Continuation Practice: Why Rich Disclosures Compound Across Applications

A long lever, anchored to the original disclosure

Continuation practice under 35 U.S.C. § 120 is one of the longest levers in U.S. patent strategy. A continuation application can claim the priority date of an earlier-filed application as long as the two share a common inventor, the continuation is filed before the parent issues or is abandoned, and the continuation’s specification adds no new matter.

The lever is long because the priority chain it creates can run for decades. A patent family that begins with a 2015 provisional and is continued through multiple non-provisionals, divisionals, and continuations-in-part can produce issued patents in 2030 that still claim priority back to 2015. Empirical work on continuation practice, most influentially Mark A. Lemley & Kimberly A. Moore, Ending Abuse of Patent Continuations, 84 B.U. L. Rev. 63 (2004), has documented patent families with continuation chains running well past a decade.

The lever is anchored because the priority chain depends on what the original disclosure contained. A continuation cannot claim invention beyond what the parent application already disclosed. New matter introduced in a continuation breaks the priority chain for any claim that depends on it.

The strategic implication is direct. Continuation practice does not write the disclosure. It rents the disclosure that already exists. The richer the original disclosure, the more the family can become.

What MPEP § 201.07 actually requires

The basic mechanics are stated in MPEP § 201.07 and the implementing rules.

A continuation is “an application for the invention(s) disclosed in a prior-filed copending nonprovisional application.” It must include at least one inventor named in the prior-filed application. It must be filed during the parent’s pendency, before issuance, abandonment, or termination of proceedings on the parent. It must include a specific reference to the prior-filed application under § 120 to claim priority. And it must contain no new matter relative to the parent’s disclosure.

Three categories sit under the same § 120 priority umbrella. A continuation pursues different claims to subject matter the parent already disclosed. A divisional is filed in response to a restriction requirement and pursues a separate invention disclosed in the parent. A continuation-in-part adds new matter and is the only one of the three that can introduce subject matter beyond the parent, but the priority date for any new-matter claim is the CIP’s filing date, not the parent’s.

The shared constraint across all three is the same: priority back to the parent extends only as far as the parent’s disclosure supports. The disclosure is the leash.

Why continuations matter operationally

A continuation strategy serves several distinct operational purposes.

It lets counsel pursue multiple claim sets from a single specification. The first non-provisional might issue with claims optimized for one infringement scenario. A continuation can pursue claims optimized for a different scenario from the same disclosure: a different claim scope, a different combination of elements, a different functional emphasis.

It lets counsel respond to competitive movement after the parent issues. If a competitor releases a product six months after the parent issues, a continuation pending at that time can be amended to capture the competitor’s product within its claim scope, as long as the parent’s disclosure supports the amended claims. The continuation has to have been kept alive.

It lets counsel rebuild claims that lost in IPR or litigation. A continuation pending at the time of an adverse PTAB decision can pursue narrower or differently-structured claims, drawing on the same disclosure record, without restarting the priority date.

It lets counsel preserve flexibility against doctrinal change. The disclosure record built in 2018 had to satisfy Berkheimer’s factual posture. The disclosure record built in 2020 had to anticipate American Axle & Manufacturing v. Neapco Holdings, 967 F.3d 1285 (Fed. Cir. 2020)‘s application of Section 101 to a method-of-manufacturing claim. The disclosure record built in 2023 has to support Amgen-grade enablement on broad scopes. A continuation in 2026 can pursue claims that respond to all three, as long as the original disclosure was rich enough to support them.

The constraint across all of these is what the disclosure preserved at the time it was filed. None of these moves work if the parent application contained a thin specification that does not support the claims the continuation now wants to pursue.

The prosecution-laches limit

Continuations are not unlimited. The Federal Circuit has held that prosecution laches, unreasonable and unexplained delay in prosecution, can render claims unenforceable.

The doctrine traces to Symbol Technologies, Inc. v. Lemelson Medical, Education & Research Foundation, LP, 277 F.3d 1361 (Fed. Cir. 2002), where the court confirmed prosecution laches was available against patents issued from continuation chains stretching back to the 1950s. The Federal Circuit reaffirmed and applied the doctrine in Hyatt v. Hirshfeld, 998 F.3d 1347 (Fed. Cir. 2021), and the USPTO has invoked it in subsequent prosecution.

