A Comparison of Reproductive Strategies Between Island and Mainland Caribbean Gesneriaceae
Total Page:16
File Type:pdf, Size:1020Kb
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/280915347 A comparison of reproductive strategies between island and mainland Caribbean Gesneriaceae Article in Journal of Ecology · July 2015 DOI: 10.1111/1365-2745.12457 CITATIONS READS 4 98 5 authors, including: Silvana Martén-Rodríguez Mauricio Quesada UNAM Universidad Nacional Autónoma de México 26 PUBLICATIONS 344 CITATIONS 129 PUBLICATIONS 3,918 CITATIONS SEE PROFILE SEE PROFILE Martha Lopezaraiza-Mikel Charles B Fenster Universidad Autónoma de Guerrero South Dakota State University 18 PUBLICATIONS 471 CITATIONS 106 PUBLICATIONS 4,957 CITATIONS SEE PROFILE SEE PROFILE Some of the authors of this publication are also working on these related projects: Floral traits and its effect on flower visitors and pollinators in La Mancha, Veracruz, Mexico View project Ecology of Tropical Dry Forest View project All content following this page was uploaded by Silvana Martén-Rodríguez on 02 March 2016. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. Journal of Ecology 2015, 103, 1190–1204 doi: 10.1111/1365-2745.12457 A comparison of reproductive strategies between island and mainland Caribbean Gesneriaceae Silvana Marten-Rodr ıguez1,2,3*, Mauricio Quesada1,3, Abel-Almarales Castro4, Martha Lopezaraiza-Mikel3,5 and Charles B. Fenster6 1Escuela Nacional de Estudios Superiores, Universidad Nacional Autonoma de Mexico, Antigua Carretera a Patzcuaro 8701, Morelia, Michoacan 58190, Mexico; 2Departamento de Biologıa Evolutiva, Instituto de Ecologıa, A.C., Carretera Antigua a Coatepec 351, Xalapa, Veracruz, 91070 Mexico; 3Laboratorio Nacional de Analisis y Sıntesis Ecologica para la Conservacion de Recursos Geneticos, Escuela Nacional de Estudios Superiores, Universidad Nacional Autonoma de Mexico, Antigua Carretera a Patzcuaro 8701, Morelia, Michoacan 58190, Mexico; 4Centro Oriental de Ecosistemas y Biodiversidad, Museo de Historia Natural Tomas Romay, Enramadas 601, esq. Barnada, Santiago de Cuba, Cuba 90100; 5Unidad Academica de Desarrollo Sustentable, Universidad Autonoma de Guerrero, Carretera Nacional Acapulco Zihuatanejo Km 106 + 900, Tecpan de Galeana, Guerrero, 40900 Mexico; and 6Department of Biology, University of Maryland, College Park, MD 20742, USA Summary 1. The evolution of self-pollination has long been considered an adaptive strategy to cope with low or variable pollinator service; however, alternative reproductive strategies, such as generalized polli- nation (>1 pollinator functional group), may also ensure plant reproductive success in environments with inadequate pollinator visitation. 2. Island–mainland systems provide ideal settings to assess the interaction between pollination and breeding systems in response to pollinator visitation because islands are often pollinator-depauperate. This study compared 28 insular and 26 mainland species of Caribbean Gesneriaceae to test the hypothesis that low diversity and possibly low pollinator service on islands would lead to a greater frequency of generalized plant–pollinator interactions and/or a higher potential for autonomous self- pollination in insular than in mainland species. We also assessed the hypothesis that epiphytic spe- cies should have greater autofertility than species occurring in other habitats. 3. Pollinator observations conducted in the field from 2004 to 2014 revealed bat, bee, butterfly, hum- mingbird, moth, and generalized pollination systems. Functional specialization in pollination systems was high in insular (71% of the species) and mainland sites (all species), but generalized and bat-polli- nated species were more common on islands. Overall, pollinator visitation rates did not differ between island and mainland; however, for hummingbird-pollinated species, visitation rate was on average three times higher in mainland than island species. Autofertility indices (fruit set of bagged/outcross flowers) ranged from 0 to 1 and did not differ between island and mainland species. Species growing on rocks (rupiculous) and trees (epiphytic) had on average higher autofertility than terrestrial species. 