Evidence
Three words that decide a safety question
Botanical names look like decoration on a supplement label, and most readers skip them. They are not decoration. They are four separate columns of information, and on one common ingredient the difference between two of those columns is the difference between a preparation a public health body calls unsafe and one it calls likely safe.
A botanical ingredient is identified by four things: the species, the plant part, the preparation and the standardisation. Drop the species and you may be reading research about a different plant. Drop the plant part and you may be reading it about a different tissue. Drop the preparation and a weight tells you nothing. Drop the standardisation and neither does a concentration.
Four layers, not one name
A complete botanical row looks like this: Turmeric (Curcuma longa) rhizome extract, standardised to 95 per cent curcuminoids, 500 mg. Five pieces of information, four about identity and one about quantity.
Each identity column exists because dropping it breaks something specific. Together they answer one question — is the material in this bottle the material anybody studied? — and the answer has to be yes before a weight is worth arguing about.
The columns are not equally important on every row. On a plant with one edible part and one traditional preparation, the plant part is nearly redundant. On a plant where one tissue is used medicinally and another is toxic, it is the only thing that matters. A reader treating the columns as a uniform checklist will miss which row carries the weight.
The species
Common names are unreliable in three ways at once: they travel between languages, they attach to unrelated plants, and they are marketing property in a way a binomial is not. A species name is a key that opens a database; a common name is a phrase that might.
Boswellia serrata names one tree among several in the genus, and it is the one the research is about: NCCIH records that “boswellia serrata extract in doses up to 1,000 mg daily has been safely used in several clinical trials lasting up to 6 months”. Because the name prints serrata, that sentence can be quoted without an argument about species.
Where the binomial is absent, the research is often still findable under the common name, and that is worth saying rather than overstating: ginger and horse chestnut are not ambiguous words in ordinary use. What is missing is the confirmation — nothing says the material in the bottle is the plant the monograph describes.
The plant part
A plant is not chemically uniform. Root, rhizome, leaf, seed, bark, resin, flower and fruit are different tissues with different constituents, and research on one does not transfer to another without an argument. Four examples make the point better than a rule does.
| Botanical | The part the research uses | What happens if the part is not stated |
|---|---|---|
| Turmeric | The rhizome. NCCIH’s file concerns the rhizome of Curcuma longa. | Little, in practice — the rhizome is what is sold as turmeric. A low-risk omission. |
| Ginger | The rhizome again, and again the usual material. | Little. The gap is one of precision rather than of substance. |
| Boswellia | NCCIH describes an extract “made from the gum resin or bark of the tree”. | Something, and it is practical rather than theoretical: gum resins are poorly soluble in water, which bears on how a liquid preparation was made. |
| Horse chestnut | The seed, and only as a standardised extract. | A great deal. This is the row where the column stops being a technicality. It is the subject of a section below. |
Read the third column downwards: minor omission, minor omission, practical question, safety question. That is why a mechanical checklist is less useful than a reader who knows which rows are dangerous to leave vague.
The preparation
Having fixed the plant and the tissue, the next question is what was made from it. Four answers are common, and they are not interchangeable.
- A powder. The dried tissue, milled. Whatever was in the plant is in the capsule at roughly the plant’s own concentration — which for most measured compounds is low single figures per cent.
- An extract. Compounds pulled out with a solvent and the solvent removed. Concentrated relative to the powder, by an amount nobody can guess without a number.
- A compound. The measured substance itself, named directly — curcumin rather than turmeric extract. This is the strongest of the four, because naming the compound is the specification.
- A complex. A phytosome, for example: the compound bound to a phospholipid, usually to improve absorption. Informative about the format, and silent about how much of the complex is the compound.
The word extract is where most of the trouble lives, because it reads as technical and specifies almost nothing. An extract without a standardisation is a weight of unspecified concentrate: the same 100 mg could be mild or twenty times stronger. That is a statement about what the word can carry, not an accusation about any label.
The standardisation
A standardisation is the number that makes extract mean something: 95 per cent curcuminoids says that in every hundred milligrams of the material, ninety-five are the compound anybody measured. It is a real disclosure and most labels in this trade print none.
It is also the column most often misread, because a percentage looks like a measurement and is not. A standardisation is a percentage of a weight, and it is only as useful as the weight it multiplies. Ninety-five per cent of a gram is 950 mg; ninety-five per cent of two milligrams is about 1.9 mg. Same percentage, nothing in common.
Standardisation is not extract ratio
A standardisation describes the finished material: what fraction of the powder is the measured compound. An extract ratio — 10:1 — describes the manufacturing: how much raw plant went in per part that came out. The first is the more useful; neither substitutes for the other, and neither substitutes for a weight.
