[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]
The text has been carefully proofread.
If you find a mistake though, let me know!



The codices of the Strategemata1 are of two classes. Of the better class, three manuscripts survive; of the second, inferior class, there are many representatives.

The three manuscripts of first class are the codex Harleianus 2666 (H) which contains the entire work, but is very carelessly written; the codex Gothanus I. 101 (G), and the codex Cusanus C. 14 (C), each of which contains excerpts only. These three have been taken from a copy which, though not free from errors and lacunae, was still most carefully written, and must be thought to have preserved with the greatest fidelity even mutilated words.

The second class is derived from a codex which the copyist had corrected in many places according to his pleasure, and the manuscripts of this class never bring any help to readings where the first class is in error. The best representative of the second class is codex Parisinus 7240 (P), which far surpasses all the rest in authority. Both classes come from the same archetype in which a leaf had been transposed.​2 A copyist of P about the  p. xxvii thirteenth century, noticing this transposition, attempted a re-arrangement which appears in some of the later codices and editions.

The archetype had many lacunae, glosses and duplicates. It seems to have contained, besides the Strategemata, a breviarium of Eutropius, since these two works are found combined in most of the oldest codices of both classes. After the thirteenth century, copyists put Vegetius and Frontinus in one codex.

The codex Harleianus, a parchment of the ninth or tenth century, is now in the British Museum. It contains many errors and omissions. Some of these were corrected by the copyist himself, others by another hand of the same period. The latter (designated by h) became the model for the second class of manuscripts, the readings coinciding in many instances with those of P. As the corrector frequently used the eraser, for readings lacking in G and C we have only the testimony of the second class. The readings of h are sometimes happy amendments, sometimes atrocious corruptions. The writer omitted, copied incorrectly, amended and changed the spelling, assimilating most prepositions.

The codex Gothanus, a parchment of the ninth century, contains a breviarium of Eutropius, a breviarium of Festus and excerpts from Frontinus.​3 It was written by two hands. The copyist of Frontinus frequently separated the words wrongly, and misunderstood signs of abbreviation.

 p. xxviii  The codex Cusanus is a parchment of the twelfth century.​4 It is written in two columns and contains many other things beside excerpts​5 from Frontinus. Few of the latter are intact, most of them being contracted or otherwise changed; but this codex contains some excellent notes.

The codex Parisinus is a parchment dating from the end of the tenth or the beginning of the eleventh century. It contains the Strategemata and a breviarium of Eutropius. The order of examples found in the early codices is retained,​6 and the passage II.IX.7-II.XII.2 (quarum metu illisecundum consuetudinem) stands at the end of Book IV, but after the words persecuti aciem in fossas deciderunt et in eo modo victi sunt (the concluding words of II.XII.2 in our present arrangement, the conclusion of II.IX.7 as it then stood), the copyist wrote duo capitula sunt requirenda.

d is the designation given by Gundermann to all the codices of the second class excluding P. These date mostly from the fourteenth or fifteenth century. The oldest and best among them are Harleianus 2729, Oxoniensis-Lincolniensis 100, Parisinus 5802, Gudianus 16. Here and there these codices show an amendment which is an improvement, but almost everywhere  p. xxix are signs of corruptions which have been afterwards corrected.

In the manuscript from which the archetype of our present manuscripts was derived, a leaf (or leaves) at some time became transposed which contained the following: IV.VII.41 (beginning with continere vellet), IV.VII.43 (now II.IX.8), 44 (II.IX.9), 45 (II.IX.10), II.X.1, 2, II.XI.1‑7, II.XII.1, 2 (through secundum consuetudinem). This leaf had stood originally after the leaf ending with the words quarum metu illi con- in II.IX.7, and before that beginning with the words adventaret recepit aciem in II.XII.2, and when transposed stood between the leaf which ends with the words cum in agmine milites, in IV.VII.42, and the leaf beginning with the words -nere vellent pronuntiavit in the same chapter.

To F. Haase​7 and E. Hedicke​8 belongs the credit of the present arrangement of the examples. Haase, having chanced on a manuscript of Frontinus in a book-shop, bought it fairly cheaply, being attracted by a note after II.IX.7, which in all the better manuscripts ended in this way: quarum metu illi cum adventarent, recepit aciem; persecuti aciem in fossas deciderunt et eo modo victi sunt. In this copy there followed in red ink the words: nota hic defectum magnum. Haase was led by this to investigate more fully and he decided, as a result of his researches, that the passages found in the codices at the end of Book IV should not be placed after II.IX.7, but within it, since the conclusion of II.IX.7 was quite unsuitable to its beginning. Hedicke later followed up Haase's work, and by the discovery of like inconsistencies  p. xxx in the two parts of II.XII.2, and IV.VII.42, as they stood, and by the use of slight emendations, restored the order and consistency of the arrangement as it stands to‑day.


As early as 1425 Poggio Bracciolini had learned that at Monte Cassino there was a copy of the De Aquis of Frontinus, but it was not until he visited the monastery in 1429 that the manuscript, C, was actually found. It was carried off to Rome, copied and returned to Monte Cassino, where it still remains. It is an original manuscript in the sense that at the time of its discovery no other manuscript of the work was known, nor has any since come to light, excepting those derived from copies made by Poggio at that time.

