One Word, Three Tests

Sort your inventory into two piles: natural and synthetic.

It sounds like a task you could finish this morning. It is the instruction sitting underneath nearly every piece of guidance a Texas operator is being handed right now. And this morning the same molecule ended up in both piles before nine o'clock, in two sources that were each being careful.

What happened when we asked about the shelf

We ask the question an operator would actually type, and we write down what the answer says rather than which pages it lists. Today's was plain: what can I still sell right now — THCA flower, delta-8, September 2026.

One answer came back holding three incompatible things at once.

It reported that smokable hemp meeting the delta-9 limit is still being sold, because a court paused the statewide sales ban. That order was lifted in early June. The answer was describing a five-month-old posture in the present tense — the same defect we [took apart yesterday](https://gethempdash.com/learn/the-record-we-will-not-summarize), arriving again.

It reported that THCA flower is not legal, because THCA is on the controlled substances list — the Texas scheduling action, effective July 31, 2026.

And then, two paragraphs later, it gave the same product a runway: the federal total-THC rule and the container cap take effect December 11, 2026 for naturally occurring cannabinoids (THCA flower, hemp Delta-9, CBD).

THCA flower as a scheduled controlled substance, and THCA flower as a naturally-occurring cannabinoid with three months to go. One answer. One shelf.

The word is the problem, not the sources

The easy read is that somebody was sloppy. We went and checked, and that is not what happened.

Here is how a Texas newsroom described the state scheduling action on July 30, 2026, fetched directly rather than summarised:

> The ban applies to all synthetically created THC forms, including delta-8, delta-10, THCP and THCA flower.

THCA — the acid the living plant actually produces, the thing that becomes delta-9 when you apply heat — inside a list headed synthetically created.

Now here is the one source in this morning's sweep that applied the federal statutory test instead of listing products. It puts delta-8, delta-10 and THCP on the delayed side of the federal window — "even where they were produced by converting CBD, because those cannabinoids do occur in the plant" — and states plainly that the test is a question about a molecular property, "not a manufacturing method or product name."

So delta-8 sits inside "synthetically created THC forms" in one, and on the naturally-occurring side in the other. Same compound. Same morning.

Neither of them is being careless. They are answering different questions, under different instruments, from different governments — and one English adjective is carrying the sorting in both.

Three tests wearing one word

Read the federal provision's own language. The exclusions are drafted as separate subclauses, and the separation is deliberate:

> (I) cannabinoids that are not capable of being naturally produced by a Cannabis sativa L. plant

>

> (II) cannabinoids that — (aa) are capable of being naturally produced by a Cannabis sativa L. plant; and (bb) were synthesized or manufactured outside the plant

Subclause (I) asks what the species can do. Could a cannabis plant, in principle, make this compound at all?

Subclause (II) assumes it could, and asks about this particular lot. Was the material in front of you made in a plant, or in a reaction vessel?

Those are not two wordings of one test. They are two tests that give opposite answers for the same molecule — which is precisely why they were drafted as separate subclauses rather than one. A compound the plant can make, produced by conversion, is outside (I) and inside (II).

And then there is the third thing the word does in ordinary coverage, which is to sort by what the product is called. "Synthetic cannabinoids" as a shelf category. That is not a test at all. It is a label, and labels do not survive contact with either of the two tests above.

Three uses. One adjective. If you are sorting stock by the word, you are using a word that means something different in each of the three places it appears.

What we are not going to tell you

We are not going to tell you which side of any of these tests your product falls on.

Whether the Texas scheduling action reaches a given delta-8 product depends on how that product was made, and our record on it holds that scope question open in both directions. It has held it open since August and it is with counsel.

Whether a given cannabinoid is "capable of being naturally produced by a Cannabis sativa L. plant" is a chemical and legal question, it is genuinely contested, and it is not one a compliance vendor settles in a blog post. Our federal records are marked scheduled and pending, and we do not state which ingredients they capture.

We are also not telling you the newsroom got it wrong, or that the peer site did. Two sovereigns wrote two instruments and each is being read against its own. The claim here is narrower and, we think, more useful: the word cannot carry the sorting, and a guide that sorts by it has quietly substituted a category for a test.

One more thing we are not doing. We have written before about [which clause an extension actually names](https://gethempdash.com/learn/which-clause-did-it-name) — we read that text in full and it names subclause (I) and not subclause (II). That an act of Congress omits a subclause is not evidence that the subclause stops operating. We refuse that inference for the same reason we refuse the confident version: neither one is a citation.

What Texas asks instead, and why it is stable

Here is the part that makes this more than a piece about confusion.

Go and read the requirements Texas actually put in force on March 31, 2026 under TRD-202601066, and count how many of them ask where a cannabinoid came from.

