Botanical Identity Testing: HPTLC, Microscopy and DNA Methods for Herb Buyers
RFQ snapshot
- A 2013 DNA-barcoding study in BMC Medicine tested 44 herbal products from 12 companies and found about 60 percent carried plant species not on the label. The method was disputed, but the buyer lesson stands: an identity line on a COA is only as good as the test behind it.
- No single method covers every format. Macroscopy suits whole and cut material, microscopy suits cut and powder, HPTLC fingerprinting suits powder and extract, and DNA methods suit raw material but struggle in processed extract.
- Under US cGMP for supplements (21 CFR 111.75), the manufacturer receiving a dietary ingredient has to run at least one appropriate identity test or examination on it.
- Silk Route Ventures (SRV) ships herbs from the Silk Foods Ceylon (SFC) Matale site against the buyer’s written spec, with lots traceable from raw material intake to shipment and a COA on request. Identity methods are buyer-specified third-party tests, so the RFQ has to name them.
- This post is for herb buyers writing the identity line of an RFQ. If you buy on a supplier’s say-so alone, no test method will rescue the spec.
A herb RFQ usually spends three lines on moisture and microbiology and one line on identity: “botanical identity confirmed.” That one line carries the most risk in the document. Moisture is easy to fix after the fact. A wrong plant is not. This piece sets out which identity method suits which format, what US cGMP asks of you, and how to write the line so a lab and a supplier read it the same way.
What does US cGMP actually ask for on identity?
Under 21 CFR 111.75(a)(1)(i), a supplement manufacturer must conduct at least one appropriate test or examination to verify the identity of any component that is a dietary ingredient, unless FDA has granted an exemption petition under 21 CFR 10.30. The wording is “at least one appropriate” test. The rule does not name HPTLC, microscopy or DNA, and it does not say which is appropriate for your herb.
That gap is where buyers get caught. “Appropriate” depends on the form of the material and on the adulterants you can reasonably expect. A single test that cannot distinguish the target plant from its likely substitute is a test the file will struggle to defend in an FDA inspection. The identity test also sits with the manufacturer receiving the ingredient, so a supplier’s paperwork supports your file but does not replace your decision. Confirm the current text and any exemption status with your regulatory adviser before relying on this summary.
Why identity testing is worth the money
In October 2013, a University of Guelph team published a barcoding survey of 44 herbal products from 12 companies in BMC Medicine (Newmaster et al., 2013). Nearly 60 percent of the products contained plant species absent from the label, 32 percent showed substitution of the labelled species, and more than 20 percent contained fillers such as rice, soybean or wheat. Only two of the 12 companies had products free of all three problems.
Industry groups challenged the study, mainly because some of the products tested were extracts, where DNA may be absent or degraded. That objection is useful in its own right. It shows why the method has to follow the format, which is the point of the next section. The consequence for procurement is practical: the herb trade has a documented substitution problem, and the RFQ is the cheapest place to address it.
Which identity method suits which format?
Four tools cover most herb RFQs. Each answers a different question, and each fails in a predictable place.
| Method | Whole or cut | Powder | Extract | What it shows | Where it fails |
|---|---|---|---|---|---|
| Macroscopic and organoleptic | Good | Poor | Not usable | Gross morphology, colour, odour, taste | Needs intact pieces and a trained examiner |
| Microscopy | Good | Good | Not usable | Diagnostic tissue: trichomes, crystals, stone cells, fibre | Needs reference slides; cannot see chemistry |
| HPTLC fingerprint | Good | Good | Good | Chemical profile matched to an authenticated reference | Needs a reference standard; profile can shift with season and extraction |
| DNA barcoding | Good | Good | Weak | Species-level genetic match | DNA can be absent or degraded in extract; added excipients can mask the target |
Source: SRV trade desk summary of the methods literature cited in Sources.
Macroscopic examination is the oldest test and the cheapest, and a 2016 review in Planta Medica notes it is not suitable for powdered or extracted material. Once a leaf is milled, the shape is gone. That is why powder buyers lean on microscopy and chemistry instead.
Microscopy reads the tissue that survives milling. Trichomes, calcium oxalate crystals and stone cells are diagnostic for many species, and a trained microscopist can often spot a foreign leaf or a bulking filler in a powder that looks fine to the eye. It cannot read chemistry, so it is poor at catching a closely related species with similar anatomy.
HPTLC compares the chemical fingerprint of your sample against an authenticated reference run on the same plate. It works on powder and extract, which makes it the workhorse for processed material. USP publishes a general chapter, <203>, on HPTLC identification of botanical articles (check the current USP-NF revision for the procedure in force). The weak spot is the reference: a fingerprint is only as authoritative as the standard it is matched to, so the RFQ should say which one.
DNA barcoding gives a species-level answer on raw and lightly processed material. The same 2016 Planta Medica review concludes that universal primers should not be the sole identification tool for processed plant material, and a 2024 study in Frontiers in Pharmacology found HPTLC combined with metabarcoding worked best for highly processed ingredients, while barcoding alone was insufficient for them. Use DNA for whole, cut and powder lots with a substitution risk. Do not lean on it for a standardized extract.
How should you match the test to the format?
