Vanilla

Marc Pignal

Abundant in tropical regions of the globe, the genus Vanilla has challenged morphologists. Although the molecular approach provides a glimpse of the phylogenetic structure of the group, the clear definition of species is fraught with difficulties. This chapter opens the “Vanilla File” and lists some technical aspects that hinder the study of the only orchid cultivated for food.

Lack of Taxonomic Revision

Although cultivated vanilla has been the subject of many studies, the wild species, and more particularly those that have no agronomic qualities (being odorless or too capricious for cultivation) have not been studied in a synthetic way. Admittedly, studies on many taxa were published, as seen among members of the Orchidaceae family. The Kew Index lists more than 250 species (Anonymous, 2009), but the taxonomic revisions are missing.

The first complete taxonomic treatment of the genus Vanilla was carried out by Rolfe in 1896. It recognizes 50 species. The last revision by Portères (1954) in Bourriquet’s book, La vanille et le vanillier dans le monde, includes 110 species. Portères’ work, although rich in biogeographic data, is unfortunately not exhaustive because several taxa published by the Brazilian botanist Hoehne between 1941 and 1944 (Hoehne, 1941, 1944) are not mentioned. These works date back to the years of the Second World War, and European libraries in the 1950s had difficulty in supplementing their collections.

The lack of adequate treatment is directly related to the state of herbarium specimens in collections, which are the basis of any taxonomic and nomenclatural reasoning.

Collections Still Poor

Collections by past botanists are scarce and illustrations have often been used in place of specimens. This is the case for the “historical herbaria” of the Muséum National d’Histoire Naturelle in Paris. In Lamarck’s herbarium, for example, only one figure illustrates vanilla, taken from the Encyclopédie Méthodique. The legend is in French: Petite vanille ou vanille musquée (small vanilla or vanilla musk). The drawing is sufficiently precise to identify Vanilla palmarum. The Jussieu herbarium is hardly any richer, with only one specimen of Vanilla aromatica. Are these illustrations derived from herbarium specimens? This is doubtful. Drawing from live specimen or from another illustration is common. This significantly contributed to confusion about the genus upon its official publication in the botanical nomenclature (Miller, 1768), even though vanilla was already widely cultivated.

Miller (1768) at first distinguishes “two or three varieties which differ in the color of their flowers and the length of their pods.” Then he recognized them as two species: V. mexicana and V. axillaris. It seems today that the specimens described by Miller included at least our existing species V. planifolia and V. pompona.

Modern collections suffer from the same flaws as their predecessors: rare collections that are often fruitless and flowerless. When flowers are present, they are poorly dried or unprepared, making determinations uncertain. Moreover, species are often represented by a single specimen.

Some major herbaria in the world possess vanilla specimens. In North America, the New York Botanical Garden is very active in New World tropical research. In Brazil, the Hoehne collection is conserved in the Rio de Janeiro Botanical Gardens. But many other structures also conserve vanilla.

This is the reason for many databases and Web sites to allow the remote consultation of specimens. Particular mention should be made of the GBIF, Global Biodiversity Information Facility (http://data.gbif.org), which brings together several hundred natural history collections and offers a gateway for the consultation of specimen data. Since 2003, the Web site of the Muséum National d’Histoire Naturelle in Paris has been providing data and photographs of all the Orchidaceae included in its collections (http://www.mnhn.fr/base:sonnerat.html). The Aluka Foundation proposes a partially paying service for consulting all types from Africa and America (http://www.aluka.org). The Swiss Orchid Foundation at the Jany Renz Herbarium, Basel (http://orchid.unibas.ch/site.sof.php), offers a very comprehensive Web site.

The poor condition of the material makes morphological approaches particularly difficult to conduct. Vanilla herbaria are therefore currently insufficient to resolve the genus taxonomy for technical and biological reasons. The difficulty of preparing suitable samples is a major handicap.

Tips for Preparing Good Herbarium Samples

Prepare the Flowers

The flowers are often thick and fleshy. Failure to prepare them in the herbarium usually leads to the deterioration of structures. It is therefore necessary to dissect fresh flowers and to dry the perianth parts separately (Figure 4.1). Petals and sepals are detached from the flower; one of the two petals and one of the two lateral sepals are presented on the lower face. The lip, which is always fused lengthwise with the column, is cut on the side in order to be able to spread it out and to present the higher face with its scales and its gibbosities. The large inflorescences must imperatively be simplified, retaining only two or three flowers.

FIGURE 4.1 Dissection of flower and presentation of the leafy stem of Vanilla trigonocarpa Hoehne.