Eight of these copies are described by Bücheler in his preface, only two of them being of any significance. According to the judgment of Poleni, in which Bücheler concurs, these two were written a little after the middle of the fifteenth century. The codex Urbinus, or Vaticanus 1345, agrees closely with the original at Monte Cassino; the codex Vaticanus 4498, which contains many errors and seems to have come from an inferior copy, was used by Pomponius Laetus and Sulpicius in bringing out their first edition in 1484‑1492. Jocundus may have had access to the original copy, since he agrees with it in certain readings not found in earlier editions. Poleni used a copy of C made by Gattola​9 and both Vatican manuscripts. Bücheler  p. xxxi consulted both Vaticans and a copy of C which Kellermann​10 had made for Schultze.

The original manuscript, C 361, is of parchment and contains, besides the De Aquis, the De Re Militari of Vegetius and a part of Varro's De Lingua Latina. It had seen hard usage before it came to the monastery, some leaves being torn and some chapters much mutilated. Poleni places its date at the end of the thirteenth or the beginning of the fourteenth century, while the catalogue of the library of Monte Cassino assigns it to the end of the eleventh or the beginning of the twelfth. Bücheler thinks it belongs to the thirteenth rather than the eleventh. It is written in minuscules which were growing dim even when copied by Gattola; to‑day some parts of the manuscript are so obscure as to be difficult to read. Various portions were copied in red ink;​11 the punctuation is erratic; sentences sometimes begin with capitals, again with small letters; no intervals are left between words except where the intention is to show that something is missing from the text; but in places where lacunae are found, the spaces left do not always seem proportioned to the words to be supplied.

There are traces of emendation by some hand of a later century; dots, originally omitted, have been placed over the letter i; marks of abbreviation have  p. xxxii been mischievously doubled and tripled in places,​12 words have been changed​13 and lacunae filled in.

In the mathematical part of the work, the numbers are often quite obviously incorrect, although it is hard to tell whether the inaccuracies are to be attributed to the author, whose interests were not primarily in the field of mathematics, and who may have used approximate figures at times, or to the copyist, who did not understand the signs or who may have erred in the substitution of the names of the figures for their symbols.

The title of the work given in C is de aqueductu urbis Romae; that of the early editions, de acquaeductibus; for these infelicities Bücheler has substituted the title suggested by Heinrich, de aquis urbis Romae.

The division into two books is indicated in C and in the codex Urbinus. The numbers of the chapters were added by Poleni. For a new collation of C, cf. Petschenig, Wien. Stud. VI (1884), p249. For a facsimile of the manuscript, cf. Clemens Herschel, The Two Books on the Water Supply of the City of Rome.​a

The Author's Notes:

1 For a full account of all the manuscripts, cf. G. Gundermann, De Iuli Frontini Strategematon libro quo fertur quarto, Commentationes Philologae Jenenses (Leipzig, 1881), p83.

2 Cf. p. xxix.

3 i.e., all of Book IV, followed by II.IX.7-II.XII.2 (from quarum metu illi through secundum consuetudinem); pref. to Book I; I.I.1‑2; pref. to Book II; II.I.1‑3; pref. to Book III; III.I.1‑3; III.III.1‑7 III.VII.1‑6.

4 Cf. Joseph Klein, Ueber eine Handschrift des Nicolaus von Cues., Berlin, 1866, p6.

5 i.e. pref. to Book I (si qui erunthostis sit, cf. p6), followed by I: i.12, 13, 14;º ii.7; v.1; vi.4; vii.2, 3; viii.8; x.4; xi.5, 8, 19; xii.1, 5, 12; II: i.1, 2, 13, 17; ii.2; v.41; vi.3; viii. 6, 7, 11; ix.6, 7; xiii. 2, 3, 8; III: vii.6; ix.6; IV: i.3, 5, 17, 29, 35, 36, 38, 42, 45; iii. 1, 2, 3, 9, 10, 12; iv.1, 2; v.12, 13; vii. 1, 4, 10, 14, 15, 37; II.XI.2.

6 Cf. p. xxix.

7 Cf. Rhein. Mus. III (1845), p312.

8 Cf. Hermes, VI (1872), p156.

9 qui satis ut illa aetate religiose Poleno librum descripsit. B.

10 quo qui epigraphica studia attigerunt, auctorem sciunt nullum esse certiorem.

11 e.g. the words at the beginning of the preface: Incipit prologus iuliifrontiniiº in libro deaqueductuº urbis; at the end of chapter iii: Explicit prologus; after chapter lxiii: liber primus explicit. liber secundus incipit; the names of the aqueducts at the beginning of chapters lxv‑lxxiii.

12 Cf. 112.

13 e.g. quarto to quinto in 11.

Thayer's Note:

a About 15% of this facsimile is onsite; I do expect to put all of it up. Those portions that are online can be accessed by clicking on the initial number of the corresponding section in the Latin text, q.v.

[image ALT: Valid HTML 4.01.]

Page updated: 4 Jun 02