The introduction of publisher's cloth bindings, a key development leading eventually to cloth edition bindings like those customarily used today, probably occurred sometime around 1821 and 1825. In his pioneering study of publisher's bindings, The Evolution of Publishers' Binding Styles 1770-1900, published in 1930, English publisher, novelist, collector and bibliographer Michael Sadleir stated on pp. 42-43 and plate 10, following the opinion of Geoffrey Keynes, that at the beginning of his career London bookseller and publisher William Pickering introduced the first publisher's cloth bindings with printed paper spine labels, beginning with his miniature edition of the works of Virgil in Latin (1821). This was the second book published in his miniature book-sized Diamond Classic Series, set in very small Diamond type, equal to 4.5 point. This set represented a novelty in publishing— not that it was in any way practical since the type was so small as to be barely legible without a magnifying glass. Nevertheless it must have pushed the skills of manual typesetters and punch-cutters to their limit, producing and setting tiny type that was so hard to read.
The first volume in Pickering's Diamond Classic series was Quintus Horatius Flaccus issued in 48mo in 1820. This Sadleir and other collectors had not seen in a publisher's cloth binding. Sadleir illustrated a copy of the Virgil in a cloth binding. The third volume in the series, Cicero's De officiis, de senectute et de amicitia also issued in 1821, was offered by Pickering bound in reddish brown calico cloth. In addition to the title the paper labels indicated the price (5s, in the case of the Cicero). I have a cloth-bound copy of the Cicero in my collection. That these books were issued in pioneering publisher's cloth bindings is not in dispute; what is uncertain is the date in which Pickering actually sold these books in publisher's cloth bindings.
Two years after Sadeir published his book, in 1932 London antiquarian bookseller and bibliographer John Carter issued Binding Variants in English Publishing 1820-1900. In his book Carter argued that it was possible, and even probable that the publisher's cloth bindings sometimes found on Pickering's Diamond Classics were put on a bit later than 1821. Pickering was known to have copies of his books bound up in fairly small quantities as demand warranted, so that the cloth bindings could have been put on later. Carter's argument for this, expressed on p. 20 of his book, was that after a very careful search, the earliest reference he could find to Pickering actually advertising any of his books in cloth bindings was in the "Spring List of 1826":
"The Prospectus of the Oxford English Classics (published in conjunction with Talboys & Wheeler, of Oxford) announces the series as 'neatly done up in extra cloth boards,' and describes the various volumes already published as 'in red cloth, lettered.' (This last, of course, means 'labelled.') "
Carter then suggested that the Pickering Shakespeare, issued as a set of 9 volumes in the Oxford English Classics in 1825 might be the earliest edition that Pickering actually issued from the beginning in publisher's cloth. He then also mentioned on p. 22 that the earliest actual mention of publisher's cloth known to him was a Pickering announcement in the Observer of July 31st, 1825, of another set in the Oxford English Classics, Johnson's Works, in "extra red cloth boards." Carter follows this statement with the following tentative conclusion:
"It is possible that some earlier mention in a newspaper may exist, though I have been unable to find one; but since tradition and external evidence give us Pickering as the introducer of cloth, and since his advertisements do not refer to it before 1825, I myself shall take a stand on that year for the great event. Nothing would please me better than to be convinced of an earlier date, and I am aware that my position is a conservative one; but I think that is all that is proved at present."
This was what I knew on February 14, 2015. Carter's evidence and tentative conclusion was essentially reiterated by Douglas Ball in his Victorian Publishers' Bindings (1985). However, after writing this much I learned that in 1935 Carter issued another shorter work on the subject of publisher's cloth: Publisher's cloth: an outline history of publisher's bindings in England 1820-1900, and duly ordered a copy. It turned out to be a small 8vo pamphlet of 48pp., which most recapped earlier research and presented it in a clearer manner. One point that I noticed in reading Carter's 1935 pamphlet was that the earliest definitive statement that Pickering was responsible for the introduction of cloth edition bindings was made by the publisher and writer Charles Knight in his article on printing presses and machinery and bookbinding in issue No.112, p. 511 of The Penny Magazine in 1833. Knight wrote, "But within the past seven years the introduction of the cheap and yet neat and substantial binding in cloth, which was first attempted by Mr. Pickering, of Chancery Lane, has created a new branch of business, of equal importance to any of the previously existing branches." One should note that Knight published this statement one year after the London bookbinder Archibald Leighton appears to have discovered how to stamp gold lettering into cloth bindings. Also, it is likely that Charles Knight's first edition of Charles Babbage's On the Economy of Machinery and Manufactures, of which three editions were issued in publisher's cloth in 1832, were Knight's first books issued in publisher's cloth.
