Pollination of Zamia incognita A. Lindstr. & Idárraga on a natural population in the Magdalena Medio, Colombia

Trabajo de grado para el cumplimiento parcial de los requisitos para obtener el grado de Bióloga de la Universidad CES

Estudiante:

Wendy A. Valencia-Montoya

Asesores:

Dino Tuberquia y Juliana Cardona-Duque

Programa de Biología

Facultad de Ciencias y Biotecnología

Universidad CES

2015

Al Magdalena Medio, para que la magia de sus bosques

sobreviva a la indiferencia, y siga contando historias

Agradecimientos

Deseo agradecer a mis padres por todo el apoyo y aliento que desde muy temprano me dieron en este camino de estudiar formalmente los seres vivos. Especialmente a mi mamá Gloria Montoya, por su colaboración desde el procesamiento de datos hasta aportes económicos para el trabajo de campo; y principalmente porque es la síntesis de la capacidad de asombro, que para mí es el legado clave para estudiar biología. A mi papá Gedwin Valencia por su enorme apoyo y entusiasmo hablando de las zamias a todas las personas que conoce. A mis asesores Dino Tuberquia y Juliana Cardona porque además de todo el apoyo académico durante estos años han sido un apoyo emocional muy importante, porque ellos son maestros que forman personas éticas y apasionadas. A Dino gracias por enseñarme a cerrar los ojos para ver los seres vivos y a Juliana porque es la personificación de la nobleza. A Arturo Aristizábal, amante de las zamias, porque de alguna manera él fue quién nos introdujo en la apreciación de éstas plantas. También quisiera agradecer a todos mis profesores especialmente a Maria José Sanín, por su sensibilidad, brillantez y apoyo durante la formulación del proyecto. También de manera muy especial a Cristina López-Gallego, gran estudiosa de las cícadas, por los consejos, la literatura, los insectos y las invitaciones a sus salidas de campo de zamias. Agradezco a todas las personas que me acompañaron durante el trabajo de campo: Cornelio Bota, Yeyson Montoya, Miguel González, Laura Isabel Marín, Laura Gómez Mesa, Eliana, Héctor Manuel Arango, Edwin Hurtado, Camilo Flórez y Sebastián Cifuentes. Especialmente quiero agradecer a Cornelio Bota y Yeyson Montoya quienes estuvieron en la mayor parte del trabajo de campo en el zamial, por la alegría y la música; que enmarcados en ese hermoso bosque tropical constituyen de los momentos más preciados en este trabajo de grado. A Henry Arenas-Castro, Cristina López-Gallego, Maria José Sanín, Edwin Hurtado, Alejandra Duque y Nicolás González-Castro por sus sugerencias al manuscrito. Agradezco muy especialmente a Gustavo Londoño y a Juan Luis Parra por el préstamo desinteresado de los equipos para registrar termogénesis. A la Beca Colombia Biodiversa de la Fundación Manuel Ángel Escobar por la financiación parcial de este estudio.

A Cornelio Bota y Camilo Flórez, mis dos naturalistas favoritos, porque para mí son el ejemplo vívido de la pasión por los insectos y el monte. A Sergio A. Muñoz-Gómez por su incondicional acompañamiento intelectual y emocional. A Henry Arenas-Castro, por todas las conversaciones y porque ojalá sigan siempre instaladas en lo que me es imprescindible. A Edwin Hurtado por todo su apoyo durante el tiempo de campo y escritura, y principalmente por sus matices de profundidad y ternura. A Yeyson Posada, la ceiba, por entender las zamias y regalármelas en cada ojo dibujado. A Michael Castaño, el Sol, por toda la alegría desprendida de su elocuente sensibilidad. A Duberney Chaverra por introducirme con paciencia en el maravilloso mundo de los insectos. A todos mis compañeros de la carrera: Vanessa Correa, Norman Echavarría, Elisa Lotero, Laura Gómez, Alejandra Duque, Natalia Duque, Valeria Zapata, Juan Diego Ospina, Lina Bolívar, Camila García, Tomás Vallejo, Pedro Gómez, Esteban Urrea, Hana Londoño y Paula Saravia por el acompañamiento tan importante durante el pregrado.

A mis amigos Yenny Correa, Isabel Restrepo, Laura Muñoz, Camila Londoño, Lorena Quintero, Carolina Torres, Eliana Pineda, Juliana Herrera, Juliana Cerón, Sheela Turbek, Nancy Chen, Matt Dickinson, Héctor Arango, Yeyson Posada, David Taborda, Daniel Monsalve, Alejandra Clavijo, Blanca Arbeláez, Sebastián Cifuentes, Laura Sierra, Manuel Sánchez, Alejandro Ospina, Giovanny Pérez, Mauricio Zapata, Alejandro Ríos, Juan Guillermo Mesa, Juan David Sánchez, Alejandra Barrada, Liliana Palacio, Samuel Monsalve y David Mechas. También a Javier Cardona, David Cardona, Dolly Montoya, Alejandra Cardona, Rosario Giraldo, Celia Ramos, Carlos Montoya, Eliana Moyano, Paulina Valencia, Sandra Ortiz, Gloria Quintero, Juan Cardona y Margarita Cardona y todos los tíos y primos que se me escapan. A Juanita mi gata, lo más querido, porque este trabajo de grado fue escrito en compañía de ella y de Pala.

A todos los integrantes del Grupo Abarco, Tejuntas y al Colectivo Bidikay por soñar conjuntamente con un país que tenga una mirada más sensible de su rica diversidad biológica y cultural.

Finalmente deseo agradecer a los habitantes de la vereda Santa Bárbara, especialmente a Doña Alba y sus hijas Darsy y Yireth, por recibirnos siempre con tanto cariño. A Don Arturo quién nos transportó hasta la vereda, y muy especialmente a Miguel González por ser un apasionado de la diversidad y la conservación.

Medellín, 1-VI-2015

Manuscript for Submission to Arthropod-Plant Interactions (Original Version)

Research Article

Pollination of Zamia incognita A. Lindstr. & Idárraga on a natural population in the Magdalena Medio, Colombia

WENDY A. VALENCIA-MONTOYA1, DINO TUBERQUIA1, 2 & JULIANA CARDONA- DUQUE1, 2

1Programa de Biología, Universidad CES-EIA, Medellín, Colombia

Calle 10 A No. 22 - 04

2 Grupo Biología CES-EIA

Correspondence to: Wendy A. Valencia-Montoya.

E-mail: [email protected]

Telephone number: (+57) (4) 444 0555 Ext. 1240

Fax number: (+57) (4) 3113505

Abstract: The genus Zamia (Zamiaceae: Cycadales) holds its greatest diversity in Colombia and most of its species are highly threatened by different factors. One of the most relevant and frequently ignored aspects for the establishment of effective conservation programs, is its highly specialized reproductive biology. Despite the importance of pollination for the viability of the Zamia populations, there are no studies about the pollination process for these cycads in Colombia. Herein we describe the pollination process of Zamia incognita A. Lindstr. & Idárraga, in a natural population from the Magdalena Medio valley in Colombia. We found Pharaxonotha beetles in the male cones where they complete all phases of their lifecycle. Cones produce heat, which generally follows a circadian pattern and its magnitude and extent, are positively correlated with the elongation of the cones and pollen shedding; therefore, the increment in cones temperature seems to play an important role in attracting the beetles to the cones. By following marked beetles with fluorescent dyes as well as by direct observations of the beetles on and into female cones, remaining at pollination droplets for a long period, we confirmed that Pharaxonotha sp. is the effective pollinator of Zamia incognita. We suggest that the pollination droplets may serve as reward to the pollinators. We also discuss the relationships of this Zamia species with other insects like Eumaeus butterflies, Atta ants, flies and Meliponinae bees.

Key words: cycad pollination, Erotylidae, plant thermogenesis, plant-insect interactions.

Introduction

The pollination in Zamiaceae was considered to be anemopholus until the 80’s end. However, this was a misconception due to a generalization of most Gymnosperms condition, and it has been reassessed by extensive evidence of insect pollination (e.g. Donaldson 1997, Norstog et al. 1986, Suinyuy et al. 2009, Tang 1987a, Terry 2001, Terry et al. 2005, Vovides 1991, Vovides et al. 1997, Wilson 2002). During most of the XX century, entomophilous pollination was thought to be restricted to Angiosperms, and it has been considered as a key interaction that had played a pivotal role in promoting the overwhelming diversity of the interacting groups (Gorelick 2001, Norstog 1987). Pioneering studies in Zamia pollination challenged this paradigm (e.g. Norstog et al. 1986, Tang 1987a) and now we know that pollination in all the Zamiaceae genera relied on highly specialized relationships with insects, most of them beetles (Donaldson 1997, Norstog et al. 1986, Suinyuy et al. 2009, Tang 1987a, Terry 2001, Terry et al. 2005, Vovides 1991, Wilson 2002) and secondarily thrips (Terry 2001).

The cycads exhibit an array of complex traits to attract insects: the female and male cones have differential attractants for feeding preferences; the pollinator life-cycles are coupled with plant phenology; and they volatilize their odors and have thermogenesis (Donaldson 1997, Norstog and Fawcett 1989). These features are shared with some beetle-pollinated angiosperm families (e.g. Cyclanthaceae: Eriksson 1994; Annonaceae: Gottsberger 1999; Arecaceae: Henderson 1986) and they are considered evolutionary convergences related to cantharophily (Norstog 1987, Terry et al. 2005). Due to the similarity of these cycad-insects interactions across different continents, and because most of the pollinator species belong to the not related beetle superfamilies Curculionoidea and Cucujoidea, it is now believed that the insect pollination has been independently evolved in different Zamiaceae genera (Suinyuy et al. 2009, Terry et al. 2004, 2005, Wilson 2002).

These specialized associations indeed have important implications in cycad species conservation, but the pollination process has been frequently undervalued in cycad conservation programs, in part due to the scarcity of information. Both insects and cycads are unable of long-term survival in nature by themselves (Norstog et al. 1986, Stevenson et al. 1998), and the overall specialists that depend on a few species are more vulnerable to extinction than generalist species (Bond 1994, Terry et al. 2004, 2005, Vovides et al. 1997). Despite the crucial role that pollination mutualism plays in conservation and its evolutionary significance, little is known about the pollination systems in cycads. Within the genus Zamia the pollination process has been only studied in Zamia furfuracea Aiton (i.e. Norstog and Fawcett 1989, Norstog et al. 1986) and Zamia pumila L. (i.e. Tang 1987a), species pollinated by beetles of the genera Rhopalotria Chevrolat (Belidae) and Pharaxonotha Reitter (Erotylidae).

The greatest diversity of the genus Zamia is in Colombia (Haynes 2009, and additional recently described species in Calonje et al. 2010, 2011, Galeano et al. 2005, Lindström and Idárraga 2009), and most of its species are highly threatened by habitat degradation and destruction (Galeano et al. 2005). In spite of the significance of the tight relationship with insects for the Zamia population’s viability, there is not any study about the pollination process for the Colombian cycads and their pollinators are entirely unknown. Herein we describe the pollination process in a wild population of Zamia incognita A. Lindstr. & Idárraga, a recently described species endemic to Magdalena Medio valley in Colombia. We present a morphological description of the larva, pupa and adult of the pollinator, and we included information of the cones thermogenesis, natural history, and interactions with other insects.

Material and Methods

Study system: Zamia incognita A. Lindstr. & Idárraga occurs in few of the Magdalena Medio valley rainforests, mainly over karst rocks. The species is diagnosed by the broad, obovate, glossy leaflets, and short obovate megastrobilus with indistinct peduncle (Lindström and Idárraga 2009). Cones emergency usually takes place between November and January (Lindström and Idárraga 2009). Its estimated population size comprises more than 4000 individuals and natural regeneration has been recorded (Aristizábal et al. 2011). This species is considered threatened (VU) (Lindström and Idárraga 2009), although it has not been officially categorized according to IUCN criteria.

Field studies were carried out on a natural population located at the Distrito de Manejo Integrado de los Recursos Naturales (DMI) “Cañón del Río Alicante”, which is located in jurisdiction of Maceo, Puerto Berrío and Yolombó municipalities (Antioquia), approximately at 06°33’51.2’’N; 74°54’38.02’’W, 500 m. The life zone is Tropical wet forest (T-wf), with karstic formations mostly of limestone, where occur endemic fauna and flora in the forests associated to the caves (Cáceres et al. 2010). Observations on this Zamia incognita population were conducted between 2010 and 2015, although systematic experiments and observations were carried out between August 2014 and April 2015. Cones development: Male and female cones were marked to follow their development. Cones height and diameter were measured approximately every 30 days in October, November and December of 2014. Descriptions of cones in different developmental stages found in the population were also made, and photographs were taken. All this information was related with other information about insect visitors and thermogenesis.

Insect visitors, behavior and pollinator abundance: Insect visitors and their behaviors were monitored during the morning (08:00- 12:00), afternoon (13:00 – 17:00), and evening (19:30-23:00). The insects found within the cones, were collected for posterior identification. These insects were collected in 70% Ethanol, and deposited in the Colecciones Biológicas de la Universidad CES (CBUCES; national record of collections 209). The kind of cone (male or female), as well as development stage, were recorded. To address the variation in the abundance of visitors/pollinators among different developmental stages of male cones, we collect two cones for each microstrobilus stage: pre-dehiscent (before pollen shedding), dehiscent (pollen release), and post-dehiscent (after pollen shedding). We dissected the cones and recorded the number of specimens and life-cycle stage (larva, pupa or adult) of the insects, for each male cone.

Dispersion Experiments: The pollen dispersion distance and the ability of insects to disperse the pollen were determined by marking pollen-shedding male cones with fluorescent dyes (Fig. 1A,1B) (Kearns and Inouye 1993). Eleven randomly selected dehiscent male cones in the zamia population (resource cones) were marked with different colors: two green, one violet, three yellow, two red, one aqua, one blue, and one orange. Male cones from the same plant or neighbor one (less than 0.2 m) were marked with the same color. To find the tracks left by dye-coated beetles, female and male cones in different stages were examined under UV light, looking for the dyes presence during four nights. When dye traces were found in male or female cones, the straight distance from the resource cone was measured. The cone developmental stage was also recorded and its height and diameter were measured.

Thermogenesis: Temperatures of 17 female and male cones were monitored for detecting changes in temperature during the day. On November, the male cones were selected randomly for taking data and the developmental stages were not conspicuously distinguishable, then it was not easy to assign them to any of the pre-dehiscent, dehiscent or post-dehiscent stage. For female cones, only the receptive ones were measured. On December, the cones were selected at three stages of development: pre-dehiscent, dehiscent, and post-dehiscent for male cones; and for female cones, pre-receptive (the cone is relatively small, closed and megasporophylls are slightly marked; in general, for Gymnosperms this stage is before seminal primordia are completely developed), receptive (female cone is open, with an upper crack exposing pollination drops), and post receptive (the cone is closed; in general, for Gymnosperms this stage pollination tube develops, fecundation occurs and seed complete its development). Temperature sensors were located within the basal third of the central axis of male cones (Fig. 1C), and above the second line of megasporophylls in female cones (Fig. 1D). The sensors, attached to Onset HOBO U12-006 4-Channel Dataloggers, recorded the temperature each minute during three days in November and five days in December 2014. The ambient temperatures (at shade) near to each cone, were also measured. Data were analyzed and plotted in R package software (R Core Team 2015). Thermogenesis periods were defined as a persistent rise in cone temperature >1.5°C relative to ambient temperature.

Fig. 1 Methodology. a Male cone marked with green fluorescent dye. b Male cone marked with violet fluorescent dye. c Male cone showing the position of the temperature sensor. d Female cone showing the position of the temperature sensor.

Systematics: Collected beetles were identified to genus following the Leschen and Skelley (2002) and Leschen (2003) taxonomic keys. Observations of external and internal morphology were made using a Nikon SMZ 745 stereomicroscope (magnification: 5.6x-50x) and Nikon Eclipse E200 compound microscope (magnification: 40-400X); dissections were made to observe internal characters like genitalia and mouthparts. Because the beetle species seems to be an undescribed species, its morphological features are being compared with types, original descriptions and it is also being corroborated by Dr. Paul Skelley. A morphological description is being prepared (Valencia- Montoya et al. in prep.), following Franz and Skelley (2008) and Chaves and Genaro (2005).

Results

Cones development

Male cone developmental stages in Zamia incognita can be clearly distinguished by single features associated with the pollination process (Fig. 2A-C). In general, there are more than one male cone per plant and they mature sequentially (Fig. 3A-C). A single adult plant with several branches can have until 20 male cones. Pre-dehiscent male cones are pale-yellowish colored and they delayed approximately 16 weeks to start microsparangia dehiscence (Fig. 2A). When pollen shed begins, male cones elongated from the base to the apex, opening in the same direction, and microsporophylls are separated exposing microsporangia. Dehiscent male cones are slightly reddish colored and they take approximately between 1-2 weeks, since elongation starts until male cone falls (Fig. 2B). Late dehiscent and post-dehiscent male cones are cream-brownish to grey colored (Fig. 2C).

Fig. 2 Cones developmental stages. Male cones: a Pre-dehiscent. b Dehiscent. c Post-dehiscent. Female cones: d Pre-receptive. e Receptive. f Post-receptive. Female plants generally have a single cone, and much less usually two or three. In the pre- receptive female cones, apical megasporophylls are slightly marked (Fig. 2D). When the female cone became receptive to pollen, the upper portion of the cone splits leaving a surrounding conspicuous fissure (Fig. 2E). Female cone receptivity lasts approximately 2-4 days. Because receptivity is a short stage, may be hard to distinguish between late pre-receptive and early post-receptive stages (Fig. 2F).

Fig. 3 Sequential maturation of male cones in a single plant. a October 2014: cones 1 and 2 are in dehiscent phase, while cones 3, 4 and 5 are in pre-dehiscent phase, and cones 6 and 7 had not developed yet. b November 2014: cones 1 and 2 had fallen, cone 3 is in dehiscent phase, and cones 4, 5, 6 and 7 are in pre- dehiscent phase. c December 2014: cones 3 and 4 had also fallen, cone 5 is in dehiscent phase, and cones 6 and 7 are in pre-dehiscent phase.

Insect visitors, behavior and pollinator abundance

The most common insects found in the Zamia incognita cones during reproductive period are listed in table 1. These insects include those that feed and breed on the cones, as well as those that apparently make only brief visits to the cones. Insects were found more frequently in male cones in different developmental stages. Adults of Eumaeus cf. godartii (Boisduval, 1870) (Fig. 4) visit pre-dehiscent male cones and oviposit on microsporophylls (Fig. 4A). When the larvae emerge they feed on microsporophylls tissues and pre-dehiscent male cone central axis making the cone non-viable (Fig. 4B). Beetles remains in male cones until they falls. The last instar larvae are more abundant in fallen cones where they feed (mainly in cone central axis) and make them hollow. Ants of the genus Atta were observed on pre-dehiscent and early dehiscent male cones cutting and carrying microsporophylls (Fig. 5A) and on new leaves (Fig. 5B, 5C); they were also observed on mature female cones while removing the seed sarcotesta, apparently not touching the megagametophyte (see discussion below), the parenchyma of the cone central axis and the megasporophyll axis (Fig. 5D, 5E).

An erotylid beetle of Pharaxonotha genus was the most abundant insect found on dehiscent male cones (Table 1). We found them in every examined mature male cone, and all the life-cycle stages of Pharaxonotha were found associated with Z. incognita male cones (Fig. 6), excepting the eggs, that were undistinguishable. Adults and larvae of Pharaxonotha sp. occur when the cones begin the elongation period, just prior to pollen shedding, and keep onto or inside the cones until the decaying period, after pollen shedding is completed. Adults feed pollen and they usually hide within the spaces between the sporophylls of male cones, and they quickly scurry away or drop off when disturbed. Beetles mated on microsporophylls, and fights between males (presumable for females) were commonly observed. Larvae feed on microsporophylls and on the parenchyma of the central cone axis. Pupae were found within microsporophylls and central axis of fallen male cones. Adults are more abundant in dehiscent male cones, and larvae in post-dehiscent ones (Table 2). Pharaxonotha sp. was the unique insect found in both male and receptive female cones (Fig. 7A). The number of Pharaxonotha sp. individuals found in receptive female cones, was fewer than the found in male cones (Maximum count = 4 individuals). The beetles go into the receptive female cone through the apical fissure and pass the pollination drops while walking throughout the megasporophylls (Fig. 7B); we observed beetles staying on a pollination droplet up to approximately 7 minutes.

Table 1 Common insects found on Zamia incognita male and female cones during reproductive time

Insect stage Insect stage Flight or Male/Female cone found on Maximum Insect found on Feeds on movement stage female count male cone period cone Male cone: Adult: pollen Dehiscent 974 adults in Coleoptera: Erotylidae: Larva, Larva: male Female cone: Adult Day and Night a dehiscent Pharaxonotha sp. pupa, adult sporangia tissue, Receptive male cone male cone rachis

Lepidoptera: 13 larvae in a Eggs, Larva: male Lycaenidae: Eumaeus Pre-dehiscent male Not found Day and Night pre-dehiscent Larva sporophyll tissue cf. godartii cone

4 adults in Hymenoptera: Apidae: Post-dehiscent Adult Not found Pollen Day only post-dehiscent Meliponinae sp. 1 male cone cone 3 adults in Hymenoptera: Apidae: Post-dehiscent Adult Not found Pollen Day only post-dehiscent Meliponinae sp. 2 male cone cone Adult in male cone: sporophyll 13 individuals Pre-dehiscent male tissue Formicidae: in pre- Seed shedding Adult Adult Adult in female Day only Atta sp. 1 dehiscent female cone: seed male cone sarcotesta, female cone rachis Post-dehiscent Larva, Larva: male cone Diptera Not found Unknown Unknown male pupa, adult rachis

Fig. 4 Eumaeus cf. godartii. a Eggs on upper part of a male cone. b Larvae eating on male cone. c Larva eating Zamia leaf. d Pupae under Zamia leaf. e Adult under Zamia leaf.

Fig. 5 Atta sp. ants. a On pre-dehiscent male cone. b and c Cutting young Zamia leaves. d and e Removing the sarcotesta of the seeds, as well as the parenchyma of the cone central axis and megasporophyll axis.

Fig. 6 Pharaxonotha sp. a Adult habitus, lateral view. b Adult on dorsal and ventral view. c Larva on dorsal and ventral view. d Pupa, ventral and lateral view.

Table 2 Abundance of Pharaxonotha sp. larvae and adults in different male cone developmental stages Male cone developmental stage Larvae Adults Pre-dehiscent 0 0 Early Dehiscent 54 974 Dehiscent 124 607 Late Dehiscent 260 636 Post-dehiscent 307 24

Fig. 7 Key aspects of reproductive process in Zamia incognita. a Pharaxonotha beetle on female cone. b Pollination drops. c Receptive female cone with dye-color marks. d Dehiscent male cone with dye-color marks.

Finally, two different bee species (Hymenoptera: Apidae: Meliponinae) were observed on the male cones: Meliponinae sp. 1 (Fig. 8A, 8C) in November 2014 and Meliponinae sp. 2 (Fig. 8B, 8D) in December 2014. Bees feed on pollen only in late dehiscent and post-dehiscent male cones, within the spaces between microsporophylls when no beetles were occupying these spaces. A fly unknown species (Diptera) was found feeding on decomposing male cones.

Fig. 8 Other insect visitors of Zamia incognita male cones, particularly bees consuming pollen. a and c Meliponinae sp. 1. b and d Meliponinae sp. 2.

Dispersion experiments

The male cones in pollen-shedding stage that were marked with fluorescent dyes, were full of Pharaxonotha beetles; these beetles were observed covered with pollen and color-dyes. The beetles foraging distances, from male source cones, were recovered in eight female cones and 11 male cones. All the recovered tracks came from dehiscent male cones, marked with yellow, green and aqua color- dyes. Traces of dyes in female cones were found only in receptive stage cone, independent of cone size. Marks on female cones were usually in the apical row of megasporophylls, near to the fissure (Fig. 7C), and were left when the beetles got into the cone (Fig. 7A). The eight female cones had tracks of dyes from two different male source cones. The maximum dispersion distance to female cones was 21.76 m and the minimum was 4.66 m. The average dispersal distance was 13.95 ± 6.85 m from dehiscent male cones to receptive female cones. The beetles also moved to other male dehiscent and late dehiscent cones (Fig. 7D). The maximum dispersion distance to male cones was 21.03 m and the minimum was 1.17 m. The average dispersal distance was 11.13 ± 7.13 m.

Thermogenesis

Data for 12 of 17 cones were included, because five temperature sensors were broken or eaten by ants. Temperature of male and female cones of Zamia incognita showed a significant elevation above ambient temperature. Dehiscent male cones and receptive female cones had the highest temperature peaks above ambient temperature, and they exhibited a similar pattern. Dehiscent male cones showed higher temperatures than receptive female cones during thermogenesis periods.

Temperature increment in dehiscent male cones generally followed a circadian pattern where both, cone and ambient temperatures, increased simultaneously (Fig. 9B). The maximum recorded temperatures coincide with the same day hour even though the measurements were taken in different months (see maximum temperatures in November – top graph, and December – bottom graph: Fig. 9A). Heat production in dehiscent male cones comprised a consistent rise in cone temperature above ambient temperature between approximately 10:30 and 15:30 hours, thereafter, male cone temperatures decrease and became more similar to the ambient (Fig. 9B). All the male cones in dehiscent developmental stage, had maximum temperatures between 12:30 and 15:30 hours. The maximum above-ambient temperature (5.35 ± 1.43°C) was observed in a dehiscent male cone beginning pollen shedding. The pre-dehiscent maximum above-ambient temperature of male cones was 1.38 ± 0.34°C and the minimum was 1.44 ± 0.34 °C below ambient temperature, generally showing inverse peaks in relation to dehiscent male cone temperatures (see December bottom graph; Fig. 9A); during November the observed pattern for early dehiscent, dehiscent and postdehiscent male cones was more synchronic with a slightly earlier maximums and minimums in dehiscent cones.. Post-dehiscent male cones showed significant heating the first two days and thereafter their temperatures oscillate around 0°C with regards to ambient temperature. After pollen shedding, post- dehiscent male cones had a relatively high peak on December 18th of 3.98 ± 0.77°C; the next three days, the increments were 2.80 ± 0.77°C, 1.15°C ± 0.77°C and 0,79 ± 0.77°C, respectively.

Female cone temperatures generally followed a periodical pattern, but there were clear differences between temperature patterns of receptive and non-receptive female cones (Fig. 9C). Heat production in receptive female cones, similar to the dehiscent male cones, increased together with the ambient temperature (Fig. 9D). Pre-receptive and post-receptive cones had similar patterns, inverse to the receptive cones behavior, with the minimum temperatures generally coinciding with maximum temperatures of receptive cones (see December bottom graph; Fig. 9C). The maximum temperature of the receptive female cone was 4.37 ± 0.78°C above ambient temperature. The pre-receptive female cones maximum temperature was 2.19 ± 1.03°C and the minimum was 6.14 ± 1.03°C below ambient temperature. Post-receptive female cones showed similar trends as pre-receptive female temperatures, with a maximum temperature of 2.87 ± 0.42°C; temperatures usually oscillate around 0°C with a minimum temperature of 0.79 ± 0.42°C below ambient temperature.

