CITES Flora and Plant Product Forensics

The Orchidaceae family contains approximately 28,000 accepted species. All of them are listed on CITES: around 800 species on Appendix I (no commercial trade permitted) and the remainder on Appendix II (trade requires permits). In practice this means any seized or undocumented orchid is a CITES exhibit, and the question for enforcement is almost always which species it is, not whether it is listed.

Two trade streams dominate orchid forensics. The first is live plant trade: artificially propagated specimens in pots are legal and constitute the bulk of orchid commerce in garden centers worldwide. Wild-collected specimens, identifiable by their irregular rhizome structure and lack of artificial substrate, are illegal for Appendix I species. The second, smaller, and more damaging stream is the salep trade in the eastern Mediterranean and central Asia. Tubers of terrestrial orchids (primarily Orchis, Dactylorhiza, and Anacamptis species) are harvested, dried, and ground into a flour used for a traditional hot drink and for ice cream thickening. Turkey consumes an estimated 10-20 million wild orchid plants per year in this trade, with some estimates based on recorded harvest volumes suggesting the figure may be considerably higher. Because the tubers are morphologically very similar across species, DNA barcoding is the only reliable method to determine which species were harvested.

For live specimens, morphological identification by a trained botanist using keys and reference herbaria is still the primary method. DNA barcoding is used when a specimen lacks flowers or when a large consignment needs rapid screening. For epiphytic tropical orchids in pots, the distinction between wild-collected and artificially propagated is often made by examining root structure (terrestrial potting mix roots vs. aerial root morphology) and leaf symmetry, but these characters require specialist knowledge. Cytological methods (counting chromosomes to detect polyploidy from tissue culture protocols) have been used in some jurisdictions to distinguish propagated from wild-collected Dendrobium and Phalaenopsis.

Agarwood (oud, oudh, agar, or gaharu in different markets) forms when Aquilaria or Gyrinops trees respond to fungal infection by producing a dense, resin-impregnated heartwood. In the wild, only a fraction of trees produce agarwood naturally. Aquilaria malaccensis (Malay eaglewood), the most-traded species historically, is listed on CITES Appendix II, as are all Aquilaria and Gyrinops species. The global market for agarwood and oud oil is estimated at several billion dollars annually, making it arguably the highest unit-value wildlife product after rhinoceros horn.

Forensic identification of agarwood uses three complementary tools. Wood anatomy (vessel arrangement and parenchyma in thin sections) confirms the genus Aquilaria or Gyrinops and distinguishes genuine agarwood from substitute woods. This works for chip-form agarwood but is limited for oil or incense products where wood structure is destroyed.

• GC-MS oil fingerprint: Genuine agarwood oil contains characteristic chromone derivatives (2-(2-phenylethyl)chromones) and sesquiterpenes (agarospirol, jinkoh-eremol). Synthetic oud and non-Aquilaria wood do not produce these chromones. GC-MS analysis of a suspected oud oil can confirm or rule out genuine agarwood resin as the source in under two hours.
• DNA barcoding: From wood chips or coarse powder, rbcL and trnH-psbA can identify Aquilaria to genus level and sometimes to species. Oil and highly processed incense forms do not retain amplifiable DNA. For these, isotope and chemical fingerprinting are the only remaining options.
• Isotope and element fingerprinting: Stable isotope ratios and ICP-MS element profiles in agarwood oil can distinguish regional origins (South Asian vs. Southeast Asian vs. Papua New Guinean). A pilot reference database for Aquilaria isotope fingerprints has been built as part of a collaboration between Singapore's Nanyang Technological University and CITES scientific committees.

The cactus family (Cactaceae) is listed almost entirely on CITES Appendix II, with several highly threatened species on Appendix I. Among the Appendix I cacti are Ariocarpus retusus and related star cactus species native to the Chihuahuan Desert in Mexico. These small, flat-bodied cacti are collected for their unusual appearance and command prices from several hundred to several thousand dollars per plant among specialist collectors. Single seizures at European and Asian ports have involved hundreds of plants packed without substrate in postal parcels.

Identification of cactus seizures is primarily morphological for intact plants because the family's diagnostic characters (spine arrangement, areole structure, growth form, flower position) are reliable and well-documented in the literature. DNA barcoding with rbcL and matK adds species-level confirmation where morphology is insufficient and is increasingly used for Appendix I verifications that may face legal challenge. For dried cactus material used in herbal preparations (peyote, Lophophora williamsii, Appendix II), alkaloid chemical profiling by HPLC or GC-MS is a faster alternative to DNA for identifying the active compounds.

