The List That Would Answer This

Ask the internet which of your products stop being hemp next month and you will get a list of compounds and a pair of dates. Ask it four different ways, as we did this morning, and you will get the same list and the same dates every time, stated without hedging.

The federal act those answers are about contains an instruction to publish that list. Nobody we can find has published it.

What we were actually looking for

We ask these questions in an operator's words on purpose — not to find out what is true, but to find out what people are being told. This morning the phrasings were about November 12, about what changes for a shop, about the 0.4 milligram figure, and about the lists themselves.

The first three came back naming compounds. Delta-8. HHC. THC-O. THCP. THCA flower. Sorted into two buckets with two dates attached.

We have written before about why we do not repeat those lists: the act names no individual cannabinoid anywhere in its text. We counted. That piece is here, and it is the direct predecessor to this one: gethempdash.com/learn/what-the-act-does-not-name

What we had not noticed, in two separate first-party reads of the same document, is that the act does something more specific than stay silent. It tells somebody to go and make the list.

Paragraph (3)

We pulled the enrolled text of the act again this morning and extracted the whole of Section 781 — 5,953 characters, end to end — rather than the definition we had quoted before.

The section does three things. It renumbers some paragraphs. It replaces the hemp definition with the new one, including the two origin exclusions and the two combined-total limbs. And then, after all of that, it adds a third instruction that is not part of the definition at all.

In its own words: within 90 days of the enactment of this act, the Food and Drug Administration, in consultation with other relevant Federal agencies, shall publish four things.

• A list of all cannabinoids known to FDA to be capable of being naturally produced by a Cannabis sativa L. plant, as reflected in peer reviewed literature.
• A list of all tetrahydrocannabinol class cannabinoids known to the agency to be naturally occurring in the plant.
• A list of all other known cannabinoids with similar effects to, or marketed to have similar effects to, tetrahydrocannabinol class cannabinoids.
• Additional information and specificity about the term container.

Read that list against the questions people are asking and the overlap is close to total. Which cannabinoids can the plant make? That is the first bullet, and it is the exact question the first origin exclusion turns on. Which other compounds count as similar enough? That is the third bullet, and it is the exact question the second limb of the container cap turns on. What is a container? That is the fourth, and it is the measurement question we wrote about here: gethempdash.com/learn/rule-watch-container-level-thc-cap

Two small notes on the quotation, because we would rather flag them than have you find them. The enrolled text spells one word "know" where it plainly means known, and drops a letter from tetrahydrocannabinol in the third item. Those are the enrolled text's, not ours, and we have not silently tidied them.

The date

The act became law on November 12, 2025. Ninety days from enactment is February 10, 2026. That was 256 days ago.

The same section's other interval is the one everybody already knows: effective 365 days after enactment, which is November 12, 2026, nineteen days from today. We mention it because the arithmetic is the same arithmetic, applied to the same sentence, in the same section. One of those two dates has been the subject of near-continuous coverage for a year. The other appears to have gone almost entirely unremarked, including by us.

What we can actually say about whether it was published

Three checks, and we want to be precise about what each one is worth, because they are not worth the same.

The one that carries weight is not ours. The Congressional Research Service maintains a short report on this definition change. Version 5, updated September 28, 2026, carries a heading of its own called FDA Publication. Under it, CRS sets out the same four items and states in a parenthesis that as of September 28, 2026, such a list has not been published.

That is CRS's statement, not ours, and it is a statement about September 28. It is not a statement about today, and we are not going to quietly promote it into one.

The second check is ours and it is a clean negative. Searching the Federal Register's full-text index across everything published on or after the date of enactment, the act's own operative phrase — capable of being naturally produced — returns nothing at all. We ran controls on the identical window to prove the search was capable of returning something: tetrahydrocannabinolic acid returns two documents, total tetrahydrocannabinols returns one, cannabinoid returns twenty-six, hemp returns eighteen, and an invented term returns nothing. We then opened both near-matches. They are Drug Enforcement Administration items — a rescheduling rule from April and an exempt-chemical-preparations notice from March. Neither is from the FDA and neither is a list of cannabinoids.

The third check is ours too, and it is worth nothing, which is why we are telling you about it. FDA's main page on cannabis regulation carries no such list. It also says, at the bottom, that its content is current as of July 16, 2024 — sixteen months before this act existed. A page that predates the statute cannot tell you anything about compliance with the statute. Finding nothing there is not a finding. We logged it as a negative result that cannot bear weight and we are not counting it.

