sixth of the total species number for the order in the Neotropics, Mexico is one of the three main centers of biological diversity in Cycadales, one of four groups of extant gymnosperms (Norstog and Nicholls, 1997; Vovides et al., 2007). As in other important seed plant groups, e. g. angiosperms (Mathews, 2009 and references therein), the systematics of cycads has advanced greatly during the present decade, especially through the use of a large number of DNA sequences with proven value for the reconstruction of phylogenetic relationships (Treutlein and Wink, 2002; Hill et al., 2003; Bogler and Francisco-Ortega, 2004; Caputo et al., 2004; Rai et al., 2004; Chaw et al., 2005; Zgurski et al., 2008). These molecular data sets have added to a relatively limited number of morphological characters, traditionally established as the basis of intergeneric and/or intraspecific classification (Stevenson, 1990, 1992). The description of new species of cycads from Mexico has also experienced notable progress recently (Schutzman and Vovides, 1998; Vovides et al., 2008a, b), but aside from a few exceptional cases (e. g. Nicolalde-Morejón et al., 2009a); the use of molecular information has not played a prominent role in the proposal of such taxonomic hypotheses. In parallel to developments in the molecular systematics of cycads and seed plants in general (Mathews, 2009 and references therein), the use of DNA as a source of evidence in comparative biology has been recently extended outside the realm of phylogenetic inference per se, bringing important changes in the disciplinary relationship between molecular biology/genomics, bioinformatics, and alpha-taxonomy. Central to these changes has been the explicit suggestion of the direct use of selected genomic regions as ‘DNA barcodes’, in close analogy to the way in which unique combinations of variable width vertical marks work for the identification of industrialized goods in commerce, etc. (Hebert et al. 2003a; Stoeckle and Hebert, 2008).

121

The proposal to employ molecular barcodes for large-scale, species-level identification in varied contexts, ranging from basic taxonomic research to clinical, agricultural, or even recreative applications (Stoeckle and Hebert, 2008), straightforwardly rests on the claim that certain regions in animal genomes behave in a practically invariant manner within members (i. e. individuals) belonging to the same species, and simultaneously vary to a clearly detectable level between species (Hebert et al., 2003a, b; for a survey of molecular systematic antecedents of this idea, see Meier, 2008). Animal DNA barcoding quickly stabilized around phenetic analysis of orthologs of a ‘single-locus barcode’, constituted by approximately 650 bp of the mitochondrial cytochrome oxidase I (COI) coding region, for which primers were designed with high PCR amplification success in a large number of animal species. In a brief period of time, several cases of reliable indication of taxonomic pertinence at the species level using this short sequence fragment accumulated in selected vertebrate and invertebrate taxa (Hebert et al., 2004; Monaghan et al., 2005; Ward et al., 2005, Hajibabaei et al., 2006; Smith et al., 2007, 2008). Nevertheless, COI-based animal DNA barcoding has not passed unquestioned by some molecular systematists (see, for instance, Brower, 2006; Cognato and Sun, 2007), a few of which have stated that the whole DNA barcoding enterprise is conceptually flawed (e. g. Ebach and Holdrege, 2005; Wheeler, 2004, 2005; Will and Rubinoff, 2004; for some of the responses made to these statements, see Gregory, 2005; Packer et al., 2009). Although harsh criticisms of DNA barcoding as a research program in plant taxonomy have been raised (e. g. Seberg et al., 2003; Spooner, 2009), a sector of the botanical community has embraced the DNA barcoding initiative with more sympathy (e. g. Chase et al., 2005; Cowan et al., 2006). As a result of a few preliminary studies and associated discussions, some of them held at international conferences (see Pennisi,

122

2007); molecular biology-oriented plant systematists decided collectively that the chloroplast genome should provide a major proportion of plant DNA barcoding data. However, botanists heavily involved in selecting the ‘definitive set’ of plant DNA barcoding regions have had a difficult time agreeing upon which single locus, or combination of loci, might perform best in the largest number of groups. Their mutual disagreements have been, in fact, evident in a number of proposals to settle the issue (Kress et al., 2005; Cowan et al., 2006; Chase et al., 2007; Kress and Erickson, 2007; Fazekas et al., 2008; Lahaye et al., 2008; Ford et al., 2009). In keeping with their largely positive attitude towards DNA barcoding, though, the central point of discussion in these papers has been how much sequence data –and from how many chloroplast (or nuclear, in some special cases such as non-green mycoheterotrophs) genome regions– should be collected in order to successfully carry out rapid, cheap and reliable molecular identification of plant species in the widest possible plant diversity. As a result of an evaluation of the accumulated evidence so far, a consensus has been reached this year by a multinational assemblage of plant DNA barcoding researchers, who settled for a two-locus ‘standard’ or ‘core’ barcode composed of two fragments of easy PCR amplification within the maturase K (matK) and the large subunit of the ribulose 1,5-bisphosphate carboxylase oxygenase (rbcL) loci, both of them chloroplast coding regions with a long history of success in plant molecular systematics (CBOL Plant Working Group, 2009). According to the proposal of this group of botanists, the core plant DNA barcode could be supplemented by a group of three non-coding regions, also from the chloroplast genome –atpF-atpH, psbK-psbI and trnH-psbA– all of which had been considered (either separately or jointly) in several of the DNA barcoding studies preceding the consensus on the standard barcode.