4. Synthesis. This study revealed that alternative reproductive strategies are used in pollinator-depau- perate environments. Pollinator visitation is lower in insular hummingbird-pollinated species (the ancestral pollination system of insular Gesneriaceae); therefore, generalized pollination may be con- sidered a reproductive assurance mechanism evolved primarily on island environments. Contrary to the long-standing tenet, however, autonomous self-pollination was similar in island and mainland Gesneriaceae suggesting that: (i) generalized pollination provides a viable alternative to selfing in pollinator-depauperate environments, (ii) autofertility as a reproductive assurance mechanism may be frequent in plant species from mainland regions in environments with unpredictable pollinator visita- tion and resource availability. Key-words: Caribbean, Gesneriaceae, island, mainland, pollination, reproductive ecology, self-pollination, specialization *Correspondence author. E-mail: [email protected] © 2015 The Authors. Journal of Ecology © 2015 British Ecological Society Pollination in insular and mainland plants 1191 transitions in pollination and breeding systems across plant Introduction lineages. Islands generally have depauperate faunas with a poor repre- The Neotropical Gesneriaceae (ca. 1200 spp.) provide a sentation of some animal pollinator groups (Carlquist 1974; suitable study system to evaluate potential drivers of evolu- Barrett, Emerson & Mallet 1996). For example, long-tongued tionary change in plant reproductive strategies because it lepidopterans and social bees are often rare in oceanic islands, exhibits great floral diversity in both island and mainland while flies are common floral visitors (McMullen 1987; Inoue environments. In the Mesoamerican region and Caribbean 1993; Newstrom & Robertson 2005). The ecological and evo- islands, there are approximately 400 species, of which some lutionary consequences of this reduced pollinator diversity 75 species are restricted to the Antilles (Skog 1976; partly depend on the pollination and breeding system of each Tropicos.org. Missouri Botanical Garden. 19 Jan 2015, plant species. For instance, vulnerability to pollination failure http://www.tropicos.org/Name/42000302?projectid=3&langid= should be greater in species with obligate outcrossing breed- 66). Antillean Gesneriaceae form a clade that includes a vari- ing systems and specialized pollination systems (e.g. one–few ety of floral phenotypes associated with pollination by hum- pollinator species, a.k.a., ecological specialization, Arm- mingbirds, bats, moths, bees or a combination of these animal bruster, Fenster & Dudash 2000; one functional group of pol- groups (i.e. functionally generalized pollination systems, linators with common behaviours and morphologies exerting Marten-Rodrıguez, Almarales-Castro & Fenster 2009; Marten- similar selection on flowers, a.k.a., functional specialization, Rodrıguez et al. 2010). Ancestral state reconstructions indi- Fenster et al. 2004). Thus, evolutionary transitions in cate that hummingbird pollination is ancestral in this insular response to limited pollinator visitation are expected in insular clade, with at least two transitions to bat pollination and three plant lineages, with shifts to novel specialized pollination sys- transitions to generalized pollination (Marten-Rodrıguez et al. tems (e.g. pollination by crickets in Angraecum, Micheneau 2010). In mainland Gesneriaceae, pollination systems are also et al. 2010), or to generalized pollination systems (Arm- diverse and inferences from comparative analyses suggest that bruster & Baldwin 1998; Smith, Ane & Baum 2008; Marten- transitions in floral syndromes are common (e.g. SanMartin- Rodrıguez et al. 2010). Alternatively, natural selection may Gajardo & Sazima 2005; Wolowski et al. 2013). There is lit- favour autonomous reproductive strategies in insular environ- tle information on the evolution of breeding systems of ments (Baker 1955). Neotropical Gesneriaceae, but there is evidence for wide- The reproductive biology of island plants has been the spread intrafloral dichogamy (i.e. the temporal separation of focus of much research (Garcıa-Verdugo & Fay 2014). For male and female phases) in both insular and mainland species instance, studies conducted on whole plant communities of (SanMartin-Gajardo & Sazima 2004; Marten-Rodrıguez & remote oceanic islands show a high frequency of self-com- Fenster 2008; Wolowski et al. 2013). Furthermore, some Ges- patible species capable of autonomous self-pollination (Far- neriaceae species are capable of autonomous self-pollination, oes, Hagerup 1951; Hawaii, Carlquist 1974; New Zealand, a condition that appears to have evolved more than once in Lloyd 1985; Galapagos, McMullen 1987; San Juan Fernan- the Antilllean clade (Marten-Rodrıguez et al. 2010). The fre- dez, Bernardello et al. 2001). However, these studies do quency of species capable of delayed autonomous self-polli- not make direct comparisons to mainland communities. At nation in mainland regions has not yet been assessed. the species level, there are few comparative studies on the We evaluated the hypothesis that low diversity and possibly reproductive ecology and evolution of plant