Start from what you are buying and work backwards to the likely adulterant.
Whole and cut herb. Macroscopic examination by a trained examiner, plus microscopy or HPTLC as the confirmatory test. Add DNA when the herb has a known look-alike.
Powder. Microscopy plus HPTLC is the standard pairing: one reads tissue, the other reads chemistry. Together they cover most substitution and bulking cases. Add DNA if the botanical is high value or a look-alike is common.
Extract. HPTLC against a reference extract, with the marker assay (the percentage the COA states) as a second line. Treat DNA as supporting evidence at best.
Blends. Identity is per component, so a blend needs either testing of each raw material before blending or a method validated for the mix. A finished-blend fingerprint alone rarely proves each herb is present.
Gotukola as a worked example
Gotukola (Centella asiatica) shows why the line has to be specific. Common names for the plant vary across the trade, so the RFQ should name the binomial, not the local or marketing name. Whole-leaf powder and a standardized extract are different ingredients, covered in the gotukola powder spec and evidence guide, and they call for different identity tests: microscopy plus HPTLC on the powder, HPTLC plus a triterpene assay on the extract.
A line a lab can run for gotukola powder might read:
Botanical identity: Centella asiatica, leaf, verified by microscopy and by HPTLC fingerprint against an authenticated reference, method and reference named on the COA. Reporting: pass or fail against the reference, with the lab and method on the COA. Third-party test, buyer-specified.
Writing the identity line into your RFQ
A good identity line has five parts, and a missing part is where disputes start.
- The botanical name, part used and form (for example, leaf, milled powder)
- The method or methods, named, not “per supplier standard”
- The reference: authenticated standard, pharmacopoeial monograph or in-house reference with provenance
- The reporting format: pass or fail, plus the lab and method on the COA
- What happens on a fail: reject the lot, retest on a retained sample at a named second lab, or escalate
The RFQ should also say who pays for and who commissions the test. For high-risk herbs, many buyers commission an independent lab on a retained sample rather than rely on the supplier’s own report. That choice is yours to make, and it belongs in the RFQ, not in a later email.
Identity sits next to contaminant testing in a complete spec. The per-lot heavy metal and pesticide testing guide and the traceability depth and COA cadence primer cover the other lines on the COA. For extract buyers, the standardized botanical extracts RFQ checklist goes further on marker assays.
Buyer’s checklist: identity testing in a herb RFQ
- Botanical binomial, part used and form stated
- At least one method matched to the format (see the table)
- Two complementary methods for powder from a high-substitution herb
- Named reference standard or monograph
- Lab and method stated on the COA
- Fail action and retest lab agreed before the PO
- Sample tested against the spec before any purchase order
- Lot batch code carried from intake to shipment
Frequently asked questions
Which identity test is best for a powdered herb?
For most powders, microscopy plus HPTLC. Microscopy reads surviving tissue such as trichomes and crystals, and HPTLC matches the chemical fingerprint to an authenticated reference. Together they cover most substitution and filler cases. DNA barcoding can be added for high-value botanicals where a look-alike is common (Planta Medica, 2016).
Does DNA testing work on herbal extracts?
Often not. DNA can be absent or heavily degraded after extraction, and the 2016 Planta Medica review advises against universal barcoding as the sole tool for processed material. A 2024 Frontiers in Pharmacology study found HPTLC combined with metabarcoding performed better than barcoding alone on highly processed ingredients.
What does 21 CFR 111.75 require for identity?
For a component that is a dietary ingredient, the manufacturer must conduct at least one appropriate test or examination to verify identity, unless FDA has granted an exemption on petition. The rule leaves the method to the manufacturer, so the choice and its rationale need to be documented. Confirm current text with your regulatory adviser.
Does Silk Route Ventures run identity testing on its herbs?
Identity methods are buyer-specified third-party tests, so name them in the RFQ. SRV supplies herbs to your written spec from the Silk Foods Ceylon facility in Matale, with a COA on request and lots traceable from raw material intake to shipment through batch coding. Microbiology, heavy metals, aflatoxin and pesticide residues are tested through an external laboratory on request.
What are the MOQ and lead time for herbs from SRV?
Dried herbs start from 25 kg per SKU, with trial quantities from 10 kg. Lead time is 2 to 3 weeks from PO to dispatch for standard SKUs, and 4 to 6 weeks for private label, custom specifications and organic orders. Samples are free for evaluation, and the buyer pays the courier. Pricing is FOB Colombo, with CFR and CIF on request.
How Silk Route Ventures can help
Silk Route Ventures supplies dried herbs and botanical powders, including gotukola, to wellness and nutraceutical brands in the US, EU and UK. Material ships from the BRCGS and FSSC 22000 V6 audited Silk Foods Ceylon facility in Matale, with USDA Organic and EU Organic on selected SKUs, confirmed per SKU. First-order MOQ for dried herbs is 25 kg per SKU, a COA is available on request, and free samples let your lab run its identity method against your spec before any PO. The full range is on the Sri Lankan herbs supplier page.
For brands moving from bulk herb to a finished product, SRV also runs private-label capsules and custom blends, with R&D from signed brief to approved formulation in 3 to 4 weeks at most. Contact SRV to send an inquiry or request a sample.