In the revised edition of his William Pickering Publisher (1969) surgeon, book collector and bibliographer Sir Geoffrey Keynes repeated on pp. 13-15 the indirect evidence for the dating of Pickering's introduction of cloth bindings which he had obtained from John Carter. The story, recorded in 1855 thirty years after the fact, cannot be depended upon for reliable dating, and mainly confirms Pickering's intent to publish half of the first edition of his Diamond Classics in cloth bindings:
"An anonymous article on 'The History of Bookbinding' published in the issue of The Bookbinders' Trade Circular for March 1855, pp. 9-120, had stated that the use of cloth for binding had been introduced by Pickering in 1823 after noticing the 'French red' lining of chintz curtains. He was said to have supplied his binder, Archibald Leighton, of Exeter Street, with the right kind of calico. At first Leighton glazed and stiffened the material with glue; later he found starch to be more suitable for the purpose. Probably the latter part of the story was true, but clearly the date 1823 for the first use of cloth was wrong. This was put right in the next issue of the Circular for May 1855, p. 22, where a correspondent, R.E. Lawson, writing from 61 Stanhope Street, contributed an amended version. He agreed that Pickering first introduced cloth into the book trade, but maintained that it was his own suggestion. Lawson was working as a binder for Charles Sully and William Greenfield ('the eminent linguist') and had known Pickering since 1809. His employers had been commissioned to bind the Diamond Classics, and he related the following incident, which presumably took place early in 1820: 'Mr Pickering came one evening—I remember, perfectly well, that the candles were slight—into the shop—I believe No. 2 Upper John Street, Golden Square—and announced to Mr. Sully that he was about to publish the works above named [the Diamond Classics], and wished a quantity done in morocco, and a portion in boards. 'Now,' said he, to Mr. Sully, 'could you suggest some neater mode in which to do the boarded portion than the present one.' I immediated handed to Mr. Pickering, from my side-board, a small oblong quarto of MS music for the guitar, which myself and Mr. Sully were studying under the same master at that time, bound in light blue glazed calico, a remnant of some my mother had been lining her window curtain with, and asked Mr. Pickering 'what he thought of THAT'. 'The VERY THING,' said he, 'and excepting the colour will do admirably.' After a little deliberation, it was decided that they should be done in couleur du puce, which was the case, while the old style of 'lettering' was retained in the now rarely-to-be-met with form of the white printed 'label' of the period! The books came in—one thousand copies; five hundred were done in morocco, five hundred in 'cloth' boards; the cloth was purchased at the corner of Wilderness-row, St. John Street, and the whole of the 'CLOTH' copies were covered by myself with glue, Richard Cross, Mr. Sully's apprentice at the time, 'squaring the boards' and 'drawing in.'
"After this beginning Pickering followed the practice throughout his publishing career. He was afterwards imitated by other publishers, but the smooth red, magenta, puce or dark blue cloth used by Pickering remained for many years the distinguishing mark of his books. Even at the present time  the external appearance of his volumes in their original state has a special attraction which arrests the eye as it passes along a shelf filled with these and other volumes issued by contemporary publishers.
"The use of cloth as a publisher's binding made little difference to the cost of the books and was a considerable economy from the buyer's point of view, since the book could be used without the necessity for re-binding entailed by the flimsy boards and paper backs with which the public had hitherto contented itself. The innovation thus quietly made by Pickering in 1820 had had its effect on the whole subsequent history of the publishing trade in England, and but for him boards or paper covers for books might now be considered to be just as inevitable in this country as they still appear to be on the continent of Europe."
In 2020 Paul W. Nash published what most probably will be the definitive study of this issue as "Two Hundred Years of Publisher's Cloth," Journal of the Printing Historical Society, Third Series, Number 1, 241-301. This long paper, that reproduced many of the earliest publisher's cloth bindings in color for the first time, argued for the date of 1821 as the date that Pickering introduced what we consider the earliest publisher's cloth bindings. Nash presented all the evidence for that date, and recognized that while we may want to specify an exact date for the introduction of this book production innovation, the key element was the development of the appropriate type of cloth by the binder Archibald Leighton, and the promotion of cloth edition bindings by William Pickering. As well as Nash collected and marshalled all the evidence, inevitably there is the problem that we may not know for sure whether or not a publisher's binding was issued exactly at the time that an edition was first sold. Some publishers stored books in sheets and had copies bound as needed over the period of time that the edition was in print. Also, they sometimes gave parts of editions to different binders, resulting in binding variants. As a result, while we know when Pickering published his books because he dated them and advertised them on certain dates, exactly when he had each copy bound and how he had it bound, without having the publisher's or the binder's detailed records, has to be inferred.
Bernard Warrington, "William Pickering and the Book Trade in the Early Nineteenth Century," John Rylands University Library of Manchester (1985).
Pickering & Chatto, Catalogue 708, William Pickering and His Successors 1820-1900 (1993).