University of Virginia Library

R. W. Chapman, "Printing with Figures," Library, 4th ser., III (1922-23), 175-176.

Lawrence C. Wroth, in The Colonial Printer (2nd ed., 1938), pp. 62-63, discusses the equipment of the colonial printing shop.

I have not thus far noted any press figures in American books earlier than this period.

A previous article which surveys patterns of press figures in a large number of books, though it uses a somewhat different method from the present one, is Kenneth Povey, "A Century of Press-Figures," Library, 5th ser., XIV (1959), 251-273; in it he surveys 111 English octavos dated from 1688 to 1797.

The reasonable assumption that figures would be more likely to occur in large books is borne out in a negative way (though of course not proved) by an examination of a collection (in the Wisconsin Historical Society) of 105 pamphlet sermons printed in the United States between 1771 and 1815; none of them contains figures.

For whatever usefulness it may have as negative evidence, Bibles printed by the following printers were checked and found to contain no figures (the numbers referring to the particular editions examined, as listed in Margaret T. Hills, The English Bible in America [1961], and the asterisks indicating that the plates were stereotyped by a firm other than the printer listed): William B. Allen (Newburyport), 278; R. Aitken (Philadelphia), 11; E. F. Backus (Albany), *300; Bible Society of New York, *303; Bible Society of Philadelphia, *213; D. & G. Bruce (New York), 279; J. T. Buckingham (Boston), 134, 155; Carey, Stewart & Co. (Philadelphia), 23; R. Cochran (Philadelphia), 106; B. & J. Collins (New York), 328; I. Collins (Trenton), 31; Cummings & Hilliard (Boston), 277; Abel Dickinson (Philadelphia), 189; Daniel Fanshaw (New York), *351, *377; Hugh Gaine (New York), 40; S. G. Goodrich (Hartford), *354; W. Greenough (Boston), 250, 260, 379; Greenough & Stebbins (Boston), 165, 177, 191, 214; Griggs & Dickinsons (Philadelphia), 259; J. & J. Harper (New York), *353; George F. Hopkins (New York), 76; Hudson & Co. (Hartford), 307, 335; Hudson & Goodwin (Hartford), 168, 179, 215, 237, 281; Jacob Johnson (Philadelphia), 116, 143; George Long (New York), 211, 229, 272; Simeon L. Loomis, Hart, & Lincoln (Hartford), 297; Mann & Douglas (Morris-Town), 122; W. Mercein (New York), 301; E. Merriam (Brookfield), 273; Munroe, Francis & Parker (Boston), 276; W. E. Norman (Hudson), *330; Sage & Clough (New York), 103; J. Seymour (New York), 175, 233; B. W. Sower (Baltimore), 217; Isaiah Thomas (Worcester), 30, 44, 90; Thomas & Andrews (Boston), 105, 115, 125; Thomas B. Wait (Boston), 171; Anson Whipple (Walpole, N.H.), 275; R. P. & C. Williams (Boston), *355; W. W. Woodward (Philadelphia), 135, 180, 240, 308. The following did contain figures: Jacob R. Berriman (Philadelphia), 53; Mathew Carey (Philadelphia), 102, 121, 132, 156, 161, 163, 208, 231, 238, 261, 345; Joseph Charless (Philadelphia), 77; Thomas Dobson (Philadelphia), 66; Hodge & Campbell (New York), 38; Thompson & Small (Philadelphia), 62; William Young (Philadelphia), 25, 32.

Another similar arrangement, used by Isaiah Thomas, was to repeat single alphabets with consecutive numbers—as 50Z, 51A, 52B, and so on.

Another version of this device occurs in the Bangs & Mason five-volume Bible, printed by J. & J. Harper (New York) in 1823; on $1r of every twentieth gathering there is an indication of the part "No." as well as the volume number. Cf. Volume I of Dobson's Encyclopaedia (1790), discussed as item 10 below.

Still other numbers that occur occasionally may simply be errors such as the "2" that appears near the lower edge of C8r in Azariah Mather's The Gospel-Minister Described (New London: T. Green, 1725).