The doctrine does not police continuation practice generally. It polices abuse: chains that stretch unreasonably without explanation, applications that sit dormant before being revived to capture intervening commercial activity. Standard continuation practice for legitimate claim refinement does not raise prosecution-laches concerns. But a strategy that depends on holding a continuation open for many years to capture future infringement should be designed with the doctrine in view.

The double-patenting cleanup

Continuation families generate obviousness-type double patenting issues that have to be cleaned up systematically.

Two patents claiming patentably indistinct inventions cannot both be enforced. The standard remedy is a terminal disclaimer, which links the later patent’s term to the earlier patent’s term and links enforceability to common ownership. The Federal Circuit’s decision in In re Cellect, LLC, 81 F.4th 1216 (Fed. Cir. 2023), clarified that patent term adjustment under 35 U.S.C. § 154(b) does not protect a patent in a family from obviousness-type double patenting. That ruling tightened the operational discipline around continuation portfolios with PTA-extended patents.

For in-house counsel managing continuation strategy, the practical implication is that terminal-disclaimer planning has to be part of the strategy from the beginning. The benefit of a long continuation chain is partially offset by the term-trimming effect of the disclaimers it requires.

What disclosure quality actually buys

The strategic case for continuation practice collapses to a single proposition: rich disclosures compound, thin disclosures do not.

A specification that captures the technical mechanism, the operating constraints, the alternatives considered, and the architectural specifics gives a continuation strategy room to operate. Future claims can pursue scopes anchored to specifics that are already in the spec. Defensive arguments can rely on examples that are already disclosed. Responses to doctrinal change can draw on disclosure that anticipated the change before it arrived.

A specification that captures only the headline invention (solution described at a high level, examples disclosed sparingly, alternatives left out) caps the family’s eventual scope. Continuations can rearrange and re-emphasize what the parent disclosed. They cannot retroactively add what the parent did not.

The empirical work on continuation practice (Lemley & Moore on chain length and litigation outcomes, and the Allison–Lemley line of papers on litigated-patent quality) has consistently shown that the patent families that produce the strongest portfolios are anchored in the deepest underlying specifications. The correlation is not surprising. The continuations are working with the disclosure they inherited.

Practical implications for in-house IP counsel

Continuation strategy starts at the original disclosure intake. Several intake practices materially change what the family can become.

• Capture more than the lead embodiment. The first product implementation is the obvious example. Variants the team considered, alternative architectures, edge cases the design accommodates: these belong in the disclosure record because they belong in the specification, and a continuation that wants to pursue claims to one of them depends on the disclosure already supporting it.
• Document the operating constraints and the technical principle. A continuation that wants broader functional claims depends on the spec articulating the general technical principle, not just the specific implementation. Amgen tightened that requirement; a strong disclosure now matters more, not less.
• Preserve the engineering record alongside the IDF. Source citations to design docs, sprint planning notes, architecture reviews, and pull requests give the eventual specification a richer evidentiary base. A continuation that wants to claim something the lead-embodiment IDF did not emphasize can rely on the engineering record the lead-embodiment IDF cited.
• Plan the family at the parent stage. What scopes might be valuable in three years, five years, ten? The answer constrains what the parent’s specification needs to disclose. Counsel cannot retroactively widen the disclosure.

None of these practices are exotic. They are the standard discipline of a strong specification. The connection to continuation practice is that the discipline pays compounding returns.

What to read next

The doctrinal anchors are stable. § 120 sets the priority-chain mechanics. MPEP § 201.07 operationalizes them. Symbol Technologies v. Lemelson and In re Cellect set the boundary conditions on prosecution laches and double patenting. Standard treatises on prosecution practice cover the routine mechanics in detail.

For the disclosure-record discipline that makes a continuation strategy viable, see our note on patent portfolio quality metrics, the metrics that reward depth at intake also reward continuation flexibility downstream.

ALID surfaces invention candidates from engineering artifacts and packages each one with the technical specifics, alternatives, and source citations that produce specifications rich enough to support a long continuation strategy. To see how the discovery and disclosure flow operates, read how ALID works, or request access to run a first discovery against your own engineering data.