How Is Kratom Dosed? Why There Is No Universally Valid Answer

Status of this article: September 2026

Hardly any question about Mitragyna speciosa is asked more often than the one about quantity. The honest answer is this: there is no universally valid dosage figure — and that is not an oversight of the trade, but a consequence of how this plant material is constituted, how it has been studied, and how it is classified in law.

This article explains why. It shows why the content of the main alkaloid can differ several-fold between two batches, how a laboratory determines that content in the first place, what a percentage figure actually says — and why even the scientific literature cannot compare its own numbers with one another.

One framing remark up front: the goods described here are not intended for consumption. This text therefore contains no quantity recommendation and no assignment between an amount and a sensation. It describes material, analytics, the state of research and the legal situation.

The short answer first

For a substance to carry a defensible quantity figure, three conditions must be met. For Mitragyna speciosa, every single one of them is missing.

A defined starting material. There is no binding monograph specifying what one gram of leaf material must contain chemically. Alkaloid content varies from tree to tree and from harvest to harvest.
A uniform reference basis. A percentage without a statement of what it refers to — dry mass or fresh weight, raw leaf or extract, single sample or composite sample — is arithmetically meaningless.
Robust human data. Controlled investigations recording amount and response under standardised conditions do not exist in a volume from which threshold values could be derived.

As long as these three points remain open, every milligram scale in circulation is false precision: it suggests an accuracy the material itself cannot deliver. That is exactly why there is no longer one in this article.

Mitragynine: what the substance chemically is

Mitragyna speciosa is a tree of the madder family (Rubiaceae) — the same plant family the coffee shrub belongs to. It grows in Southeast Asia, above all in Thailand, Malaysia, Indonesia and New Guinea, and under favourable conditions reaches heights of over twenty metres.

More than forty alkaloids have been described in its leaves. Mitragynine is quantitatively the most important of them. It belongs to the indole alkaloids of the corynanthe type, carries the molecular formula C₂₃H₃₀N₂O₄ and has a molar mass of roughly 398.5 g/mol. Structurally it consists of an indole skeleton to which, via a ring system, an ester and an enol ether moiety are attached; a methoxy group at position 9 of the indole ring distinguishes it from related alkaloids of other Mitragyna species.

Further alkaloids regularly detected are speciociliatine, speciogynine and paynantheine — all of them diastereomers or close relatives of mitragynine. 7-hydroxymitragynine (C₂₃H₃₀N₂O₅, roughly 414.5 g/mol) arises through oxidation at position 7 and is usually found in dried raw leaf far below 0.05 percent. It is therefore a trace alkaloid, not a main constituent — even though it receives disproportionate attention in the literature.

Alkaloids are secondary metabolites: the plant does not produce them in order to grow, but as a response to its environment. That single sentence already explains the core of the problem — what arises as a response to environmental conditions is not a constant.

Why the content varies from batch to batch

In dried raw leaf the mitragynine share typically lies between 0.5 and 2 percent of dry mass. That is a factor of up to four between the lower and the upper edge — for nominally identical material. Individual analyses fall outside this range as well. Six causes interlock here.

Genetics and strain labels

Individual trees differ in their alkaloid pattern even when standing side by side. The designations common in trade, such as green, red or white, are neither botanical varieties nor analytical statements; they describe regional traditions and processing routes. A fuller discussion is given in the article Kratom strains explained simply.

Growing region, soil and climate

Nutrient supply, water availability, light exposure and soil chemistry alter secondary metabolism. Material from different regions of Southeast Asia therefore regularly shows divergent alkaloid patterns, even under the same trade name.

Harvest time and leaf age

Drought stress, insect pressure or a delayed rainy season shift alkaloid formation measurably. Young and older leaves from the same tree do not yield the same profile. The same tree, harvested twice in a year, produces two different numbers.

Drying

Whether drying happens in the sun, in the shade or mechanically determines the residual moisture. And residual moisture shifts every percentage figure purely arithmetically, because the percentage refers to total mass: moister material dilutes the content without a single molecule being lost.

Storage and oxidation

Light, oxygen and heat promote oxidation processes. Mitragynine is not indefinitely stable; over storage time the alkaloid profile shifts, among other things towards 7-hydroxymitragynine and further oxidation products. An analysis therefore only ever describes the state on the day of measurement, not the state one year later.

Processing, particle size and extracts

Unevenly milled material segregates inside the container and yields no representative sample. With extracts, concentration is added: when solvent is removed, the alkaloid share per gram rises without a single molecule having been newly formed. A percentage figure for an extract and a percentage figure for raw leaf therefore sit on entirely different scales.

How a laboratory actually measures the content

Anyone who wants to know what is in a sample cannot avoid instrumental analytics. The standard route runs through high-performance liquid chromatography (HPLC), frequently coupled with a tandem mass spectrometer (LC-MS/MS).

