"There are no letters larger than the rest at the beginning of sentences, though the continuity of the text is much broken by a line being left incomplete (sometimes it will contain only two or three letters), in which case the first letter in the next line mostly stands out of the range of the column, encroaching on the margin (see Facsimile 2, 11. 5, 6).
This manuscript must have been derived from one more ancient, in which the lines were similarly divided, [i.e., narrow columns 12-16 cpl] since the writer occasionally omits just the number of letters which would suffice to fill a line, and that to the utter ruin of the sense; as if his eye had heedlessly wandered to the line immediately below. Instances of this want of care will be found Luke xxi. 8 ; xxii. 25, perhaps John iv. 45 ; xii. 25, where complete lines are omitted : John xix. 26 ; Heb. xiii. 18 (partly corrected) ; Apoc. xviii. 16 ; xix. 12 ; xxii. 2, where the copyist passed in the middle of a line to the corresponding portion of the line below.
It must be confessed, indeed, that the Codex Sinaiticus abounds with similar errors of the eye and pen, to an extent not unparalleled, but happily rather unusual in documents of first-rate importance ; so that Tregelles has freely pronounced that "the state of the text, as proceeding from the first scribe, may be regarded as very rough" (N. T. Part ii. p. 2). Letters and words, even whole sentences, are frequently written twice over, or begun and immediately cancelled : while that gross blunder technically known as Homoeoteleuton, whereby a clause is omitted because it happens to end in the same words as the clause preceding, occurs no less than 115 times in the N. T., though the defect is often supplied by a more recent hand. We have thought it right to record all such clerical errors in their proper place for the reader's information ; hut while they must he admitted to deform the face of this exquisite relique of the primitive ages of our faith, they need not he held to detract materially from its intrinsic value, much less ought they to militate against our conviction of its very high antiquity."
- F.H.A. Scrivener,
A Collation of Codex Sinaiticus., (1864) p. xiv-xv
Of course many Hortians have complained that Scrivener's count here is skewed by his use of the TR as a reference. But it must be acknowledged that Scrivener is in the main talking about singular and nonsensical readings not shared by Vaticanus, and that are traceable to the copyists who made this manuscript, 'prima manu' (1st hand, 1st generation errors). Scrivener is hardy dismissing possible variant readings that have other textual support and which preserve the sense. He highly prizes Sinaiticus, in spite of its errors, and would not ignore or abandon important readings in the process of collating plain errors.