What is a means-plus-function claim limitation?

A means-plus-function claim limitation is a claim element that recites a function to be performed without expressly naming the structure, material, or acts that perform it. Congress authorized this drafting technique in 35 U.S.C. § 112(f), which provides that such an element "shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof."

This is not a broadening device. The scope of a means-plus-function limitation is strictly limited to the corresponding structure disclosed in the specification and its equivalents, a result that is often narrower in practice than the functional language alone might suggest.

What language triggers this format

Using the word "means"

Using the word "means" in a claim element, coupled with functional language, creates a rebuttable presumption that § 112(f) applies. The presumption is overcome only when the claim language also recites sufficient structure, material, or acts to perform the function.

Example that triggers 112(f):

"...a means for filtering the output signal..."

Example that overcomes the presumption:

"...a low-pass RC filter for filtering the output signal..."

The second example names a specific structure (a low-pass RC filter) and therefore falls outside § 112(f), even though it describes what the structure does.

Nonce words without "means"

A claim element that omits the words "means" and "step" carries a rebuttable presumption that § 112(f) does NOT apply. That presumption falls, however, when the element instead uses a generic placeholder term such as "mechanism for," "module for," "device for," "unit for," or "element for," coupled with functional language, if the term fails to convey a sufficiently definite meaning as the name for structure to a person of ordinary skill in the art.

The controlling standard, drawn from the Federal Circuit's en banc decision in Williamson v. Citrix Online, is: "whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure."

Terms like "circuit," "filter," and "brakes" carry structural meaning and typically avoid § 112(f) treatment even in a functional context. Bare "module," "component," or "mechanism" used without a structural modifier typically do not convey structural meaning to a skilled artisan and invoke the statute. The decisive question is always whether the skilled artisan reads the term as a structural descriptor or as a functional stand-in.

How scope is determined

The two-step construction

Once a limitation is treated under § 112(f), courts and examiners construe it in two steps: first defining the function as stated in the claim language; then identifying in the specification the structure corresponding to that function, which requires the specification or prosecution history to clearly link or associate the disclosed structure to the claimed function.

The scope covers only what the specification discloses as performing the function, plus equivalents. A "means for sorting records" does not cover every conceivable sorting approach; it covers only the structures the specification ties to that sorting function, and approaches that are equivalent to those specific structures.

The corresponding-structure requirement

A structure is "corresponding" only if the specification or prosecution history clearly links or associates it to the claimed function.

Computer-implemented claims

Software patent claims present a recurring difficulty. For a computer-implemented means-plus-function limitation, the specification must disclose an algorithm, expressible as a flowchart, pseudocode, mathematical formula, or prose description, that shows how a general-purpose processor is transformed into a special-purpose device performing the claimed function; a bare recitation of a processor or computer alone is insufficient.

An exception applies for basic operations: receiving or storing data are examples where a general-purpose processor may serve as sufficient corresponding structure without a separate algorithm disclosure.

The indefiniteness trap

If the specification discloses no corresponding structure at all for a means-plus-function limitation, the claim is indefinite under 35 U.S.C. § 112(b). This is one of the sharpest hazards in functional claiming. A broad "means for" element can look attractive in the claims but is fatally deficient if the written description provides no structural anchor. The practical implication: using means-plus-function format demands careful investment in specification quality, not less.

Prior art during prosecution

When an examiner applies prior art to a means-plus-function claim, the analysis extends beyond whether the reference performs the claimed function. An examiner who identifies a prior art reference performing the claimed function must also demonstrate that the prior art element is the same as, or equivalent to, the corresponding structure disclosed in the specification. The examiner must provide an explanation and rationale for that equivalence; bare assertion is insufficient.

Once the examiner establishes a prima facie case of equivalence, the burden shifts to the applicant to rebut it with substantive reasoning grounded in specification language, prior art disclosures, or affidavit evidence.

Even after a successful rebuttal showing nonequivalence, the examiner may still pursue a § 103 obviousness rejection. Defeating equivalence forecloses anticipation but not obviousness.

Step-plus-function: the method-claim analog

Method claims can contain a step-plus-function limitation using the format "step for performing [function]," governed by the same § 112(f) scope-limiting rule. The analogy is direct: the "corresponding acts" in the specification play the same role as "corresponding structure" in apparatus claims.

Trigger language at a glance

Practical notes for drafters

• Anchor every means element to named structure in the specification. For each means-plus-function element you intend to use, confirm the specification names the performing structure and explicitly ties it to the recited function. Vague or missing disclosures are not curable by prosecution argument after filing.
• Disclose multiple embodiments to widen the equivalents range. The scope of equivalents is anchored to what you disclose. Multiple disclosed embodiments performing the same function provide a broader baseline for equivalents analysis and harder-to-design-around coverage.
• Audit nonce words before filing. After Williamson, terms like "module," "unit," "component," and "mechanism" paired with functional language carry real risk of unintended § 112(f) treatment, which narrows the claim scope and imposes algorithm-disclosure obligations for software claims. Add structural modifiers where you want to avoid it.
• Software claims need algorithm disclosure. A "sorting module for sorting records" will almost certainly trigger § 112(f) and will require an algorithm in the specification. Verify that any term you intend as structural actually carries structural meaning to a skilled artisan in that technology area.
• Account for the litigation scope gap. The § 112(f) equivalents test is narrower than the doctrine of equivalents. A competitor who builds a different structure performing the same function may escape literal infringement entirely. Draft apparatus claims with that scope gap in mind.