Separation and detection

The sample is first extracted, filtered and diluted, then sent through a separation column. The individual alkaloids pass through the column at different speeds and leave it at different times — this is how they are separated from one another. A UV detector registers them via their light absorption, a mass spectrometer additionally via their mass-to-charge ratio. The latter is why LC-MS/MS distinguishes the diastereomers of mitragynine more reliably than a pure UV measurement.

Reference standard and calibration curve

An instrument initially delivers only a signal, not a quantity. The signal is assigned via a reference standard — a substance of known identity and purity. From several dilutions of this standard a calibration curve is produced, against which the sample signal is read. Without a certified standard, a numerical value is not defensible. Added to this are parameters such as limits of detection and quantification, recovery rate and the control of matrix effects, often via an isotope-labelled internal standard.

Sampling and homogenisation

The underrated step comes right at the beginning. One gram out of a 25-kilogram container stands for the whole only if the material was properly homogenised beforehand and sampled according to a defined plan. Two analyses of the same batch can differ simply because they were drawn from two different places. A usable certificate of analysis therefore names method, reference standard, reference basis, sampling and measurement uncertainty — not merely a number.

What a percentage figure says — and what it does not

Two percent, taken by itself, is not information. The value only becomes a statement once the reference basis is supplied with it. In practice, batches become comparable through a uniform figure in milligrams per gram of dry mass.

Figure given	What it settles	What remains open
Percentage without reference basis	nothing defensible	Dry mass or fresh weight? Raw leaf or extract?
mg per gram of dry mass	comparison between batches	purity, contaminants, minor alkaloids
Method and reference standard	traceability of the measurement	representativeness of the sample
Batch and harvest date	traceability of origin	storage conditions after measurement
Complete alkaloid profile	share of the minor alkaloids	any statement about human beings

And a content figure says nothing about impurities. Heavy metals, pesticide residues and microbiological load are separate test parameters with their own methods; how these tests work is described in the article Heavy metals & pesticides: why quality testing matters.

Why even the specialist literature cannot compare its numbers

A frequent objection runs: but studies do contain numbers. True — except those numbers refer to such different things that placing them side by side yields nothing.

Different preparations. One paper examines milled raw leaf, the next an aqueous infusion, the third isolated mitragynine as a pure substance. Those are three different objects of study.
Different extraction procedures. Aqueous, alcoholic, acid-base: every procedure pulls out a different alkaloid spectrum. An extract is chemically not the same thing as the leaf it came from.
Missing characterisation. Older work often describes the material only in general terms, without stating the measured alkaloid content. What exactly was studied can then no longer be reconstructed after the fact.
Different models. Cell culture, rodent and human respond differently; results from one model do not translate seamlessly into another.
No uniform reporting format. Sometimes material is given in grams, sometimes alkaloid in milligrams, sometimes a ratio to body weight — without a basis for conversion.

This inconsistency is not a marginal problem of the field but one of the principal reasons why no threshold values can be derived from the existing literature. Anyone wishing to read the current state of research will find review articles via PubMed; a German-language overview is given in the article Kratom research 2025/2026: where do we stand?.

What has been observed at receptors in laboratory experiments

This section describes preclinical findings: measurements on cell cultures and isolated receptor systems, plus animal experiments. It expressly does not describe any effect in human beings.

In binding and functional assays, mitragynine shows affinity for the µ-opioid receptor and behaves there, in several investigations, as a partial agonist — it activates the receptor only partly. Notably, activation in these systems proceeds predominantly via the G-protein pathway, while β-arrestin is barely recruited. At δ- and κ-opioid receptors, antagonistic behaviour has been described in several approaches. Alongside this there are findings on α₂-adrenergic receptors, on serotonin receptors and on adenosine receptors; the numerical values differ considerably depending on assay and cell system.

7-hydroxymitragynine shows markedly higher affinity and potency at the µ-receptor than mitragynine in the same systems. In rodents it additionally arises as a metabolite of mitragynine, with the involvement of the enzyme CYP3A4; to what extent this happens in humans is assessed differently across the literature. The differences between the two molecules are set out in more detail in the article Mitragynine vs. 7-hydroxymitragynine, and the basics of receptors in the article Receptors explained simply.

Why no quantity figure follows from this can be stated precisely:

A receptor assay measures binding and signal in an isolated system. Absorption, distribution in the body, breakdown and excretion do not occur in it.
The concentrations used in the assay cannot be converted into a material quantity, because it remains unknown what share of a substance ever reaches the receptor.
There are pronounced species differences in metabolism; results from rodents are not readily transferable to humans.
Animal studies point to non-linear pharmacokinetics: twice the amount does not lead to twice the blood concentration there. A linear scale would be wrong for that reason alone.
Mitragynine binds strongly to plasma proteins and undergoes pronounced first-pass metabolism; both further shift the relation between material quantity and substance actually available.

