Markush Claims: 6 Smart Rules for Drafting Alternatives

Sometimes an invention is not a single compound but a family of them, any one of which works. Writing a separate claim for each would be wasteful. Markush claims solve that: a single claim recites a closed list of alternatives — “selected from the group consisting of” A, B, and C — so one limitation covers every member at once. The device has been part of U.S. practice since Ex parte Markush in 1925, and it remains indispensable in chemistry, pharmaceuticals, and materials science. But the format carries traps that can leave a claim indefinite. This guide explains how the grouping works, the improper-Markush rejection, election of species, and six rules for drafting alternatives that survive examination.

What Markush Claims Are

Markush claims recite a closed list of named alternatives from which a selection is made. The technique is named after the applicant in Ex parte Markush, 1925 Dec. Comm’r Pat. 126, whose claim the Patent Office allowed to define a component by a group of specific alternatives rather than a single species. The listing itself — for example, “an alcohol selected from the group consisting of methanol, ethanol, and isopropanol” — is called a Markush group.

The point is efficiency. Instead of filing dozens of parallel claims, one for each workable variant, you capture the entire set in a single limitation. That keeps the claim set compact, keeps prosecution costs down, and signals to a competitor that swapping one listed member for another will not design around your patent. The USPTO explains the practice in MPEP § 2117 and the definiteness rules in § 2173.05(h).

Although the format grew out of chemical practice, it is not limited to chemistry. Any invention with interchangeable alternatives — a fastener that may be a rivet, a screw, or a bolt; a signal that may be optical or electrical — can use the same construction, as long as the members truly belong together.

When to Use a Markush Group

Reach for a Markush group when your data support several alternatives that each achieve the same result, and you cannot fairly claim them with a single generic term. Chemistry is the classic case: a genus like “a halogen” may be available, but often no clean generic word exists for the exact set you tested, so you list the members instead.

The construction earns its keep in a few recurring situations:

• No adequate genus. The alternatives share a function but no accepted class name covers them precisely, so an enumerated list is the clearest option.
• Selective coverage. A broad genus would sweep in inoperative or prior-art members; naming only the ones that work keeps the claim both valid and defensible.
• Efficient prosecution. One claim replaces many, trimming excess-claim fees and giving the examiner a single focused set to search.

Breadth by itself is not a problem. As the court put it in In re Gardner, 427 F.2d 786 (CCPA 1970), “breadth is not indefiniteness.” A long list is fine so long as a skilled reader can tell exactly which alternatives are covered. Pair the grouping with a strong specification — see our guide to patent specification writing — so every listed member has real support.

The “Selected From the Group Consisting Of” Format

The sanctioned phrasing is “selected from the group consisting of,” followed by the members. The words “consisting of” make the group closed — the claim covers those members and nothing else. That closure is the whole point, and it is why you must not blend in open transitional language.

Using “comprising” inside the group breaks it. In In re Kiely, 2022 USPQ2d 532, the Board found a Markush-style limitation indefinite because “comprising” left it unclear what other alternatives were intended. The Office also accepts equivalent phrasings — “wherein the material is A, B, or C” is treated the same as the formal “group consisting of” wording, because treatment turns on substance, not the exact template.

Two refinements are worth knowing. First, “optionally” can extend a list — “containing A, B, and optionally C” was upheld in Ex parte Cordova when the alternatives stayed unambiguous. Second, some overlap between a broad member and a narrower one (listing both “halogen” and “chloro”) is tolerable when scope remains clear, as the Federal Circuit held in Eli Lilly & Co. v. Teva Parenteral Medicines, 845 F.3d 1357 (Fed. Cir. 2017).

The Improper Markush Grouping Trap

Since 2018 the MPEP has authorized a distinct rejection: the improper Markush grouping. It targets lists whose members do not really belong together. To be proper, the alternatives must share two things — and missing either one invites the rejection:

1. A single structural similarity, meaning the members belong to the same recognized physical or chemical class, or an art-recognized class; and
2. A common use or property that flows from that shared structure and is mainly responsible for the members’ function in the claimed invention.

The roots run back to In re Harnisch, 631 F.2d 716 (CCPA 1980), which allowed a broad Markush group of dyes because the members shared both a structural core and a common use. Turn that around — a list of chemically unrelated compounds thrown together only because they each happen to work — and an examiner can reject it as an improper grouping, forcing you to split the claim.

The practical defense is disclosure. Explain in the specification why the members form a class and what property they share, so the record answers the two-part test before the examiner asks. A grouping built on a genuine common structure and function is far harder to attack, in prosecution and later in an enablement or validity dispute.

Election of Species and Restriction

Even a proper Markush group can draw a restriction requirement. Under MPEP § 803.02, if the listed members are patentably distinct and searching all of them would burden the examiner, the Office may require you to elect a single species — one member — for initial examination.

The mechanics are predictable. You make a provisional election of one species, the examiner searches and examines the elected member, and prosecution of the full group waits. If the elected species is found allowable and the prior art does not otherwise divide the group, the examiner rejoins and examines the remaining members. If the elected species is rejected over prior art, that rejection usually reaches the related members too.

So a Markush claim can quietly narrow to whichever species you elect. Choose the elected member strategically — typically your lead compound or the one with the cleanest prior-art picture — and understand how the process compares with an ordinary restriction requirement. The election shapes what actually gets examined, and therefore what you can enforce first.

6 Smart Rules for Drafting Markush Claims

Pull the threads together and a short checklist emerges for drafting alternatives that survive examination and hold up later:

1. Keep the group closed. Use “selected from the group consisting of” and never let “comprising” creep inside the list.
2. Prove the class. In the specification, state the structural similarity and common property that tie the members together, pre-answering the improper-grouping test.
3. List only operative members. Drop alternatives that do not work or that read on prior art; every member should be supported by your data.
4. Mind new matter. You cannot add members later that the original disclosure never supported, so capture the full set at filing.
5. Plan your election. Decide in advance which species you would elect if a restriction issues, and make sure it has the strongest support.
6. Back the group with dependent claims. Claim key individual members separately so a single failed alternative cannot sink the whole group.

Handled with care, a Markush claim is one of the most powerful tools in a drafter’s kit — broad, efficient, and hard to design around. Handled carelessly, it collapses into an indefiniteness or improper-grouping rejection. The difference is disclosure discipline and a clear grasp of the format, the same instincts that drive sound patent claims drafting generally.

How PerspireIP Can Help

Alternative-based claims reward precision and punish guesswork. PerspireIP helps applicants structure Markush groups that pass the improper-grouping test, search the prior art behind each listed member, and prepare specifications that support the full set of alternatives. Contact our team to strengthen the claims that protect your invention.

Frequently Asked Questions