As I wander around Cambridge University Botanic Garden, a tree called the Wollemi pine often catches my eye. It’s one of our rarest trees, and a distinctive looking pine, with broad needles and bark that reminds you of coco pops. First discovered in 1994 in a ravine in the Wollemi National Park in western Australia, only a few hundred survive in the wild. Although it has been on planet earth for hundreds of thousands of years, it is close to extinction. This tree species, like many others, represents a paradox: a rare and threatened species thriving in cultivation while its wild counterparts are just about hanging on to existence. As a curator of one of the world’s largest university botanic gardens, I often talk about the power of living collections. I also recognise their limits. The world’s botanic gardens hold an extraordinary diversity of plants. But, they are struggling to keep up with the accelerating biodiversity extinction crisis.Botanic gardens are often seen simply as peaceful retreats from the daily rat race or living museums where species are catalogued and displayed. But they are far more than that. Collectively, the world’s gardens form an extensive network of living plant collections, acting as refuges for biodiversity, sources of genetic material for research, and hubs for ecological restoration.Our recent study, published in the journal Nature Ecology & Evolution, analysed 50 of the world’s largest living plant collections, currently growing 41% of all species in cultivation, and 500,000 individual plants. Our research spanned a century of digitised data and the findings are striking.Programmes like the International Conifer Conservation Programme, led by the Royal Botanic Garden Edinburgh, have successfully safeguarded conifer species at risk of extinction. And Missouri Botanic Garden has changed how it manages its collection to prioritise threatened species, embedding conservation into its core activities. For example, they are increasingly only growing plant species that will naturally fit their own climate. Yet, despite these successes at specific gardens, our new research suggests that our current global system of botanic gardens is not keeping pace with the biodiversity crisis.We have hit “peak capacity” in botanic gardens – both in the number of plants grown and in the diversity of species held. This means we are growing as many individual plants as we possibly can, and our collections are as diverse as they can possibly be. While this may seem like a success, it reveals an uncomfortable truth: we are running out of space and resources to add more species.We have already passed “peak wild”. This means that we are collecting and sourcing less plants directly from the wild. Since 1992, the proportion of wild-collected plants entering botanic gardens has declined, alongside a decrease in material sourced across international political boundaries. This shift coincides with the Convention on Biological Diversity, which aims to regulate the trade of wild animals and plants and the use of genetic resources. While intended to promote fair sharing of the benefits of biodiversity, it has seems to have negatively effected the cultivation of plants outside of their native environment, even when this is being done to protect them. It has done this by limiting the movement of plant material.Collections are also becoming less globally diverse. Since the early 1990s, they have become increasingly regionalised, potentially limiting their capacity to act as global conservation networks.These trends expose a crucial challenge: if botanic gardens are to play a serious role in conservation, their curators must rethink how they collect, share and manage plant diversity.Some gardens are already adapting, exemplified by the global charity Botanic Gardens Conservation International’s (BGCI) Global Conservation Consortia, which are forming networks to safeguard specific tree genera. Central to their efforts is the concept of the “meta-collection” – a coordinated network of living collections that steward global plant diversity. Collaboration is essential, as no single institution has the capacity or expertise to conserve every threatened species alone. BGCI is leading efforts to collate data from thousands of collections worldwide. Its searchable platforms, such as PlantSearch and ThreatSearch, are leading the way in terms of the data tools institutions need to identify conservation priorities and track the status of threatened species.Smart next stepsTo save endangered plants, we need to focus on three key actions that make a real difference, much like how we protect the Wollemi pine.The rules around plant protection need to be fixed. Right now, complicated legal barriers can make it harder, not easier, to save plants. We need clear guidelines that help gardens and conservationists share and protect rare species responsibly, without getting stuck in red tape.We need to make better use of what we already have. Many botanic gardens are running out of space, so rather than collecting more and more species, we need to focus on preserving strong, genetically diverse populations of the most endangered plants – like ensuring the Wollemi pine has a secure future in multiple locations around the world.A global data system would allow scientists to see, in real time, where rare plants like the Wollemi pine are being grown, how well they’re doing, and where help is needed most. Better information means smarter conservation decisions.Botanic gardens have a long history of adaptation. They have evolved from medicinal gardens to scientific institutions, and now they must become conservation leaders on a global scale. The extinction crisis demands bold action, strategic collaboration and a willingness to rethink traditional approaches.Don’t have time to read about climate change as much as you’d like?Get a weekly roundup in your inbox instead. Every Wednesday, The Conversation’s environment editor writes Imagine, a short email that goes a little deeper into just one climate issue. Join the 40,000+ readers who’ve subscribed so far.Samuel Brockington does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.