123

While fully recognizing that the matK+rbcL pair will now be recognized as the standard, universal barcode in land plants, in the present plant DNA barcoding study we have decided to address the discussion topic that was at the base of the consensus decision taken by the CBOL Plant Working Group. In the language chosen by the CBOL Plant Working Group, this topic amounts to the definition of three straightforward criteria: (i) universality; (ii) sequence quality and coverage; and (iii) discrimination. Our study focused on the three genera of cycads that occur in Mexico, Ceratozamia Brongn., Dioon Lindl., and Zamia L., and involved comparison of the performance of several different genome regions, all of them potentially useful for unique species identification in plants by CBOL standards. In a manner similar to the multinational group, we have explicitly considered the proposals resulting from the Second International Barcode of Life Conference (held in Taipei, 2007; see Pennisi, 2007) as well as relevant previous work on plant DNA barcoding, including taxonomically restricted studies in selected plant genera and/or familes (Kress et al., 2005; Cowan et al., 2006; Chase et al., 2007; Little and Stevenson, 2007; Kress and Erickson, 2007; Fazekas et al., 2008; Lahaye et al., 2008; Ford et al., 2009; Seberg and Petersen, 2009). In contrast to the CBOL Plant Working Group, though, we have also followed the pioneer DNA barcoding study conducted by Sass et al. (2007) on a set of selected cycad species representative from all biogeographic centers of diversity. As a result of all of the above considerations, the definitive group of regions assayed here includes four chloroplast coding loci (the genes matK, rpoC1, rpoB and rbcL), three non-coding intergenic spacer regions from the same plastid genomic compartment (atpF/H, psbK/I and trnH-psbA) and, finally, a non-plastid genome region, the nuclear ribosomal internal transcribed spacer (ITS).

124

Initial processing of the resulting data sets included standard phenetic tests (e. g. neighbor-joining phenogram construction), following previously published DNA barcoding work (e. g. Hebert et al., 2003). However, for advanced data analysis, we have employed the recently proposed “Character Attributes Organization System” approach (CAOS; Sarkar et al., 2008), a character-based approximation to DNA barcoding implemented in the software package of the same name. To our knowledge, this is the first report in which a plant DNA barcoding data set has been analyzed under these methodological assumptions (for examples of CAOS analyses with animal DNA barcoding data, see Kelly et al., 2007; Rach et al. 2008; and Naro-Maciel et al., 2009). Under the CAOS analytical regime, which basically looks for diagnostic characters instead of relying on phenetic distance measurements, we have calculated that psbK/I, one of the non-coding chloroplast intergenic segments, provide the highest number of ‘DNA diagnostic’ sites overall. We have also found that a three-gene combination with 100% amplification success, involving two non-coding and one coding regions – namely, atpF/H+psbK/I+rpoC1– from our selected chloroplast working set, allows 79 and 67 percent unique species identification in the genera Dioon and Zamia, respectively. For Ceratozamia, however, >70 percent (78%) identification could only be achieved through the addition of two regions (matK and ITS) to the two-gene group that worked adequately in Dioon, whereas for Zamia, addition of a fourth genomic element (the nuclear ITS) was needed to increase the identification percentage above 70 percent (up to 75%). We discuss our findings in the light of recent observations, also made in a botanical, taxonomically restricted context, which point out to a potential ‘intrinsic limit to resolution’ for low-number combinations of candidate DNA barcoding regions in plants (Seberg and Petersen, 2009). However, we additionally observe that even considering this explicit acknowledgment of potential boundaries of reliable

125

identification in plant DNA barcoding, the optimal gene set of DNA barcodes in two of the three Mexican cycad genera studied here does not coincide at all with the matK+rbcL core barcode, nor does it match any of the three-gene combinations entertained for implementing DNA barcoding in plant species in the relevant publications previous to the CBOL Plant Working Group consensus. We therefore suggest that ‘taxonomically local’ combinations of plant genome loci for molecular species identification, which work best for relatively restricted phylogenetic assemblages, should be seriously considered besides the ‘2+3’ standard chloroplast set of barcoding regions. In our opinion, this alternative needs to be entertained if molecular reference databases for species identification are to be successfully applied in either floristic, conservation biology, or strictly taxonomic-nomenclatural research contexts, particularly in certain Neotropical plant groups.

