University of Virginia Library

ALTERATION OF LEADING WITHIN EDITIONS
by
Gillian G. M. Kyles

When, in his 1965-66 Lyell Lectures, Simon Nowell-Smith spoke of Victorian books in which “the leads had been removed, the type reimposed,”[1] he seemed to assume widespread awareness of those characteristics of reprinted editions. The occurrence of such features seems not, however, to have been widely remarked. I offer here some examples of changes in leading between impressions in order to draw attention to the practice and to stimulate others to identify more instances and to explore the nature and extent of the phenomenon.

Changing the spacing between the lines is one way of addressing the frequent desire of publishers to bring out a book in several forms. At times this has been achieved by printing copies on different surfaces—on finer, thicker, or larger paper, for instance, or on vellum or some form of cloth. Complete resetting is always possible, especially when a different bibliographical format is needed, though sometimes reimposing the existing type serves equally well. Such reimposition takes place on a small scale when margins are adjusted in the transition between copies on dissimilar sizes of paper, but it occurs in a more extensive manner when type-pages are moved around to enable the printing of a different number of pages on a sheet. In Pope's Dunciad of 1728, for instance, the original duodecimo was immediately reimposed as an octavo; Scott's Life of John Dryden (1808) was first set up for octavo printing as volume 1 of Dryden's Works and then reimposed for production as an independent quarto. The practice of nineteenth-century American plate-makers who included signatures for more than one imposition likewise points to the custom of printing the same setting in different formats. A more complicated situation arises when the type metal is reused in units other than the original type-pages. In Pope's Works of 1735 and the Foulis Press's edition of Caesar's works of 1750, for instance, the same setting of individual lines is employed for both the folio and the quarto, though the number of lines on each page has been adjusted to compensate for the difference in leaf size. More intricate yet are the situations where the length of the original lines is changed as well. After the quarto 1762 edition of Fielding's Works was produced, the type was rearranged in a shorter measure for printing in octavo; the folio version of the 1733 tract The Case of the Planters of

Tobacco in Virginia was also reimposed as an octavo, with each folio line divided into two.[2]

The practice Nowell-Smith mentions is one way of adjusting existing type-pages—in this case, by reducing or completely removing intralinear leading. A scientific classic from the period of his interest offers a clear example of the procedure. Alfred Russel Wallace's Malay Archipelago was first published in two volumes by Macmillan in 1869. Printed by “R. Clay, Sons, and Taylor,” this set was reprinted once the same year but then, in 1872, the text was released as a “Third Edition” in a single volume. (It was reprinted six more times in London, and printed twice in the United States, before being replaced by another one-volume edition, the “Tenth,” in 1890.) To compress the 478 and 524 pages of the original books (and the 24 and 4 pages of preliminaries) into 16 preliminaries and 653 text pages, Clay and Taylor removed most of the leading; the number of lines per page subsequently rose from 26 to 41, and the 10-line measurement (for type with a face height of 3.4 mm) correspondingly decreased from 54 mm to 36 mm. Although the preliminaries were reset and the text of the Appendix (2.465-501 in the two-volume printings; pp. 599-633 in the later ones) was not modified at all, a sample comprising the first four chapters (or hundred pages in the first printing) shows that the type here is almost always the original type but with the leading reduced. The reasons for this kind of compression are obvious. The paper and binding costs of the one-volume edition would be considerably less than for several volumes, and the adjustment could be carried out more quickly than setting the text anew—and by an unskilled employee at lower wages.

That the type is the same and not a line-for-line resetting is made clear from damage that occurs throughout. Examples from the first of the original volumes and from the single-volume reprint (I have checked the 1883 printing) include the following (listed, if applicable, by volume, page, and line):

     

1.1.1 / 1.1		IF] top arm of “F”
1.8.15 / 6.8		posi-] bottom center of “o”
1.11.7 / 7.37		life] bottom center of “e”

       

1.20.14 / 13.32		100] bottom left of second “o”
1.25.16 / 17.1		toucans] top right of “o”
1.92.23 / 59.27		though . . . the] top right of “o”; bottom center of “e”
1.95.17 / 61.17		hands . . . and soon] both “d”s battered

Four minor resettings occur in the first four chapters, sometimes for reasons that are apparent. The original 29.18-20 is reset into four lines (here the cause is not evident); 47.25-49.3 are redone in order to accommodate a twenty-five-line wood-engraving occupying the right-hand side of the text column; 43.21-22, originally set loosely, are tightened into a single line; and 93.19 seems to have been reset with the deleterious loss of a comma at its end. In the latter instance it is clear that the adjacent lines were not reset, for the “y” of “They” in the previous line remains, as does the thickening of the “l” in “fleshly” below. In the pages beyond this sample, it appears that occasionally lines have been reset to avoid widows or to equalize facing pages.[3]

