University of Virginia Library


Footnotes for Chart IV


Henry S. Van Duzer, A Thackeray Library (1919; rpt. 1965), pp. 123-132.


Proofs from an unknown auction catalogue in the Lilly Library, Indiana University.


By "first edition" I mean, of course, all printings of Vanity Fair using the type set for the book as it appeared in parts from 1 January 1847 through 1 July 1848. (See Fredson Bowers, Principles of Bibliographical Description [1949], pp. 379ff.) Some collectors use the term to mean only the "first printing," as seems to be the case in David A. Randall's "Notes Towards a Correct Collation of the First Edition of Vanity Fair," PBSA, 42 (1948), 95-109.


W. M. Thackeray, Vanity Fair (London: Bradbury and Evans, 1848) p. 50, line 11 up. Further references to Vanity Fair are made in the text within parentheses giving page and, if appropriate, line number in the form (77.25), i.e. page 77, line 25.


Randall, p. 96.


Mr. Pitt (453.31), which was not changed to Sir Pitt until the second edition (London, 1853), seems to be a favorite point, perhaps because its presence in every copy of the first edition makes every owner happy. Other "points" such as the use of roman type for the heading on page 1 and the absence of the Steyne illustration on page 336 first occurred in the third printing of the numbers containing them.


The terms "sheet," "signature," and "gathering" all refer to the same basic unit of the book produced at the printing press. In Vanity Fair, first edition, each gathering consists of sixteen pages (eight leaves) and is signed at the foot of the first and third pages.


Publishers' records are used by permission of Bradbury, Agnew, Ltd., London.


Stereotyping with plaster-of-Paris produces a slightly smaller type-page due to shrinkage of the plaster mold. For further details see P. L. Shillingsburg, "Detecting Stereotype Plate Usage in Mid-Nineteenth Century Books," Editorial Quarterly, 1 (1975), 2-3.


In the Berg Collection, New York Public Library, copy 5 in parts (part 1 only), and copy 1 in bound form (formerly belonging to Charles Dickens—numbers 1, 7-8, and 10-11), and the one copy in the Mitchell Memorial Library, Mississippi State University (numbers 1, and 7-11).


See the similar circumstance noted in P. L. Shillingsburg, "The First Edition of Thackeray's Pendennis," PBSA, 66 (1972), 35-49, esp. pp. 36-42.


The records do show that "correction and alteration" of numbers 19-20 cost £3.10.00 on 15 July 1848; that unspecified plate "repair" cost £1.10.00 on 24 February 1849; and that unspecified "mending" of plates cost 15 shillings on 28 February 1857. These charges probably refer to stereotyped plates of text rather than the steel engraved plates of full-page illustrations, for in a separate record of reprints there are standard charges for "bringing up plates and cuts." But it seems unlikely that the record of plate alteration is complete or that the cost of altering stereotyped plates totalled only five or six pounds, particularly when original corrections and night work on parts 19-20 alone came to £11.6.00.


When an alteration in standing type involves the substitution of only one character for another the effect on the rest of the line is usually too small to be noticeable even under careful examination, but larger changes, particularly if the total number of characters in the line is affected, cause alterations in the arrangement of words contiguous to the change which are readily seen in a collating machine. Often the corrector changes the spacing of contiguous words to achieve an even balance of spacing along the line. When similar alterations are made in stereotyped plates, such aesthetic considerations are a luxury generally out of reach. The line (indeed the page) is a solid piece of metal which cannot be moved and pushed about; the new reading replaces the old reading and contiguous words remain frozen in place. If the resulting disproportionate spacing is too obvious, the whole line or perhaps two or three lines are reset, the original plate being cut away to make way for the new material. Most of the alterations in numbers 2-6 and 14-20 of Vanity Fair have an appearance compatible with the effects of changes in plates. Those that do not (confined to numbers 2, 3, and 5) are discussed further in the text and identified with an asterisk in Chart III.


It appears to have been usual practice with publishers then, as now, to post-date works published late in the year; so the Simon Fraser copy probably belongs to the November 1865 printing. I have not actually seen that book which is non-circulating.


My own attempt to provide a critical assessment of the evidence will appear in a special Thackeray issue of Studies in the Novel in 1981.


Copies examined: Univ. of South Carolina, Parts; Berg Collection, NYPL, Parts copies 1, 2, 3, and 4; Univ. of North Carolina, Whittaker copy; Pierpont Morgan Library, copies 1 and 2; and two personally owned copies.


Univ. of South Carolina, V28/1849.


Univ. of South Carolina, V28/c2/1849.


Univ. of South Carolina, V2/1848.


Duke Univ. Library, (only copy).


Pierpont Morgan Library, copy 3.