University of Virginia Library

See especially his Printing and Proofreading of the First Folio, 2 vols. (1963).

See Gordon Lindstrand, "Mechanized Textual Collation and Recent Designs," Studies in Bibliography, 24 (1971), 204-214; Vinton A. Dearing, "The Poor Man's Mark IV or Ersatz Hinman Collator," PBSA, 60 (1966), 149-158; Richard Levin, "A Poor Man's Collating Machine," Research Opportunities in Renaissance Drama, 9 (1966), 25-26; Gerald, A. Smith, "Collating Machine, Poor Man's Mark VII," PBSA, 61 (1967), 110-113.

Hyder Rollins, ed., A New Variorum Edition of Shakespeare: The Sonnets, (1944), II, 1. However, his reproduction on I, pp. xx,1 of the two title pages seems to have been retouched since it shows no period after 'SONNETS' in the Aspley title. In fact, the period plainly present in Rollins in the reproduction of the Wright title is also found invariably in the Aspley, although not always so well inked. Collation of the 13 surviving copies of Q1 reveals no variants in the titles save for the imprints. I have examined 12 copies in microfilm. For the 13th, I am much indebted to Nicolas Barker, Head of Conservation, the British Library, for his generous assistance in checking against my photographs the British Library's Aspley copy sigs. [A]1r, K2v, L1v, and L2v, since this copy may not be photographed.

I used "Xerox Transparencies". These come in a 100-sheet box, reorder no. 3R459 for use in Xerox 2400, 3100, 3100LDC, 3600 1, 4000, 4500, and 7000, but not in 3600-111. According to advertising enclosed Xerox 914, 720, 1000, 813 or 660 use order number 3R163. Transparencies also come in colours. The help of an operator is required to stock the machine.

In the Huntington Wright copy handwritten notations, the initials of Narcissus Luttrell Sr., and the signature of George Steevens have been opaqued in order to avoid confusion, and a notation '5' in the upper right corner of the Aspley copy has also been removed. All plates are altered in size to fit the present page.

See W. L. Williamson, "An Early Use of Running-Title and Signature Evidence in Analytical Bibliography," Library Quarterly, 40 (1970), 245-249, for details of an analysis of 1867.

See the early work of Fredson Bowers, "Notes on Running Titles as Bibliographical Evidence," The Library, 4th ser., 19 (1938), 315-338; the definitive account of headline analysis is his "The Headline in Early Books," The English Institute Annual for 1941 (1942), 185-205, repub. in Bowers, Essays in Bibliography, Text, and Editing (1975), 199-211. His "Running Titles," in Principles of Bibliographic Description (1949), reissued 1962, treats the proper description of the portion of the headline dealt with in this essay. See also Charlton Hinman, "New Uses for Headlines as Bibliographical Evidence," Engl. Inst. Annual for 1941 (cited above), 207-222.

D. F. McKenzie, "Printers of the Mind: Some Notes on Bibliographical Theories and Printing-House Practices," Studies in Bibliography, 22 (1969). 1-75.

Bowers, "The Headline in Early Books," loc. cit., p. 198.

John Smith, The Printer's Grammar (1755), p. 41 (facsimile reprint in English Bibliographical Sources, Series 3: Printers' Manuals, ed. D. F. Foxon (1965).

MacD. P. Jackson, "Punctuation and the Compositors of Shakespeare's Sonnets, 1609," The Library, 5th ser., 31 (1975), 1-24, see pp. 2-3.

There are variations in the punctuation of the headlines between formes and within some formes. The present paper ignores all headline punctuation for the sake of simplicity.

See Bowers, "Running-Title Evidence for Determining Half-Sheet Imposition," SB, 1 (1948-49), 199-202, reprinted in Essays in Bibliography, Text, and Editing (1975), pp. 254-257.

When the last signature in an otherwise normal quarto is printed by the method known as "two half sheets printed together" or "twin half-sheet imposition," it is usual to print preliminaries to the same volume as its forme mates. Early printers' manuals, such as Stower, The Printer's Grammar (1808), p. 172, for reasons unstated, show the inner forme of one half sheet conventionally imposed with the outer of the other. We can extend our surmise, then, to the following order through the press: K(o), K(i), L(o)/[A](i), L(i)/[A](o). This implies that the variation of imprint was made during the perfecting of the last sheet for the volume. Strictly speaking the evidence shows only that L and K(i) derive from K(o); there is no absolute reason why K(i) must be printed before L(o) or L(i), though this would be normal (to clear a pile of unperfected sheets).

McKenzie, op. cit., has shown that once we assume cast-off copy there is no necessity for setting and printing sheets in alphabetical order. Although this means we cannot prove that setting or printing of other material did not intervene between K and L, it does not throw doubt on the direction of derivation, from K to L; derivation from L to K does not account, for example, for why the 'Complaint' of L is not found in K.

Note, for example the titling capitals 'A' (Sonnets 23, 37-C1v and C4v), 'O' (39, 54-D1r and D4r), 'S' (65, 75-E2r and E4v), (78, 93-F1r and F4r); and the 'g' (55.3, 57.9-D4r and D4v). These suggest only a probability because it is conceivable that copy was set seriatim and the composition halted at the end of the seventh page of these formes while the inner forme was printed off and its type distributed prior to the setting of the eighth page. The same 'g' reoccurs, however, in 115.13, 121.11-H1r and H2r, and an italic colon is found in 141.8, 145.3-I2r and I3r, two of some nine occurrences of an italic colon (there never being more than one per forme). These latter type reocurrences allow the confident surmise that copy was cast off and set by formes in Sonnets as well as "A Louers Complaint."