One supported claim does not carry the whole disclosure back

How to read the ledger below. Each numbered point carries the kind of evidence behind it. ESTABLISHED is stated in the authority's own text, with a locator. IMPLEMENTED is the rule as the authority actually applied it. OBSERVED is carried by a source outside this one. ASSUMED is the drafter's inference, flagged as such and not the authority speaking. Quotations are checked against a cached copy of the primary source by script.

In one paragraph

The Federal Circuit held in In re Riggs on 24 March 2025 that a prior-art published application does not receive its provisional filing date for everything it teaches merely because one claim of the reference is supported by that provisional, and that the specific portion of the reference relied on in the rejection must be supported as well, so the pre-AIA test has 2 requirements rather than 1. The court affirmed the finding on the first requirement and vacated because the Board never reviewed the second.

Reader summary

Dynamic Drinkware is usually remembered as a one-step gate: before a reference publication gets its provisional's date as a prior-art date, someone must show the provisional supports at least one claim of the reference. The Board here read that as sufficient, treating a single supported claim as carrying the provisional's date across the reference's whole disclosure.

The Federal Circuit disagreed, and was explicit that this scenario had not been decided before. Showing support for one claim is necessary but not enough. The provisional must also support the specific portion of the reference's specification the examiner actually cited. The court's reasoning is blunt: a claim usually does not recite everything in a written description, so a supported claim says nothing about whether some other paragraph was disclosed in the provisional at all.

The court nonetheless affirmed that substantial evidence supported the Board's finding on the claim-support half. What it vacated was the second half, because the Board never reviewed whether the passage relied on in the rejection had provisional support.

The contrast with penumbra-v-rapidpulse is the reason both records exist. Under the AIA the Board dropped the claim-support step entirely and asks only whether the subject matter relied on was described. Under pre-AIA law, after this case, there are two requirements. The pre-AIA route to an early prior-art date is therefore the harder one, which is the opposite of what a practitioner extrapolating from a single half-remembered rule would guess.

Why this matters when you are drafting

Two checks, not one, against a pre-AIA reference. Claim 2 is the operative rule. When a reference publication is cited against a continuation and its provisional date is what makes it prior art, the question is not only whether some claim of it is supported. It is whether the paragraph being quoted at you was in the provisional.

This is a live argument, not a theoretical one. Claim 5 describes the Board doing exactly what the court rejected, and claim 6 shows the consequence: vacatur for failing to reach the second requirement. The same gap is available wherever an examiner has dated a reference to a provisional without addressing the cited passage.

Know which statute governs before you pick a test. The pair with penumbra-v-rapidpulse is the point of this tier. Pre-AIA gives you two requirements to attack; the AIA gives you one, and it is the easier one for whoever is asserting the reference. Getting the regime wrong means arguing the wrong test, and nothing in either decision flags the error for you.

The asymmetry cuts against continuations specifically. A continuation's value often depends on an early priority chain, and the references most likely to be cited against it are publications in the same technology with their own provisional chains. This case gives a concrete way to push a pre-AIA reference's date forward.

What the authority establishes

[ESTABLISHED, the court says this is new ground] On whether a supported claim suffices to carry the rest of the disclosure back, "we have not addressed the scenario at issue here". (P. 12.) The court then answers it: "We conclude it is not." (P. 12.)

[ESTABLISHED, the holding] "Even if one demonstrates that a provisional application provides written description support for one claim of the non-provisional application or patent, the provisional application must also provide written description support for the specific portions of the patent specification identified and relied on in the prior art rejection". (P. 12.)

[ESTABLISHED, the same rule stated operationally] "to claim priority to the provisional filing date, the portion of the application relied on by the examiner as prior art must be supported by the provisional application". (P. 12.)

[ESTABLISHED, the reasoning in one sentence] "It makes no sense to suggest that if a single claim is supported by the provisional application, then everything in the later filed application gets the benefit of the provisional date whether supported or not". (P. 12.)

