Historians of astronomy name four great celestial atlases: Bayer's, Hevelius's, Flamsteed's, and Bode's. To this, they add one great work that could have been: Bevis's Uranographia. While the creation of this work achieved much notoriety and the publication was greatly awaited, the printer Neale's bankruptcy derailed the project. Fortunately, the plates had already been made and separately issued copies, in addition to thirty completed works, were made.
The atlas comprised 51 plates, the same number as Bayer's. Further, each plate analyzes the same celestial region. However, Bevis greatly added to the detail of Bayer's work, drawing on his own astronomical knowledge. Copies of these printings which survived were of great public and scientific interest at the time.
Careful cross-referencing of the dedications on each work allows the date for the creation of the plates to be constrained to between 1747-1749. By comparing the titles suggested in the work to Royal Society and clergy records, upper and lower bounds of the date of creation can be made.
References to a posthumous 1786 Bevis Atlas Celeste prove difficult to follow. Academic debate as to the nature of a paper residing in the British Library copy of the Uranographia advertising a 1786 publication suggests that there was an effort after Bevis's death to resell the work, without crediting him. Other "title page" editions, including one from 1818 held at Cambridge's Whipple Library, advertise a similar thing. Ashworth concludes that several later entrepreneurs tried to resell the original copy under their own name, with the 1786 copy being a prime example of a "ghost work."