Survey data and self-reports: what they are worth methodologically

A considerable share of the quantity figures circulating publicly stems not from controlled investigations but from online surveys and forum posts. Such data have their place — they show what people report. They simply answer a different question from the one they are usually invoked for.

Self-selection. Whoever takes part in a survey on a topic is rarely representative of the general population. Those who are satisfied answer differently, and more often, than those who are not.
No tested material. Nobody analysed what the respondents actually held in their hands. A stated number of grams says nothing about the quantity of alkaloid without a content analysis.
Recall bias. Retrospective statements about amounts and timing are systematically imprecise.
No control group, no blinding. Expectancy effects cannot be factored out when there is nothing to compare against.
Combined use stays invisible. Substances used at the same time are frequently not recorded.

Methodologically, therefore: a self-report is an account of a perception, not a measurement. It can prompt hypotheses and describe frequencies within a group. It cannot establish a quantity figure — and it is not used for that purpose in this article either. An example of how far self-reports and the research record can diverge is given in the article Kratom & opioid withdrawal: what the research shows.

Documented risks and adverse effects

What can indeed be read off the existing literature are risks. They are documented in case reports, case series, poison centre data and registry analyses, and they are named clearly here on purpose.

Frequently reported: nausea, vomiting, constipation, dry mouth, dizziness, headache, sweating, itching, rapid heartbeat and elevated blood pressure.
Liver injury. Several case series describe a predominantly cholestatic or mixed injury pattern, usually with a latency of days to a few weeks. The cases are rare but well documented.
Seizures and disturbances of consciousness. Individual case reports, predominantly at very high amounts or with the simultaneous use of further substances.
Dependence and withdrawal syndrome. After prolonged regular use, restlessness, irritability, muscle and limb pain, sleep disturbance and mood swings are documented.
Pregnancy. Case reports exist of neonatal withdrawal symptoms in newborns.
Interactions. In vitro, mitragynine inhibits CYP2D6 and CYP3A4 among others. This matters because numerous medicines are broken down via these enzymes. The literature judges combination with sedating substances to be particularly critical.
Fatalities. In toxicological analyses, further substances were detected almost throughout. Monocausal cases are the rare exception, which complicates assessment — exculpatory it is not.
Adulteration and contamination. The leaf blend traded in Sweden as krypton was laced with O-desmethyltramadol; in the United States there were recalls over salmonella. Heavy metal findings such as lead and nickel have been published repeatedly.

A fuller account is given in the article The risks and safety of kratom.

How authorities classify the state of the evidence

The missing standardisation running through this entire article reappears in official assessments as an argument in its own right: a health assessment presupposes a defined product. Where composition varies from batch to batch and is not documented, the assessment lacks its object.

EU novel food law. Under Regulation (EU) 2015/2283, foods not consumed to a significant degree in the EU before 15 May 1997 may only be placed on the market with explicit authorisation. No such authorisation exists for Mitragyna speciosa. The connection is set out in the article Kratom and novel food: what the EU regulation means; the official basis is published by the European Commission.
Risk assessment. Authorities such as the German Federal Institute for Risk Assessment regularly point out, for botanical preparations, that without a defined composition and without sufficient human data no defensible assessment is possible.
European monitoring. The EUDA, formerly the EMCDDA, covers the plant in its reporting on new psychoactive substances.
WHO. The Expert Committee on Drug Dependence concluded in 2021 that the available data do not justify international control, but expressly recommended continued monitoring — that too is a statement about the gaps in the evidence.
National rules. These differ considerably within Europe and do change. What applies is always the rule in force at the given place; an overview is given in the article Is kratom legal?.

What this means for these goods

From all of this follows the classification that applies to the goods offered here: they are not intended for consumption and are supplied as collector and research material. A quantity figure would, for this purpose, not only be legally inadmissible; it would also be impossible to justify factually — the evidence simply does not support one.

What can sensibly be documented is the material side: batch number, origin, analytical method, reference standard, reference basis and tested purity parameters. That is the part of the original question which can be answered honestly.

In brief

No universally valid quantity figure exists, because defined starting material, a uniform reference basis and human data are all missing.
Mitragynine content in raw leaf typically varies between 0.5 and 2 percent of dry mass — depending on genetics, location, harvest time, drying, storage and processing.
Measurement is done by HPLC or LC-MS/MS against a certified reference standard; without a reference basis and a sampling plan, a percentage is meaningless.
Figures from the specialist literature are not comparable with one another, because preparations, extraction procedures and reporting formats diverge.
Preclinical receptor findings describe cell systems and animal models, not quantities for human beings.
Self-reports are accounts of perceptions, not measurements.
Documented risks — from liver injury through withdrawal symptoms to interactions — are the most defensible part of the existing literature.

Legal notice

This article is for informational purposes only and does not constitute legal advice. Its content is not intended to encourage consumption. Laws may change; the applicable regulations and information from official bodies are authoritative. Image source: https://www.kratoein.com/

Back to blog