None of them do.

25 TAC §300.301(d) names twelve fields on a certificate of analysis as a condition of sale. Laboratory identification and contact information. Sample and lot identification. Analytical methods and instrumentation, with limits of detection and quantitation. An expiration date. Measurement-of-uncertainty analysis parameters. A QR code verifying the authenticity of testing at an accredited laboratory. And at paragraph (12), results reporting delta-9 THC, total delta-9 THC, and total THC per container — all three. A certificate missing any one of them is not a passing certificate.

Not one of those twelve is an origin question. They are identification and measurement.

The threshold is arithmetic. It binds total delta-9 THC — not delta-9 alone, not total THC — and total delta-9 THC is delta-9 plus 0.877 times THCA, at 0.3% or less on a dry weight basis, inclusive, with an allowance where the accredited laboratory's measurement of uncertainty produces a range including 0.3% or less. That is a calculation on numbers a laboratory reported. Its enforcement posture is under review and we assert nothing about it in either direction — the distinction we set out in [effective is not enforced](https://gethempdash.com/learn/evidence-gap-effective-is-not-enforced).

25 TAC §300.402(a) requires batch number, batch date, product name, the manufacturer or processor's name with telephone number and email address, recommended serving size in milligrams and servings per container, and a URL providing or linking to a certificate of analysis — conspicuously marked, reaching the certificate in three or fewer steps. The URL is mandatory and a QR code does not substitute for it; DSHS declined a comment proposing that swap, citing HB 1325. Five warnings under §300.402(b), on the outer packaging under §300.402(c).

25 TAC §300.502(g) sets $5,000 per retail location for registration and renewal; §300.202(c) sets $10,000 per facility for a consumable hemp product licence. Fees are not prorated.

That is the whole shape of it. The document requirements sort by measurement. The category fight sorts by adjective. A number a laboratory reported does not change when somebody renames the molecule, and that is not a small property — it is the reason these are the requirements you can act on this afternoon while the other question sits with counsel.

Worth noticing where the dates live

The DSHS Consumable Hemp Program page, read this morning, carries dated items going back to 2022. Every one of them is a Texas date: the rules effective March 31, 2026, the scheduling action effective July 31, 2026, the emergency age rules of October 2025, the hearing in January.

There is no federal date on it. Not November, not December, no public law, no section number.

That is a statement about what is in the record and nothing more. It is not a claim about what the agency thinks, intends, or is about to do, and turning it into one would be the same move this piece has spent six sections objecting to. The page also still carries no last-updated stamp, which is worth knowing when you are reading it.

What is worth doing this morning

If someone has handed you a two-pile sorting instruction, ask which of the three tests they are using. If the answer is a list of product names, you have a category, not a test, and the two come apart exactly where your inventory is.

Then go and do the part that does not move. Pull a certificate and check it against §300.301(d), field by field, including all three THC values at paragraph (12). Check a label against §300.402(a) and count the steps from the URL to the certificate.

Those requirements are numbered, verified, and indifferent to what anybody calls the molecule.

Source: Federal provision: P.L. 119-37 §781, signed November 12, 2025, scheduled effective November 12, 2026 — a future requirement, not a rule in force today; subclause language quoted from the statutory text read in full on September 6, 2026. A subsequent act of Congress narrows the application of §781 for a defined window; our review records the text read and quoted and is expressly not a settled interpretation, and the three federal records in our snapshot are unchanged and remain pending. Which ingredients the synthesized-cannabinoid exclusion captures is not stated. Texas scheduling of manufactured and synthesized cannabinoids: 51 TexReg 4597, published July 10, 2026, effective July 31, 2026; scope as applied to particular delta-8 products is pending counsel verification and asserted in neither direction. Texas newsroom classification quoted from a report of July 30, 2026. Texas provisions, all TRD-202601066 and effective March 31, 2026: COA field set 25 TAC §300.301(d) and (d)(12), verified August 16–17, 2026, dual-source; total delta-9 THC threshold 25 TAC §300.301(a)(4), (b)(4) and §300.302(b)(1), (c)(3), verified August 16, 2026 against the DSHS adopted-rule text and the Texas Register adoption notice, computation method resolved August 17, 2026; label content 25 TAC §300.402(a), (a)(6)(A)–(B), (b), (c), verified September 8, 2026, dual-source across authorities; fees 25 TAC §300.502(g) and §300.202(c), verified August 16, 2026, triple-source. DSHS Consumable Hemp Program page read September 13, 2026. Enforcement posture: under review, asserted in neither direction, on every record named here.

The full record set is at [the Texas Rule Center](https://gethempdash.com/texas-rules).