A plate count is the number of colonies grown from a sample of the extract, calculated per gram. “Absent” means not found in the quantity that was tested. Both describe a batch on the day of the test.
Near the end of a botanical extract specification come the microbiological rows: a total plate count, a count of yeasts and moulds, and two or three bacteria marked “absent” or “negative”. They are short and easy to pass over. An extract is made from a plant, and a reflection paper of the European Medicines Agency’s committee on herbal medicines starts from that fact: “Being of natural origin herbal substances generally have a higher microbial content compared to chemical drug substances.”1
Here are six checks to make on these rows. They draw on European guidance for herbal medicines, on food law in the European Union and the United States, and on a published study of supplements. Our earlier articles covered the specification, the extract ratio, the Certificate of Analysis, the solvent and carrier, the heavy metal and pesticide rows and the assay method; this one stays with the micro-organisms.
01 / THE ROWS
Which organisms does the specification name?
The rows are of two kinds. Two are counts: the total aerobic microbial count, or TAMC, which a specification may list under a name such as total plate count, and the total combined yeasts and moulds count, or TYMC. The name alone does not tell you the method, so ask which method was used; the sixth check returns to methods. The rest are named organisms, most of which must not be found in a stated quantity. The reflection paper concludes that specifications for herbal substances, herbal preparations and herbal medicinal products “should include tests for TAMC and TYMC and absence of certain specified micro-organisms, unless otherwise justified”.1
The European Pharmacopoeia describes tests for a short list of such organisms. Its chapter for herbal medicinal products taken by mouth and the extracts used in them (2.6.31) covers Escherichia coli, bile-tolerant gram-negative bacteria and Salmonella; its general chapter on specified micro-organisms (2.6.13) adds Pseudomonas aeruginosa, Staphylococcus aureus, clostridia and Candida albicans.1 The list is not closed. European guidance on specifications says that the source of the plant should be taken into account “when considering the inclusion of other possible pathogens (e.g. Campylobacter and Listeria species)”.2
Not every named organism is a matter of present or absent. The 2019 study described in the next check determined the “presence and count” of bile-tolerant gram-negative bacteria.3
Check: that the specification gives a limit for each of the two counts and names each of the other organisms, with a limit or the word absent for each, rather than a single line that says “complies”.
02 / THE COUNT
What does a count in CFU/g include?
A count is not a weight of bacteria. In the 2019 study described below, for example, each sample was prepared and diluted, mixed with a culture medium and incubated; the colonies that grew were counted, and the result was calculated as colony-forming units per gram, written CFU/g, or per millilitre for a liquid.3 A count covers what grows under the conditions of the test: bacteria and fungi “that may grow under aerobic conditions”.1
A result above zero is not in itself a sign of a fault, even for an extract made with alcohol. Ethanol acts on growing cells, but it is “ineffective against bacterial spores”, and because the test supplies nutrients and warmth, spores of certain bacteria, mostly aerobic Bacillus species, are counted along with the bacteria. What remains after extraction with alcohol is “represented mainly by bacterial endospores”. The paper adds that hydroalcoholic extraction with heating usually yields products with a TAMC below 10,000 CFU/ml; that figure is for the liquid, and the level may rise again when the alcohol is evaporated to make a dry extract, as the fifth check explains.1
A study published in 2019 shows the range in practice. Its authors analysed batch-release tests on 122 samples of supplements with plant ingredients and tests on 30 samples of plant-based raw materials. Of all the samples, 92.1% contained bacteria, and spore-forming Bacillus species were found in almost all. Against the limits the authors applied, 100,000 CFU/g for the aerobic count and 10,000 CFU/g for yeasts and moulds, six supplement samples and two raw-material samples exceeded the first, and five supplement samples and six raw-material samples the second.3
Check: the unit, CFU/g, and a limit for each of the two counts.
03 / ABSENCE
Absent in how much?
No test can show that an organism is absent from a whole batch. It shows that the organism was not found in the quantity examined, so the row is complete only when it gives that quantity. European Union food law writes the quantity into the limit: for Salmonella in ready-to-eat sprouted seeds the criterion is “Not detected in 25 g”, in each of five sample units.4 Its definition of a microbiological criterion has the same shape: such a criterion is based on “the absence, presence or number of micro-organisms, and/or on the quantity of their toxins/metabolites, per unit(s) of mass, volume, area or batch”.4
Test quantities differ. The 2019 study, which followed European Pharmacopoeia methods, tested for Escherichia coli in 1 g and for Salmonella in 10 g.3 The food criterion above uses 25 g. The larger the quantity, the more demanding the test, so “absent in 25 g” and “absent in 1 g” are different limits. A row that says only “negative” does not show which was applied.
Check: for each organism marked absent or negative, the quantity that was tested and the number of samples.
04 / THE REFERENCE
Where do the limits come from?