Fig. 9 Thermogenesis graphs. a Difference of male cone temperatures in relation to ambient in November 2015 (top) and December (below). b Male cone temperatures in different developmental stages. c Difference of female cone temperatures in relation to ambient in November 2015 (top) and December (below). d Female cone temperatures in different developmental stages.

Discussion

The sequential development of male cones could increase the viability of the pollinator populations through the entire reproductive season, because it extends the male cone resource availability for all the life cycle stages. Female cones have slow growth and they persist almost three times the male cones durability, while seeds develop; therefore, female plants invest more and it could be related with the low number of female cones per plant in relation to male cones. The short time of female cones receptivity and the low number of them would be a determinant factor for conservation efforts.

The results of this study, strongly suggest that Pharaxonotha sp. is the main pollinator of Zamia incognita in natural populations. Behavioral observations of Pharaxonotha sp. individuals going into female cones and staying at pollination drops (see below), is the strongest evidence of cantarophily in Z. incognita; in addition, the pollen dispersion experiments showed that the pollen is transported from male to female cones. Beetles of the genera Pharaxonotha Reitter (Cucujoidea: Erotylidae) and Rhopalotria Chevrolat (Curculionoidea: Belidae) have been considered the regular visitors and pollinator of other Zamia species. Insect pollination in cycads was firstly demonstrated in the Fairchild Botanical Garden in ex situ conditions for Zamia furfuracea, which is pollinated by the single species Rhopalotria mollis (Norstog and Fawcett 1989, Norstog et al. 1986). Nevertheless in natural populations of Z. furfuracea in Mexico, Pharaxonotha beetles have also been found together with R. mollis (Vovides 1991). The interaction with Pharaxonotha beetles has been studied by Tang (1987a) in Zamia pumila where Pharaxonotha zamiae and Rhopalotria slossoni are co-pollinators.

These findings coincides with other Zamia species, because Pharaxonotha seems to be the principal beetle genus related with pollination: i.e. 22 species of Zamia have information about insect visitors, four of them are visited by Rhopalotria and Pharaxonotha beetles (e.g. Calonje 2009, Norstog et al. 1986, Tang 1987a, Taylor et al. 2008, Taylor and Holzman 2012, Vovides 1991); other four species are apparently visited only by Rhopalotria (e.g. Stevenson 2001); and 14 species seems to be visited only by Pharaxonotha (e.g. Calonje et al. 2010, 2011, Chávez and Genaro 2005, Franz and Skelley 2008, González 2004, Lindström et al. 2013, Pakaluk 1988, Schutzman and Vovides 1998, Stevenson et al. 1998, Taylor et al. 2008, Valencia-Montoya et al. in prep.). The genus Pharaxonotha is also associated with the Zamia sister group, the monotypic genus Microcycas (Chaw et al. 2005, Nagalingum et al. 2011), where Pharaxonotha esperanzae is considered the unique pollinator of Microcycas calocoma (Chaves and Genaro 2005, Vovides et al. 1997). The remaining Neotropical cycad genera Dioon and Ceratozamia, which are not closely related to the Microcycas + Zamia clade (Nagalingum et al. 2011), are also visited by Pharaxonotha species (Vovides 1991); therefore, the relationships between these erotylid-beetles and neotropical cycads could be considered as a convergent interaction.

This study comprises the first description of the pollination process in a zamia species apparently pollinated only for Pharaxonotha beetles (co-pollination by other beetle genera can be seen in Norstog et al. 1986, Tang 1987a). The general process and pollinator behavior are similar to those described by Tang (1987a) about P. zamiae in Zamia pumila. Symbiosis between Pharaxonotha sp. and Zamia incognita seems to be very specific and highly specialized, because Z. incognita is the unique known resource for this Pharaxonotha species, and the entire beetle life cycle depends of the zamia, because it is coupled to the plant phenology. Zamia incognita offers food, protection and breeding sites for the beetles. This is similar to other pollinator–cycad mutualisms where pollinator mate and lay eggs on the microsporophylls, where the larvae develop (e.g. Donaldson 1997, Hall et al. 2004, Norstog and Fawcett 1989, Norstog et al. 1986, Tang 1987a, Terry et al. 2005, Wilson 2002). High abundance of adults and larvae were found in a single dehiscent male cone (Table 2), compared with other zamia species like Z. pumila, in which the maximum count was approximately 300 adults per male cone, including Rhopalotria slossoni beetles (Tang 1987a); or with Z. amblyphyllidia and Z. portoricensis, with five adults in average per dehiscent male cone (Franz and Skelley 2008). Although larvae and adults of Pharaxonotha sp. shared the same niche, they have different feeding preferences and it could be account to the coexistence of large amounts of individuals in the same male cone.

Dispersion experiments conducted had not been performed in cycads. Suinyuy et al. (2009) and Terry et al. (2005) also marked insects with dye-colors to estimate the ability of different insect pollinators to deliver pollen in the micropyle and the pollen loads, but traveled distances for Zamia pollinators had not been recorded before. However, our data are not conclusive because the number of marked cones and the time we wait to recover the dyes, were not enough to determine maximum distances; our aim was to determine whether the fluorescent dyes laid on male cones, can reach the female cones due to the ability of the insects to disperse the pollen. Dye-color marks left by beetles on the female cones, reinforces that Pharaxonotha beetles are the pollen vectors in natural conditions for this Zamia species. Marked insects also visited others dehiscent male cones, suggesting that a single beetle may carry pollen from different male plants. On the female cones, we always recovered dye-colors from two different plant sources; this implies that they were visited by beetles coming from various male cone sources, which indeed increase the genetic variability of the seeds. These observations have important implications for the population viability (although other threats as habitat lost, will be discussed below), especially considering that gravity has an important role in seed dispersal and most of the seeds germinate close to the mother (Hall and Walter 2013, Tang 1989). Average and maximum traveled distances were similar between male and female cones and the both cones type seem to have attractants for the beetles. Even though it would be important to obtain additional data from polled dispersion experiments, the low traveled distance found may have significant consequences in the Zamia populations connectivity and gene flow.

One of the most significant results of the dispersion experiments is that beetles seems to only visit receptive female and dehiscent male cones, although there are other available cones in different developmental stages, even in the same plant, just few centimeters away. This supports the presence of key signals to attract insects, that are linked to the pollen release and receptivity stages in the cones.

