In Kellogg Co. v. National Biscuit Co., 305 U.S. 111 (1938), the Supreme Court ruled unanimously that National Biscuit could not stop Kellogg from selling a pillow-shaped biscuit under the name “shredded wheat.” Justice Louis Brandeis, writing for the Court, held that the name was generic and the shape was functional, so trademark law could not be used to extend a monopoly that had ended when the underlying patents expired. The decision is still the touchstone for how courts handle generic product names, functional product features, and competition after a patent runs out.
How the Case Reached the Supreme Court
Henry D. Perky introduced shredded wheat in 1893 and secured patents on the machinery that made it, including U.S. Patent 548,086, granted in 1895. Under the patent law of that era, those patents gave him a 17-year monopoly on the manufacturing process. The earliest patents expired around 1912, putting the machines and the biscuit itself into the public domain.
Perky’s business went through Natural Food Company and then The Shredded Wheat Company before the National Biscuit Company acquired it in 1930. When Kellogg started producing a competing pillow-shaped biscuit and calling it shredded wheat, National Biscuit sued for unfair competition.
National Biscuit conceded it had no exclusive right to manufacture the cereal. Its argument was narrower: it claimed a trademark in the name “Shredded Wheat” and in the pillow shape. If that theory worked, no competitor could sell the cereal under the name customers actually asked for, or in the form customers recognized. The patent monopoly would effectively continue forever.
Why “Shredded Wheat” Was Generic
The Court held that “shredded wheat” was a generic term for the product itself, not a brand name identifying the producer. Throughout the patent period, the cereal had been known only by that name, and when the patents expired, the right to use the name passed to the public along with the right to use the process.
National Biscuit invoked the doctrine of secondary meaning, arguing that decades as the sole manufacturer had trained consumers to associate “shredded wheat” with its company. Justice Brandeis rejected the argument. Some consumers did associate the cereal with the Niagara Falls factory, but only because National Biscuit had been the only maker for a long time. To claim trademark rights in a generic term, a company must show that “the primary significance of the term in the minds of the consuming public is not the product but the producer.” National Biscuit could not clear that bar.
The practical stakes were obvious. If the only common name for a product could be trademarked, consumers would have no way to ask for the product without infringing. Trademark law does not permit that kind of linguistic monopoly.
Why the Pillow Shape Was Functional
National Biscuit also claimed the pillow-shaped biscuit as its trade dress, a visual identifier of its brand. The Court rejected this too. Evidence showed that making the biscuit in any other form would raise production costs and reduce quality, so the shape was functional.
That reasoning became the foundation of what trademark lawyers now call the functionality doctrine. If a product feature is essential to how the product works, or affects its cost or quality, the feature cannot be a trademark. The rule prevents an end-run around patent law. Patents are deliberately temporary; converting a patented shape into a permanent trademark the day the patent expires would collapse the bargain at the heart of the patent system.
The Supreme Court reinforced the same principle in TrafFix Devices, Inc. v. Marketing Displays, Inc. (2001), holding that when an expired utility patent claimed particular features, a party seeking trade dress protection over those features carries “the heavy burden of showing that the feature is not functional.” As the Court put it in TrafFix, “The Lanham Act does not exist to reward manufacturers for their innovation in creating a particular device; that is the purpose of the patent law and its period of exclusivity.”
The Duty to Prevent Confusion
Kellogg did not walk away with unlimited freedom. The Court held that competitors are entitled to share in the goodwill of a product no longer protected by patent or trademark, and that sharing, in Brandeis’s words, “is the exercise of a right possessed by all, and in the free exercise of which the consuming public is deeply interested.” But the same opinion required Kellogg to use “every reasonable means to prevent confusion,” meaning its packaging and labeling had to make clear the product was not National Biscuit’s.
The obligation is to take reasonable steps to distinguish, not to guarantee that no shopper anywhere is ever mistaken. That balance is what allows generic competition to exist without becoming a license to deceive.
What the Ruling Means for Trademark Law Today
The framework Brandeis laid out still controls modern disputes. Descriptive marks can be protected only if they acquire secondary meaning; an applicant typically needs to show substantially exclusive and continuous use of the mark in commerce for at least five years, or offer other persuasive proof that consumers connect the term with the applicant rather than the product. Generic terms cannot be registered at all. Federal law allows cancellation of a registration at any time if the mark “becomes the generic name for the goods or services” it covers, with the primary significance of the mark to the relevant public as the test.
That test has ended the life of trademarks that once seemed untouchable. In Bayer Co. v. United Drug Co. (1921), Judge Learned Hand asked what buyers understood the word “aspirin” to mean; when the answer was the drug rather than Bayer, the trademark was gone. “Thermos” fell the same way after a court found consumers used the word for any vacuum-insulated bottle. “Escalator” was cancelled after fifty years of registration once the Trademark Trial and Appeal Board concluded the public read it as the name of a moving stairway.
The risk explains why companies spend heavily reminding consumers to say “adhesive bandage” rather than “Band-Aid” or “facial tissue” rather than “Kleenex.” A brand name that becomes the common word for the product loses its protection permanently, and no amount of advertising can win it back.
The functionality half of Kellogg does the same work on the product-shape side. A company that held a utility patent on a design cannot, after expiration, use trade dress law to fence off the same features. Anyone claiming protection for a feature that appeared in an expired patent has to prove the feature is not functional, and that is a hard showing to make.
Read together, the two holdings mark the outer limits of trademark protection. A company can own a name that identifies its goods, but not a name that is the goods. It can own a look that signals its brand, but not a shape that the product needs to work. And when a patent ends, the technology, the standard name, and the standard form all pass to the public, subject only to the competitor’s obligation to label honestly. Those lines, drawn in 1938 over a pillow-shaped biscuit, still decide how far a trademark can reach.