Materials and Methods

Sampling of biological materials

We collected leaf samples from all Mexican cycad species known to date from the three genera that occur in Mexico –Ceratozamia, Dioon y Zamia– as published in the World Cycad List by Hill et al. (2007), plus three new species recently published (see Vovides et al., 2008a, b; Nicolalde-Morejón et al., 2009a). We also collected leaf material from at least one individual from each non-Mexican cycad genus (Table 2). All materials were obtained from living plants included in the National Cycad Collection at the Jardín Botánico ‘Francisco Javier Clavijero’ (JBC), which houses materials from the Mexican species as well specimens in cultivation from several

126

countries, covering the entire order Cycadales. We also performed ex profeso field collections to complement the set of sampled materials for the present study. Leaf tissue from Chigua restrepoi D. W. Strev., Zamia standleyi Schutzman, Z. tuerchkeimii Donn. Sm. and Z. prasina W. Bull was kindly donated by the Montgomery Botanical Center (MBC).

Leaf genomic DNA extraction and PCR amplification (including DNA sequencing)

With the exception of the leaf samples transported from the field to the lab, fresh materials were always used for the total leaf genomic DNA extractions of material collected at the greenhouses of the JBC. For the extractions, we used either the DNAeasy Plant Mini kit (QIAGEN) or a user-tailored protocol based on a widely known CTAB DNA extraction procedure (Doyle & Doyle, 1987). PCR amplification and automated sequencing includes all loci proposed at the Second International Barcode of Life conference (Taipei, 2007; see Pennisi, 2007; Fazekas et al., 2008; Ford et al., 2009 and CBOL, 2009). Nucleotidic variability in four chloroplast coding regions (the genes matK, rpoC1, rpoB and rbcL), three non-coding intergenic spacer regions from the same plastid genomic compartment (atpF-atpH, psbK-psbI and trnHpsbA) and the nuclear ribosomal internal transcribed spacer (ITS) as a complement nonchloroplast locus (see Table 1 for primers used in each type of PCR reaction) was evaluated here (see Table 2 for an overview of amplification success for each gene assayed for each species). Polymerase chain reaction (PCR) amplification experiments were performed as reported in recent plant DNA barcoding publications (e. g. Sass et al., 2007). Amplification products were visualized through gel electrophoresis in 1% agarose gels

127

stained with ethidium bromide. In all cases where single bands were clearly detected, PCR products were directly purified using QIAquick® PCR Purification Kit (QIAGEN). Automated sequencing was carried out in Macrogen (South Korea; http://dna.macrogen.com).

Sequence analysis

Electropherograms were edited and contigs were assembled using the software program Sequencher 4.8 (Gene Codes Corp., Ann Arbor, Michigan, USA). Sequences were aligned in BioEdit 7.0.9 (Hall, 1999), through its implementation of the Clustal X (Thompson et al., 1997) multiple alignment mode. Alignments were imported into MacClade (Sinauer Associates, Sunderland, Massachusetts, USA), further edited by eye, and saved in Nexus format for ulterior character analysis.

Character-based analysis/identification of ‘DNA diagnostics’ and determination of DNA barcodes in Mexican cycad species

Neighbour-joining (NJ) guide trees were estimated for each matrix/amplified locus, using a Kimura-2-parameter distances model in PAUPv4.0b10 (Swofford, 2002). These trees were stored in Nexus format and edited in MacClade. The tree topology that was ultimately selected and used in the “character attributes organization system” analyses corresponds to Stevenson’s 1992 previously published phylogenetic hypothesis for the cycads. Program P-GNOME from the CAOS software package (Sarkar et al., 2008) was executed according to the authors’ instructions (http://sarkarlab.mbl.edu/CAOS). Actual determination of DNA diagnostics involved the manual revision of the “CAOS-

128

attribute file” and “CAOS-group file” archives generated by P-GNOME. Only characters (‘attributes’) with confidence value of 1.00 were selected. Corroboration of these attributes was performed by eye, observing and comparing the information of the “CAOS-group file” archives with the original, MacClade-edited matrices.