Dehiscent male cones and receptive female cones showed higher temperatures above ambient, relative to the cones in others developmental stages, bearing that thermogenesis has a valuable role attracting insects. Thermogenesis is widespread phenomenon in cycads and has been considered to be an adaptation for insect-pollination (Tang 1987b), because it help to volatilize odors that attract pollinators such as in flowering plants (Meeuse 1975). Heat production in Z. incognita followed a circadian pattern, being consistent with the general pattern described for cycads (Tang 1987b). The maximum temperature (5.35 ± 1.43°C) for a male cone starting elongation and pollen shedding, coincided with the time when beetles started to arrive. The maximum found in Z. incognita was lower than the maximum found for Z. furfuracea (6.6°C) and higher than in Z. lodiggesii (3.3°C) and Z. fischeri (1.2°C), which are the only species with thermogenesis information in situ conditions (Tang 1987b).

The magnitude and extent of heat production generally were higher in male (at least four days) than in female cones (from two to three days, coinciding with the receptivity period). Lower temperatures in female cones are a trend for Zamia species and generally the known maximums do not exceed 1°C above ambient temperature (Tang 1987b). Temperature maximums in receptive female cones were inverse to pre-receptive and post-receptive cones, and the last two showed temperatures below ambient at the warmest times of the day. These values below ambient temperature suggest no intrinsic thermogenesis in pre-receptive and post-receptive female cones.

The higher temperatures in cones, related to the ambient, could be regarded as a reward to pollinators, especially considering that insects are mainly poikilothermic (Chapman 1998). It is also particularly important because Pharaxonotha beetles, as well as other heat attracted beetles, have small bodies, and the insect ability to elevate the temperature depends on the body size, with smaller insects producing less heat and losing it faster (Chapman 1998). The palatable pollen and the breeding sites, seems to be the main rewards than male cones offers, and therefore the beetles generally spend more time on male plants. However, it is not yet clear why insects visit female cones, because it is known that female cones have a higher toxicity (Stevenson et al. 1998). Norstog and Fawcett (1989) proposed that beetles entry to female cones looking for sheltering, in response to similarities of shape, color and perhaps odor of male and female cones. In dioecious plants where only one sex provides a reward or brood site, chemical similarity between sexes can be considered to reflect mimicry by the sex which is visited by mistake (Ashman 2009). This has been described to phytelephantoid palms in which chemical mimicry between sexes, ensures visits of pollinators to both inflorescences (Ervik et al 1999). We observed beetles remaining long time within the female cones, which could contradict the idea that the beetles visit female cones only by mistake, because it seems they are staying within the cones due to possible rewards.

The pollination drop system is a novelty associated to seed evolution (Simpson 2006) and was widespread during Carboniferous, playing an important role in the diversification of the earliest seed plants (Nepi et al. 2009, Nepi et al. 2012). This droplet is mostly composed of water, sugar and aminoacids, and it is formed by the breakdown of cells at the distant end of the nucella (Simpson 2006). This secretion is conserved in some living Spermatophyta and works as the landing site for the majority of gymnosperm pollen (Jin et al. 2012, Nepi et al. 2009, Nepi et al. 2012, Simpson 2006, Wagner et al. 2007), except for the Araucariaceae family (Nepi et al. 2012). Insect-pollinated species in the orders Cycadales and Gnetales only have one type of pollination droplet, although some gnetophytes additionally possess nectaries on male buds and elsewhere (Nepi et al. 2012). After the pollen lands, the pollination droplet is withdrawn, thus transporting the pollen onto the surface of the nucellus, where it germinates (Nepi et al. 2009); therefore, the contact of the pollen grain with the drop is essential for the pollination process. Although pollination drops and nectar perform different main functions, they have been related due to similarities in chemical composition, the anatomy of secretory tissues and the role as attractants to pollinators (Nepi et al. 2012). Insect-pollinated gymnosperms have been generally split into those offering pollen rewards (cycads) and those offering sweet pollination drops and nectar (gnetophytes), mostly because sugar concentration in pollination droplet is lower in Cycadales than in Gnetales and is not considered as a reward (Nepi et al. 2012).

Tang (1987a) analyzed chemical composition of pollination droplet in Zamia pumila and found sugars like fructose, glucose and sucrose, as well as amino acids (Tang 1987a). Sugar concentration in pollination drops of Z. pumila was higher than in conifers which are wind-pollinated (Nepi et al. 2012). Although in Z. pumila, the observed feeding behavior of the weevil does not overlap with pollination droplet, the chemical composition of the secretion led Tang (1987a) to suggested that this secretion could also be acting as a reward for pollinators, analogous to the flowering plants nectar (Tang 1987a). Our observations of beetles remaining at pollination drops for a long period, could be considered as evidence for Tang (1987a) hypothesis, besides confirming the role of Pharaxonotha sp. beetles as the pollinators of Zamia incognita.

Insect interactions in Zamia incognita were more frequent with male cones and generally occur in different cone developmental stages: the Eumaeus butterflies oviposited in pre-dehiscent male cones, Pharaxonotha beetles were found in early-dehiscent and dehiscent cones, and persisted even when the cones have fallen. Bees visited early post-dehiscent and post-dehiscent cones. The relationships with beetle pollinators and Eumaeus are the most commonly reported for zamias (Contreras-Medina et al. 2003, González 2004, Koi 2008), but despite the Tang’s (1987a) reports of ants (Formicidae), mealy-bugs (Pseudococcidae) and spring-tails (Collembola) occasionally found in Z. pumila cones, there are no descriptions of other insect-interactions for Zamia species. Meliponinae sp. 1 and Meliponinae sp. 2 do not occur simultaneously and these relationships seems to be non-random, because they were found collecting pollen on every late dehiscent and post-dehiscent examined cone. Meliponinae sp. 1 and Meliponinae sp. 2 have the same feeding behaviors, which are similar to that described for Trigona sp. in Cycas media, where the bees collected pollen and moved between male cones and not between male and female cones (Forster et al. 1994). Similarly, the larvae and pupae of Diptera sp. 1 were found in all the decomposing fallen male cones, even in different months, which also suggests that this relationship could be not occasional.

The preference of these insect species for feeding on male cones rather than on female cones, may be due to higher toxicity reported for Zamia female cones. Norstog and Fawcett (1989) suggested for Z. furfuraceae that the idioblast of male tissues store toxins (reducing the probability of the insect to be exposed to these toxins) and the equivalent toxins might not be similarly avoidable in megasporophylls. Pharaxonotha sp. was the unique insect species found within female cones and they were not observed feeding tissues or ovipositing. Higher toxicity levels in female’s tissues have been proposed as an adaptation to prevent seed predation (Norstog and Fawcett 1989, Tang 1987a).

One of the most frequently observed interactions was with Atta ants; they were observed on different structures and some of these events could be considered as merely opportunistic. For example the observation of the ants cutting microsporopylls of a pre-dehiscent male cone, could be opportunistic because it was observed in a low proportion of the population’s male cones; in addition, these ants are highly generalists because they use the cut material to farm the fungi from which they feed (Fernández 2003). Leaf-cutter ants were also found removing the seed sarcotesta, apparently without affecting the megagametophyte. Although the ants eventually can cut on megametophytous, comprising the viability of the seed, they were just observed removing soft tissues of the megastrobilous and the sarcostesta, this could be occurring because to cut the megagametophyte could also imply an addition energy waste. However, it is necessary more field observations in order to clarify whether or not these ants use the megagametophyte. This interaction could be very important for the seed viability and the germination time, because in ex situ conditions expert growers emphasize the importance of removing the sarcotesta (which is likely to cause rot because it can easily acquire fungi) to assure the seed germination (Aristizábal in prep.).