One more case belongs here. For a complex, the analogous figure is the complexation ratio, which says how much of the complex is the compound and how much is carrier. A phytosome named without its ratio is making a bioavailability claim in a single word with no number behind it.
Four rows, read across
Now take four botanicals as a label might print them, and read what each row does and does not license.
| As printed | Species? | Part? | Preparation? | What a reader can do |
|---|---|---|---|---|
| Turmeric (Curcuma longa) rhizome extract, 95% curcuminoids | Yes | Yes | Yes | Everything. Find the right research, and convert a weight into milligrams of the measured compound. |
| Curcumin | n/a — a compound | n/a | Yes, implicitly | Most things. The compound is its own specification; only the weight is missing. |
| Ginger extract | No | No | Partly | Find the research, cautiously. No conversion is possible, because an extract with no standardisation is concentrate of unknown strength. |
| Horse Chestnut | No | No | No | Read the next section. |
Three of those rows fail in ways that cost precision. The fourth fails in a way that costs something else.
The horse chestnut row
Here are two consecutive sentences from the relevant public body, quoted in full and in order.
“The raw seeds, bark, flowers, and leaves of horse chestnut are unsafe when used orally because they contain a toxic component. Standardized horse chestnut seed extracts, from which this component has been removed, are likely safe for short-term use and have been used safely in research studies for up to 12 weeks.”
NCCIH — Horse Chestnut: Usefulness and Safety
Everything separating unsafe when used orally from likely safe lives in three words: standardised, seed, extract. One of them is the plant part. Two of them are the preparation and the standardisation. All three are columns from the framework above, and on this row they are not conveniences — they are the whole distinction.
A label printing “Horse Chestnut” and nothing else has not said which side of that line its material falls on. That is not evidence that it falls on the wrong side, and nobody should read it so. The question is simply unanswerable from the document — and where there is no Supplement Facts panel either, there is no second document to check.
What the processed extract is studied for is narrower than the marketing around this plant suggests. NCCIH says it is “promoted for chronic venous insufficiency” and that “a 2012 systematic review of 17 studies suggested that horse chestnut seed extract can improve symptoms of CVI” — a vein condition, and again the seed extract.
What this licenses you to say
This framework is useful precisely because it is modest, so its limits are worth stating.
What a missing column does license
- Saying that a specific question cannot be answered from the label.
- Asking the seller a specific question, in one sentence, that would answer it.
- Declining to apply a monograph about one plant part to a row that does not state its part.
- Treating a weight of “extract” as a weight of unknown concentrate.
- Noticing which rows the omission matters on, and saying so rather than treating all six alike.
What it does not license
- Saying the product contains the wrong form. A blank column is not evidence of anything.
- Saying the ingredient does not work. Absence of specification is not absence of effect.
- Saying the manufacturer broke a rule. Most of these columns are voluntary.
- Filling the blank with the most likely answer and then reasoning from it.
- Recommending anything medical. That belongs with a clinician.
The second column is longer than the first, and that is the right shape. A method concluding more than the documents support is not a method; it is a guess with a framework attached.
And the bottle this site is about
Lymph Flow names six ingredients in marketing artwork: boswellia serrata, curcumin, horse chestnut, gotu kola, quercetin phytosome and ginger extract. Against the four columns: one species, no plant parts at all, two preparations named without numbers, and no standardisation on any row.
Two rows do genuinely well. Curcumin is named as the compound rather than as turmeric, the strongest of the four preparation answers, and boswellia serrata carries its species. Four of the six ingredients have real NCCIH monographs, a better hit rate than most products this desk covers.
The row that matters is the third. “Horse Chestnut”, with no plant part, no statement that it is an extract and no standardisation — and there is no Supplement Facts panel on this product to check it against. It is finding 5 on this site, and it is the first question for the order desk: is it the standardised seed extract? One sentence answers it.
And if you are reading this because of swelling you have noticed, the order of operations matters more than the label does. The National Cancer Institute advises: “Contact your doctor as soon as you notice heaviness, swelling, or other signs of lymphedema.” Keep taking or wearing anything you have been prescribed or fitted with.
Ask for the panel, and ask about the horse chestnut
A 2 fl oz bottle of liquid drops, two droppers a day, and no Supplement Facts panel on anything supplied with it. The only quantity that exists — 600 mg of blend a serving — is printed in a marketing graphic. No price was supplied either; the live figure and the 60-day return terms are on the seller’s checkout screen.
60-day window · Questions first? Call +1 (302) 200-3480