The orchidaceous subtribe Pleurothallidinae forms one of the most diverse natural groups of this family (Dressler 1993a). The ~30 (sensu Luer 1986a) resp. 35 genera (sensu Pridgeon & Chase 2001) comprise more than 3500 species (Luer, pers. commun.), i.e. ~15% of the world’s orchid flora.
The ecological centre of diversity of these exclusively neotropical orchids lies in the (sub)montane rain- and cloudforests of the Central American Sierra Madre chain and the South American Andes. Yet pleurothallids are found from Florida to Bolivia and N Chile, including the Antillean Arc. Growing primarily in epiphytic habitats, many taxa can be found in rupicolous and (pseudo)terrestrial niches as well.
The subtribe consists mostly of microphytic plants. Due to the minor horticultural role, these plants have long been excluded from detailed studies in every respect. Only recently pleurothallids have been the target of morphological and studies (Pridgeon & Williams 1979; Pridgeon 1981a,b,c; Pridgeon & Stern 1982, 1983, 1985; Stern & al. 1985; Neyland & Urbatsch 1993; Stenzel 2000). Two features are usually considered to delimit the subtribe (Fig. 1): the nonbulbous secondary stems (ramicauls) and the persistent pedicel with an abscission layer just below the ovary and not at the base of the pedicel as in all other orchids (Luer 1986a). A third notable macromorphological feature is a ring-like structure above the abscission layer, the annulus (Fig. 1). This character of still unknown function (Stern & al. 1985, Luer 1986a) is found only in the more derived genera (Pridgeon & al. 2001).
Within the subtribe there are the world’s smallest orchids (Luer 1990) as well as species measuring several meters in height. All stages from densely caespitose to long creeping habits can be found. The ramicaul may be reduced or may be elongate. Except for Frondaria Luer, it bears a single terminal leaf. The erect or pendent inflorescence, usually inserted below the abscission layer, arises from the rhizome in some groups, while it is merged into the leaf blade in others, resembling a phylloclade as in Ruscus. Floral organs are very diverse. There is a wide array of ornamentations and appendages such as warts, scales, hairs etc. Sepals show various degrees of connation. Both sepals and petals may carry osmophores at their tips. The lip may be simple or lobed and often forms a complex, [page 6↓]3-dimensionally elaborated structure. It is hinged either to the base of the column or to a foot, which is formed by an extension of ovary and column. The labellum is often mobile or even actively motile in some cases. The column carries the anther, stigma and rostellum at its distal end. The anther may be hooded and ventral (facing the lip) or apical. The stigma is usually entire; only in some genera we find stigmatic lobes or even bi-partitioned stigmas.
|Fig. 1: General morphology of a pleurothallid orchid. Adapted from Luer (1986a).|
A widespread feature of pleurothallid orchids is succulence, although the classic storage organs in orchids, the stems, are not thickened into pseudobulbs. Succulent tissues may be found in all organs, i.e. they are not confined to vegetative parts of the plant. This is the result of adaptation to xeric conditions of the epiphytic and epilithic habitat.
The enormous richness of vegetative and reproductive features in Pleurothallidinae is equally reflected in microscopical characteristics. Such an amazing variety of palynological differentiation as in Pleurothallidinae (Stenzel 2000) can be found nowhere else in higher orchids (Epidendroideae s.l. Dressler 1993a). The number of pollinia, 2, 4, 6, or 8, differs substantially between genera, with the majority of taxa having only one pair. It is widely agreed now that reduction in number proceeded from 8 to 2 pollinia (Dressler 1993a, Stenzel 2000, Pridgeon & al. 2001). However, this evolutionary process is not necessarily accompanied by a sporodermal differentiation. Thus, there can be taxa with 8 (Octomeria) or 4 (Barbosella) pollinia with a highly elaborated sporoderm in the outer tetrads, while other genera with 2 pollinia still show an ancient state of sporodermal sculpture (Stenzel 2000). These “inconsistencies” illustrate once more the complex morphological evolution within the subtribe.
Although clearly heterobathmic constellations complicate the picture of palynological evolution, individual features were found to be usually consistent with generic boundaries. Yet there is one striking exception: Pleurothallis. This large genus (sensu Luer 1986b) shows virtually all sculpture types found elsewhere in the subtribe (Stenzel 2000), strongly suggesting polyphyly in Luer’s system.
