A block of type without a raised letter was used for spacing between words or sentences. Occasionally these quads are worked up to the level of the printing service and may leave an inked impression on the paper. (These raised quads are sometimes called 'raised space' or 'blacks'.) They reflect, as G. Thomas Tanselle wrote, more than 'the state of the type when it was composed', as 'impressions from quads and bearers, even when they remain the same, are often the result of faulty make-ready and locking up of the forme or the action of the press itself' (see G. Thomas Tanselle, 'The treatment of typesetting and presswork in bibliographical description', in: Studies in bibliography, vol. 52, 1999).
Some copies of In the key of blue show such impressions of quads.
|Inked impression of a quad visible between 'quite' and ';' in J.A. Symonds, In the key of blue and other prose essays, 1893, first edition, cream coloured cloth (p. 150, line 7)|
|Inked impression of a quad visible between 'words' and '?' in J.A. Symonds, In the key of blue and other prose essays, 1893, first edition, cream coloured cloth (p. 294, last line)|
In some copies, however, there are no visible inked impressions of quads on these pages.
|No inked impression of a quad visible between 'words' and '?' in J.A. Symonds, In the key of blue and other prose essays, 1893, first edition, blue coloured cloth (p. 294, last line)|
Both are present in:
a) the proof copy;
b) the large paper copy;
None are present in:
c) copies with the edition statement: 'Reprinted July 1893';
d) copies with the edition statement: 'Third Edition January 1896';
e) copies with the edition statement: 'Third Edition (Unaltered Reprint), October, 1918'.
The regular copies of the first edition, be they bound in blue or cream cloth, show a more varied pattern. There are:
1) copies with both inked impressions of a quad visible on pages 150 and 294: bound in cream cloth;
2) copies with an inked impression of a quad visible on page 150, but none visible on page 294: bound in cream or blue cloth;
3) copies with no inked impressions of a quad visible on pages 150 or 294: bound in cream or blue cloth.
Copies c, d and e have been printed from electrotypes. The other copies have been printed before these plates were made.
After correcting the proofs - and after removing the asterisks from the signatures - the type may have been damaged slightly during the printing process, causing the inked impressions of the quads on page 150 and 294. While regular copies of the first edition were printed, the unintended visibility of the inked impressions of the quad on page 294 may have been noticed, and this was corrected. Later, the other one was noticed and another correction took place. All blue copies show only one or no traces of these quads. This might indicate that they belong to the later copies that were printed, not to the earliest ones. This corresponds with the large-paper copies that were usually printed after the regular ones: these also show both impressions.
However, how then is it possible that the proof copy shows them already? Was more than one forme used for printing the whole of the edition? We have not yet reached our conclusion.