The differences in the print of the old books confirm that proof-reading was a common practice when printing Lithuanian books in the 16th century already. The differences can be classified into technical and proofing-related. The former occurred due to typographical actions or materials, the latter were caused by corrective proofing. They allow reconstructing the course of the printing processes and suggest that the presswork would not be ceased after the press proof was taken. While corrector was at work, certain copies of sheets would be printed and the unrevised sheets would later be bound rather than destroyed. They would end up in different copies, resulting in slight differences of print between some of them. Later on, any mistakes that were noticed would be included and printed in a list of errata. There were no lists of errata in the very first Lithuanian books. The oldest 16th century issue with clear evidence of revision was Evangelijos bei Epistolos (1579) by Baltramiejus Vilentas, with corrections made by pasting bits of paper on top of printed words with errors. The first Lithuanian book that had the list of errata was Postilė (1591) by Jonas Bretkūnas, which was printed at the same printing house of Georg Osterberger. Bretkūnas’s Postilė possesses a number of proofing differences in the print that show, with an almost absolute degree of accuracy, whether the printer’s sheet was printed before, or after proofing. A comparison of some of the differences present in the copies (accounting for a fraction of all copies known to exist) allows making a cautious statement that the unrevised sheets had been printed in a smaller number.
Some major differences between copies might have been caused by the printer’s decisions or could have occurred as a result of changing a part of the run to fit the target audience. The last half-sheet signature of the Lithuanian grammar in German (1654) by Daniel Klein was composed twice, allowing a more efficient use of the press and cutting of the printing time by half. The forewords to Klein’s hymnal were removed from some of the copies by the printer (1667). The forewords to the 1701 New Testament were removed to accommodate the target audience. To distribute the remaining run of Konstantinas Sirvydas’s postil (1629), the forewords of the first part were reprinted when publishing the second part of the book in 1664. The proofing differences in the print of the books by Bretkūnas, Sirvydas, Klein, and others were discovered by accident. After the second copy of the first issue of Suma Evangelijų, a postil from Knyga Nobažnystės was identified in Krakow, it was carefully compared to the copy that had been known to exist in Uppsala. Computer algorithms aided to discover four proofing differences, all of them in the headings of chapters. The Krakow copy contained printing errors (mixed-up order of words, mistakes in references to the Bible), which had been corrected in the Uppsala copy; still several headings had errors in both copies. One thing that the proofing revisions have in common is that they have to do with references to the Gospel of John. The postil was prepared by two translators. The distribution of variance of the references in the other parts of Knyga Nobažnystės and the proofing revisions thereof suggest that the translator of the middle part of the postil made the corrections of the part he had translated or that it was revised by the printing house’s proof-reader based on the translator’s manuscript.