In 2013, the Trademark Manual of Examining Procedure (TMEP) was amended to specify that a mark including the hash symbol (#) is registrable “only if it functions as an identifier of the source of the applicant’s goods or services”; the inclusion of a hashtag at the front of a mark, just like that of a top-level domain at the end, does not make the mark more distinctive than it would be without. But the TMEP does not go far enough — not only does the addition of a hash symbol fail to transform a descriptive term into a protectable mark, it can render an otherwise registrable mark unregistrable. The symbol can transform a phrase from a viable trademark into a mere hashtag: consumers perceive it as a branding effort that directs them to social media to discuss, follow, or learn more about a particular campaign, or invites them to participate in an online conversation. Since the advent of the hashtag on Twitter, approximately six hundred hashtag marks (“hashmarks”) have been registered successfully on the Principal Register; approximately two thousand more use- or intent-to-use-based applications are currently pending, under examination, or have been abandoned or refused. The time is ripe for the Trademark Office to adopt a uniform approach in reviewing these applications and treat them consistently to ensure putative marks actually function as marks, which includes requiring distinctiveness, nonfunctionality, and use as a mark in interstate commerce in connection with the identified goods and services. Courts will inevitably need to adjudicate claims of infringement or dilution from hashmark owners as an issue of first impression. This article sets forth a taxonomy of hashmarks, explores the problems in granting them protection under the Lanham Act, and proposes a framework for ensuring their treatment serves the goals of trademark law, including protecting consumers, maximizing economic efficiency, and providing incentive and reward for producer creation of goodwill.