696 COSA. were settled there. (Id. zzxii. 2, xxziii. 24.) The chief importance of Goea waa derived from its port, known as the Portus Cosanns, which became a frequent point of departure for the Roman fleets and squadrons, from its ready communication with the islands of Ilva, Corsica, and Saniinia. (Lir. xxii. 1 1, zzx. 39.) It was from thence that Lepidns em- barked for Sardinia, when driven from Italj bj his colleague Gatuhis in b.c. 78. (Rutil. Jtin, i. 297.) It was in the neighbomhood of Cosa also that during the Olvil War of B.a 49, Domitius assembled a small force and a squadron, with which he proceeded to occupy Massilia. (Caes. B. C i. 34 ; OicoJ AU. ix. 6, 9.) The town of Cosa is not again mentioiied in history, but its name is found in all the geographers, and inscriptions prove it to have been still in exist- ence in the third century. Rutilius, however, speaks of it as in his time utterly desolate and lying in ruins, and relates a ridiculous legend as the cause of its abandonment (/tin. i. 285—290.) The city does not appear to have been ever again inhabited, and the origin of the name of AntedotUaf now given to its ruins, is uncertain. The remains of Cosa are of much interest, and present a very striking specimen of ancient fortifica- tions. Strabo correctly describes the city as stand- ing on a lofty height above the bay, at a short dis- tance from the 8ea(v. p^225). A steep ascent of above a mile lead* to the gates; and reroams of the ancient road are visible all the way. The waUs, which are preserved more or less perfectly, in their whole ex- tent, enclosed a rude quadrangle, hardly a mile in circuit, forming the level summit of the hill, which rises about 600 feet above the sea. They vary from 12 to 30 jfeet in height, and are composed of poly- gonal blocks of hard limestone, fitted together with great nicety: the upper course of the masonry pre- sentiog a marked approximation to a horizontal and r^ular style. They are moreover strengthened at intervals by square towers, projecting from the front of the walls, 14 of which are still standing or dis- tinctly to be traced, forming a continuous chain of towers round the W. and S. portions of the city. No other instance of this regular emplojrment of towers is known in the Etruscan cities, or tjie massive poly- gonal walls of so many cities in Latium: while it precisely resembles that adopted by the Romans at Falerii and Alba Fucensis. It therefore furnishes a strong argument for supposing that the walls now standing, were either erected, or at least in great measure rebuilt, when Cosa became a Roman colony. Dennis, however, from whom the above description is taken, strenuously mamtains their high antiquity and Pelasgic origin. (Dennis's Etruria^ vol. il pp. 269—289 ; Micali, Anticki PopoU Italiam, vol. i. p. 152, iii. p. 6.) The small extent of the space enclosed within the walls sufficiently proves that Cosa could never have been a veiy powwfnl dty. The Itineiaiy of Antoninus pbioes Cosa on the Via Aurelia, uid gives also another line of route passmg through Tarquinii to Cosa (Itkk AwL pp. 292, 300); but it is clear that the ^gh road could never have ascended the hill to the city itself: and the Tab. Pent, gives the name of Suocoea (Subcosa), which appears to have been a station or Mutatio at the foot of it. The port of Cosa, called by livy Portus Cosanus, is evidently the same which is termed by Strabo and Rutilius the Portus Herculis, and is still called Porto dErcole: it is on the opposite side of the bay from Cosa itself, under the shoulder of the Mons Argentarios, the whole of which remarkable COSSA. p romontory appears to have been included in tfai territory of CJosa. Henoe it is termed by Tadtna
- Cosa, a pramontary of Etruria" (^Afm. iL), where
he is oertunly spealong of the MonU Argmtaro. 2. A town of Lucania, mentioned by Caesar, who calls it ""Cosa in agro Thurino" (^.C. iiL 22), and relates that Milo laid si^ to it and was killed under its walls. Velleias, however, refers the same event to Compsa in the Hirpmi (iL 68), and Pliny speaks ci the death of HUo as oocozring '^ juxta castellnm Carissanum" (iL 56), for wbich SUlig would read Compsanmn. But tin reading in Caesar is well supported, and there is no xeason to reject it: the Cosa there mentianed would appear, however, to have been but an dbscniB place, a mere Castellum in tiie territory of Thnxii, and there is clearly no ground for supposing the Roman odony of B.a 273 to have been settled here instead of at Cosa in Etraria. It is not improbable that we should lead in Pliny ' Cossanum' or * Cassanum' fbr ' Carissa- num,' and that the name is still retained by the modem town of Cossano, near which is a place called Cttnto, where the nuns of an ancient tHj are said to be still visible. (Cluver. Ital. p. 1205; Ro- manelli, vol. L p. 288). Stephanua of Byzantium dtes from Heeataeus a city of Cossa (Kiavu)^ as existing in the interior of Oenotria, which may pro- bably be identical ^th the preceding. [E. H. B.] COSA, a town m Gallia, is plaod in the Tabk on a road from Tokaa (^Taulomae) to Divena (Co- hors). The distance of Cosa from Divona is marked 20 Gallic leagues ; which is too much, if the place is Coi or Coz^ — as it seems to be, — on the river Jvetron, which flows into the ram, a branch of the COSci'NU (r& KoffK'ma) or GO^SCINUS (Plin. V. 29), a place in Caria. Strabo (p. 650) speaks of Cosdnia and Orthosia as considerable places (m- roiicfm), by which he means, periiaps, something less than towns. In another passage (p. 587) he says that the river which flows fi^ Cosdnia to Alabanda has many forda, by which he seems to mean that a traveller must cross it many times. We may probably infer that Cosdnia was higher up the stream than Alabanda. Leake says {Ana Minor, p. 234), "if Alabanda was at Andnsfor, Tshma, where Pococke found conddecable remains, may be the site of Cosdnia, and its modern name may possibly be a corruption of the andeni." [G. L.] COSET^IA, a pboe in Gallia, in the conntiy of the Unelli. The Antonine Itin. places it oo a road from Alauna (ilfeottme) to Condate (iiemies). The Table gives a route from Coriallum {CkeAourff) to Condate through Cosedia, which is the next place to Coriallum. D'Anville discusses the site of Cosedia without detennlning its position, fbr there is great difficulty about the distances. Some geographen take Cosedia to be X*a Coutmiire f and there are other guesses. fQ. L.] COSETA'NI (Ko<nrravo(, PtoL iL 6. § 17; Plin. iii. 3. s. 4 ; Cositani, Inscc ap. Gniter, pw 499), a small people of Hispania Tarraoonensis, along Uie coast, from the mouth of the Iberus {Ebro') north- wards to the LALvrANL Their territory, called Cosetania or Cossetania, contained the capital dty I^RRAoo and the river Subub. [P. S.] COSSA (Kifirira or K»(ra, PtoL viL 1. § 65), a town in India, mentioned by Ptolemy as a place fiunous for the diamonds found tJiere. It has been conjectured by Forbiger to be the present CoOo, on the river ZesuL [V.]