Succulents beyond cacti, particularly Aloe, Agave, and stem-succulent Euphorbia, are covered by CITES Appendix II. Aloe ferox (bitter aloe from South Africa) and Aloe vera are both listed but at different regulatory levels. The enforcement challenge in the aloe trade is that processed products (aloe gels, capsules, juice concentrates) may not declare the species used, and CITES documentation requirements apply to the raw plant, not necessarily to processed derivatives depending on the level of transformation. DNA metabarcoding of commercial aloe products has revealed species substitution in multiple market studies.

Herbal medicines and dietary supplements represent a major forensic challenge because they combine multiple plant species in processed, dried, or extracted form, often in a single tablet. Conventional single-species PCR cannot screen for all potential CITES-listed ingredients at once. Metabarcoding provides a solution: amplify a short universal barcode region from all DNA in the sample, sequence everything in parallel on a high-throughput platform, and query each unique sequence against a reference database.

The workflow is borrowed directly from environmental DNA (eDNA) studies. The key steps are: bulk extraction of total DNA from the processed product, PCR amplification with universal plant primers (usually targeting ITS2 or trnL for herbals, as these are short and amplifiable from degraded material), next-generation sequencing on an Illumina or similar platform, bioinformatic filtering of reads, and BLAST or custom database matching. A study by Newmaster et al. (2013) published in BMC Medicine reported that a substantial proportion of commercial herbal products in North American markets contained species not listed on the label, including some potentially allergenic or toxic plants; however, the paper was retracted by BMC Medicine in July 2024 after a University of Guelph investigation found evidence of data fabrication, and its findings are no longer considered reliable.

• Detected unlisted CITES-listed ingredients: Metabarcoding of Chinese traditional medicine products has identified Appendix I and II plants (Paris polyphylla, Fritillaria spp.) in tablets where they were not declared. This creates both a CITES enforcement issue and a consumer safety concern.
• Species substitution: Expensive or hard-to-source CITES-listed plants are frequently substituted with non-listed species. Genuine Panax ginseng may be replaced by cheaper Siberian ginseng (Eleutherococcus senticosus), which has different pharmacological properties.
• Reference database coverage: The accuracy of metabarcoding identification is limited by database completeness. ITS2 in GenBank has better plant coverage than rbcL for many medicinal families (Apiaceae, Lamiaceae), but significant gaps remain, particularly for Asian medicinal plant diversity.

The Consortium for the Barcode of Life (CBOL) Plant Working Group published the case for a two-locus plant barcode in 2009. The recommendation was rbcL plus matK. Neither alone gives sufficient resolution: rbcL is easier to amplify from degraded material but is less variable, typically resolving to genus or family. matK is more variable and gives better species-level discrimination but is harder to amplify reliably from processed material and has PCR primer mismatches in some plant families.

In practice, forensic plant laboratories use rbcL + matK as the starting point, add ITS2 for herbal and processed products where nuclear DNA gives better recovery, and use trnH-psbA as a supplemental marker when both primary loci fail. For the most commercially important CITES-listed species (Aquilaria, Swietenia, Dalbergia, Panax), species-specific PCR assays with diagnostic restriction enzyme digestion have been developed as rapid screening tools, allowing a laboratory to process large numbers of samples without full sequencing.

TRAFFIC's periodic plant trade reports synthesize CITES trade database exports with seizure data to identify trafficking routes, source and destination countries, and which species and product types are moving in the largest volumes. The 2020 TRAFFIC report on plant trade found that flowering plants (dominated by orchids and cycads) represented the largest number of CITES plant seizures by count, while timber and agarwood products represented the highest estimated values. The two streams require different enforcement responses: orchids need morphological and molecular identification capacity at major postal and air freight hubs; agarwood requires chemical fingerprinting and GC-MS capability at sea ports.

A consistent enforcement gap identified by TRAFFIC across multiple reports is the lack of forensic laboratory capacity in source countries. Countries with the highest plant biodiversity (and therefore the greatest trafficking exposure) often lack the molecular biology infrastructure to conduct DNA barcoding casework, and must send samples to reference laboratories in Europe, North America, or Australia. This creates delays, chain-of-custody complexity, and cost barriers that mean many seizures result in administrative penalties rather than prosecution. Capacity-building programs run by UNODC, CITES, and bilateral aid agencies have expanded laboratory access but have not yet closed the gap.