Major literary works of the Victorian era also provide evidence of reprints in which the leading has been changed. The late Josiah Q. Bennett, Rare Book Cataloguer at the Lilly Library, pointed out to me leading reduction in Henry James's The Bostonians, likewise published in London by Macmillan but printed by R. and R. Clark in Edinburgh. In 1886 the novel was made available first in three volumes and then in one. The 702 pages of the main text were reduced to 449; the number of lines per page rose from 25 to 40, and the 10-line measurement (for type with a total face of 3.2 mm) shrank from 58 to 36 mm. The same phenomenon occurred in James's The Princess Casamassima that year, also printed by the Clarks and published by Macmillan: the 751 text pages in three volumes became 596 pages in one; the number of lines per page grew from 31 to 40, and the 10-line measure (for the same type as in The Bostonians) dropped from 47 to 36 mm. Both novels, moreover, show similar type damage in the sample passages checked.[4] But the procedure was not universal for books even from this same author, printer, and publisher: in 1888, the single-volume editions of James's The Reverberator and The Aspern Papers were instead reset after their initial two-volume appearance.

Two ornithological works formerly in the H. Bradley Martin collection also display variation in leading. One of these is again from late nineteenth-

century England; the other was produced much earlier in the century, in Germany. The first example, Thomas Littleton Powys Lilford's Notes on the Birds of Northamptonshire and Neighbourhood (2 vols.; London: R. H. Porter, 1895), appeared as a large-paper issue (limited to 100 copies) and a trade issue, both printed by Taylor and Francis. The large-paper copies have plates in two states and the wood-engravings on separate leaves; the trade issue has only one state of the plates, with the wood-engravings set into the text. Because the larger pages have more leading, each comprises only 31 lines of text, instead of the 34 lines on each page of the trade issue.

The German publication offers an instance in which leading was clearly increased for the second impression. Teutsche Ornithologie (Darmstadt, 1800-11 and ca. 1837-41) is customarily attributed to Maritz Balthasar Bor[c]khausen, though it was the work of several authors—and is sometimes catalogued under the name of the illustrator, Johann Conrad Susemihl. Copies of the two printings in the Martin collection had identical type, with the exception of the first fascicle (where the text was completely reset) and fascicle 15 (where French text is added to the original German). The first printed page of fascicle 15 (containing the description of “Falco subbuteo”) is now accompanied by a French version on the facing page; because the French translation is longer, the leading in the German text has been increased to balance the size of the type pages. That adjustment results in seven fewer lines on the German page than in the first printing. The subsequent pages of the fascicle are treated differently: the French text is simply set below the existing German text, with no alteration of leading. No printer has been identified for the new impression; the single front wrapper bound in the Martin copy does not list one, and no copies in parts (with potentially revealing wrappers) are recorded.[5]

Adjustment of leading also occurred in America, and in the twentieth century. I am indebted to Josiah Bennett and to Dr. Cecil K. Byrd, former Indiana University Librarian and later a consultant at the Lilly Library, for information about the practice at the Warner Press in Anderson, Indiana. In the 1930s, when Byrd worked as a printer's devil there, the press regularly kept standing the linotype slugs of every book and pamphlet, paged but in galley trays. If there was no further call for the type for nine months, it was then sent to the melting room. If, on the other hand, a reprint in a smaller format was warranted, a printer's devil took a long pair of tweezers and removed the three- or two-point leading, after which he repaged and transferred the type to regular formes. (The task obviously was easier with solid linotype pieces than with foundry type.) The same firm also did a great deal of job printing and would commonly print handbills in two sizes—2000 copies of the larger ones, and then, after removing the leading, 5000 copies of the small version.

Another American printer known to have used leading reduction was Carl Purington Rollins. John S. Kebabian, who worked as an assistant to Rollins in 1930s, recalls the removal of leading as a common practice. Kebabian's personal collection of antique tools includes many used in the printing trade, but he has been unable to identify the kind of tweezers that might have been used by the Warner Press. In any event, it may be that different printers had different ways of removing leading and that there was no tool manufactured specifically for the purpose.

These glimpses of leading alterations in several countries over a couple of centuries suggest that systematic scrutiny would produce many more examples. The investigation would help fill gaps in the historical understanding of book production, but the inquiry also would undoubtedly raise new questions. If it is correct that in the nineteenth century first impressions of American editions, unlike English ones, tended to be printed from plates, one would anticipate turning up fewer examples of changed leading in American publications during that period—one of a number of hypotheses that analysis might test.

Notes