Clearly recognizing and naming species that are exploited and commercialized is a long standing and widely shared question. It is not solely a problem in the timber industry but also in food, medicine and cosmetic (Andrianoelisoa et al. 2012, Oketch-Rabah et al. 2018). In these sectors, apart from the problem of traceability of the product and its economic value, confusions in trade names may result in sometimes serious health problems (Small 1996, Carson and Riley 2001, Bussmann et al. 2015). As for timber species, confusions may arise 1) from new taxonomic separation of previously combined species which has not been transferred from scientists to productors or traders (Iglésias et al. 2010, Rivera et al. 2013) or 2) lack of interest from productors or traders in separating species which may present close properties such as taste, medicinal or wood properties (Lamb 1963, Mariaux 1963).
In particular, confusion in the use of commercial names for timber species, discussed for the first time with regard to Gabon nearly half a century ago by Détienne (1979), persists today. If reference documents in the nomenclature of tropical woods (ATIBT 2016, Gérard et al. 2016) have made some progresses in clarifying confusions of species within commercial wood names, these progresses are still insufficient. Despite the adoption of new legislation emphasizing the sustainable use of forest resources and the development of management plans for forestry concessions based on the concept of sustainability, species of Bikinia and Aphanocalyx corresponding to andoung sensu stricto are still regularly confused with one another and with taxa that belong to other genera. Just over 10% of the trees referred to as andoung by the professional spotters surveyed in our study were correctly identified to one of these nine species. The basic training in botany that some spotters had received does not appear to provide any guarantee of their being able to make accurate identifications; on-the-job training and experience in the field appear to be more effective. While the overall level of accurate species identification of andoung sensu stricto is very low, it is worth noting that the only tree spotters who made correct identifications work for forestry companies that have a commitment to engage in sustainable management.
The low rate of accurate identification of andoung using a commercial name with a qualifying epithet may be linked to a lack of attention on the part of the Ministry responsible for Gabon’s forests, which does not require that timber companies distinguish among the various types of andoung. Moreover, poorly known or unknown species (Bikinia spp) are also exploited as andoung (ATIBT 2016). This has added to the confusion and reinforced the tendency of many spotters to take advantage of the lack of administrative clarity and simply identify trees as andoung rather than making the additional effort to assign each tree to a species using a name with a qualifying epithet. Moreover, the various species of true andoung (i.e., andoung sensu stricto) are all of the same commercial value and are therefore sold under the same name. It is thus not surprising that most forestry companies have no particular interest in the accurate identification of trees using qualifying epithets that correspond to species and have not required their tree spotters to apply these names.
Nearly 63% of the trees inventoried were identified simply as andoung (without a qualifying epithet), indicating that the tree spotters studied were, as a group, either not aware that true andoungs are including several taxa, or they were unable to differentiate among the taxa or saw no use in doing so. The fact that other species (false andoungs) were recorded in the inventories underscores the spotters’ lack of knowledge and the need to improve their skills. Nearly two decades after the publication of the field identification manual of forest trees by Wilks and Issembe (2000), the spotters included in our study continued for the most part to base their work on a considerably older work in which the name andoung was applied in the vernacular sense of the Fang people (Saint Aubin 1963), and was used to refer to taxa placed in the genera Didelotia, Julbernardia and Tetraberlinia according to taxonomic concepts that are no longer accepted. It is thus not surprising that some spotters did not distinguish between strictly vernacular names (in Fang) and commercial names. Some vernacular names have been adopted as commercial names, and several species of true andoungs (such as Bikinia coriacea, B. durandii and B. le-testui) share the Fang vernacular name andoung. The fact that the Gabonese forestry administration and the timber industry have failed to integrate the taxonomic revision of Wieringa (1999) into regulation and practice has contributed significantly to maintaining the level of confusion revealed by our study.
Among the commercial names with a qualifying epithet, andoung rouge was the most often confused. Two different types or confusion plague this name, intra-generic (between the two species of Aphanocalyx and among the eight species of Bikinia) and inter-generic (between the members of these two genera). This situation reflects the fact that members of all species of true andoungs can have red trunks, depending on where they grow, although as indicated above, only A. heitzii has an evident dark red bark slash. As mentioned above, these errors reflect either the lack of training of the tree spotters or a lack of interest in separating various species.
Three species of andoung sensu stricto (Bikinia aciculifera, B. coriacea and B. durandii) are endemic to Gabon and three others (Aphanocalyx heitzii, B. grisea and B. media) are nearly endemic (Wieringa 1999, Sosef et al. 2006,). Among the species of false andoung present in Gabon, four are likewise subendemic [Hymenostegia pellegrinii, Julbernardia brieyi, Tetraberlinia longiracemosa and T. moreliana]. However, all species of andoung, whether endemic, subendemic or more widely distributed, are harvested, without regard to their geographic range or rarity. As a result, forestry companies and the tree spotters they employ likely contribute to the over-exploitation of rare and/or range-restricted species and thereby unknowingly or unintentionally exacerbate their risk of extinction.
