John the Lydian was an antiquarian writer of the 6th century AD, whose career flourished under Justinian. His three works, De magistratibus Romanis, De Ostentis, and De Mensibus, all are full of information about Roman origins. John wrote in Greek but knew Latin, and sought to transmit to the future information that was already fading in his own day.
The work of John the Lydian, like most other texts from antiquity, has reached us through the medium of a single copy. This is referred to in the editions as the Codex Caseolinus, and is today in the French National library under the shelfmark Ms. Paris supplementi graeci 257. It dates around 900 AD. It has currently 100 folia, but various leaves have been lost, and the whole codex was disarranged before Hase sorted out the order of the leaves for the first time, in order to use it. It was written on thin, good quality parchment.
The manuscript has suffered considerably from damp, which has given it purplish wine-coloured stains, sometimes to the point of illegibility. The most recent editor, Anastasius Bandy, made use of infra-red light to read more of the text than his predecessors; but it seems likely that the use of multi-spectral imagining would recover more.
The “Caseolinus” name is thus not from a library, as one might suppose. Instead it refers to M. Choiseul-Gouffier, French Ambassador to Constantinople in the late 18th century, whose ancestors bore the title “Comites de Caseolo”.
When Choiseul-Gouffier was sent on his embassy in 1784, he was accompanied by Jean-Baptiste-Gaspard d’Ansse de Villoison, a noted scholar whose instructions were to seek out manuscripts. He had been sent by the French government to Venice in 1779, where he had discovered the Marcianus codex of Homer, and printed an Anecdota Graeca in Venice in 1781 based on other finds. It was Villoison who located the manuscript in the house of a Greek named Konstantinos Slouziares, who possessed the remnants of the library of Nicholas Mavrokordatos. The latter ruled Bucharest for the Sultan in 1722 and amassed a collection of codices from Greek monasteries in the East, especially Thessaly and Athos. After much negotiating, Choiseul-Gouffier was able to purchase the manuscript of John the Lydian.
The manuscript thus came into the hands of Choiseul-Gouffier in 1785, and he brought it back with him to Paris when he returned in 1791. But in 1793 he was obliged to flee from the Revolution, and took the manuscript with him to Russia. But it returned, and he still owned it in 1817, when his collection passed into the Bibliothèque Nationale, by agreement with his heirs. 
As well as the codex Caseolinus, another manuscript existed of De magistratibus. This is the lost Codex Atheniensis, written at Trebizond at the start of July 1765. In 1879 von Lingenthal went to stay with the owner, Georgios A. Rhalles, a Greek professor in Athens, and wrote that it contained book II of De Caeremoniis aulae Byzantinae and an incomplete copy of John the Lydian’s De magistratibus. Unfortunately its present whereabouts are unknown. It seems to have been last seen in 1909, and may have been destroyed in a fire in Thessalonika in 1917. The text seems to have been inferior to that in the Caseolinus, and it is not clear whether it was a copy, or merely an inferior relation. No proper collation seems to have been made.
- I owe all of these details to the much more detailed introduction in the most recent text and translation, A. Bandy, Ioannes Lydus: On Powers, 1983, p.xxxix f.↩