Concerning to the identity of the Zamia incognita pollinator, it is highly probable that it could be an undescribed species. In spite the little morphological variation of the Neotropical Pharaxonotha species, this species holds a unique set of attributes that can distinguish it from the other described species of the genus. Although Colombia is the country with the highest diversity of zamias, there are not described pollinator species for the country, and given that this Zamia species was recently described, its associated pollinator could certainly be also unknown (Skelley pers. comm.) and it is being described (Valencia-Montoya et al. in prep.). Currently the morphological and molecular identity of the pollinator is being checked for further confirmation.

The information about the pollination process may contribute significantly to further conservation efforts. The results of this study show a complex net of interactions never described before for a Zamia species, this is probably because the study was carried out in a natural and healthy population of Zamia incognita. Sorrowfully, this population is seriously threatened by mining limestone and could disappear promptly because it is located in a karstic cone, which is going to be exploited shortly. If the Zamia incognita population of the Cañón del Río Alicante become extinct, not only plants and specialist pollinator insects would be extinguished, but also a complex network of direct and indirect interactions, which has long been thought as impossible for non-flowering plants.

Acknowledgements

We are grateful to Gloria Montoya for unconditional support and her help in processing data. To Cornelio Bota, Camilo Flórez, Laura Gómez Mesa, Miguel González, Laura Marín and Yeyson Montoya for invaluable field assistance. Yeyson Montoya and Cornelio Bota also provided field observations, and Cornelio let us use some of his beautiful photographs. Gustavo Londoño and Juan Luis Parra kindly lent us the temperature recording equipment. To Pablo Guzmán for his valuable help with the statistical analyses. Maria José Sanín, Cristina López-Gallego, Henry Arenas-Castro, Sergio A. Muñoz-Gómez, Alejandra Duque, Nicolás González-Castro and Edwin Hurtado gave constructive input to the manuscript. To the Fundación Alejandro Ángel Escobar for financial support through the Colombia Biodiversa Award, and to the Biology program from Universidad CES for supporting field and laboratory work. Diana María Carmona and Ahída Paulina Herrera provided logistical support while working in the Animal Biology lab from Universidad CES. Marta Wolff and Grupo de Entomología from Universidad de Antioquia (GEUA) kindly provided laboratory resources during the initial stages of this work. Finally, we express our gratitude to all the people in La Hacienda Santa Bárbara, from Maceo, for their very kind help with logistics.

References

Aristizábal A, Tuberquia DJ, Martínez D, Parra G (2011) Caracterización de poblaciones naturales de dos especies amenazadas de Zamia (Zamiaceae), en el Magdalena Medio Antioqueño. Memorias del Congreso Colombiano de Botánica. Cali, Colombia.

Ashman T-L (2009) Sniffing out patterns of sexual dimorphism in floral scent. Funct Ecol 23:852- 862.

Bond WJ (1994) Do Mutualisms Matter? Assessing the Impact of Pollinator and Disperser Disruption on Plant Extinction. Philos Trans R Soc B Biol Sci 344:83-90.

Cáceres RA, Berrío MA, Hincapié Posada JE (eds.) (2010) Atlas de Áreas Protegidas Departamento de Antioquia. Sistema Departamental de Áreas Protegidas de Antioquia. Antioquia, Colombia.

Calonje M (2009) A New Cliff-dwelling Species of Zamia (Zamiaceae) from Belize. J Bot Res Inst Tex 3:23-29.

Calonje M, Stevenson D, Calonje C, Ramos YA, Lindstrom A (2010) A new species of Zamia from Chocó, Colombia (Cycadales, Zamiaceae). Brittonia 62:80-85.

Calonje M, Esquivel HE, Stevenson D, Calonje C, Pava D (2011) A new arborescent species of Zamia from the Central Cordillera of Tolima, Colombia (Cycadales, Zamiaceae), with comments on the Z. poeppigiana species complex. Brittonia 63:442-451. Chapman RF (1998) The Insects: Structure and Function. Cambridge University Press. 770pp.

Chaves R, Genaro JA (2005) A new species of Pharaxonotha (Coleoptera: Erotylidae), probable pollinator of the endangered Cuban cycad, Microcycas calocoma (Zamiaceae). Insecta Mundi 19:143-150.

Chaw S-M, Walters TW, Chang C-C, Hu S-H, Chen S-H (2005) A phylogeny of cycads (Cycadales) inferred from chloroplast matK gene, trnK intron, and nuclear rDNA ITS region. Mol Phylogenet Evol 37:214-234.

Contreras-Medina R, Ruiz-Jiménez CA, Luna Vega I (2003) Caterpillars of Eumaeus childrenae (Lepidoptera: Lycaenidae) feeding on two species of cycads (Zamiaceae) in the Huasteca region, Mexico. Rev Biol Trop 51:201-203.

Donaldson JS (1997) Is There a Floral Parasite Mutualism in Cycad Pollination? The Pollination Biology of Encephalartos villosus (Zamiaceae). Am J Bot 84:1398-1406.

Eriksson R (1994) The remarkable weevil pollination of the neotropical Carludovicoideae (Cyclanthaceae). Plant Syst Evol 189:75-81.

Ervik F, Tollsten L, Knudsen JT (1999) Floral scent chemistry and pollination ecology in phytelephantoid palms (Arecaceae). Plant Syst Evol 217:279-297.

Fernández F (2003) Introducción a las Hormigas de la Región Neotropical. Instituto de Investigaciones de Recursos Biológicos Alexander Von Humboldt. Bogotá, Colombia. XXVII + 398 pp.

Forster PI, Machin PJ, Mound L, Wilson GW (1994) Insects Associated with Reproductive Structures of Cycads in Queensland and Northeast New South Wales, Australia. Biotropica 26:217-222.

Franz N, Skelley PD (2008) Pharaxonotha portophylla (Coleoptera: Erotylidae), new species and pollinator of Zamia (Zamiaceae) in Puerto Rico. Caribbean J Sci 44:321-333.

Galeano G, Bernal R, Calderón E, García N, Cogollo A, Idárraga A (2005) Zamias (familia Zamiaceae: géneros Zamia y Chigua). In Libro Rojo de Plantas de Colombia, (Bogotá: Instituto Alexander von Humboldt - Instituto de Ciencias Naturales de La Universidad Nacional de Colombia - Ministerio del Medio Ambiente, Vivienda y Desarrollo Territorial), 454 pp.

González F (2004) Herbivoría en una gimnosperma endémica de Colombia, Zamia encephalartoides (Zamiaceae) por parte de Eumaeus (Lepidoptera: Lycaenidae). Rev. Acad. Colomb. Ciencias. XXVIII:233-244.