Results and discussion

A DNA barcode for land plants: difficult roads toward a consensus

Although still unexplored in its full extent for a wide array of taxa, analyses of molecular evolutionary processes taking place in certain plant mitochondrial genomes indicate that the cytochrome oxidase I (COI) coding region, which has been adopted by consensus as the ‘universal animal DNA barcode’ in animals (Hebert et al., 2003a, b; Stoeckle and Hebert, 2008), is not suitable for analogous use in plants (Adams and Palmer, 2003; Spooner, 2009). A central interest in the international effort to develop DNA barcoding in plants has therefore been directed to find a different individual locus, or a combination of loci, that could fulfill the set of pragmatic criteria that during the early stages of this international research initiative have been recognized to constitute the mark of an adequate DNA barcode (for a concise description of these criteria, see Sass et al., 2007; Ford et al., 2009; for a description of their mature version, see CBOL Plant Working Group, 2009). The comparative performance of several different combinations of loci in plant genomes –particularly, from the chloroplast– has correspondingly received special attention, and the selection of the best combination among these has been discussed in recent meetings specifically devoted to plant DNA barcoding initiatives, as well as in

129

recent publications derived from these meetings (see, for instance, Pennisi, 2007; Fazekas et al., 2008; Ford et al., 2009). Only recently, an international consensus on a standard botanical DNA barcode has been reached (CBOL Plant Working Group, 2009). It is worth noting, though, that some prominent plant DNA barcoding studies initially promoted the idea that empirical evidence was enough to support either psbAtrnH, a non-coding chloroplast region of about 400 base pairs (Kress et al., 2005; Kress and Erickson, 2007) or an N-terminal approximately 800 base pairs-long segment of the coding region for the maturase K gene (matK), also from the plastid genome (the latter claim was made as recently as last year; Lahaye et al., 2008), as sufficiently good candidates to achieve the status of individual, standard plant DNA barcodes. The fact that the final decision on a ‘core’ and a supplementary set of DNA barcodes ultimately involved a total of five (two coding and three non-coding) regions from the chloroplast genome, clearly shows that the early hopes to reach a botanical analog of the ‘singlelocus DNA barcode’ scheme that is accepted in the zoological community were unwarranted.

DNA barcoding in the cycads redux I: surprising results under the assumptions of a character-based method

Within the context of exploratory work in different plant groups carried out in order to select a universal set of plant DNA barcodes, the order Cycadales has occupied a visible place (Sass et al., 2007; Little & Stevenson, 2007). This situation is not surprising; in taxonomic terms, cycads are particularly well suited for exhaustive DNA barcoding due to the relatively small number of genera and species that constitute the group (Norstog and Nicholls, 1997; Hill et al., 2007). At the same time, cycad

130

specialists have already realized that DNA barcoding data collected for all genera might potentially contribute to molecular systematic studies at that taxonomic level. At present, a reliable phylogenetic classification of cycad genera is not available, despite the growing number of works devoted to the subject in recent years (Treutlein and Wink, 2002; Hill et al., 2003; Rai et al., 2004; Chaw et al., 2005; Zgurski et al., 2008). The genus Dioon, arguably the most difficult taxon to place in a phylogenetic context within the order as a whole (Hill et al., 2003), is a peculiar case in this respect within the context of the Mexican cycads not only because it is basically endemic to the Mexico, but also because of the potential that molecular matrices of chloroplast DNA barcode candidates might have as sources of informative characters in cladistic analyses (Nicolalde-Morejón et al., unpublished observations). Given that the trnH-psbA region was the first chloroplast locus to be suggested as a universal DNA barcode in plants (Kress et al., 2005; Kress and Eriksson, 2008), during the development of this study we were particularly interested in the degree of nucleotidic variability at the species level that this chloroplast intergenic spacer could show in the three cycad genera that occur in Mexico. Previously, Sass et al. (2007) had already found that, with the exception of Cycas, all amplifications of cycad genomic DNA with the primers suggested by Kress et al. (2005) for this region yielded two distinct bands, even when the annealing temperature is raised to 62°C. We replicated this result for all the leaf genomic DNA samples tested. Working with the smallest of these bands in all cases where amplification was successful (all but one Dioon species, several Ceratozamia and Zamia species, and all outgroups; see Table 2), which corresponds to the amplification that was not sequenced by Sass et al. (2007), we have unequivocally observed that trnH-psbA is not variable between species in either Ceratozamia, Dioon, or Zamia. This result is compatible with the conclusion reached