[ESTABLISHED, what the Board got wrong] The Board treated a single supported claim as applying the provisional's date across the reference. "We disagree with this approach." (P. 14.) The court explains that "A reference constitutes prior art for all that it teaches as of its earliest accorded filing date", and that this does not follow from one claim being supported. (P. 14.)

[ESTABLISHED, what was affirmed] The claim-support half survived: the court agreed the Board found support for each limitation of the reference's claim 1, and "we conclude that substantial evidence supports these findings". (P. 14.) The defect was that "the Board did not review the Examiner" on the second requirement. (P. 14.)

[ESTABLISHED, the default treatment of a published provisional] "it is best to treat a published provisional application" like a published non-provisional or patent, as prior art for all that it teaches as of its filing date. (P. 13.) The two-requirement rule is about what a LATER publication may claim from the provisional, not about the provisional's own effect.

[IMPLEMENTED, the rule is a conjunction and the burden sits with the challenger] The court leaves Dynamic Drinkware and Amgen intact as the first requirement and adds the second; it does not replace one with the other. The party asserting the earlier date must make both showings, which is how the Office's own MPEP 2136.03 already framed the first.

[ASSUMED, drafter's inference, flagged as such] Read against penumbra-v-rapidpulse, the pre-AIA and AIA regimes now diverge in a way that favours the patentee under pre-AIA law and the challenger under the AIA. Neither decision says this; it follows from putting the two tests side by side.

Quotations, verified against the source

All from the Federal Circuit's opinion. Page locators computed from the docket stamp.

"we have not addressed the scenario at issue here" (P. 12.)
"Even if one demonstrates that a provisional application provides written description support for one claim of the non-provisional application or patent, the provisional application must also provide written description support for the specific portions of the patent specification identified and relied on in the prior art rejection" (P. 12.)
"to claim priority to the provisional filing date, the portion of the application relied on by the examiner as prior art must be supported by the provisional application" (P. 12.)
"It makes no sense to suggest that if a single claim is supported by the provisional application, then everything in the later filed application gets the benefit of the provisional date whether supported or not" (P. 12.)
"it is best to treat a published provisional application" (P. 13.)
"We disagree with this approach" (P. 14.)
"A reference constitutes prior art for all that it teaches as of its earliest accorded filing date" (P. 14.)
"we conclude that substantial evidence supports these findings" (P. 14.)
"the Board did not review the Examiner" (P. 14.)

Tensions and open questions

How much of the cited passage must be supported is not calibrated. Claim 2 requires support for "the specific portions" relied on. The opinion does not say whether that means verbatim support, support for each limitation the passage teaches, or something looser.

Claim 7 sits slightly awkwardly with the holding. The court says a published provisional is prior art for all it teaches as of its own filing date, while a later publication gets the provisional date only for supported portions. Both are coherent, but a reader skimming will conflate them, and the practical difference is whether the provisional itself is available as a reference.

The AIA boundary is stated by the other side of the pair, not by this case. This opinion is about pre-AIA 102(e) and does not say what happens under the AIA. That comes from penumbra-v-rapidpulse, which is Board precedent rather than court precedent, so the two sides of the pair do not carry equal authority.

Unresolved by this record: the outcome on remand. The court vacated for the Board to address the second requirement. Nothing here says how it came out.

What this record does and does not cover

The priority and prior-art-date analysis was read end to end, including both footnotes that bear on it, because that analysis is the whole reason this record exists.

What was NOT read closely, and is therefore outside what this record can support: the anticipation and obviousness holdings that depend on the date question, and the technical detail of the reference. This record says nothing about whether the claims were ultimately patentable. Dynamic Drinkware, Amgen and MPEP 2136.03 are known here only as this opinion describes them, and the Board decision and joint appendix only through the court's characterization.

Currency

This record was verified against its primary source on 2026-08-15. That is a fidelity check: the quotations, locators and attribution are accurate to that source. Whether the authority has since been narrowed, distinguished, abrogated or reversed has not been checked here. Confirm it still stands before relying on it.