So: a dated statement from Congress's own research service, plus an independent and controlled absence from the register where such a thing would ordinarily appear. That is what we have. What we do not have is a statement from the agency, and we have not found one either way.

Two different missing things, which is easy to merge and wrong to

Our record has carried a related gap for six weeks: the container cap's second limb reaches other cannabinoids with similar effects "as determined by the Secretary of Health and Human Services," and we have never located any such determination.

It is tempting to treat today's finding as the same gap. It is not.

Paragraph (3) is a publication duty, it names the Food and Drug Administration, and its clock ran from enactment. The (bb) limb is a determination, it names the Secretary of Health and Human Services, and the text attaches no deadline to it at all. Same subject matter, different instrument, different actor, different timing. Neither one closes the other.

We say this because the two get collapsed constantly, and the collapsed version sounds more conclusive than either half. Counting the elements is the whole job here.

Why this explains the confident lists

Here is the part worth taking away.

For two months we have been writing down a particular failure shape: an answer that arrives as operations rather than argument. A date. A range. A lookback period. A compound list. Nobody argues with a list; they act on it. And specificity is exactly what gets manufactured when there is a gap to fill.

This is the gap. The statute asks which cannabinoids the plant can naturally produce, and makes a product's status turn on the answer. It then directs an agency to publish the answer. If that publication is not on file, then every list you are reading — ours would be no better — is somebody's reconstruction of a document that was supposed to exist and that we cannot find.

That does not make the lists wrong. We want to be careful here, because the inverse inference is just as unsupported as the original. An absent list is not evidence that delta-8 is in, or out, or anything else. It is evidence that the question has not been answered by the body the act assigned it to, on the surfaces we can read.

And it does not mean nothing happens on November 12. The effective date is in the text and is not in dispute.

What about the stock already on the shelf

This was the loudest thing in the video lane this week — what happens to inventory you already own, with a fair amount of advice about buying more of it now.

We can tell you one narrow thing. We scanned the complete text of Section 781 for twenty-two terms that would signal a transition: sell-through, inventory, stock, grace, disposal, transition, grandfather, possession, safe harbor, wind down, manufactured before, prior to the effective date, and a dozen more. All of them return zero. The only near-hit is the phrase stream of commerce, which appears once, inside the definition of industrial hemp, about research plants.

The controls on that same extract return cannabinoid twenty-three times, hemp twenty, container five, Secretary twice. The scan works. The section simply does not use that vocabulary.

Scope, and it matters: that is a statement about the text of Section 781. It is not a statement that no transition provision exists anywhere — in the later act, in an agency action, in guidance we have not read. We looked at one section and we are reporting what is in it.

The date everyone has settled on, including CRS

One more thing, and it is uncomfortable to write.

All four phrasings this morning said the container cap takes effect December 11. On Thursday, three phrasings split two to one on that same question. The disagreement closed this week, and it closed on the reading we do not publish.

The later act narrows the first act by naming two specific paragraphs. It does not name the paragraph the container cap is in. Our position has been that this does not resolve what happens to an unnamed paragraph, in either direction, and so we do not state a date for the cap.

The CRS report says otherwise. On its second page it writes that the other portions of the new hemp definition do not take effect until December 11, 2026.

We are not going to pretend that is a careless source. It is nonpartisan congressional staff, and it is the most authoritative reading available to us. We are reporting that it reads the split that way, with the date of the version we read, and we are leaving our own record where it is — because a reading of a statute, however well sourced, is still a reading, and ours is a record of what instruments say.

What we will say plainly is that the quality of a source predicts whether it transcribed the text correctly. It does not predict whether the inference drawn around the text is sound. A law firm, a retailer blog, a compliance vendor and the Congressional Research Service have now landed in the same place, and that is interesting but it is not the same thing as the statute saying it.

The short version

Section 781 directed the FDA to publish, within 90 days of enactment, a list of the cannabinoids a cannabis plant can naturally produce, a list of the THC-class cannabinoids that occur in it, a list of other cannabinoids with similar effects, and more detail on what counts as a container. Ninety days was February 10, 2026. CRS reported on September 28 that no such list had been published; the Federal Register contains nothing using the act's own phrase; FDA's own page predates the act and proves nothing.

The effective date of November 12 is nineteen days away and is not in doubt. The lists that would tell you which of your products that date reaches are, as far as we can establish, not on file. If you are being handed a compound list this week, the useful question is not whether it sounds right. It is where its author got it.