131

by Sass et al. (2007) on the basis of the species-level variability of the larger band in at least one species per genus in the Cycadales. However, we noticed that the distribution of four indels varies consistently between genera (data not shown), and therefore might be useful for systematic purposes at the genus level. We specifically suggest that characters corresponding to indels in the trnH-psbA region sequenced in this paper might contribute with information to clarify the phylogenetic position of Dioon, possibly the most contentious subject in current cycad molecular systematics (Hill et al., 2003; Bogler and Francisco-Ortega, 2004; Rai et al., 2004; Chaw et al., 2005; Zgurski et al., 2008). However, our results do not lend support to the status of ‘potential’ core DNA barcode that this non-coding chloroplast region still had just prior to the selection of the two-locus standard plant DNA barcode (CBOL Plant Working Group, 2009: 12795). The chloroplast coding region for the matK and rbcL genes, both of them wellestablished as important sources of characters for molecular systematics in angiosperms, are two loci whose variability we also decided to explore in detail. Our decision was evidently taken in the face of recent claims of the relative superiority of the former as a DNA barcode in plants (Lahaye et al., 2008), and the ultimate selection of both loci as the two core DNA barcoding regions (CBOL Plant Working Group). It is important to notice, though, that rbcL had already been discarded as a DNA barcoding region by Sass et al. (2007) for not complying with basic reproducibility criteria. With an interest in checking if that negative result could be reverted, a random sample of five Ceratozamia, six Dioon and five Zamia species was selected, for which we obtained complete amplification success. This result was, however, again linked to an absolute lack of variability at the nucleotide level within genera (data not shown), leading us to confirm the judgement of unsuitability for rbcL as a DNA barcoding region in cycads.

132

In contrast to rbcL, Sass et al. (2007) had not ignored matK in their cycad DNA barcoding study, but the region was not considered beyond Step 2 (i. e. testing of selected primer pairs) of their optimized flowchart. The elimination of matK for further testing in that study was due to failure of amplification in selected species of eight cycad genera (Ceratozamia, Chigua D. W. Stev., Dioon, Encephalartos Lehm., Lepidozamia Regel, Macrozamia Miq., Microcycas (Miq.) A. DC. and Stangeria T. Moore) and to the fact that in the remaining two genera, Cycas L. and Zamia, amplification was only partially successful (products were not obtained in Cycas platyphylla K. D. Hill and Zamia variegata Warsz., one out of three species tested for each genus). In our hands, it was also impossible to obtain good quality amplifications of matK in the genus Dioon, and only a few Zamia samples behaved successfully as templates in the PCR reactions (Table 2). In contrast, leaf genomic DNA from all the species of Ceratozamia, tested with the set of primers that were not tested by Sass et al. (2007), supported sufficient product amplification with high associated quality of sequences. Despite the partial amplification success, matK data from Zamia species was not variable enough to be further considered useful for DNA barcoding purposes (data not shown). On the other hand, analysis of this Ceratozamia matrix with our preferred analytical regime, the “characteristic attributes organization system” (CAOS) algorithm, implemented in the software of the same name (Sarkar et al., 2008), retrieved ‘DNA diagnostics’ for only five out of 24 species. Our results do not necessarily rule out the use of matK in DNA barcoding, though. It remains to be seen if the use of different universal primers for this locus, which amplify larger segments of the gene (the primers that were successful in the present study yield an approximately 800 base pairs-long, Nterminal fragment of this coding region) is more suitable due to a higher concentration

133

of CAOS informative sites/DNA diagnostics in other regions of matK. However, since this caveat could hardly apply to rbcL data, we explicitly consider inadequate to use the CBOL Plant Working Group core DNA barcode as the main source of information for DNA barcoding in cycads. The locus psbK/I is a chloroplast region that has only been included recently as part of two of the sets of candidate loci for DNA barcoding (Kim et al., 2007 (abstract at the Taipei conference); Pennisi, 2007). We selected this locus as the third region of interest in the present DNA barcoding study mainly because no information was previously available for it, neither in gymnosperms nor in cycads in particular, and also due to the fact that this locus was included in the set of non-coding regions supplementary to the core DNA barcode of the CBOL Plant Working Group. Pragmatic criteria for the selection of a good DNA barcoding region in the flowchart in Sass et al. (2007) were easily fulfilled starting with the fact that amplification success was 100% (Table 2). In view of its levels of variability for species in the three genera tested here under the assumptions of our preferred character-based analytical approach (see Figure 1, and Table 3), we propose that –contra Fazekas et al. (2009, p. e2802), who argued against its use due to “its higher failure rate in amplification and sequencing”– the psbK/I locus should be seriously considered as a candidate for inclusion in any final DNA barcoding gene combination used in cycads. Presumably, performance of this chloroplast genome region for DNA barcoding in other gymnosperms also might be acceptable. By itself, psbK/I allows variable, and not very high (i. e.