Freeze-dried and locked away in liquid nitrogen–filled vaults around the world are hundreds of thousands of microbial cultures. In the U.S. these living libraries include 91,000 bacterial and fungal strains in Peoria, Ill., as well as 13,000 fungal strains in Ithaca, N.Y., and 7,500 wild yeasts at the University of California, Davis. Researchers have used strains from such publicly accessible culture collections to develop penicillin as well as the CRISPR gene-editing systems and the PCR (polymerase chain reaction) DNA-amplifying technique.

But as government priorities apparently change, and as the federal money that historically created and preserved many of these biological resource repositories in the U.S. dries up, their infrastructure is at risk of deteriorating. With less funding for daily curation—such as occasionally reviving strains to maintain viability, in addition to licensing and distributing them—the less accessible these collections become. Plant pathologist David Geiser oversees The Pennsylvania State University’s collection of 20,000 samples of species in the genus Fusarium, a ubiquitous soil fungus that can be pathogenic. When the collection’s previous curator retired, Geiser found that grant funding for a full-time replacement was practically nonexistent. “There’s a national crisis that’s going on with culture collections,” he says. “It’s been going on for a long time; it’s just gotten worse lately.”

One way to help maintain at least some microbe libraries would be raising user fees. Private U.S. collections set up to generate revenue, however, often charge researchers $250 to $350 per strain—which can deter early-stage investigations and historically underserved institutions. And privatization removes obligations to distribute strains publicly, making it easier to keep them secret. Sourcing strains overseas poses legal and logistical problems, and isolating them from nature or getting them from peers runs the risk of using nonidentical or mutated microbes—potentially compounding reproducibility problems in research.

Unlike some other countries, the U.S. has no centralized facility to store strains. And there is no microbial equivalent of the giant Arctic vault in Svalbard, Norway, where the nonprofit Global Crop Diversity Trust preserves crucial food-crop seeds in the event of natural or man-made disasters. But Kevin McCluskey, a researcher at Kansas State University who oversees the school’s Fungal Genetics Stock Center (a collection of 25,000 filamentous fungi, including mostly mutant strains), says some microbes are also important for agriculture because they fend off pathogens and harmful compounds produced by fungi. “There’s all sorts of things we’re learning how to do with microbes,” he says. “We need those publicly available—not behind a paywall.” To that end McCluskey has been organizing a National Science Foundation–funded project called the U.S. Culture Collection Network, which aims to preserve vulnerable collections, sponsor related workshops and training sessions, and implement a long-range management plan to keep collections open.

Part of the value of living microbe collections is that they can be used to map genetic changes a strain undergoes over time as a result of selection pressure from fungicide or antibiotics. As such, microbes may have applications that scientists have yet to discover or envision. McCluskey says these collections are priceless; he can make a perfect copy of a given strain—almost as if microbe “librarians” were freely lending out exact copies of the sole surviving first edition of a book—while safely keeping the original in perpetuity. This, McCluskey says, is critical for reproducibility and future scholarship. Andrew Carnegie once advocated for public libraries, and McCluskey thinks it is high time for funders—such as Jeff Bezos, the billionaire entrepreneur and founder of Amazon—to invest in an open network of public microbe collections. “There’s a lot of seeds in Svalbard,” McCluskey says. “But we want a microbe vault.”