The accurate field identification of species of andoung remains a significant challenge for forestry companies in Gabon and the tree spotters working for them. Wieringa’s (1999) revision and description of new taxa greatly improved our understanding of species limits but let to further confusion within the forestry industry. The broadly defined commercial names currently being used, many of which refer to several taxa, are insufficient to distinguish species from one another, a situation that is likely to compromise Gabon’s efforts to reconcile economic development with sustainable use of natural resources and biodiversity conservation. The fact, for example, that Bikinia aciculifera and B. coriacea both have highly restricted distributions yet are being harvested under the same commercial name casts doubt on whether they are indeed being managed sustainably and diminishes the prospects for their long-term conservation.
The current application of commercial names with qualifying epithets to distinguish among the nine species of andoung sensu stricto is problematical, with the exception of the easily recognized species Aphanocalyx heitzii. All of the remaining species of andoung were incorrectly named by the tree spotters surveyed in our study, which points to the inadequacy of the spotters’ current knowledge and training. For a country such as Gabon, which has chosen to make sustainable forestry a cornerstone of the national economy, the ability of field staff to identify accurately all species that are marketed or potentially marketable should be a sine qua non for exploitation in forest concessions. The application of a governmental decision to prohibit the export of raw logs, starting on May 15, 2010 (Manciaux 2011), has forced the industry to strengthen capacity to transform products locally, as required since 2001 by the forestry code. However, to develop this new dimension properly, the ability to identify felled trees with a high degree of accuracy is essential in order to guarantee that wood sold under a particular commercial name is uniform in quality and corresponds to a single species.
Some species of true andoungs are rare or have a restricted distribution (Wieringa 1999, Mboma 2012) and are likely threatened according to the IUCN Red List risk of extinction criteria (IUCN 2012). Other species that are more common and widespread are probably not threatened by timber harvesting, at least not today, although the example of moabi should serve as a reminder of what can happen as a result of abusive over-exploitation. The same could be predicted for Okan, Cylicodiscus gabunensis Harms, a species of Fabaceae subfam. Mimosoideae whose export volume quadrupled in just four years (Fig. 5). In the absence of both adequate capacity to distinguish reliably among the taxa of true andoungs and a regulatory system that requires doing so, these species are being exploited without any consideration of the consequences to their conservation status or their long term survival.
To address the twin problems of confusion in the commercial names applied to species of andoung and the inability of professional tree spotters to recognize them reliably, the following measure are suggested:
-
Identify easily observable morphological characters that can be used, even on sterile trees, to enable spotters and others involved in forestry inventories make accurate field identifications of species of andoung sensu stricto.
-
Produce and distribute to tree spotters and other potential users a practical field guide (including distribution maps) of andoung species occurring in Gabon in order to facilitate identification.
-
Update Gabon’s list of marketable tree species produced by the Ministry responsible for forestry so as to bring it in line with the current taxonomic framework. Some current trade names are still standing for two botanic species and should be changed. A solution could be to add to the generic name andoung an epithet highlighting a morphological characteristic of each species. Another solution could be to add as epithet the specific botanical name; for example, andoung de Morel could be changed in andoung aciculifera (for Bikinia aciculifera) and andoung coriacea (for B. coriacea). This could have the advantage of linking more closely trade and botanical names.
-
Organize a training program and refresher courses on the identification of commercial tree species, with a emphasis on andoung sensu stricto. This could be coupled with an effort to increase the level of professionalism for those involved in tree spotting, which would benefit the spotters themselves, the forestry companies for which they work, and Gabon’s regulatory agencies.
-
Pursue and expand botanical inventory work to improved our knowledge of the geographic distribution and ecology of the species of andoung, especially those that are rare and/or have a restricted range, in order to inform the evaluation of their conservation status and risk of extinction.
-
Initiate a research program on andoung species, focusing on phenology, reproductive biology, regeneration, population characteristics, growth rates, diameter and its relationship to minimum flowering size, etc., in order to inform and improve management policies so that they fully consider the unique aspects of each species.
In order to evaluate and verify the reliability and accuracy of identifications made by tree spotters and others professionals in the forestry industry, forest inventory of timber trees should be monitored before, during and after harvesting. This could be facilitated, including for species of andoung, by using DNA barcoding, which has proven to be effective for commercial timber species elsewhere in the world (Gonzalez 2009, Gonzalez et al. 2009). A unique genetic fingerprint would be established for each of the nine taxa of andoung, making it possible to identify living trees before harvest as well as logs once they have been cut. This could also help strengthen the conservation and rational exploitation of andoungs by ensuring that threatened species are not harvested, while enabling accurate identification of those that can be cut as part of a sustainable forest management program. Toward this end, Gabon should develop the expertise and technical capacity needed to establish a DNA barcoding program for its entire forestry industry, with an ultimate goal of ensuring that all timber trees can be identified reliably and that only authorized species are harvested.
Finally, as mentioned earlier, our work on the confusion of timber species is undoubtedly of a wider importance within and outside the timber sector itself. Most of the above recommendations might also be valid in other contexts.