For herbal medicines in Europe the limits come from the European Pharmacopoeia. As the reflection paper describes it, chapter 5.1.8 groups finished herbal medicines into three categories, A, B and C, according to how they are made and used. Herbal teas, which are prepared with boiling water, are allowed relatively high counts. Extracts for oral use, the paper adds, should meet category C where it is shown that processing would not reduce the micro-organisms enough to reach the criteria of category B.1 European guidance on specifications asks that herbal preparations for oral use, extracts among them, be “in-line with Ph. Eur. chapter 5.1.8”.2
For the plant material itself, the same guidance notes, the pharmacopoeia’s chapters specify no limits. Acceptance criteria “should be established”, and the limits in section A of chapter 5.1.8 “may be considered acceptable for herbal substances as starting material”.2
Food law says less. The European Union’s regulation on microbiological criteria sets its limits food category by food category, and no category in its annex is named for food supplements, herbs or botanical extracts. Its only rows that are not tied to one kind of food are those for ready-to-eat foods, and they set a criterion for a single organism, Listeria monocytogenes.4 The authors of the 2019 study note that the annex does not identify supplements as a separate group, add that responsibility for their quality “rests with the manufacturers”, and suggest that criteria for them should be “not less than for plant medicines”.3 In the United States the rule for dietary supplements gives no figures either: the company that makes the supplement must establish, for each component, “limits on those types of contamination that may adulterate or may lead to adulteration of the finished batch of the dietary supplement”.5
Outside medicines, the microbiological limits on an extract specification are therefore largely a choice, and the specification should say what they follow: a pharmacopoeia chapter and category, a criterion of food law, or a figure set by the supplier or the buyer. Ask which. Ask also whether they suit the product you will make, since the limits for a finished herbal medicine depend on its form and on how it is taken.1
Check: the standard each limit is taken from and, for a pharmacopoeia limit, the chapter and the category.
05 / THE TREATMENT
How was the count brought down?
A low count can come from clean plant material and quick drying, or from a treatment applied afterwards. The reflection paper wants the first: contamination should ideally be limited “by prevention rather than by use of decontamination methods”, and a treatment may not be used “to disguise poor microbiological quality”.1
Extraction is itself a treatment, up to a point. The paper credits ethanol at 60 to 95% with “marked bactericidal and fungicidal effects against vegetative forms”. Once the alcohol is evaporated, though, the level can rise again, because the water left in the concentrate “may facilitate microbial growth”.1 Heat and steam are among the other methods the paper describes.
Two treatments call for a statement of their own. The first is ethylene oxide gas. Its use for decontaminating herbal substances, the paper records, “has been prohibited in the European Union since 31 December 1989”.1 A food directive of 1999 notes the same of dried herbs and spices, whose contamination “can no longer be treated with fumigants such as ethylene oxide because of the toxic potential of their residues”.6
The second is irradiation. The paper calls irradiation of herbal preparations “not advisable”.1 In food law, the same directive of 1999 sets the Union’s own list of foods that may be irradiated, and the list has a single entry: “Dried aromatic herbs, spices and vegetable seasonings”.6 Until the list is completed, Member States may maintain authorisations that already existed, under conditions set by the framework directive of the same date, Directive 1999/2/EC.7 Where a food or an ingredient has been irradiated, “the indication of treatment shall in all cases be given on the documents which accompany or refer to irradiated foodstuffs”.7
Most treatments do not undo what the micro-organisms have already left behind: they lower the living count “but do not reduce the levels of mycotoxins or endotoxins”. And where a treatment is used, the pharmacopoeia’s monograph on herbal drugs, as the paper quotes it, requires proof “that the constituents of the plant are not affected and that no harmful residues remain”.1
Check: whether the plant material or the extract was treated to reduce the count, by which method, and what the documents say about irradiation and ethylene oxide.
06 / STORAGE
Will the result still hold when the extract reaches you?
The microbiological rows are among those that can change after the test. Micro-organisms need water to grow.1 It is generally recognised that “in products with water activity below 0.60, moulds and yeasts do not proliferate”.8 The lowest water activity at which the vast majority of bacteria will grow is higher, about 0.85, and the paper puts the water activity of dried herbal materials stored under normal conditions at usually 0.50 to 0.60. A powder that absorbs moisture in storage can rise above that: hygroscopic materials, the paper notes, “are more prone to support microbiological growth”.1 Warmth adds to the risk. Some micro-organisms, human pathogens among them, grow best at 30°C to 40°C, particularly at high relative humidity.8
So read these rows together with the moisture or loss on drying row, the packaging and the storage conditions. A moisture figure helps to judge the risk, but such testing “cannot replace a test on TAMC and TYMC”.1
If you retest a delivery, agree three things first: the method, the quantity and number of samples, and the time. For the method, European guidance asks for “pharmacopoeial procedures (2.6.12, 2.6.31) or other validated procedures”.2 The food regulation applies the analytical methods listed in its annex as reference methods.4 The quantity and the number of samples are those of the third check. As for time, conventional methods are slow: results are not available “before an incubation period of 5-14 days”.1
Whether a result on a certificate was measured on that batch is a separate question. Our article on the Certificate of Analysis explains periodic testing and what a certificate shows for the batch in hand.
Check: the moisture or loss on drying row, the packaging and storage conditions and, before you retest, the method, the quantity tested and the number of samples.
WHAT TO SEND
State your microbiological limits with your enquiry
When microbiological limits matter to your product, tell us:
- the ingredient, and the plant part if more than one is offered;
- your intended application and destination market;
- the limits your quality team applies for the two counts, and the standard they come from;
- the organisms that must be absent, and in what quantity;
- whether you need microbiological results for each batch;
- whether you need a statement on irradiation or ethylene oxide treatment;
- which technical documents you need, such as a specification sheet or Certificate of Analysis;
- your name, company and contact details.
We offer technical documentation and COA support, and the available document set is confirmed for each ingredient and project. Specification ranges in our catalogue are references, and feasibility, availability and price are confirmed with your quotation.
This article describes published guidance and law and gives no microbiological values for YOHOLIFE® ingredients; the specification and documents for a selected grade are confirmed when you enquire. Sources 1, 2 and 8 were written for herbal medicines and are cited here for their definitions and approach. The European Pharmacopoeia is described as they describe it; consult its current edition for the text and the limits themselves. Source 3 is a published study; sources 4, 6 and 7 are European Union food law; source 5 is United States law. Great Britain has its own versions of the food rules cited, which are not covered here. Sources were checked on 12 November 2026 and are given for orientation, not as regulatory advice.
