This statement of Boswell's is taken from a note in Malone's handwriting prefixed to the copy in question, which we transcribe.
"This copy of King Lear differs in some particulars from the two others in Vol. IV.
"The title-page of it is the same as the second of those copies, that is, it has no direction to the place of sale, and the first signat. is B,—notwithstanding which there are minute diversities; thus, in this copy in H3 verso, we have 'A foole vsurps my bed'; in the other whose first signature is also B, we find—'My foote usurps my body', and in the copy without any direction to the place of sale (whose first signature is A) 'My foote usurps my head'."
Now it is a little remarkable that at present the copy has no title-page at all, and there is no trace of the title-page having been removed since the volume has been in its present condition. The probability is that the title was originally wanting and that one had been supplied from a copy of Q1 before it came into Malone's hands, and that while it was in this condition he wrote the above note upon it. It was then sent to be bound in a volume with other quartos, and the title may have been lost at the binder's, or may have been intentionally removed as not belonging to the book. That alterations were made by the binder is evident from the fact that the copy to which Malone refers as the second of those in Vol. IV. is in reality the first. Malone, writing his note when Vol. IV. was arranged for binding, described the then order of the plays, which must afterwards have been altered. In any case, however Malone's statement is to be accounted for, it is quite clear that Boswell must have described the Quarto after it was bound, when the title could not have existed.
We have said that Boswell quotes the three Quartos of Lear, now in the Bodleian, by the letters A, B, C, respectively. In doing so, however, he is not consistent. We record his mistakes that others may not be misled by them. Bearing in mind therefore that A = Q2 (Bodl. 1), B = Q1, and C = Q2 (Bodl. 2), we find in Act II. Scene 2 (Vol. X. p. 97) 'Quarto B, ausrent; Quarto A, reads unreverent.' Here B and A should change places. In Act III. Scene 7 (p. 188), 'Quarto A omits roguish:' for A read C. In Act IV. Scene 2 (p. 199), for 'Quartos B and C, the whistling,' read 'Quarto C' alone. In Act IV. Scene 6 (p. 220) B and A should again be interchanged. In Act V. Scene 3 (p. 277), 'Quarto A omits this line'; for A read B. It will be seen from these instances that A has been in turn made to represent three different copies.
The differences in various copies of Q2 are accounted for by supposing that the corrections were made before the sheets were all worked off, and that the corrected and uncorrected sheets were bound up indiscriminately. It will be observed that the readings of the uncorrected sheets of Q2 agree for the most part with those of Q1, and this led us to the conclusion which had previously been arrived at by Capell and also by J. P. Kemble, that the edition which we have called Q1 was the earlier of the two printed in the same year. But upon collating a copy of Q2 in the Bodleian, which we have called Q2 (Bodl. 1), we found evidence which points to an opposite conclusion. In Kent's soliloquy (II. 2. 160) that copy, as will be seen in our notes, reads,
nothing almost sees my rackles
But miserie, &c.
which of course is an accidental corruption, by displacement of the type, of 'myrackles' (i.e. 'miracles') the true reading. In the corrected copies of Q2 this is altered, apparently by the printer's conjecture, to 'my wracke', which is also the reading of Q1. Throughout the sheet in which this occurs the readings of Q1 agree with the corrected copies of Q2, and had it not been for the instance quoted, we might have supposed that the corrections in the latter were made from Q1. But the corruption 'my rackles' for 'miracles' must have come from the original MS., and 'my wracke' is only a conjectural emendation, so that the order of succession in this sheet at least appears to be the following. First the uncorrected copy of Q2, then the same corrected, and lastly Q1. On the other hand it is remarkable that Q1, if printed from Q2 at all, must have been printed from a copy made up, with the exception just mentioned from II. 1. 128 to II. 4. 133, and another containing from IV. 6. 224 to V. 3. 64, of uncorrected sheets. Another hypothesis which might be made is that Q1 and Q2 were printed from the same manuscript, and that the printer of Q1 corrupted 'miracles' into 'my wracke', while the printer of Q2 made it 'my rackles', which was afterwards altered by a reference to Q1. The question, however, is very difficult to decide, and at most is one rather of bibliographical curiosity than of critical importance. We may mention that, without giving the reasons for his conclusion, Jennens, in his edition of Lear in 1770, quotes as the 1st Quarto that which we have called Q2 and vice versa.
A third Quarto, which we have called Q3, was printed very carelessly page for page from Q1 and published in 1655.