The letters contained in this volume cover a large and important period in Cicero’s life and in the history of Rome. They begin when he was 38 years of age; and at first they are not very numerous. There are only two of that year (68 B.C.), six of the following year, one of the year 66, when he held the praetorship, and two of 65. Then there is a gap in his correspondence. No letters at all survive from the period of his consulship and the Catilinarian conspiracy; and the letters to Atticus do not begin again until two years after that event. Thereafter they are sufficiently frequent to justify Cornelius Nepos’ criticism, that reading them, one has little need of an elaborate history of the period. There are full—almost too full—details, considering the frequent complaints and repetitions, during the year of his banishment (58–57 B.C.), and the correspondence continues unbroken to the year 54. Then after a lapse of two years or more, which Atticus presumably spent in Rome, it begins again in 51, when Cicero was sent to Cilicia as pro-consul, much against his will; and the volume ends with a hint of the trouble that was brewing between Caesar and Pompey, as Cicero was returning to Rome towards the end of the next year.

The letters have been translated in the traditionary order in which they are usually printed. That order, however, is not strictly chronological; and, for the convenience of those who would read them in their historical order, a table arranging them so far as possible in order of date has been drawn up at the end of the volume.

For the basis of the text the Teubner edition has been used; but it has been revised by comparison with more recent works and papers on the subject. Textual notes have only been given in a few cases where the reading is especially corrupt or uncertain; and other notes too have been confined to cases where they seemed absolutely indispensable. For such notes and in the translation itself, I must acknowledge my indebtedness to predecessors, especially to Tyrrell’s indispensable edition and Shuckburgh’s excellent translation.

There remain two small points to which I may perhaps call attention here in case they should puzzle the general reader. The first is that, when he finds the dates in this volume disagreeing with the rules and tables generally given in Latin grammars and taught in schools, he must please to remember that those rules apply only to the Julian Calendar, which was introduced in 45 B.C., and that these letters were written before that date. Before the alterations introduced by Caesar, March, May, July and October had 31 days each, February 28, and the other months 29. Compared with the Julian Calendar this shows a difference of two days in all dates which fall between the Ides and the end of the months January, August and December, and of one day in similar dates in April, June, September and November.

The second point, which requires explanation, is the presence of some numerals in the margin of the text of letters 16 to 19 of Book IV. As Mommsen pointed out, the archetype from which the existent MSS. were copied must have had some of the leaves containing these letters transposed. These were copied in our MSS. in the wrong order, and were so printed in earlier editions. In the text Mommsen’s order, with some recent modifications introduced by Holzapfel, has been adopted; and the figures in the margin denote the place of the transposed passages in the older editions, the Roman figures denoting the letter from which each particular passage has been shifted and the Arabic numerals the section of that letter.

The following signs have been used in the apparatus criticus:—

M = the Codex Mediceus 49, 18, written in the year 1389 A.D., and now preserved in the Laurentian Library at Florence. M¹ denotes the reading of the first hand, and M² that of a reviser.

? = the reading of M when supported by that of the Codex Urbinas 322, a MS. of the 15th century, preserved in the Vatican Library.

E = Codex Ambrosianus E, 14, a MS. probably of the 14th century, in the Ambrosian Library at Milan.

N = the Codex ex abbatia Florentina n. 49 in the Laurentian Library, written in the 14th or 15th century.

P = No. 8536 of the Latin MSS. in the BibliothÈque Nationale at Paris, a MS. of the 15th century.

R = No. 8538 of the same collection, written in the year 1419. These four MSS. E, N, P, R, with some others form a separate class; and

S = the reading of all the MSS. of this class, or of a preponderant number of them.

C = the marginal readings in Cratander’s edition of 1528, drawn from a MS. which is now lost.

Z = the readings of the lost Codex Tornaesianus, Z^b denoting the reading as preserved by Bosius, and Z^l that testified to by Lambinus.

I = the reading of the editio Jensoniana published at Venice in 1470.

Rom. = the edition published at Rome in 1470.