Prioritize Collections in Alcohol

Collections in 70% alcohol are valuable additions to the herbarium. Higher levels that make the tissue brittle should be avoided. Ensuring that the structures remain flexible and unpressed makes them easier to study (Figure 4.2). Alcohol also makes it easier to observe vascularization. The alcohol collections are, however, difficult to maintain since large volumes of alcohol require strict safety conditions and the level of the liquid must be constantly monitored owing to evaporation. Furthermore, alcohol specimens cannot replace dry samples, but they make a very useful addition.

FIGURE 4.2 Dissection of a flower of Vanilla bahiana Hoehne preserved in 70% ethanol.

Photographs taken on the field and the fine details of flowers can also be very useful. However no picture, even the best possible one, can replace a collection specimen. Botanists observe many minute details, such as pilosity on the labellum, the structures on the column or organ vascularization, which photography cannot capture.

Phylogenetic analysis, phenetic, and then cladistic methods helped considerably with the comprehension of the plant groups. The use of molecular characters, today very widespread, made it possible in numerous cases to confirm the morphological hypothesis. A combination of molecular genetics and taxonomy, is a very powerful tool. Vanilla herbaria collections must therefore be accompanied by specimens in silica gel to allow further molecular studies, for example, pieces of leaf rapidly dehydrated in absorbent silica crystals.

The herbaria are used as tools in the identification of the species. For all the aforementioned reasons, this is not always easy to obtain. When a regional flora exists, or if a genus has been revised, species determination can be based on the keys of identification defined. But some characters nonetheless require certain knowledge of the group.

The specimen used for the reference to a species name is the holotype (Figure 4.3). In theory, any specimen should be compared with this holotype in order to be suitably identified. It does not, however, represent the variability of a species. Holotypes are used for the stability of names. But it is necessary to examine all the specimens to have an idea of the taxonomy, the number of species, and the morphological relationships.

FIGURE 4.3 Vanilla ochyrae Szlach. et Olsz. (holotype). [Reproduced from the Herbier National (MNHN) Web site. With permission.]

The duplicates of the holotypes are the isotypes (Figures 4.4 and 4.5). These are different parts of the same individual collected on the same day by the same botanist. They are often deposited in other institutions. Isotypes are important for ease of consultation and for the security of the material. An unhappy incident was observed after the Second World War, when the Berlin herbarium was largely destroyed, including the type collections. The isotypes conserved in other herbaria are the only reference material available to the scientific community. The International Code of Botanical Nomenclature (McNeill et al., 2006), which sets the rules for giving and using plant names, envisaged the replacement of a holotype that disappeared with an isotype. If there is no isotype, one of the other specimens (paratypes) quoted in the original publication will be used. This replacement specimen is called the lectotype. If all the original material is lacking, the botanist who revised a species can choose a specimen (neotype) that best corresponds to the description published.

FIGURE 4.4 Vanilla tahitiensis Moore (isotype). [Reproduced from the Herbier National (MNHN) Web site. With permission.]

FIGURE 4.5 Vanilla humblotii Rchb.f. (isotype). [Reproduced from the Herbier National (MNHN) Web site. With permission.]

Scarcity of Flowers in Nature

Vanillas are seldom observed as flowers in the field. In floristic inventories, it is much more common to observe stems and leaves than inflorescences or fruits.

For these reasons, ex situ collections must be created to make it possible to monitor the bloom and properly prepare the samples. If information on initial collections in the field is suitably preserved, the specimens will appear rich and ready to be studied.

Herbarium specimens are difficult to prepare, expensive to maintain, and require specific skills for analysis. They are, however, absolutely indispensable in order to revise the taxonomy of the genus as long as molecular tools, based on DNA analyses in particular, have not been established.

References

Anonymous, 2009. The International Plant Names Index (2009). Published on the Internet http://www.ipni.org [accessed October 28, 2009].

Hoehne, F.C. 1941. Arq. Bot. Estado Sao Paulo new ser. 1, fasc. 3: 61, tab. 81.

Hoehne, F.C. 1944. Arq. Bot. Estado Sao Paulo n. ser. form. maior, 1: 125, tab. 136. McNeill, J., Barrie, F. R., Burdet, H. M., Demoulin, V., Hawksworth, D. L., Marhold, K., Nicolson, D. H., et al. (eds.). 2006. International Code of Botanical Nomenclature (Vienna Code) Adopted by the Seventeenth International Botanical Congress Vienna, Austria, July 2005. Gantner Verlag, Ruggell, Liechtenstein.

Miller, P. 1768. Gard. Dict., ed. 8. n. 2.

Portères, R. 1954. Le genre Vanilla et ses espèces. In: G. Bouriquet, ed. Le vanillier et la vanille dans le monde. Editions Paul Lechevalier, Paris, 94–290.

Rolfe, A. 1896. A revision of the genus Vanilla. Journal of the Linnean Society (Botany) 32:439.