good care to avoid some of the worst of them: the readings of the second hand are usually, though not always, banished to the margin, their number on the whole is increased, gross errors are corrected, omissions supplied, and the Vatican chapters are given faithfully and in full. But Mai’s whole procedure in this matter is so truly unfortunate, that in a person whose fame was less solidly grounded, we should impute it to mere helpless incapacity(148). Not only did he split up the paragraphs of his quarto into the modern chapters and verses (in itself a most undesirable change, _see above_, p. 70), but by omitting some things and altering others, he introduced almost as many errors as he removed. When Dean Burgon was permitted to examine the Codex for an hour and a half in 1860, on consulting it for sixteen passages out of hundreds wherein the two are utterly at variance, he discovered that the quarto was right in seven of them, the octavo in nine: as if Mai were determined that neither of his editions should supersede the use of the other. Dean Alford also collated numerous passages in 1861(149), and his secretary Mr. Cure in 1862, especially with reference to the several correcting hands: “in errorem quidem et ipse haud raro inductus,” is Tischendorf’s verdict on his labours. Thus critics of every shade of opinion became unanimous on one point, that a new edition of the Codex Vaticanus was as imperatively needed as ever; one which should preserve with accuracy all that the first hand has written (transcriptural errors included), should note in every instance the corrections made by the second hand, and, wherever any one of the previous collators might be found in error, should expressly state the true reading.