When it comes to books that present the maps of a single library or museum—take, for example, Debbie Hall’s Treasures from the Map Room, about the Bodleian, or Tom Harper’s Atlas: A World of Maps from the British Library—there seems to be a standard, curatorial template, one that focuses on full-colour reproductions of the maps, each of which is accompanied by a short explanatory text. The maps, as objects, are the point.

The Library of Lost Maps (Bloomsbury, 2025) is about the maps held in the Map Room of University of College London. But author James Cheshire is doing something quite different here. Partly this is because UCL is neither the Bodleian nor the British Library. Their holdings are not remotely similar. There’s no equivalent of the Gough Map or the Selden Map here; the maps are more modern (19th and 20th century) and less rare and singular. UCL’s Map Library was a working map library, used by its staff for teaching and research, rather than something more curatorial.

But what the maps at UCL do have is stories attached—about how they were made, and about how they came to be in UCL’s hands. The Library of Lost Maps is simultaneously a story of the early days of UCL and its role in broadening education in Britain, its role as a repository for so many maps being produced during the twentieth century’s bloodiest conflicts, and its uncertain future as that role of map repository is increasingly seen as obsolete.

You may remember James Cheshire as the co-author, with Oliver Uberti, of books like Atlas of the Invisible or Where the Animals Go. He’s also Professor of Geographic Information and Cartography at the University College London Department of Geography and the first director of the UCL Social Data Institute. He works with digital data, which doesn’t much call for old maps, so it took him a full decade to find himself in the map library, looking for maps he could use as teaching materials. That led him down the rabbit hole that led to this book.

UCL is by British standards a fairly recent institution. Its origins date to the 19th century, when it was established as the first secular (i.e., non-Anglican) university in England as part of a movement to improve public education. The maps in this collection are contemporary to its existence from that beginning to the present day, and in many cases UCL’s history intersects with the maps being produced. In the 19th century UCL was in the same mix as the Society for the Diffusion of Useful Knowledge, whose maps are the focus are the focus as chapter four; and toward the end of the book, when Cheshire discusses digital mapping, we learn that Roger Tomlinson and Steve Coast were UCL students.

But other maps don’t have such a direct link, though they were very much of the era in which UCL was operating: geological maps, ethnological maps, propaganda maps, wartime maps, German maps seized at the end of the Second World War, even Marie Tharp’s seafloor charts. The kind of maps a working university map library would be expected to have on its shelves over the course of the 20th century. The library also benefited from being in receipt of reams of slightly out-of-date maps being cast off by the Ministry of Defence. The end result is a vault of prints rather than originals, not produced for durability or posterity. The kind of maps that run the risk of being lost or forgotten precisely because at one time they were common.

What Cheshire is doing with The Library of Lost Maps is essentially to take a set of these maps off the shelf, blow off the dust, and say, now what do we have here? It turns out that there’s a lot of story behind the maps: we know about their provenance and their production. And it turns out that there’s a lot of story behind the mapmakers, too: They include characters like the “obstinate” geologist George Bellas Greenlough; Karl Haushofer, the professor who tutored Hitler; and Sándor Rádo, the Hungarian cartographer turned Soviet spy.

The Library of Lost Maps is in one sense an act of remembrance—a rage against the dying of the light, as map libraries are shuttered and their holdings discarded or deaccessioned. It’s certainly a compelling argument against their closure. It also happens to be an absolute delight, one of the best general-interest map books I’ve read in a long while. Anyone with an interest in 19th- and 20th-century mapping will enjoy this book.

(One odd thing about The Library of Lost Maps is the decision to provide the endnotes in a separate PDF file downloadable from the accompanying website. I can see the logic behind it: hardly anyone reads the endnotes, a good chunk of the references are to websites, and the publishers could reduce the size of the physical book by 10 percent. But I can’t help think that a century from now, copies of the physical book will exist in some forgotten corner of a map library, long after its website has gone dark. For a book that focuses on the persistence of working maps, maps made for a purpose rather than for posterity, I can’t help but noice the irony.)

I received electronic and physical review copies from the publisher.

The Library of Lost Maps: An Archive of a World In Progress
by James Cheshire
Bloomsbury, 9 Oct 2025 (UK), 4 Nov 2025 (US), £30/$40
ISBN 978-1-5266-7661-0 (UK), 978-1-63973-428-3 (US)
Amazon (Canada, UK) | Bookshop