An evaluation of the species boundaries of two putative taxonomic entities of Euterpe (Arecaceae) based on reproductive and morphological features

Abstract

Past taxonomic treatments have classified Euterpe espiritosantensis as a synonym of E. edulis. However, both continue to be identified as two distinct species and are enumerated in the threatened species list. The goal of this study was to compare the reproductive biology of sympatric populations of these two morphs so as to identify morphological characters and mechanisms of reproductive isolation that could help to clarify species boundaries. Individuals of E. edulis and E. espiritosantensis show differences in size and duration of the peduncular bract, and differ in regard to inflorescence and immature fruit coloration. With an overlapping flowering period, but with distinctly different flowering peaks, the two taxa share the main visitors, viz. different Apidae and some Vespidae. Fruit and viable seeds are produced after self-, cross- and inter-specific cross-pollination in both morphs. Seed germination is significantly higher in E. edulis than in E. espiritosantensis. Differences in morphological and reproductive features suggest that these sympatric populations of two Euterpe morphs are to a reasonable degree reproductively isolated, which supports the recognition of E. espiritosantensis as a distinct species from E. edulis. The potential for hybridization does not reject the hypothesis of species distinctness, but points to a potential case of sympatric speciation that merits further investigations. Given that natural populations of Euterpe are nowadays fragmented and reduced in area of occurrence, retaining the high conservation status for E. espiritosantensis will help to safeguard this taxonomic entity under considerable threat.

Introduction

Despite the seemingly incessant discussion about “species concepts”, most plant species continue to be defined based on morphological data (Coyne and Orr, 2004, Levin, 2000). Whether one likes it or not, this situation will persist, because one usually does not have sufficient information on plant reproductive behavior to allow the biological species concept to be applied successfully (Stuessy, 1989). Recognisable morphological discontinuities usually reflect biological limits such as reproductive isolation and genetic divergence. Thus, there is most frequently harmony between species differentiated by morphological and by other criteria. Under every species concept, the division of a continuously evolving lineage into named species is a pure subjective exercise, although it is essential for scientific communication (Coyne and Orr, 2004, Stuessy, 1989).

The boundaries between putative palm species are often obscured by a lack of well defined morphological differences, by paucity of relevant information and by a rather frequent potential for hybridization (Henderson, 2006). Plant hybridization is a common event in nature (Rieseberg and Carney, 1998), especially among related species that share pollinators (e.g., Esfeld et al., 2009, Parrish et al., 2003, Wendt et al., 2001). Hybrid development does not mean that the taxa involved are only one species (Coyne and Orr, 2004). Disagreements over whether local forms should be classified as species, infraspecific taxa or results of natural intraspecific variation have led to varying estimates of species numbers in the palm family (Henderson, 2004, Henderson and Martins, 2002, Roncal et al., 2007). One example of such a taxonomic controversy involves the Euterpe genus.

Euterpe Mart. contains seven species distributed in South and Central America, five of which are found in Brazil (Henderson and Galeano, 1996). Some Euterpe species produce the palm-heart, locally called palmito, which is an exotic gourmet food of high monetary value. The palm-heart includes the apical meristem of the plant, and harvesting of the heart from this single-stemmed palm requires its destruction. Euterpe edulis Mart. was for many years the most important species for the extraction of palm-heart. Even today, the harvest of palm-heart frequently comes from wild plants, and extensive exploration can affect the regeneration of this species’ natural populations (Guilherme et al., 2004, Reis et al., 2000, Silva Matos and Watkinson, 1998). The fruit of E. edulis is an important food resource for a wide assemblage of vertebrate frugivores and the plant represents therefore an ecological keystone species (Genini et al., 2009).

Boudet-Fernandes (1989) described Euterpe espiritosantensis H.Q.B. Fern., which occurs sympatrically with E. edulis in the Santa Teresa municipality, Espírito Santo State, southeast Brazil. In the most recent taxonomic treatment of this genus, E. espiritosantensis was considered as a synonym of E. edulis (Henderson and Galeano, 1996). These authors considered that most species in the Euterpe genus contain much variation and thus present many local races and forms. They proposed that characters used to define E. espiritosantensis fall into the range of variation of E. edulis. Although these two palm names are valid synonymies, the plants continued to be treated as two distinct species in subsequent publications (e.g., Martins et al., 2007a, Martins et al., 2007b) and by the local people in Espírito Santo. A critical revaluation of the species boundary is necessary by the fact that both are currently listed as endangered species (Kollmann et al., 2007).

Different morphological entities may occur in sympatric populations under similar environmental conditions, situations commonly observed in several genera of neotropical palms (Borchsenius, 1997, Borchsenius, 2002). Mixed populations of sympatric species have the potential for increased interspecific pollen transfer. Therefore, we would expect that species limits within regions of sympatry are maintained by a variety of isolation mechanisms (Grant, 1981). Reproductive isolation can be achieved by differences in pollination syndromes, flowering time and/or interspecific incompatibility systems (Levin, 1971, Ortigosa and Gómez, 2010). So, speciation involves the evolution of reproductive barriers between populations, and those barriers ultimately must be maintained if incipient species are to remain distinct entities (Grant, 1981, Levin, 2000). The biggest challenge related to the “species problem” is to understand how sympatric, sexually reproductive organisms fall into discrete clusters (Coyne and Orr, 2004). Comparative studies of reproductive biology between sympatric congener taxa might help to elucidate such taxonomic problems (e.g., Frolov et al., 2007, Wendt et al., 2000, Wendt et al., 2001, Yang et al., 2007).

Available studies on the reproductive biology of E. espiritosantensis (Bovi et al., 1994) and E. edulis (Castro et al., 2007, Reis et al., 2000) where not designed to elucidate species circumscription. The goal of our study was to investigate whether E. edulis and E. espiritosantesis could be confidently recognized morphologically in narrow sympatric occurrence in Santa Teresa regions at Espírito Santo, and if this morphological distinction could reflect reproductive isolation. Thus, we compared their morphology and reproductive biology, and also conducted artificial treatments to investigate potential hybridization between them.