Gorelick R (2001) Did insect pollination cause increased seed plant diversity? Biol J Linn Soc 74:407- 427.

Gottsberger G (1999) Pollination and evolution in neotropical Annonaceae. Plant Species Biol 14:143-152. Hall JA, Walter GH (2013) Seed dispersal of the Australian cycad Macrozamia miquelii (Zamiaceae): Are cycads megafauna-dispersed “grove forming” plants? Am J Bot 100:1127-1136.

Hall JA, Walter GH, Bergstrom DM, Machin P (2004) Pollination ecology of the Australian cycad Lepidozamia peroffskyana (Zamiaceae). Aust J Bot 52:333-343.

Haynes JL. World List of Cycads: A Historical Review. IUCN/SSC Cycad Specialist Group. c2009 - [modified 2012; cited 2014 feb 10] Available in: http://www.cycadsg.org/publications/Haynes- Historical-Review-of-World-List-of-Cycads-2012.pdf

Henderson A (1986) A review of pollination studies in the Palmae. Bot Rev 52:221-259.

Jin B, Jiang X, Wang D, Zhang L, Wan Y, Wang L (2012) The behavior of pollination drop secretion in Ginkgo biloba L. Plant Signal Behav 7:1168-1176.

Kearns CA, Inouye DW (1993) Techniques for pollination biologists. University Press of Colorado.

Koi S (2008) Nectar Sources for Eumaeus atala (Lepidoptera: Lycaenidae: Theclinae). Fla Entomol 91:118-120.

Leschen RAB (2003) Erotylidae (Insecta: Coleoptera: Cucujoidea): phylogeny and review. Fauna of New Zealand 47, 108 pp.

Leschen RAB, Skelley PE (2002) Languriidae Wiedeman, 1823. In: Arnett Jr. RH, Thomas MC, Skelley PE, Frank JH (eds) American Beetles, Volume 2, Polyphaga: Scarabaeoidea to Curculionoidea. Florida: CRC Press, pp 343-347.

Lindström AJ, Idarraga A (2009) Zamia incognita (Zamiaceae): the exciting discovery of a new gymnosperm from Colombia. Phytotaxa 2:29-34.

Lindström AJ, Calonje M, Stevenson D, Husby C, Taylor A (2013) Clarification of Zamia acuminate and a new Zamia species from Coclé Province, Panama. Phytotaxa 98:27-42.

Meeuse BJD (1975) Thermogenic Respiration in Aroids. Annu Rev Plant Physiol 26:117-126.

Nagalingum NS, Marshall CR, Quental TB, Rai HS, Little DP, Mathews S (2011) Recent Synchronous Radiation of a Living Fossil. Science 334:796-799.

Nepi M, von Aderkas P, Wagner R, Mugnaini S, Coulter A, Pacini E (2009) Nectar and pollination drops: how different are they? Ann Bot 104:205-219.

Nepi M, von Aderkas P, Pacini E (2012) Sugary Exudates in Plant Pollination. In: Vivanco JM, Baluška F (eds) Secretions and Exudates in Biological Systems. Springer Berlin Heidelberg, pp 155- 185.

Norstog K (1987) Cycads and the Origin of Insect Pollination. Am Sci 75:270-279.

Norstog KJ, Fawcett PKS (1989) Insect-Cycad Symbiosis and its Relation to the Pollination of Zamia furfuracea (Zamiaceae) by Rhopalotria mollis (Curculionidae). Am J Bot 76:1380-1394. Norstog KJ, Stevenson DW, Niklas KJ (1986) The Role of Beetles in the Pollination of Zamia furfuracea L. fil. (Zamiaceae). Biotropica 18:300-306.

Pakaluk J (1988) Review of the New World species of Pharaxonotha Reitter (Coleoptera: Languriidae). Rev Biol Trop 36:447-451.

Schutzman B, Vovides AP (1998) A New Zamia (Zamiaceae, Cycadales) from Eastern Chiapas, Mexico. Novon 8:441-446.

Simpson MG (2006) Plant Systematics. Elsevier Academic Press.

Stevenson D (2001) Cycadales. Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogotá.

Stevenson DW, Norstog KJ, Fawcet PKS (1998) Pollination biology of cycads. In: Owens SJ, Rudall PJ (eds) Reproductive Biology. Royal Botanic Gardens, Kew, pp 277-294.

Suinyuy TN, Donaldson JS, Johnson SD (2009) Insect pollination in the African cycad Encephalartos friderici-guilielmi Lehm. South Afr J Bot 75:682-688.

Tang W (1987a) Insect Pollination in the Cycad Zamia pumila (Zamiaceae). Am J Bot 74:90-99.

Tang W (1987b) Heat Production in Cycad Cones. Bot Gaz 148:165-174.

Tang W (1989) Seed dispersal in the cycad Zamia pumila in Florida. Can J Bot 67:2066-2070.

Taylor Blake AS, Holzman G (2012) A New Zamia Species from the Panama Canal Area. Bot Rev 78:335-344.

Taylor B AS, Haynes JL, Holzman G (2008) Taxonomical, nomenclatural and biogeographical revelations in the Zamia skinneri complex of Central America (Cycadales: Zamiaceae). Bot J Linean Soc 158:399-429.

Terry I (2001) Thrips and Weevils as Dual, Specialist Pollinators of the Australian Cycad Macrozamia communis (Zamiaceae). Int J Plant Sci 162:1293-1305.

Terry I, Moore CJ, Walter GH, Forster PI, Roemer RB, Donaldson JD, Machin PJ (2004) Association of cone thermogenesis and volatiles with pollinator specificity in Macrozamia cycads. Plant Syst Evol 243:233-247.

Terry I, Walter GH, Donaldson JS, Snow E, Forster PI, Machin PJ (2005) Pollination of Australian Macrozamia cycads (Zamiaceae): effectiveness and behavior of specialist vectors in a dependent mutualism. Am J Bot 92:931-940.

Valencia-Montoya WA, Cardona-Duque J, Tuberquia DJ (in prep.) Estado del conocimiento de la polinización en Zamia L. (Cycadales: Zamiaceae).

Vovides AP (1991) Insect Symbionts of Some Mexican Cycads in Their Natural Habitat. Biotropica 23:102-104. Vovides AP, Ogata N, Sosa V, Peña-García E (1997) Pollination of endangered Cuban cycad Microcycas calocoma (Miq.) A.DC. Bot J Linn Soc 125:201-210.

Wagner RE, Mugnaini S, Sniezko R, Hardie D, Poulis B, Nepi M, Pacini E, von Aderkas P (2007) Proteomic evaluation of gymnosperm pollination drop proteins indicates highly conserved and complex biological functions. Sex Plant Reprod 20:181-189.

Wilson GW (2002) Insect Pollination in the Cycad Genus Bowenia Hook. ex Hook. f. (Stangeriaceae). Biotropica 34:438-441.