Taxonomy and Systematics
The earliest revision of pleurothallid orchids was published by J. Lindley in several fascicles of his Folia, with the most extensive one covering the genus Pleurothallis (Lindley 1859). F. Kränzlin treated some masdevallioid genera in his “Monographien der Gattungen Masdevallia, Lothiania, Scaphosepalum, Cryptophoranthus, und Pseudoctomeria” (1925). Another 60 years passed until C. Luer started his Icones Pleurothallidinarum with a generic survey (Luer 1986a). All these classifications were morphologically based. Pridgeon (1982b) and Neyland & al. (1995) tried to liberate pleurothallid systematics from subjective weighting by using numerical analysis, still being based on morphological characters. Finally, Pridgeon, Solano & Chase (2001) presented molecular data and radically changed the pleurothallid system in a subsequent paper (Pridgeon & Chase 2001, with nomenclatural corrections in Pridgeon & Chase (2002). Yet, even the new system, introduced by Pridgeon and Chase with about 500 taxonomic changes, i.e. transfers for the most part, seems to be partially questionable. It was completely turned down by Luer (2002) and critical reviews have been published both in papers and on the Internet (Hammel & al. 2002, Jost & Endara 2002).
Morphological traits that have been traditionally employed in pleurothallid systematics comprise number of pollinia, presence or absence of the annulus, shape of the ramicaul and the transition area with the leaf base, degree of sepalous connation, floral [page 8↓]appendages and osmophores, shape and adnation of the lip as well as special floral structures, e.g. motile lips. In most cases it is not just one synapomorphic character, but a combination of different traits, that distinguishes taxa. This high level of parallel evolution and the subsequent “homoplasy rife” (Pridgeon & al. 2001: 2286) has made so difficult the distinction between homologous and analogous characteristics. Consequently, distinct features used in taxonomy form often merely “key characters” instead of true synapomorphies. This applies to both macro- and micromorphological and anatomical traits (Pridgeon 1982a, Stenzel 2000, Pridgeon & al. 2001).
Palynological data has already been successfully applied earlier in phylogenetic investigation in orchids (reviewed in Stenzel 2000) and has been found to be useful in delimiting pleurothallid genera (Stenzel 2000). However, it must be used with care when discussing relationships. The molecular based system proposed by Pridgeon & Chase (2001) has made palynological patterns much more consistent with generic boundaries, especially in the re-defined genera Pleurothallis and Stelis (Stenzel 2004b). Yet, the new concept comprises again new or resurrected genera that show palynologically discordant traits, as was found in initial studies at the outset of this thesis. With additional DNA sequences and further palynological data, especially from the “genus” Pleurothallis, I investigate whether the molecular based system introduced by Pridgeon & Chase has provided pleurothallid systematics a convincing frame work reflecting natural evolution.
The Greater Antilles form the northernmost boundary of pleurothallid orchids. Only a few species have been found further N at the S tip of Florida.
The first Pleurothallis described, Pleurothallis ruscifolia (Jacq.) R. Br., was based on Antillean material, i.e. Martinique (Jacquin 1763). It is the type of the largest natural orchidaceous group world-wide. The next species of this genus were published in 1788 by the Swedish botanist O. Swartz and, like P. ruscifolia, they were placed in the genus Epidendrum. After R. Brown had established the genus Pleurothallis (in Aiton & Aiton, 1813), J. Lindley from the Royal Botanical Gardens Kew united the genera Lepanthes, Octomeria, Pleurothallis (incl. Specklinia) and Stelis in the subtribe Pleurothallidinae Lindl. The majority of the Antillean taxa described in the 19th century were published by Lindley. Apart from a few taxa that had been described by him earlier, he benefited from the extensive field work by Charles Wright in Cuba in the late 50ies (Lindley 1858: 9 epithets). H. G. Reichenbach added another 5 species based on Wright’s set that was sent to W. J. Hooker (Reichenbach 1865). Parallely, Grisebach had also worked on Wright’s material. However, Reichenbach’s publication made most of his pleurothallid epithets “Makulatur” (i.e. superfluous; Griseb. in lit. according to Howard 1988: 251). The next century started with several new species from Jamaica described by W. Fawcett and A. B. Rendle [page 9↓](1909a,b), after Fawcett had already described P. uncinata (Myoxanthus uncinatus (Fawc.) Luer) in 1895. A. C. Cogniaux published several Antillean taxa of Pleurothallis (1909-1910), mainly based on Wright’s collections, that either had been overlooked or misinterpreted by Lindley and Reichenbach. Meanwhile, the Swede E. Ekman had started his extensive field work in Cuba. His material served as a base for many orchid species published by R. Schlechter, then curator in the Botanical Garden Berlin-Dahlem. However, merely one species of Pleurothallis was published from Ekman’s material, the Cuban P. ekmanii. Another 50 years passed until D. D. Dod started investigations on Hispaniola, a work that resulted in the publication of no less than 17 epithets attributed to Pleurothallis from this island (Dod 1976, 1977, 1978, 1984b, 1989a). H. Dietrich added another epithet for Cuba (1984a). As the most recent contribution, Luer (1998c, 1999a) and Stenzel (2001, 2002) described 13 species from Cuba and Hispaniola referable to Pleurothallis. This shows that the Greater Antilles still host a treasure of taxa new to science, despite intensive scientific work in this neotropical region for several hundred years now.