For the view that unfigured formes represent work by the same press which printed the preceding or succeeding figured formes, see F. B. Kaye's edition of Mandeville's The Fable of the Bees (1924), II, 394-395; and Philip Gaskell, "Eighteenth Century Press Numbers: Their Use and Usefulness," Library, 5th ser., IV (1950), 249-261. For the view that unfigured formes are the work of a separate press, working without a number, see Walter E. Knotts, "Press Numbers as a Bibliographical Tool: A Study of Gay's The Beggar's Opera, 1728," Harvard Library Bulletin, III (1949), 198-212. D. F. McKenzie, in "Press-Figures: A Case History of 1701-03," Transactions of the Cambridge Bibliographical Society, III (1960), 32-46, believes that a single unfigured forme is the work of the press which printed the other (figured) forme of the sheet but that an entirely unfigured sheet is the work of a different press; William B. Todd, in "Patterns in Press Figures: A Study of Lyttleton's Dialogues of the Dead," SB, VIII (1956), 230-235, believes that the normal pattern was to use one figure for both formes, unless speed was important, in which case two different figures would be used.

For a statement of the position that figures stand for the pressmen, not the machines, see William B. Todd, "Observations on the Incidence and Interpretation of Press Figures," SB, III (1950-51), 171-205; Todd, "Press Figures," Library, 5th ser., VII (1952), 283; McKenzie, loc. cit. For the opposite view, that they represent presses, not men, see Philip Gaskell, "An Early Reference to Press-Figures," Library, 5th ser., VII (1952), 211; and J. D. Fleeman, "Eighteenth-Century Printing Ledgers," TLS, 19 December 1963, p. 1056. For additional discussion of press figures, see William B. Todd, "Press Figures and Book Reviews as Determinants of Priority," PBSA, XLV (1951), 72-76; Todd, "Bibliography and the Editorial Problem in the Eighteenth Century," SB, IV (1951-52), 41-55; and Giles Barber, "Catchwords and Press Figures at Home and Abroad," Book Collector, IX (1960), 301-307.

In future articles I plan to pursue in greater detail various aspects of American press figures suggested here. For the present discussion only one copy of most works was examined, since the aim, at this preliminary stage, was not to discover previously unrecognized printings through variations in press figures but simply to locate some of the books in which the figures could be found and to ascertain some of their characteristics.

I do not mean to imply that I have found evidence which proves that the figures designate the men instead of the presses; but it is convenient to be able to refer to them as standing for one or the other, and I shall speak of them as men in this article. A pressman was often identified with a press, and references to the number of a press in nineteenth-century printing books may have taken for granted the equation of the two. There does not seem to be a great deal of point in indicating on the printed sheets themselves the number of the press they were printed on, except as a record of the work done by a particular man at that press, in which case the distinction between man and press has little significance.

In this distinction in terminology between "half-sheet imposition" and "two half-sheets worked together," I am following the analysis by Kenneth Povey, "On the Diagnosis of Half-sheet Imposition," Library, 5th ser., XI (1956), 268-272.

See Wroth, The Colonial Printer, p. 164; the minimum pay per week for composition was eight dollars, at 25&c.nt; per 1000 ems, and a man on a six-day week of about nine or ten hours a day could earn his minimum by setting about 600 ems per hour. In London at this time a compositor had to be capable of setting 1500 ens (750 ems) per hour; see Ellic Howe, The London Compositor (1947), p. 59. Charlton Hinman, discussing an earlier period, uses the figure 1000 ens per hour (in a twelvehour day); see The Printing and Proof-Reading of the First Folio of Shakespeare (1963), I, 44-45.

It has become conventional, following Moxon, to set this figure for presswork, and Wroth accepts it for early American printing (p. 80), suggesting however that a ten-hour day resulted in only eight tokens printed on one side. See Herbert Davis and Harry Carter's edition of Moxon's Mechanick Exercises (2nd ed., 1962), pp. 292, 484-486; cf. Hinman, I, 41-42.

Quoted in Wroth, pp. 162, 181.

It may also be pointed out that in Young's 1790 Bible the signature pages often have a cross or dagger (τ) in the lower margin under the first column; this is true of all gatherings except E, N, Dd, Ff, and Gg (in Dd the cross comes on 5r, since 1r is the New Testament title page; in Gg there is an asterisk on 2r).

Cf. William B. Todd, "Concurrent Printing: An Analysis of Dodsley's Colleclection of Poems by Several Hands," PBSA, XLVI (1952), 45-57: "whenever books contain press figures their very presence implies unsystematic piecework engaged in conjunction with other miscellaneous endeavors. For labor which is predetermined, controlled, and properly recorded by the overseer . . . the figures become superfluous and accordingly disappear."

