previous next


* constructed in 144-140 B.C. by Q. Marcius Rex, the water being brought to the Capitol in the latter year. He had been commissioned by the senate to repair the Appia and Anio (Plin. NH xxxvi. 121, who wrongly adds the Tepula, constructed in 127 B.C.). The total cost was 180,000,000 sesterces or £1,800,000 sterling (Frontinus, de aquis i. 4, 7, 9, 12, 13, 14, 18, 19; ii. 67, 68, 72, 76, 81, 87, 89, 91-93, 125; Prop. iii. 2. 14; Strabo v. 3. 13. p. 240; Vitruv. viii. 3. 1; Tac. Ann. xiv. 22; Plin. NH cit. and xxxi. 41; Martial vi. 42. 18; ix. 18. 6; Stat. Silv. i. 3. 66; 5. 27 ; Not. app.; Pol. Silv. 545, 546; CIL vi. 1245-1251, 31559-31563; xiv. 4074-4078, 4081; Mon. Anc. iv. 11, 12).

Two arches of this aqueduct may be represented on a coin of C. Marcius Censorinus (circa 87 B.C.; BM Rep. i. 301. 2419), and five arches on coins of L. Marcius Philippus (ib. 485. 3890-5).

It was repaired by Agrippa in 33 B.C. and again by Augustus, with the rest of the aqueducts, between 11 and 4 B.C. (rivos aquarum omnium refecit, in the inscription (CIL vi. 1244) of the latter year on the monumental arch by which it was carried over the via Tiburtina, later incorporated in the Aurelian wall as part of the PORTA TIBURTINA (q.v.); see BC 1917, 207-215).

Numerous cippi belonging to this restoration (CIL vi. 1250, 1251 (= 31562); add 509 (unpublished) 803 (CIL vi. 31570 c) 1 and 816 (NS 1892, 152-EE ix. 966)) have been found, including the 1215th from Rome, which stood only about 3.5 kilometres from the springs, and about 86.6 from Rome. This corresponds fairly closely with Frontinus' measurement of 61,7101 paces (91.4 kilometres) : whereas the distance from the springs by road was 38 miles along the via Sublacensis, from which they lay 200 paces to the left, or 3 miles to the right of the thirty-sixth milestone of the via Valeria (Plin. and Strabo loc. cit. are in error). Augustus also added another spring, the aqua Augusta, which lay 800 paces further up (see AQUA CLAUDIA), which doubled the volume of the aqueduct (Mon. Anc. iv. 11, 12). The supply at the springs was 4690 quinariae, or 194,635 cubic metres in 24 hours.

The same group of springs are still in use for the modern water supply of Rome, and are now, as then, famous for coldness and purity; though, owing to the fact that the floor of the Anio valley has risen since Roman times, it is impossible to identify them exactly.

Nero outraged public opinion by bathing in its springs: but the aqueduct itself seems to have yielded but little to the city in his day, owing to the depredations of private persons (Frontinus cit.; Plin. NH xxxi. 42), and a further restoration was carried out by Titus in 79 A.D. (CIL vi. 1246): there is evidence of repairs by Hadrian; and others were probably made by Septimius Severus in 196 A.D. (CIL vi. 1247); while in 212-3 Caracalla cleared the springs, made some new tunnels, and added another spring, the fons Antoninianus, in connection no doubt with the construction of the branch to his thermae (ib. 1245).

The aqua Marcia was joined by the AQUA TEPULA (q.v.) and the AQUA IULIA (q.v.) before the point where it emerged from its underground course, near the sixth mile of the via Latina; and their channels were carried above it on the same arches, and are to be seen in section in the Aurelian wall, just to the right of the PORTA MAIOR (q.v.). From this point they have been made use of by it as far as the PORTA TIBURTINA (q.v.), soon after which they begin to run underground once more, and reach their terminal castellum just inside the porta Collina, at the north angle of the thermae Diocletiani.

For all this stretch there were cippi of Augustus bearing the names of the three aqueducts (CIL vi. 1249=3156 ; add No. 71, BC 1905, 289; CR 1905, 330, and No. 82, BC 1899, 39).

The regions served by the main channel of the aqua Marcia were in the neighbourhood of the castellum; numerous lead pipes were also found near the porta Viminalis, which served for its distribution (LF 10, 17, 24, etc.). It also ran to the Quirinal (Mart. ix. 18. 6; see DOMUS MARTIALIS).

The water was brought to the Capitol by Marcius himself in 140, 2 and where it issued forth must have stood his statue mentioned in the diploma of Nero of 64 A.D. (descrip. et. recognit. ex tabula aenea quae fixa est Romae in Capitolio post aedem Iovis O.M. in basi Q. Marci Regis pr(aetoris) CIL iii. p. 846; cf. Bull. d. Inst. 1845, 119; the last word makes it unlikely that CIL vi. 3825=31613 =i 3. 660, Q. Marcius Q. f. Rex Cos. is the base in question).