Although the subtribe Pleurothallidinae represents essentially a continental taxon, it comprises at least ~150-200 species in the Greater Antilles. There are ~10 genera on the islands of the arc, the diversity of which is distributed very unevenly. Only a few species rich genera (Pleurothallis, Lepanthes, Lepanthopsis) contrast a majority of poorly represented taxa (Barbosella, Brachionidium, Myoxanthus, Octomeria, Platystele, Stelis, Trichosalpinx, Zootrophion). Pleurothallis seems to be by far the most diverse genus. It may be dethroned, however, once Lepanthes is thoroughly revised on all islands.
In any case, Pleurothallis is the morphologically most diverse pleurothallid group in the Arc. While the other pleurothallid genera, except Lepanthes, were revised in the last decade in Luer’s Icones, Pleurothallis has been treated thoroughly only for Puerto Rico (Ackerman 1995, 1997) and Jamaica (Adams 1972). Concerning the species rich, large islands Cuba and Hispaniola, however, “only” several new descriptions were published by H. Dietrich for Cuba and D. D. Dod for Hispaniola. Thus, a taxonomic revision is still pending.
This status quo does not concern Pleurothallis alone. Cuba’s latest revision of the orchid family dates back to the 40ies of the last century (León & Schweinfurth 1946) and there are many other plant families that have not been treated since then either. To create a modern Flora, Cuba, in collaboration with the GDR and Hungary, had launched therefore the ambitious Nueva Flora de la República de Cuba project in the early 70’s (Lepper 1992). This collaboration is based mainly on scientific exchange between the Jardín Botánico Nacional de La Habana, the Friedrich-Schiller-University Jena, the Humboldt-University Berlin, and, since the 1990’s, the Botanical Museum Berlin-Dahlem. Since then, a substantial stock of herbarium material as well as phytogeographical and ecological [page 10↓]data has been gathered in numerous field trips. Both, physical and non-physical material and data on Orchidaceae (Dietrich 1979, 1980, 1982, 1983, 1984c, 1985, 1988, 1992) were considered to provide an excellent base for a revision of the genus Pleurothallis as well as further studies. While several plant families have been treated as part of the Flora Cuba Project since the mid 90’s, Orchidaceae have not been tackled yet. The revision of the main orchid genus Pleurothallis, as part of this study, can be considered the starting point of a series of taxonomic treatments of this important family in one of the floristic hotspots of the world. It will form the base for all further studies that make up the present thesis.
Island biogeography has fascinated evolutionary scientists since Darwin, who dedicated a whole chapter to the “inhabitants of oceanic islands” (Darwin 1859]. Some general traits of island biotas, like low diversity and high endemism compared to similar continental areas, were already stated by him. Other questions that arise when it comes to issues of island floras or faunas and their evolution depend heavily on the archipelago and the taxon in concern. Questions as origin, dispersal and speciation cannot be detached from the specific geological and climatic history, as well as their present constellation, dispersal capabilities of the taxon in question, ecological demands and reproductive traits.
Islands are often chosen as a geographical unit, because they combine some specific features not found in continental areas: “small, isolated, and relatively simple systems to most continental situations” (Baldwin & al. 1998 quoted from Carlquist 1965). Although the prevalence of these traits is questionable and may be heavily influenced by our anthropogenic view, islands do have some advantages for biogeographical studies: (1) they are geographically clearly defined, (2) political borders are mostly identical with island limits, i.e. the geographical limits of local Floras are naturally defined, (3) the age can often be easier determined than in continental areas, especially in the case of volcanic islands (Baldwin & al. 1998).
However, in some cases not even these advantages may be present. The West Indies, for instance, are an archipelago that has a complex geological history (Hedges 2001, Iturralde-Vinent & MacPhee 1999]. In the recent past, Caribbean palaeogeography and biogeography were discussed in close connection by aligning one with the other. In this attempt, biogeographical aspects were studied most intensively in animals. In contrast, the number of Antillean phytogeographical works is comparably small (Samek 1988, Howard 1974). Other studies are confined to single islands (Alain 1978, Samek 1973, Borhidi 1996) or form little more than introductory chapters of floral works. The two most recent and comprehensive publications on West Indian biogeography (Woods 1989, [page 11↓]2001) contain merely one botanical study each. Thus, phytogeographical issues are mainly dealt within floristic studies, which focus naturally on a very limited geographical scope.
Similar to the insufficient taxonomic treatment of the Orchidaceae, the phytogeographical patterns of this family in the Antilles are only rarely dealt with (Dietrich 1989a, Trejo-Torres & Ackerman 2001, Ackerman & al. in press). Zooming further into the family, the situation gets even worse: there are no phytogeographical studies dealing with any of the Antillean orchid groups. Considering the poor taxonomic knowledge that notoriously accompanies this family this is not surprising. Even the introduction of molecular techniques has not triggered further biogeographical studies. Most studies using genetic markers are aimed at the “re”-revision of the orchidaceous system, with little or no attention being paid to geographical issues. Ironically, the introduction of this new and powerful tool may even postpone its application in a phytogeographical context, since both the new systematic revision as well as the introduction of nomenclatural consequences are very time-consuming. Moreover, the discussion on morphological vs. molecular systems, further distract the attention from the fact that molecular methods have proven to be very powerful in the phytogeographical field, too (Baldwin & al. 1998). Pleurothallidinae provide an excellent example for this barren conflict (Pridgeon & Chase 2001, Luer 2002).
The phytogeography of the Greater Antilles has long focused on comparing species richness and endemism, and assessing putative channels of floristic exchange based on floristic similarity. Only recently have been introduced cladistic methods (Judd 2001, Trejo-Torres & Ackerman 2001: PAE and PAD) to make floristic affinities between areas more lucid. However, apart from Judd (2001) no phylogeographical study has been published to my knowledge, so far. While molecular data has already been used in the study on the genesis of other island floras (reviewed in Baldwin & al. 1998) there are virtually no data available for the Antilles.
Patterns of orchidaceous phytogeography in the Greater Antilles are still poorly known too. Orchids serve often as a “political tool” employed in conservation issues when defining the floristic value of an area. Detailed knowledge about distribution, host specifity and other ecological correlation rarely exists, unfortunately. This is especially the case in endemic taxa with a limited distribution. The endemic portion in the genus Pleurothallis was estimated at ~60% for Cuba (Acuña Galé 1939, León & Schweinfurth 1946); however, distribution data given in these treatments and more recent data from field work within the Flora Project often do not coincide. Moreover, it is a well-known fact, that endemicity, local or regional, is often a collection artefact or is simply overestimated. Dietrich (Dietrich 1989a), referring to some highly endemic pleurothallid genera, takes the view that “every more or less isolated mountain chain has its own species”. Initial [page 12↓]sampling and collecting as well as herbarium studies prior to this study had indicated a probably much wider distribution than actually considered in many taxa. Likewise, field observations prior to this study had indicated that taxa of wider, e.g. Circum-Caribbean, distribution grow in a great array of different habitats, i.e. spatial distribution seems to be correlated with less ecological specialisation. Unfortunately, there is almost no data on ecological preferences of orchids (Freiberg 1992). The predominance of pleurothallids in moist habitats is the only correlation that is usually stated. H. Dietrich (pers. commun.) reported the observed association of orchids with certain petrologic features (limestone). To test these hypotheses, Cuba with its diverse ecological and geological patterns, provides an excellent background.
It is surprising that both an area and a plant family that have been attracting scientists for centuries now, still show blank areas. As was shown in the previous chapters, several factors have influenced the decision to choose the genus Pleurothallis for a study on the genesis of the Caribbean orchidaceous flora. The main objectives of the present study are the following:
|© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.|
|DiML DTD Version 3.0||Zertifizierter Dokumentenserver|
der Humboldt-Universität zu Berlin