It has been suggested that differing sizes of press figures, when more than one size of a given number appears in the same work, may indicate different pressmen or presses—see J. D. Fleeman, "William Somervile's 'The Chace,' 1735," PBSA, LVIII (1964), 1-7. By referring to William Bowyer's ledgers, Fleeman is able to determine which presses and pressmen were responsible for each forme of this book; but the significance of the sizes of type is not clear, since two sizes of figure 3 were used by each of two presses and do not therefore provide a means for distinguishing between the work of the two presses. In the present instance, the 1 in Nn, the 2 in E and Gg, and the 4 in F, G, and H are appreciably smaller than those same numbers used as press figures elsewhere in the volume. Since it seems unlikely, however, that a system promising such confusion would have been preferred to the more obvious and sensible procedure of continuing the numbers in order, I have disregarded size in tabulating the totals for various figures.

A 1799 printing of this work which appears to have been produced from the same setting of type is the one which reads "Philadelphia: Printed by A. Bartram" on the title page.

Wroth, p. 294; other discussions of the Dobson Encyclopaedia are in Evans 22486 (the source for information about the plans and dates of publication) and in Herman Kogan, The Great EB: The Story of the Encyclopaedia Britannica (1958), p. 25.

Cf. Povey, "A Century of Press-Figures," pp. 254-256. Povey cites evidence that some pressmen prefer to figure the near side of the forme, some the off side; in the Encyclopaedia the off side is consistently the choice.

It should be remembered that half-sheets are not included in any of the summary statistics.

See David Kaser, Joseph Charless: Printer in the Western Country (1963), pp. 27-31. Charless also printed for Carey Choice Tales (1800) and Conductor Generalis (1801). Cf. Letter XI (5 February 1834) of "Autobiography of Mathew Carey," New England Magazine, VI (1834), 230-234: "In 1801, I published a quarto edition of the Bible, (of three thousand copies,) . . ." (p. 230). On Carey's Bibles, see also Mason Locke Weems: His Works and Ways, ed. E. E. F. Skeel (1929), II, 133-157.

Information about Riley is taken from G. L. McKay's directory of the New York book trade from 1633 to 1820 (1942), the imprint catalogue at the American Antiquarian Society, and the imprints of Riley's books.

When 3 placed his figure on N4v, it was perhaps because N3v had a short text.

As in the previous statistics, only regular gatherings are counted—in this case, only the complete half-sheets. Gatherings of two leaves (quarter-sheets) are listed but are not tabulated, since they would unfairly affect the results. Also as before, numbers following an oblique line in the summaries refer to figures occuring nearer the right edge of a page; the normal position is nearer the left edge.

Many of the reprinted legal volumes are paged so that reference can be made to the first edition. Thus the number in the headline corresponds to the page in the first edition on which the text beginning the present page fell; wherever the next page of the first edition began, there will be an asterisk in the text and in the margin indicating the new page. Under this system a recto headline may carry an even page number, and two consecutive headlines may have the same number (or skip over a number).

Figure 4 on 23 v of the Wisconsin copy is not present in the University of Chicago copy. This is the only instance of a variation in figuring among the books which I examined in multiple copies, and the significance of the variation is not clear —whether it represents a second impression of gathering 2 or an accidental loss of the figure.

Kenneth Povey's observation, in "A Century of Press-Figures," p. 254, that the preference for rectos or versos often depends on the predilection of individual pressmen does not seem to hold true for these American books; in these examples all figures appear most often on those pages ($4r and $4v in quarto, $3v in half-sheet octavo) which must have been established by shop policy as the pages for figures.

It should be understood that the policy for assigning entry numbers in the Shaw-Shoemaker checklist sometimes makes it difficult to refer to one particular volume. The multi-volume legal reports for various states, for example, are often given only one entry, under the year of the first volume. Even though inclusive dates are given, the dates of individual volumes are obscured, and the names of later printers do not appear at all. Cf. Shaw 20243, 27411. Entry 16644 lists four volumes of the Virginia Reports under the date of the first volume, 1808; since the printer for that volume is Smith & Maxwell, Riley's name never appears. Sometimes the name of the compiler of the reports is missing (cf. the omission of McHenry in 25956 and of Caines in 29344). Nor is the policy consistent, for three volumes of Caines' Term Reports are given separate entries (29344, 32323, 32324), as are the two volumes of Chitty (25064, 25065). When an index of the Shaw-Shoemaker list is prepared, therefore, certain volumes, authors, and printers may not be represented in certain years as accurately as they should be.