The rivus Herculaneus (not to be confused with the stream of the same name, cf. ANIO NOVUS) diverged from the aqua Marcia (Pliny (NH xxxi. 42) is entirely wrong in associating it with the aqua Virgo) post hortos Pallantianos; the castellum is incorporated in the Aurelian wall, in the fifth tower south of the porta S. Lorenzo (BC 1874, 53-55; LF 24) and ran across the Caelian, though at too low a level to supply it (Frontinus i. 19. Cf. RL 1888, 301; BC 1886, 406; 1888, 400; 1889, 130; 1914, 199; 1917, 242; NS 1888, 59; 1889, 66; 1917, 179; Mitt. 1889, 235, for traces of it and especially of a conduit formed of solid stone blocks with a circular orifice through them, which may have belonged to it), to its terminal castellum over the PORTA CAPENA (q.v.), which was therefore called madida (Juv. 3. 11 ; Mart. iii. 47; see also ARCUS STILLANS; HCh 287).

Hulsen attributes a cippus of Augustus, found near the Lateran, bearing the name Marcia and the number 3 (CIL vi. 31560) to this aqueduct: but there is another hypothesis admissible in regard to it. Before the construction of the aqua Claudia, so Frontinus tells us (ii. 76), the Caelian and Aventine were supplied by the Marcia and Iulia; and it is quite possible that the cippus, and both the ARCUS DOLABELLAE ET SILANI (q.v.) and the ARCUS LENTULI ET CRISPINI (q.v.) (CIL vi. 1384, 1.385 ; cf. p. 3125) belonged to this conduit. But afterwards these hills only received water from the Claudia by the ARCUS NERONIANI (q.v.) until Trajan took the Marcia amplo opere to the Aventine (Frontinus ii. 87).2 It appears as AQUA HERCULEA (q.v.) in Not. app. and Pol. Silv. 545. This conduit may be the forma of Eins. II. 4; 13.27 (Mon. L. i. 515). See LF 35, 41.

Another branch, the starting-point of which is uncertain, though it may have been near the third mile of the via Latina, was constructed by Caracalla to supply his thermae (q.v.). It crossed the via Appia by the so-called ARCO DI DRUSO (q.v.) and thence led to the great reservoir to the south-west of the thermae. It is mentioned as a separate aqueduct (aqua Antoniniana) in Not. app.; Pol. Silv. 545.

The supply was increased by Diocletian, from whom it took the name forma Iovia (Iobia, Iopia): cf. Eins. 11. 2; 13. 22: ibi (at the porta Appia) forma Iopia quae venit de Marsia, et currit usque ad ripam. The conduit was restored by Hadrian I (LPD i. 504), Sergius II (ii. 91) and Nicolas I (ii. 154, where it occurs under the form Iocia; Tocia, which some writers have referred to the aqua Appia, is a wrong reading; cf. Duchesne, ib. 167 n. 12). And the name forma Iovia is found in documents of the tenth century relating to the territory of Tivoli (Reg. Subl. p. 36 (973), 30 (998)), though the forma quae appellatur Iovia ... foris porta maiore, via Lavicana milliario ab urbe Roma p. m. IIII in loco quae dicitur IIII 2 (quarta), ib. p. 151, must be the aqua Alexandrina (LA 315-319; LF 46; Jord. ii. 228, 229; HJ 190, 195; PBS iv. 14, 15).

See Jord. i. 1. 468; LA 270-293, 298-318; LR 49-52; RM 1889, 235 ; HJ 220; Delbruck, Hellenist. Bauten in Latium, i. 1, 2; BC 1912, 228-234; NS 1913, 6-8; TF 137-139. Livellazione and Builder cited under Anio Novus; CIL vi. 8496 = 33729.

1 Identical with CIL vi. 1250 a; xiv. 4082.

2 Liv. Epit. Oxyrhync. 188-190: aqua Anio aqua [Marcia in Capi]tolium contra Sibyllae carmina [perductae]. It was probably a small supply, brought by a syphon; it certainly did not cross the depression between the Capitol and the Quirinal (see FORUM TRAIANI, p. 238) on arches, or something would have been said (or found) of them.

3 See p. 23. n. 1.

hide Dates (automatically extracted)
Sort dates alphabetically, as they appear on the page, by frequency
Click on a date to search for it in this document.
79 AD (1)
64 AD (1)
196 AD (1)
87 BC (1)
33 BC (1)
127 BC (1)
November (1)
April (1)
hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: