Grave with a machaira from Most na Soči
Last versions of single-edged swords with a bent hilt
Andrej GASPARI and Miha MLINAR
Izvleček
Članek obravnava žgan grob 25 z ledine Repelc na Mostu na Soči, v katerem je bil pokopan bojevnik z ukrivljenim eno-reznim mečem in pilumu podobno ostjo. V zasutju močno poškodovane grobne jame je bila najdena še keramična črepin-ja z glavničenim okrasom, na njegovi površini pa bronasta fibula srednjelatenske sheme, ki verjetno prav tako pripada grobni celoti. Meč sodi v skupino najmlajših izvedb železnodobnih mahair, znanih izključno z območja osrednje in zahodne Slovenije, od katerih dve izvirata iz Ljubljanice oz. Ljubljanskega barja, dve pa nimata sporočenih najdiščnih okoliščin. Dataci-ja mečev variante Ljubljanica ostaja odprta, ker sestava groba z Repelca ni povsem zanesljiva, ostalih mečev pa ni mogoče časovno opredeliti. Na podlagi podobnosti z inventarjem grobov latenske idrijske skupine, domnevanega tipološkega razvoja mahair, ki kaže tendenco k podaljševanju rezila ter drugih primerov uporabe halštatskih oblik orožja v mlajši železni dobi jugovzhodnoalpskega prostora se avtorja nagibata k opredelitvi groba v zadnji dve stoletji pr. n. št.
Ključne besede: Slovenija, Most na Soči, Repelc, grob, železna doba, svetolucijska skupina, idrijska skupina, orožje, mahaira
Abstract
The article discusses the cremation grave 25 from the Repelc fallow at Most na Soči, which contained the remains of a warrior equipped with a curved single-edged sword and a pilum-like weapon. The fill of the badly damaged grave pit yielded a pottery fragment with combed decoration, while a bronze fibula of the Middle La Tene scheme was found on the grave's surface and probably also formed part of the grave unit. The sword belongs to the group of the latest versions of Iron Age machairas, known exclusively from central and western Slovenia. Two of these were found in the Ljubljanica River, at the Ljubljana Moor, and two are of unknown provenance. The composition of the Repelc grave not being entirely reliable and other swords escaping chronological determination, both result in the date of the Ljubljanica type swords remaining open. Based on the similarity with the grave contents of the La Tene Idrija group, the presumed typological development of the machairas with a tendency towards prolongation of the blade, and other instances of the Hallstatt weapon forms being used during the Late Iron Age in the south-eastern Alpine area, the authors are inclined towards dating the grave in the last two centuries BC.
Keywords: Slovenia, Most na Soči, Repelc, grave, Iron Age, Sveta Lucija group, Idrija group, weapons, machaira
INTRODUCTION
The team of the Tolmin Museum conducted rescue archaeological excavation on the western edge of the Repelc fallow, on the left bank of the Idrijca River in Most na Soči in 2000. The excavation yielded 30 graves dating from the Hallstatt, La Tene, and Roman periods.1 The work at Repelc, a previ-
ously virtually unresearched area,2 continued in 2002 when a further 35 graves were documented at the easternmost area of the first terrace of the Idrijca. Approximately two thirds of the cemetery remain unresearched.
The La Tene and Roman graves represent an exception within the Repelc necropolis, otherwise predominantly Hallstatt in date. Repelc may be
1	The rescue excavations preceding the construction of the purification plant and a new sewage system were conducted by Miha Mlinar under the guidance of Drago Svoljšak.
2	Gabrovec, Svoljšak 1983, 33. Construction of a private house at Repelc (Most na Soči 70a) revealed prehistoric cremation graves, the material from which is not preserved. Some covering slabs were preserved by eng. Milan Mikuž from Most na Soči 61 (D. Svoljšak, Excavation journal 1973).
Fig. 1: Most na Soči-Repelc. Ground-plan of a part of the cemetery (excavations in 2000) (drawing A. Hawlina). Sl. 1: Most na Soči-Repelc. Tloris dela grobišča (izkopavanja v letu 2000) (risba A. Hawlina).
considered a specific element within the complex of the Sveta Lucija cemetery, with the burial area that terminates at the eastern edge of the terrace. East of the mentioned graves, the 2002 investigation also uncovered traces of an offering place with highly fragmented metal and pottery material dating to the Late La Tene and Early Roman periods.3
The investigated graves were in a poor state of preservation. This is a consequence of the terrace being used as a cultivated field until recently, while the First World War with its activities behind the lines of the Soča Front also left strong traces. The total number of sixty-five graves therefore included only few that were undisturbed, mostly those that were cut deeper in the sterile clay or even limestone bedrock. The extent of damage can be discerned also from the stone grave slabs, found mostly overturned and scattered across the terrain, as well as from the damage to the top sections of grave pits.
Among the later graves, the male grave 25 stood out already during excavation. It was presented at
the exhibition at the Tolmin Museum and published in the accompanying catalogue, together with other important grave units.4 The grave formed part of a group of four graves (graves 18, 23, 24, and 27), that partially overlapped in ground-plan (Fig. 1) though their pits were clearly discernible in section (Fig. 2).
The oval pit of grave 25 was cut into a layer of brown-grey clayish silt (Fig. 3). It measured 115 cm along the longer axis, preserved in the depth of 15 to 20 cm. Its western edge could not be reliably determined. In the centre, at the bottom of the pit, burnt remains were found together with very small unburnt bone fragments and charcoal,5 surrounded at the southern and eastern edges by the spearhead and the curved iron sword. Of the latter, the upper third of its blade was broken off and laid underneath the sword. The fill of the grave pit was composed of grey-black clayish silt that yielded a fragment of a hand-made vessel with combed decoration. A heavily deformed bronze La Tene fibu-
3	The material is currently under research by Miha Mlinar for his master's thesis under the mentorship of Dr. Mitja Guštin.
4	Mlinar 2002, 26, fig. 18, cat. no. 30.
5	Radiocarbon analysis of the wood charcoal, conducted at the Leibniz Labor für Altersbestimmung und Isotopenforschung Christian-Albrechts-Universität in Kiel (KIA24524) at the end of 2003, gave the conventional age of BP 2501 ± 20 (2 sigma calibrated result: cal BC 785-522).
Fig. 2: Most na Soči-Repelc. Section on the area of grave 25 (drawing Ana Hawlina). Sl. 2: Most na Soči-Repelc. Profil na območju groba 25 (risba Ana Hawlina).
la was found on the surface of the fill.
Though the appurtenance of the fibula to the grave unit was not clear upon discovery, this is nevertheless confirmed by the following observations. Underneath and beside it lay the burnt remains with human bones and bits of charcoal that
continued uninterrupted to the level of the weapons. It was ritually bent similarly to the iron weapons. Furthermore, the similarity in objects distribution with the Late La Tene grave 14 containing agricultural implements has to be underlined.6 There the pieces of jewellery lay among the burnt remains
Fig. 3: Most na Soči-Repelc. Ground-plan of grave 25 (drawing Ana Hawlina). Sl. 3: Most na Soči-Repelc. Tloris groba 25 (risba Ana Hawlina).
6 Ib., 26, fig. 19, cat. no. 31.
Fig. 4: Most na Soči-Repelc. Objects from grave 25. 1,2 iron; 3 bronze; 4 pottery. Scale 1,2 = 1:4; 3,4 = 1:2 (drawing N. Grum). Sl. 4: Most na Soči-Repelc. Predmeti iz groba 25. 1,2 železo; 3 bron; 4 keramika. M. 1,2 = 1:4; 3,4 = 1:2 (risba N. Grum).
in the centre of the grave pit while the intentionally damaged iron tools were laid at the perimeter!
DESCRIPTION OF GRAVE'S CONTENTS
1. Bent single-edged sword with a narrow blade and a bent hilt plate. The gently curved blade with a wedge-shaped cross-
section is slightly wider in the middle. A wide guard is forged at the transition between the blade and the hilt. Rec. l. 69.2 cm; tang l. 11.5 cm; max. blade width 3.2 cm; inv. no. TM 101 (Fig. 4: 1).
2. Bent iron spearhead with a short deltoid blade with a ridge, rhombic in cross-section, and a long socket. The edges and point of the blade as well as end of the socket are damaged. Pres. l. 35 cm, rec. l. 37 cm; pres. blade w. 3.5 cm; socket l. 22 cm; max. socket diameter 1.9 cm; inv. no. TM 102 (Fig. 4: 2).
3.	Deformed bronze fibula of a Middle La Tčne scheme with a solid semicircular bow that widens towards the head, a spring with six coils, and a recurved foot. The foot was attached to the bow with a band clip. The upper part of the foot bears a ribbed knob, surrounded on both sides by a thickening with smaller ribs in the interspace. The catch-plate is decorated with incisions. The foot of the fibula is bent sideways; the spring coil is also damaged. Rec. l. 10.4 cm; bow l. 4 cm; h. 3.1 cm; inv. no. TM 103 (Fig. 4: 3).
4.	Wall fragment of a hand-made clay pot. Exterior and interior surfaces are brick red to grey in colour and bear combed decoration. Fragment size 5.7 x 4 cm; wall th. 0.6-0.8 cm; inv. no. TM 104 (Fig. 4: 4).
TYPOLOGICAL AND CHRONOLOGICAL DETERMINATION
The chronologically most tangible object of the presented unit is the bronze fibula of the Middle La Tene scheme with a six coil spring, solid bow, and one large barrel-shaped and two smaller thickenings on the foot (Fig. 4: 3). The only known analogy, though it has a longer spring with thirteen coils, is a grave find from Vinica (gr. 86/195). The two fibulae are formally close to later examples of the Kastav type/Idrija variant from graves 7/8, 9, 3, and 16 from Idrija pri Bači. These have two spherical thickenings and incised decoration of parallel lines on the termination of the foot.7 Grave units show these to be the leading forms of the Middle La Tene period that remained in use throughout the 1st century BC.8 The decoration is reminiscent of fibulae with three knobs on the bow and a round plate on the foot extension. These are mainly concentrated in the Posočje area,9 but appear also at Este, Aquile-ia, Jereka in Bohinj, Kranj, and Vinica.10 Characteristic of the Middle and the early part of the Late La Tene periods, these fibulae also usually have a spring with three coils that is rarely wrapped around the foot. Common to all three types, however, is the catch-plate incised decoration.
The spearhead with a short blade with a ridge (Fig. 4: 2) is a relatively rare form and is, in the
ratio between the length of the blade and length of the socket (almost two thirds of total length) close to pilum-like spears from the central and eastern Alpine region.11 The heads measure between 35 and 65 cm and appear at the end of the Early La Tene period. They have biconvex or bayonet-shaped blades with ridges, common also on the contemporary Celtic examples with a shorter socket.
Two short and two long examples of pilum-like spears with bayonet blades were found within the group of ritually damaged weapons on a hill near Borče (Förker Laas Riegel) in the Zilja (Gail) Valley,12 dated to around 300 BC. At least four such heads were found at the cemeteries near Šmihel, two of which with preserved grave units.13 An approximately 46 cm long head with a classically shaped blade was found in grave 99 from the Za Polšno, which also yielded a bronze knob with an incised motif of a spiral.14 M. Guštin ascribed the latter to a Certosa fibula. It is more likely, though, that it is the knob on the foot of a fibula of the Early or Middle La Tene scheme.15
Grave 127 from Šmihel revealed a 54.4 cm long head with a bayonet blade and a shaft of square cross-section, found together with a nail that probably served to fasten a shield boss of La Tene scheme, but also with a spear butt, a bronze torque with spirally curved terminations, fragments of a solid bronze bracelet, and glass beads.16 The torque is of one of the leading forms of the Late Hallstatt Notranjska VI phase. However, based on the example with a spectacle pendant worn together with the Kastav type fibula by the deceased from the eastern abyss of the Gorenja jama near Povir, it allows for the assumption that it was still in use at least in the Middle La Tene period.17
Pilum-like heads with a narrow blade of rhombic cross-section belong to the Middle or the Late La Tene period, appearing mostly in the Alpine valleys18 and the western Celtic area.19 Another head of this type, over 55 cm long, intentionally
7	Guštin 1991, 37, pl. 9: 10; 10: 9; 12: 1; 14: 9,13.
8	Ib., 37.
9	Ib., pl. 3: 15; 6: 6; 12: 2; 13: 1; 21: 13; 35: 1: 4; 40: 4; 41: 6.
10	Ib., 37.
n Schaaff 1990, fig. 11.
12	Ib., 20, fig. 10.
13	The upper part of the head with the bayonet blade and the socket with square cross-section (Guštin 1979, pl. 80: 26) as well as a 45.5 cm long example with a biconvex blade (Hoernes 1888, 234, pl. 5: 8) are without provenance.
14	Guštin 1979, 42, pl. 51: 1,2.
15	A similarly shaped knob with a relief spiral decoration can be found on a foot of a bronze fibulae from Vinica (Gabrovec 1966b, pl. 15: 8) and Karaburma (gr. 14; Popovic 1996, fig. 12: 14).
16	Guštin 1979, 44, pl. 57: 1-6.
17	Guštin 1973, 479; Crismani, Righi 2002, 93, fig. 52.
18	Nothdurfter 1979, 79-89, pl. 70, 1192-1195,1198-1200; Lang 1998, 126, pl. 26: 498-500,502,503.
" Sievers 2001, 163-164, pl. 69: 359; Wyss, Rey, Müller 2002, pl. 45: 145-147; 46: 145-147.
bent and broken in two places, was found at Re-pelc as a surface find.20
A later date for the spear-head from grave 25 could further be indicated by the deltoid shape of the blade with its broadest part in the lower third, which is frequent in the contexts from the Late La Tene and Early Roman periods. At that time, examples with a ridge of rhombic or round cross-section are widely spread,21 with sockets measuring approximately a third of the total length, comparably to similar heads with blades of flattened rhombic cross-section.22 Heads with a deltoid blade are characteristic of Herzegovinian and Dalmatian sites from the second half of the 1st century BC, known in the literature as typical Dalmatian weapons.23 This type is also depicted on the tropaeium from the legionary fortress Tilurium/Gardun.24
Single-edged sword with a bent hilt
The sword with a bent hilt from grave 25 (Fig. 4: 1), for which the old Greek term ^axaip« -machaira is used in Slovene literature,25 belongs to the group of curved single-edged weapons used by the peoples on the western and northern coasts of the Mediterranean throughout the Iron Age. Machairas were supposed to be very effective weapons, appreciated for their balanced and manageable character, important particularly in mounted battle. They were doubtlessly used as a sabre, contrary to falcatas, the other typical form of a single-edged sword, known mostly as the weapon of Iberian tribes. The approximately 60 cm long falcata represented a weapon for thrusting and cutting, similarly to the Roman gladius.26
The use of large knives with curved blades is a tradition that reaches back into the Late Bronze Age in the eastern Adriatic coast and its hinter-
land. Iapodic and Liburnian groups of this time were familiar with virtually all the basic forms of machaira and falcata swords.27 They can be traced in the Balkans area throughout the Iron Age. The outline of the typological and chronological development of the iron machaira swords is based on the basic study of the finds in the eastern alpine region and the Balkans, published in 1974 by Mitja Guštin.
The machaira swords of the Basarabi variant, characterized by a triangular blade and a transverse T-shaped iron plate at the termination of the flang hilt, appear in the 8th century BC in the area of the Basarabi Culture in Oltenia.28 To the west of it, they are known from Ilijak on Glasinac (IVb phase), Legen near Slovenj Gradec (Podzemelj 1), and Breg (Frög) near Rožek.29 Close to the Basarabi variant is an approximately 50 cm long sword with a solid bronze hilt and preserved scabbard. It was found in a rich grave 16/I at Kapiteljska njiva in Novo mesto which represents a characteristic grave unit of the Podzemelj 2 phase.30
Swords of the Tržišče-Donja Dolina variant differ from the Basarabi variant for the most part only in the wider blade with a straight or only gently curved back, distinctly convex edge of the blade, and a simpler flanged or tanged hilt. They are spread in the area between the eastern Alps and Bosnian Posavina.31 The swords of this group show a considerable variety, as indicated by metric and formal characteristics of nine machairas from the Tržišče hoard, which measure from 35 to 65 cm in length and 3.3 to 5.1 in blade width.32 They have a relatively sharp angle between the blade and the hilt (ca. 135-143°) as well as a flat-forged back with a T-shaped cross-section, only exceptionally encountered in later variants. Almost all swords have preserved rivet holes or rivets for fastening the grip. Related to the Tržišče-Donja Dolina variant are
20	The unpublished find is held at the Tolmin Museum; inv. no. TM 28.
21	Idrija pri Bači, grave 5 (Guštin 1991, 15, pl. 5: 1).
22	Polhov Gradec (Ložar 1938, 95, fig. 22 a,b), Štalenska gora/Magdalensberg (Dolenz 1998, 65-66, pl. 6: M 41).
23	Škegro 1992, 91, pl. 2: 13; 3: 14.
24	Abramic 1937, fig. 2; 3.
25	The use of the terms machaira and kopis, most frequently mentioned in connection with curved single-edged swords, is questionable. The term machaira first indicated ritual knives for slaughtering sacrificial animals and surgical tools, from the 4'h century BC onwards, it was occasionally used also as a common name for swords. Xenophon's treatise on cavalry (Xen., De re Eq. XII, 11) is the only source that clearly distinguishes between the terms machaira or single-edged sabre and xiphous, signifying a short straight sword (see Quesada Sainz 1997, 171).
26	Quesada Sanz 1997, 169-171.
27	Batovic 1983, 314, fig. 20: 4,5; pl. 46: 6-8; Drechsler-Bižic 1983, 383-384, pl. 53: 9.
28	Guštin 1974, 81-83, fig. 4; 5; Tomedi 2002, 112-113.
29	Guštin 1974, 78-79, 81-83, fig. 2; 6: 3; Tomedi 2002, 112-113.
30	Knez 1993, 23-24, pl. 19.
31	Guštin 1974, 80, list.
32	Guštin 1979, pl. 31: 1-4; 32: 1-5.
Fig. 5: 1 Donja Dolina, gr. 126 (from Fiala 1899, fig. 140); 2 Prozor (from Drechsler-Bižic 1956, pl. 1: 2); 3 Šmihel-Za Polšno, gr. 135 (from Guštin 1979, pl. 56: 2); 4 Šmihel-Za Polšno (from Guštin 1979, pl. 85: 5). Scale = 1:5.
Sl. 5: 1 Donja Dolina, gr. 126 (po Fiali 1899, sl. 140), 2 Prozor (po Drechsler-Bižic 1956, t. 1: 2), 3 Šmihel-Za Polšno, gr. 135 (po Guštinu 1979, t. 56: 2); 4 Šmihel-Za Polšno (po Guštinu 1979, t. 85: 5). M. = 1:5.
the machairas of the Novilara variant from Pice-num, dated to the first half of the 7th century (the Novilara Ilb-IIIa phase),33 and from Istria, attributed to phase III by K. Mihovilic.34 Most of the 10 grave finds of the Tržišče-Donja Dolina variant from the eastern Alpine region as well as the Tržišče hoard belong to the Ha C2 horizon,35 while
the machairas from Donja Dolina and Sanski Most date to the 6th century BC (phases 2b and 2c).36 In the later Iron Age phases, the use of swords with a bent hilt remains characteristic for the tribes of the north Adriatic hinterland. In central parts of the Balkans, on the other hand, other forms of curved single-edged weapons begin to appear.37
33	Examples from graves 30, 64, 65 at the Novilara-Servici necropolis (Beinhauer 1985, pl. 77: 897; 127: 1401,1409) have a flanged hilt, the excellently preserved example with scabbard from grave 14 has a tang (ib., pl. 63: 751). The origin of the south Italy-Picenum variant of short machairas with hilt plate that bear three rivets in triangular position, can most probably be sought already in the 8th century (Tomedi 2002, 113; Beinhauer 1985, pl. 66: 781; 68: 806; 86: 973). In the eastern subalpine area these characteristics can be observed on the sword from Črnomelj (Gabrovec 1992, 250, fig. 2: 18).
34	Mihovilic 1991, pl. 1: 8; 3: 7-10; ead. 1995, 284-285, fig. 14; pl. 20: 1-3; Moretti 1983, 155, pl. 43: 5.
35	Guštin 1974, 87.
36	Čovic 1987, 246, 250-251, fig. 15: 3,20.
37	One group is constituted by 30 to 42 cm long knives from sites in western Bosnia and in Serbia, dated to the end of the 5th and the 4th centuries BC. They are characterized by a gently curved back, symmetrically convex edge, and hilt plate with three
A transitional form to the latest variants of machairas is represented by two swords from the Za Polšno necropolis at Šmihel, dated to the Late Hallstatt period. One was found in grave 135 (Fig. 5: 3), which also contained a shaft-hole axe and two spearheads with classically formed blades,38 while the other find is of unknown grave unit (Fig. 5: 4).39 The two swords of the Šmihel variant measure slightly over 50 cm in length and have a stout, up to 3.6 cm wide blade with an almost straight back, distinctly convex central part of the edge, a thinned point, and a convex transition from the back to the flanged hilt.40 The sword from the unknown grave unit had a preserved short chape made of thin sheet metal and a scabbard's mount with a semicircular suspension loop: it differs from the other sword in a flat-forged T-shaped back.41 Similar in form is the machaira from the Iapodian necropolis at Prozor near Otočac in Lika, somewhat wider and also of unknown provenance (Fig. 5: 2).42 Shorter than all of the above is a 41.2 cm long example with a missing hilt tang and a damaged blade point from grave 10 at Jelenšek above Godovič, a necropolis of the Sveta Lucija group, where burial took place exclusively during the IIc
phase.43
Machairas of the Ljubljanica variant, named after the site where the sword with a decorated hilt had been found, are known also from Repelc, another one from Šmihel, one from an unknown site "in Carniola", and two from the Ljubljana Moor. They differ from the previously enumerated forms in a longer blade, measuring between 65 and 80 cm, but share the ration of particular parts as well as the direction of the blade, i.e. the angle between
the back and the hilt (Fig. 6). They are marked by a thin curved blade of a wedge-shaped cross-section as well as a sharp transition into the hilt with a flanged guard. The edge is more or less parallel to the back, being widest at the middle of the blade. The rounded part of the iron guard part is turned to the edge while the guard itself had an organic or metal covering, fitted vertically from the hilt side. The same goes for the termination of the hilt plate, while the grip, rectangular in cross-section, was made of organic material or a set of discs.
Geographically closest comparison for the Repelc machaira originates from the Za Polšno cemetery near Šmihel (Fig. 6: 1) where A. Burger excavated 96 graves in the years 1885/1886 under the leadership of either J. Szombathy or M. Hoernes.44 The sword with the damaged point is of unpre-served grave unit. Its date, though, is indicated by the existence of both Late Hallstatt as well as Middle La Tene burials at the necropolis. Graves from the Late La Tene and Early Roman periods, which could be expected on the basis of the material from the settlement, are not known.
A well preserved, 79 cm long sword from an unknown site "in Carniola" (Fig. 6: 2) is held at the Peabody Museum at Harvard University and came to the United States of America together with a part of the heritage left by the Duchess of Mecklenburg.45 The Treasures of Carniola auction catalogue mentions it as a typical sword of the Illyrian tribes but states neither the exact provenance nor the way in which it was acquired.46
The "unusually formed iron sabre" (Fig. 6: 4) was discovered while scooping the water on the V Blekah fallow at the Ljubljana Moor, between the
to five rivets. Later examples within the above-mentioned time frame represent examples of the Sanski Most type with a groove under the back of the blade and a saddle-shaped guard (Parovic-Pešikan 1982, 35-39, pl. 3). Čovic sees them as characteristic of phase Illb (350-300/275 BC; Čovic 1987, 258, 262, fig. 16: 25). The second group is constituted by falcatas, i.e. swords with an evenly curved back, an S-shaped edge with the centre of gravity in the upper third, and a specifically formed hilt in the form of a horse's or bird's head. They appear in the Balkans in the area between Bosnian Posavina, Pomoravje, and Macedonia, where they are dated between the 6th and the 3rd centuries BC (Parovic-Pešikan 1982, 25-35, pl. 2: 1-3; 5: 1-3; see also the recent find from the vicinity of Prilep - Staro oružje, 11, cat. no. 103). Related to these are swords with a straight back and a slightly convex edge (Parovic-Pešikan 1982, pl. 1: 1,2).
38	Hoernes 1888, 230, pl. 4: 4; Guštin 1974, 81, fig. 9: 1; id. 1979, 44, pl. 56: 2.
39	Hoernes 1888, 230, pl. 4: 5; Guštin 1979, 48, pl. 85: 5.
40	These characteristics can already be seen in the 54.2 cm long sword with missing hilt from grave 126 from Sanski Most (Fig. 5: 1), dated by Čovic into the Donja Dolina-Sanski Most II c phase, i.e. to the end of the 6th century. The grave also contained two laurel-leaf spearheads, part of a bowl, and a pot (Fiala 1899, 39, fig. 140; Čovic 1987, 251-252, fig. 15: 21).
41	The manner of attaching the scabbard to the belt directly at the saddle-shaped transition of the guard to the edge is clearly visible on the machairas from Novo mesto (Knez 1993, 23-24, pl. 19) and Cvetež near Vače, where two ringlets hang from loops on a figurally decorated bronze sheathing (Stare 1955, 14, pl. 1).
42	Ljubic 1889, 147, pl. 29: 215; Drechsler-Bižic 1956, pl. 1: 2.
43	Bratina 1997; for the received information we thank the excavators Patricija Bratina (Institute for the Protection of Cultural Heritage of Slovenia, Regional Office Nova Gorica) and Drago Svoljšak (National Museum of Slovenia).
44	Hoernes 1888, 230, pl. 4: 6; Guštin 1974, fig. 9: 6; id. 1979, 48, pl. 85: 1.
45	Guštin 1974, 81; we thank Dr Mitja Guštin for the permission to publish.
46	Treasures of Carniola, 81.
Fig. 6: 1 Šmihel (after Guštin 1979, pl. 85: 1), 2 "Carniola", 3 Most na Soči-Repelc/gr. 25, 4 the Ljubljana Moor-V Blekah, 5 River Ljubljanica near the Zrnica outflow. Scale = 1:5.
Sl. 6: 1 Šmihel (po Guštinu 1979, t. 85: 1); 2 "Kranjska"; 3 Most na Soči-Repelc, gr. 25; 4 Ljubljansko barje-V Blekah; 5 Ljubljanica pri izlivu Zrnice. M. = 1:5.
outflow of the Iščica stream and the Preproščica ditch, around 1500 m south of the Špica in Ljubljana. It was given over to the Provincial Museum in Ljubljana in 1846 by landowner Anton
Breskvar. The well preserved sword with an almond-shaped pommel, made of bronze alloy with a roughly worked surface, was lying on the surface of the clay and beneath the peat.47 Based on Hoernes's
47 National Museum of Slovenia - inv. no. B5051; "Verzeichniß der im Jahre 1946 dem Museum in Laibach verehrten Geschenke. - Illyrisches Blatt 1847, 188, no. 38: "Vom Herrn Anton Einsiedler Breskquar, Realitäten-Besitzer in Laibach: - ein alter, eigenthüm-lich geformter Eisensäbel, stark vom Roste zersressen, mit einem messingenen Griffe, im Moorgrunde am Volar zwischen dem Isch-zaflusse und dem Perproshza-Graben v' Blekah unter dem Torfe am Lehmboden beim Wasserschöpfen aufgefunden."; Deschmann 1888, 142; Müllner 1894, 149, pl. 12: 18; id. 1899, 151-152, fig. 3; Stare 1952-1953, 72, fig. 1; Gabrovec 1966, 251, pl. 8: 1; we thank the Archaeological Department of the National Museum of Slovenia for the drawings made and for the permission to publish.
writing on the swords from Šmihel, A. Müllner ascribed the "yatagan" to Hallstatt forms and correlated its provenance to the stratigraphic position of the cargo vessel from the near-by Lipe.48
A recent addition to the group of machairas treated above is a sword with a decorated hilt, found in the bed of the Ljubljanica at the outflow of the Zrnica stream, southeast of Blatna Brezovica (Fig. 6: 5). It was found by diver Miro Potočnik in 1992. The data on the direct find circumstances of the sword are limited to the fact that it was discovered atop of clay, in a layer of fine-grained sand at the left bank. It is nevertheless necessary to stress that the section of the bed and bank between the outflow of the Bistra and Kamin, particularly the area of the outflow of the Zrnica stream itself, is an archaeological site of extreme interest. Particularly outstanding documented finds from water are a group of Middle and Late Bronze Age metal objects,49 a Late Hallstatt ribbed bracelet and an iron collared axe50 as well as finds from the Late La Tene period. They all indicate the existence of a ritual place of a long duration. Discovered only several ten metres apart were the Posočje type fibula, the bronze jug of Gallarate type,51 and two bronze votive statuettes.52 The approximately 50 m long section of the stream also yielded a well-known group find of Celtic and Republican silver coins, dated soon after 147 BC. P. Kos and A. Šemrov suppose it to represent accidentally lost money along an important communication route.53
The sword's description is given on the basis of a photo, which served in the making of the drawing. The sword is preserved in the length of 72 cm with the upper third of the blade broken off. Strongly bent, the 16 cm long hilt tang of rectangular cross-section has a wide tang at the transition to the blade with a guard made of bronze alloy attached
to it. The guard of this example is, contrary to other machairas of the same variant, straight towards the blade and roundly curved towards the hilt. The guard is fastened to the tang with seven rivets of a circular cross-section, four of which are placed at the front edge of the guard, two at the centre, and one towards the hilt. The hilt terminates in a mushroom-shaped pommel with three semicircular prongs, the side ones being higher and the central one lower. Underneath the prongs, four decorative rivets are visible. The prongs were fastened with an additional rivet at the beginning of the pommel. Three relatively wide oval metal ringlets with obliquely incised edges are fitted in equidistance between the pommel and the guard. These indicate that the grip was originally probably composed of alternating metal and organic elements. Each ringlet was fastened to the hilt tang with two rivets. Such a composition of the grip is known already on the swords from the Late Bronze Age,54 on Hallstatt short swords and daggers from the Ha C2 and Ha D phases,55 and also on swords of the Middle56 and the Late La Tene scheme.57 Of further interest in this respect is a short single-edged sword with knee-shaped back and double curved edge from an Early Imperial cemetery at Starigrad (Argyruntum). It has a band bronze guard positioned on the hilt tang of circular cross-section, followed by a decorated cylindrical mount, two circular plates, and a solid bronze knob of a biconic shape, originally probably placed at the edge of the tang.58 Analogies for the pommel of the machaira from the outflow of the Zrnica stream are not known, except for the simpler version on another example from the Ljubljana Moor. The guard, on the other hand, has a good analogy in the sword from Gorenje Vrhpolje near Šentjernej in Dolenjska.59 A sword with a wedge-shaped cross-section and maximum width of 3.6 cm, without
48	Müllner 1894, 149.
49	Gaspari 2004.
50	Held at the National Museum of Slovenia; unpublished.
51	Potočnik 1987, fig. 4; Breščak 1995, fig. 4: 20.
52	Istenič 2003; Gaspari, Krempuš 2003.
53	Kos, Šemrov 2003.
54	Reinecke 1965, pl. 2: 642.
55	See iron and bronze discs on short swords with "antenna" handle and daggers of the Hallstatt type (Sievers 1982, 18-24, pl. 5-11).
56	The hilt tangs usually have two or three circular plates with incised, usually slightly thinned edge (Keltoi, 78; Lejars 1994, 48).
57	The sword from grave 225 from Povegliano near Verona, for example, belongs to the Lt D1 phase. It has eight thin bronze plates as well as two bone ringlets preserved on the hilt (Restituzioni 1993, 11, 15). Thickly fitted round or oval metal plates are typical also for swords hilts from northeastern Gallia and the Germanic area, dated to the later part of the Lt D2 phase (Verwers, Ypey 1975; Lejars 1996, 90).
58	Abramic, Colnago 1909, 110, fig. 82; Stipčevic 1960, 89, fig. 1: 3.
59	Guštin 1974, 81, fig. 3; Knez 1993, 24; Dular 2003, 178, pl. 13: 5. The authors thank Dr. Dragan Božič for the information on the provenance of the sword.
exact provenience data, is attributed to the Tržišče-Donja Dolina variant by M. Guštin. Its blade is broken off while the remainder bears traces of wooden panels of a scabbard. The bronze hilt tang of rectangular cross-section is fitted onto the extension of the blade and the stability of the junction is provided by a bronze guard with roughly finished edges. The thinned termination of the tang was originally probably mounted with a metal bar.
CONCLUSION
The sword from grave 25 from Repelc is a long machaira of the Ljubljanica variant. Previous researchers supposed a La Tene date for it, based on the typological development or the tendency towards prolongation of the blade.60 The date of the Ljubljanica variant swords therefore remains open, since the composition of the grave is not entirely reliable. The head with a deltoid blade and a long socket, lying next to the machaira, can be dated approximately between the end of the 4th
and the 1st century BC. The fragment of the vessel with combed decoration speaks for the last two centuries BC or the Early Roman period. This date is valid also for the deformed bronze fibula of a Middle La Tene scheme found atop of the grave fill. Radiocarbon dating of the charcoal from the bottom of the pit, on the other hand, indicates an Early Iron Age date of the grave.
The supposition of a La Tene date for grave 25 from Repelc, when the Posočje warriors used weapons and other military equipment of the Celtic type,61 is supported by the Late Hallstatt forms being accepted into La Tene armour, a fact treated several times already. The Late La Tene grave 1 from Idrija pri Bači contained, beside parts of attire, a spearhead, and predominantly agricultural implements, also two large knives.62 One of them is formally reminiscent of a machaira, since it has a curved narrow blade and bent hilt tang, as noted already by M. Hoernes.63
The so-called "Hallstatt renaissance" is attested to in Posočje and Notranjska also by the Idrija variant of the Negova helmets from the 1st century BC64
Site / Najdišče	Sword length / Dolžina meča (cm)	Blade length / Dolžina rezila (cm)	Blade width / Širina rezila (cm)	Hilt length / Dolžina ročaja (cm)	Tang width / Širina trna (cm)	Angle between blade's back and the hilt tang / Kot med hrbtom rezila in ročajnim trnom
Šmihel-Za Polšno	50,4+	38,7	3,2	9,6 cm	3,9+	ca. 130°
Šmihel-Za Polšno, gr. 135	53,0+	40,0+	3,5	10,0+	4,0+	ca. 140°
Šmihel-Za Polšno	70,5+	58,5+	2,7	12,0+	4,4+	ca. 137°
"Kranjska" / "Carniola"	78,0	63,2	3,1	11,0+	5,0	ca. 141°
Ljubljansko barje- V Blekah	75,6	64,4	3,2	13,9	4,8	ca. 134°
Ljubljanica- izliv Zrnice / the Zrnica outflow	82,0+	72,0+	3,0	10,0	5,1	ca. 141°
Most na Soči- Repelc, gr. 25	69,2	56,7	3,2	11,5	5,3	ca. 139°
Fig. 7: Metric data for the Late Hallstatt and La Tene machaira swords. Sl. 7:. Metrični podatki poznohalštatskih in latenskih mahair.
60	Gabrovec 1966a, 255; Guštin 1974, 81, note 16.
61	The unpublished finds include fragments of shield-bosses, scabbaros and swords from Repelc (the Tolmin Museum), a spearhead with a bayonet blade from Žaga near Bovec (unpubl., held at the Museum of Gorenjska, Kranj; we thank Mija Ogrin for information), and part of a scabbard of the Late La Tene scheme from Grad near Reka (unpublished; information provided by Dr. Dragan Božič).
62	Guštin 1991, 13-14, pl. 1: 9.
63	Hoernes 1888, 232-233.
and by pilum-like spearheads, in the area of the Mokronog group with shaft-hole axes65 and a bronze helmet of the Filottrano type from Bela Cerkev.66 The use of single-edged swords or long knives is characteristic of Vinica and the Iapodian area,67 at the end of the La Tene period also for the Mokronog group otherwise considered as entirely Celti-cized, at least as far as the weaponry is concerned. Early Roman variants of the curved knives of the "Iapodian type",68 that represent a derivation of the single-edged swords of the Sanski Most type,
reach to the west as far as the Ljubljana Moor, where they are attested to with at least 5 finds from the Ljubljanica. The existence of tight connections of the population of the southern Ljubljana basin, thought to be the western edge of the area populated by the Taurisci, with their non-Celtic neighbours in the hinterland of Caput Adriae, is confirmed also by a considerable number of the latest variants of Cer-tosa (type XIIb and the Notranjska variant of type VIIf) as well as Picugi and Kastav fibulae types, found in the bed of the Ljubljanica.69
ABRAMIC, M. and A. COLNAGO 1909, Untersuchungen in Norddalmatien. - Jh. Österr. Arch. Inst. 12, 12-113.
ABRAMIC, M. 1937, O predstavama Ilira na nekim antiknim spomenicima. - Čas. zgod. narod. 32, 7-19.
BATOVIC, Š. 1983, Kasno brončano doba na istočnom Jadranskom primorju. - In: Praist. jug. zem. 4. Bronzano doba, 271-374.
BEINHAUER, K. W. 1985, Untersuchungen zu den eisenzeitlichen Bestattungsplätzen von Novilara (Provinz Pesaro und Urbino/Italien). - Frankfurt.
BOŽIČ, D. 1987, Keltska kultura u Jugoslaviji. Zapadna grupa. - In: Praist. jug. zem. 5. Željezno doba, 855-897.
BOŽIČ, D. 1999, Die Erforschung der Latenezeit in Slowenien seit Jahr 1964. - Arh. vest. 50, 189-213.
BRATINA, P. 1997, Godovič. - Var. spom. 36, 146.
BREŠČAK, D. 1995, Roman Bronze Vessels in Slovenia. New Finds 1982-1991. - In: Acta of the 12th international congress on Ancient Bronzes. Nijmegen 1992, Nederlandse ar-cheologische rapporten 18, 15-21, Njimegen.
CRISMANI, A. and G. RIGHI 2002, Considerazioni sul materiale protostorico di San Servolo. - In: La necropoli di San Servolo. Veneti, Istri, Celti e Romani nel territorio di Trieste, 89-94,Trieste.
ČOVIC, B. 1987, Grupa Donja Dolina-Sanski Most. - In: Praist. jug. zem. 5. Željezno doba, 232-286.
DESCHMANN, K. 1888, Führer durch das Krainische LandesMuseum Rudolfinum in Laibach. - Laibach.
DOLENZ, H. 1998, Eisenfunde aus der Stadt auf dem Magda-lensberg. - Kärnt. Musschr. 75.
DRECHSLER-BIŽIC, R. 1956, Gradine u Lici. - Ves. Voj. muz. 3, 36-51.
DRECHSLER-BIŽIC, R. 1983, Japodska kulturna grupa. - In: Praist. jug. zem. 4. Bronzano doba, 374-389.
DULAR, J. 2003, Halštatske nekropole Dolenjske. - Opera Instituti Archaeologici Sloveniae 6, Ljubljana.
FIALA, F. 1899, Das Flachgräberfeld und die prähistorische Ansiedlung in Sanskimost. - Wiss. Mitt. Bos. Herz. 6, 62128.
GABROVEC, S. 1966a, Latensko obdobje na Gorenjskem. -Arh. vest. 17, 243-270.
GABROVEC, S. 1966b, Srednjelatensko obdobje v Sloveniji. -Arh. vest. 17, 169-242.
GABROVEC, S. 1992, La regione a sud-est delle Alpi e la civilta picena. - La Civiltä Picena nelle Marche, Studi in onore di Giovanni Annibaldi, 242-254, Ripatransone.
GABROVEC, S. and D. SVOLJŠAK 1983, Most na Soči (S. Lucia) 1. Zgodovina raziskovanj in topografija. - Kat. in monogr. 22.
GASPARI, A. 2004, Bronzezeitliche Funde aus der Ljubljanica. Opfer, Überreste von Bestattungen oder zufällige Verluste? - Arch. Korrbl. 34, 37-50.
GASPARI, A. and R. KREMPUŠ 2002, Bronze "donor" from the votive site in the River Ljubljanica. - In: A. Giumlia-Mair (ed.), I bronzi antichi. Produzione e tecnologia, Atti del XV Congresso Internazionale sui Bronzi Antichi, Gra-do-Aquileia, 22-26 maggio 2001, 446-449, Montagnac.
GUŠTIN, M. 1973, Kronologija notranjske skupine. - Arh. vest. 24, 461-506.
GUŠTIN, M. 1974, Mahaire. Doprinos k povezavam Picena, Slovenije in Srednjega Podonavja v 7. stol. pr. n. š. - Situla 14-15, 77-94.
GUŠTIN, M. 1979, Notranjska. K začetkom železne dobe na severnem Jadranu. - Kat. in monogr. 17.
GUŠTIN, M. 1990, Poznolatenska čelada iz Ljubljanice. - Arh. vest. 41, 121-134.
GUŠTIN, M. 1991, Posočje. Posočje in der jüngeren Eisenzeit. -Kat. in monogr. 27.
HENCKEN, H. 1978, The Iron Age Cemetery of Magdalenska gora in Slovenia. - Bull. Amer. Sch. of Prehist. Res. 32.
HOERNES, M. 1888, Die Gräberfelder an der Wallburg von St. Michael bei Adelsberg in Krain. - Mitt. Anthr. Ges. 18, 217-249.
ISTENIČ, J. 2001, Un bronzetto di Appolo (Beleno?) dal fi-ume Ljubljanica (Slovenia). - Aquil. Nos. 72, 57-70.
KELTOI, Kelti in njihovi sodobniki na ozemlju Jugoslavije. -Ljubljana, 1983.
64	Guštin 1990, 128-129; id. 1991, 52.
65	Gabrovec 1966, 183, pl. 8: 2; Windl 1975, fig. 2; Križ 2001, 83, fig. 63.
66	Guštin 1990, 128; Božič 1987, 870.
67	Here we want to draw attention to several over 50 cm long swords with straight or curved back and a convex blade that continues into the hilt plate with several rivets (Hencken 1978, fig. 99: a,f; 365: i). Hilts of some better preserved swords from Vinica terminate in a narrowed part, onto which an oval plate was fitted (graves 59/48, 96/239, 110/311, 110/312). We kindly thank Dr. Dragan Božič for the explanation of the origin of the swords in the Mecklenburg Collection.
68	Božič 1999, 199.
69	Stare 1952-1953, 73, pl. 1: 1,2.
KNEZ, T. 1993, Novo mesto III. Knežja gomila. Kapiteljska njiva. - Carniola Archaeologica 3, Novo mesto.
KOS, P. and A. ŠEMROV 2003, Skupna najdba keltskih in rimskih novcev v reki Ljubljanici. - Arh. vest. 54, 381-395.
KRIŽ, B. 2001, Kelti v Novem mestu. Katalog razstave. - Novo mesto.
LANG, A. 1998, Das Gräberfeld von Kundl im Tiroler Inntal. -Frühgeschichtliche und provinzialrömische Archäologie. Materialien und Forschungen 2, München.
LEJARS, Th. 1994, Gournay III. Les fourreaux d'epee. Le sanc-tuaire de Gournay-sur-Aronde et l'armement des Celtes de La Tene moyenne. - Archäologie aujourd'hui, Paris.
LEJARS, Th. 1996, L'armement des Celtes en Gaule du nord ä la fin de l'epoque gauloise. - Revue archeologique de Picar-die 3/4, 79-103.
LJUBIC, Š. 1889, Popis arkeologičkoga odjela Nar. zem. muzeja u Zagrebu. Odsjek 1. - Zagreb.
LOŽAR, R. 1938, Rimska najdba iz Polhovega gradca. - Glas. Muz. dr. Slov. 19, 85-108.
MARIC, Z. 1964, Donja Dolina. - Glas. Zem. muz. 19, 5-128.
MIHOVILIC, K. 1991, Nalazi prahistorijskih ostava na području Istre. - Arh. vest. 42, 207-218.
MIHOVILIC, K. 1995, Reichtum durch Handel in der Hallstattzeit Istriens. - In: B. Hänsel (ed.), Handel, Tausch und Verkehr im Bronze- und Früheisenzeitlichen Südosteuropa, Südosteuropa-Schriften 17, Prähistorische Archäologie in Südosteuropa 11, 283-329.
MLINAR, M. 2002, Nove zanke svetolucijske uganke. Arheološke raziskave na Mostu na Soči. 2000 do 2001. - Tolmin.
MORETTI, M. 1983, Necropoli del catelliere mediano di Piz-zughi. - In: Preistoria del Caput Adriae, 153-158, Udine.
MÜLLNER, A. 1894, Reiseskizzen aus Italien. -Argo 3, 145-145.
MÜLLNER, A. 1899, Yatagan aus dem Laibachflusse. - Argo 7, 151-152.
NOTHDURFTER, J. 1979, Die Eisenfunde von Sanzeno im Nonsberg. - Röm. Germ. Forsch. 38.
PAROVIC-PEŠIKAN, M. 1982, Grčka mahajra i problem krivih mačeva. - God. Cent. balk. isp 20/18, 25-51.
POPOVIC, P. 1996, Early La Tene Between Pannonia and the Balkans. - Starinar 47, 105-125.
POTOČNIK, M. 1986, Ljubljanica. - Arh. preg. 27, 170-172.
QUESADA SANZ, F. 1997, El armamento iberico. Estudio ti-pologico, geogräfico, funcional, socialy simbolico de las armas en la Cultura Iberica (siglos VI-I a. C.). - Monographies Instrumentum 3/1, Montagnac.
RESTITUZIONI 1993, Restituzioni'93. Opere restaurate. - Vi-cenza.
REINECKE, P. 1965, Mainzer Aufsätze zur Chronologie derBronze-und Eisenzeit. - Bonn.
SCHAAFF, U. 1990, Keltische Waffen. - Mainz.
SIEVERS, S. 1982, Die mitteleuropäischen Hallstattdolche. -Prähistoriche Bronzefunde 6/6.
SIEVERS, S. 2001, Les armes d'Alesia. - In: M. Redde, S. v. Schnurbein (eds.), Alesia. Fouilles et recherches franco-alle-mandes sur les travaux militaires romains autour du Mont-Auxois (1991-1997) 2. Le materiel, Memoires de l'Academie des inscriptions et belles-lettres 22, 121-241.
STARE, F. 1952-1953, Nekaj železnodobnih predmetov iz Ljubljane. - Zgod. čas. 6-7, 71-77.
STARE, F. 1955, Vače. - Kat. in monogr. 1.
STARO ORUŽJE, Staro oružje. Istorija-kultura-tradicija. - Skopje 1998.
STIPČEVIC, A. 1960, Latenski predmeti iz okolice Obrovca.
-	Diadora 1, 87-94.
ŠKEGRO, A. 1992, Ostava iz Matijevica i njezino značenje za klasifikaciju ofanzivnog oružja s delmatskih prostora. - Opusc. arch. 16, 89-101.
TOMEDI, G. 2002, Das hallstattzeitliche Gräberfeld von Frög.
-	Archaeolingua 14.
TREASURES OF CARNIOLA, Treasures of Carniola. The unique collection of Prehistoric antiquities excavated by H. H. the LateDuchess PaulFriderich of Mecklenburg. - New York, 1934.
WINDL, H. J. 1975, Zur Chronologie des Gräberfeldes von Mihovo, Unterkrain (Dolenjsko). - Ann. Naturhist. Mus. Wien 80, 877-894.
VERWERS, J. and J. YPEY 1975, Six iron swords from the Netherlands. - Analecta Praehistorica Leidensia 8, 79-91.
WYSS, R., T. REY and F. MÜLLER 2002, Gewässerfunde aus Port und Umgebung. Katalog der lat ene- und römerzeitlichen Funde aus der Zihl. - Schr. d. Bern. Hist. Mus. 4.
Grob z mahairo z Mosta na Soči
Zadnje izvedbe enoreznih mečev z odklonjenim ročajem
UVOD
Leta 2000 je med zaščitnimi arheološkimi raziskavami na zahodnem robu ledine Repelc na levem bregu Idrijce na Mostu na Soči ekipa Tolminskega muzeja izkopala 30 halštatskih, la-tenskih in rimskih žganih grobov.1 Izkopavanja na Repelcu, ki je bil doslej arheološko praktično neraziskan,2 so se nadaljevala tudi v letu 2002, ko je bilo na skrajnem vzhodnem predelu prve idrijske terase dokumentiranih še 35 grobov. Območje po-
kopov se ob vzhodnem robu terase zaključi, približno dve tretjini grobišča pa ostajata neraziskani.
Latenski in rimski grobovi z Repelca predstavljajo v okviru skoraj povsem halštatske svetolucijske nekropole posebnost. Med raziskavami v letu 2002 so bili vzhodno od omenjenih grobov odkriti tudi sledovi daritvenega mesta z močno frag-mentiranim kovinskim in keramičnim gradivom iz poznega latena in zgodnjerimskega obdobja.3
Stanje ohranjenosti raziskanih grobov je zelo slabo, saj je
1	Zaščitne raziskave pred gradnjo čistilne naprave in novega kanalizacijskega sistema je pod mentorstvom Draga Svoljšaka vodil Miha Mlinar.
2	Gabrovec, Svoljšak 1983, 33. Med gradnjo stanovanjske hiše na Repelcu (Most na Soči 70 a) so bili najdeni žgani prazgodovinski grobovi, katerih gradivo ni ohranjeno. Nekaj krovnih plošč je shranil ing. Milan Mikuž z Mosta na Soči 61 (D. Svoljšak, Dnevnik izkopavanj 1973).
3	Gradivo v okviru magistrskega študija pod mentorstvom dr. Mitje Guština obdeluje Miha Mlinar.
bila na terasi še do nedavnega njiva, močne sledove pa je zapustila tudi prva svetovna vojna oz. dogajanje v zaledju soškega bojišča. Med skupno petinšestdesetimi pokopi je bilo tako nepoškodovanih le nekaj grobov, predvsem tistih, ki so bili vkopani globje v sterilno glino ali celo v apnenčasto geološko osnovo. Na obseg poškodb kažejo kamnite grobne plošče, ki so bile večinoma prevrnjene in razmetane po celotnem terenu, ter uničenost vrhnjih delov jam.
V skupini mlajših pokopov je že ob odkritju izstopal moški grob 25, ki je bil skupaj z ostalimi pomembnejšimi celotami prikazan na razstavi v Tolminskem muzeju in objavljen v spremnem katalogu.4 Obravnavana celota tvori del skupine štirih grobov (gr. 18, 23, 24 in 27), ki so se v tlorisu deloma prekrivale (sl. 1), vendar so bile njihove jame dobro prepoznavne v profilu (sl. 2).
Ovalna jama žganega groba 25 je bila vkopana v plast rjavo sivega glinenega melja (sl. 3). Jama je po daljši osi merila 115 cm, ohranjena globina pa je znašala 15 do 20 cm. Zahodni rob grobne jame ni bil zanesljivo določen. Na sredino dna je bila nasuta žganina z drobnimi nedogorelimi kostmi in ogljem,5 ki sta jo na južni in vzhodni strani omejevala ukrivljena železen meč in sulična ost. Odlomljena konica rezila meča je bila položena pod meč. Polnilo ostalega dela grobne jame je sestavljal sivo črn glinast melj, v katerem je bil odkrit odlomek prostoročno izdelane posode z glavničenim okrasom. Na površini zasutja jame je ležala močno deformirana bronasta latenska fibula.
Čeprav ob odkritju pripadnost fibule grobni celoti še bila jasna, pa o tem govorijo naslednja dejstva. Pod in ob njej je ležala žganina s človeškimi kostmi in drobci oglja, ki so se neprekinjeno nadaljevali do nivoja orožja. Tako kot železno orožje iz groba je bila ritualno zvita. Poudariti velja tudi podobnost s poznolatenskim grobom 14 s poljedelskim orodjem,6 ki je bilo v grobu razvrščeno na enak način. Nakitni predmeti so ležali med žganino na sredini grobne jame, namerno poškodovano železno orodje pa je bilo razporejeno ob obodu!
OPIS GROBNEGA INVENTARJA
1.	Zvit enorezni meč z ozkim rezilom in odklonjeno ročaj-no ploščo. Rahlo usločeno rezilo s klinastim presekom je na sredini nekoliko razširjeno. Na prehodu rezila v ročaj je izko-van širok trnast branik. Rek. d. 69,2 cm; d. trna 11,5 cm; največja š. rezila 3,2 cm; inv. št. TM 101 (sl. 4: 1).
2.	Zvita železna sulična ost s kratkim deltoidnim listom z rebrom rombičnega preseka in dolgim tulastim nasadiščem. Ostrini in konica lista ter konec tula so poškodovani. Ohr. d. 35 cm, rek. d. 37 cm; ohr. š. lista 3,5 cm; d. tula 22 cm; najv. pr. tula 1,9 cm; inv. št. TM 102 (sl. 4: 2).
3.	Deformirana bronasta fibula srednjelatenske sheme z masivnim polkrožnim lokom, ki se širi proti glavi, peresovino
s šestimi navoji in nazaj zavito nogo. Ta se je pripenjala na lok s trakasto objemko. Na zgornjem delu noge je večji rebrast gumb, ki ga na obeh straneh obdaja manjša odebelitev, vmes pa so še manjša rebra. Nosilec za iglo je okrašen z vrezi. Noga fibule je zakrivljena vstran, poškodovana je tudi peresovina. Rek. d. 10,4 cm; d. loka 4 cm; v. 3,1 cm; inv. št. TM 103 (sl. 4: 3).
4. Odlomek ostenja prostoročno izdelanega keramičnega lonca. Zunanja in notranja površina opečnate do sive barve sta okrašeni z glavničenjem. Vel. odlomka 5,7 x 4 cm; deb. ostenja 0,6-0,8 cm; inv. št. TM 104 (sl. 4: 4).
OPREDELITEV IN DATACIJA
Kronološko najbolje oprijemljiv predmet v domnevni celoti predstavlja bronasta fibula srednjelatenske sheme s pereso-vino s šestimi navoji, z masivnim lokom in eno večjo svitkasto ter dvema manjšima odebelitvama na nogi (sl. 4: 3). Neposredno primerjavo poznamo le v grobni najdbi iz Vinice (gr. 86/195), ki pa ima daljšo peresovino s trinajstimi navoji. Fibuli sta oblikovno blizu mlajšim primerkom tipa Kastav/variante Idrija iz grobov 7/8, 9, 13 in 16 z Idrije pri Bači, ki imajo na nogi dve kroglasti odebelitvi in vrezan okras vzporednih črt na objemki.7 Grobne celote kažejo, da gre za vodilne oblike srednjega latena, ki ostanejo v uporabi še celo 1. stoletje pr. n. št.8 Okras spominja tudi na fibule s tremi gumbi na loku in okroglo ploščo na podaljšku noge, ki se poleg glavne koncentracije najdb v Posočju9 pojavljajo še v Estah, Akvileji, na Jereki v Bohinju, v Kranju in Vinici.10 Tudi te fibule, značilne za srednji laten in starejši del poznega latena, imajo praviloma peresovino s po tremi navoji, ki je le redko zavita okoli noge. Vsem trem tipom je skupno tudi z vrezi okrašeno ležišče za iglo.
Sulična ost s kratkim listom z rebrom (sl. 4: 2) pripada razmeroma redki obliki, ki se po razmerju med dolžino lista in tula (ta obsega skoraj dve tretjini celotne dolžine) približuje pilumom podobnim ostem iz osrednje- in vzhodnoalpskega območja.11 Med 35 in 65 cm dolge osti, ki se pojavijo ob koncu zgodnjelatenskega obdobja, imajo bikonveksno ali bajonet-no oblikovane liste z rebri, običajne tudi pri sočasnih keltskih primerkih s krajšim tulom.
Po dva kratka in dva daljša primerka pilumom podobnih osti z bajonetnimi listi sta bila najdena v okviru skupne najdbe ritualno uničenega orožja na vzpetini pri Borčah (Förker Laas Riegel) v Ziljski dolini,12 datirani v čas okoli 300 pr. n. št. Vsaj štiri osti tega tipa izvirajo iz grobišč pri Šmihelu, od katerih imata dve ohranjeni grobni celoti.13 Okoli 46 cm dolga ost s klasično oblikovanim listom je bila pridana v grob 99 iz ledine Za Polšno, v njem so našli še bronast gumb z vrezanim motivom spirale.14 M. Guštin ga je pripisal certoški fibuli, vendar je verjetneje, da gre za gumb na nogi fibule zgodnje- ali sred-njelatenske sheme.15
4	Mlinar 2002, 26, sl. 18, kat. št. 30.
5	Radiokarbonska analiza lesnega oglja, opravljena na Leibniz Labor für Altersbestimmung und Isotopenforschung Christian-Albrechts-Universität v Kielu (KIA24524) konec leta 2003, je pokazala konvencionalno starost BP 2501 ± 20 (2 sigma kalibriran rezultat: cal BC 785-522).
6	Ib., 26, sl. 19, kat. št. 31.
7	Guštin 1991, 37, t. 9: 10; 10: 9; 12: 1; 14: 9,13.
8	Ib., 37.
9	Ib., t. 3: 15; 6: 6; 12: 2; 13: 1; 21: 13; 35: 1-4; 40: 4; 41: 6.
10	Ib., 37.
11	Schaaff 1990, sl. 11.
12	Ib., 20, sl. 10.
13	Brez najdiščnih podatkov sta zgornji del osti z bajonetnim listom in tulom kvadratnega preseka (Guštin 1979, t. 80: 26) in 45,5 cm dolg primerek s klasično oblikovanim listom (Hoernes 1888, 234, t. 5: 8).
14	Guštin 1979, 42, t. 51: 1,2.
15	Podobno oblikovan gumb z reliefno izvedenim spiralnim okrasom najdemo na nogi bronastih fibul iz Vinice (Gabrovec 1966b, t. 15: 8) in Karaburme (gr. 14; Popovic 1996, sl. 12: 14).
V	grobu 127 iz Šmihela je bila 54,4 cm dolga ost z bajo-netnim listom in steblom kvadratnega preseka najdena skupaj z žebljem, ki je verjetno služil pritrjevanju ščitne grbe latenske sheme, nadalje s suličnim kopitom, bronasto ovratnico s spiralno uvitimi zaključki, fragmenti masivne bronaste zapestnice in steklenimi jagodami.16 Ovratnica sodi med vodilne oblike poznohalštatske stopnje Notranjska VI, vendar lahko na podlagi primerka z očalastim obeskom, ki ga je skupaj s fibulo tipa Kastav nosil pokojnik iz vhodnega brezna Gorenje jame pri Povirju, sklepamo, da so bile v uporabi še vsaj v srednjem latenu.17
V	srednji ali pozni laten sodijo pilumom podobne osti z ozkim lečasto oblikovanim listom rombičnega preseka, ki se pojavljajo predvsem v alpskih dolinah18 in na zahodnokelts-kem območju.19 Več kot 55 cm dolga, namerno zvita in na dveh mestih prelomljena ost tega tipa je znana tudi med površinskimi najdbami z Repelca.20
Za mlajšo datacijo osti iz groba 25 bi lahko govorila tudi deltoidna oblika lista s težiščem v spodnjem delu, ki je pogosta v kontekstih iz konca latena in zgodnjerimskega obdobja. V tem času so razširjeni tako primerki z rebrom rombičnega ali okroglega preseka,21 pri katerih zavzema nasadišče okoli tretjino celotne dolžine kot podobne osti z listi sploščenega rom-bičnega preseka.22 Osti z deltoidnim listom so značilne za hercegovska in dalmatinska najdišča iz druge polovice 1. st. pr. n. št., zato se v literaturi omenjajo kot tipično delmatsko orožje.23 Ta tip je upodobljen tudi na tropeju iz legijskega tabora Tilurium/Gardun.24
Enorezni meči z odklonjenim ročajem
Meč z odklonjenim ročajem iz groba 25 (sl. 4: 1), za katerega uporabljamo v naši literaturi uveljavljen starogrški izraz ^axaipa - machaira,'2'' sodi v skupino ukrivljenih enoreznih orožij, ki so jih ljudstva ob zahodnih in severnih obalah Sredozemskega morja uporabljala vso železno dobo. Mahaira naj bi bila zelo učinkovito orožje, odlikovano z uravnoteženostjo in obvladljivostjo, pomembnima posebej pri bojevanju s konja. Za razliko od druge značilne oblike enoreznega meča s
povprečno dolžino okoli 60 cm, ki je pod imenom falcata znana predvsem kot orožje iberskih plemen in je tako kot rimski gla-dius predstavljala kombinirano orožje za suvanje in sekanje, se je mahaira nedvomno uporabljala kot sablja.26
Tradicija uporabe velikih nožev z zakrivljenim rezilom sega na vzhodni Jadranski obali in njenem zaledju v pozno bronasto dobo. Japodska in liburnska skupina tega časa poznata praktično že vse osnovne oblike mahair in falkat, ki jim nato v balkanskem prostoru sledimo v železni dobi.27 Pri orisu tipološ-ko-kronološkega razvoja železnih mahair se naslanjamo na temeljno študijo o najdbah iz vzhodnopredalpskega prostora in Balkana, ki jo je leta 1974 objavil Mitja Guštin.
Mahaire variante Basarabi, ki jo označujeta trikotno oblikovano rezilo ter prečna železna ploščica v obliki črke T na koncu jezičastega ročaja, se v 8. stoletju pr. n. št. pojavijo na območju Basarabi kulture v Olteniji,28 proti zahodu pa so znane z Ilijaka na Glasincu (stopnja IVb) ter Legna pri Slovenj Gradcu (Podzemelj 1) in Brega/Fröga pri Rožeku.29 Varianti Basarabi se približuje tudi okoli 50 cm dolg primerek s polnim bronastim ročajem in ohranjeno nožnico iz bogatega groba 16/I na Kapiteljski njivi v Novem mestu, ki sodi med značilne celote stopnje Podzemelj 2.30
Meči variante Tržišče-Donja Dolina, ki se od variante Basarabi v osnovi razlikujejo le po širšem rezilu z ravnim ali le rahlo usločenim hrbtom, izrazito izbočeni ostrini ter po preprostejšem jezičastem ali trnastem ročaju, so razširjeni na prostoru med vzhodnimi Alpami in bosansko Posavino.31 Precejšno raznolikost mečev te skupine nakazujejo že met-rične in oblikovne značilnosti devetih mahair iz depoja Tržišče, pri katerih sega dolžina med 35 in 65 cm, širina rezila pa od 3,3 do 5,1 cm.32 Imajo razmeroma oster odklon med rezilom in ročajnim delom (ca. 135-143°) ter sploščeno izkovan hrbet s presekom v obliki črke T, ki ga srečamo pri poznejših variantah le izjemoma. Skoraj vsi meči imajo ohranjene predrtine oz. zakovice za pritrjevanje ročajne obloge. Sorodne so ma-haire variante Novilara iz Picenuma, datirane v prvo polovico 7. stoletja (stopnja Novilara IIb-IIIa),33 ter primerki iz Istre, ki jih je K. Mihovilic pripisala stopnji III.34 Večina od okoli 10 grobnih najdb iz vzhodnoalpskega prostora ter depo s Tržišča sodi v horizont Ha C,35 medtem ko so mahaire iz Donje
16	Guštin 1979, 44, t. 57: 1-6.
17	Guštin 1973, 479; Crismani, Righi 2002, 93, sl. 52.
18	Npr. Nothdurfter 1979, 79-80, t. 70, 1192-1195, 1198-1200; Lang 1998, 126, t. 26: 498-500,502,503.
19	Sievers 2001, 163-164, t. 69: 359; Wyss, Rey, Müller 2002, t. 45: 145-147; 46: 145-147.
20	Neobjavljeno najdbo hrani Tolminski muzej; inv. št. TM 28.
21	Npr. Idrija pri Bači, gr. 5 (Guštin 1991, 15, t. 5: 1).
22	Npr. Polhov Gradec (Ložar 1938, 95, sl. 22 a,b), Štalenska gora/Magdalensberg (Dolenz 1998, 65-66, t. 6: M 41).
23	Škegro 1992, 91, t. 2: 13; 3: 14.
24	Abramic 1937, sl. 2; 3.
25	Uporaba izrazov machaira in kopis, ki se največkrat omenjata v zvezi z ukrivljenimi enoreznimi meči, je vprašljiva. Izraz mahaira se je sprva nanašal na obredne nože za ubijanje žrtvenih živali in kirurško orodje, od 4. stoletja pr. n. št. naprej pa občasno tudi kot enotno poimenovanje za meče. Ksenofontov traktat o konjenici (Xen., De re Eq. XII, 11) je edini vir, ki jasno razločuje med izrazoma machaira oz. enorezno sabljo ter xiphous, ki pomeni kratek raven meč (glej Quesada Sanz 1997, 171).
26	Quesada Sanz 1997, 169-171.
27	Batovic 1983, 314, sl. 20: 4,5; t. 46: 6-8; Drechsler-Bižic 1983, 383-384, t. 53: 9.
28	Guštin 1974, 81-83, sl. 4; 5; Tomedi 2002, 112-113.
29	Guštin 1974, 78-79, 81-83, sl. 2; 6: 3; Tomedi 2002, 112-113.
30	Knez 1993, 23-24, t. 19.
31	Guštin 1974, 80, lista.
32	Guštin 1979, t. 31: 1-4; 32: 1-5.
33	Primerki iz grobov 30, 64, 65 na nekropoli Novilara-Servici (Beinhauer 1985,
ročaj, odlično ohranjen primerek z nožnico iz gr. 14 pa trnast nastavek (ib., t. 63:
t. 77: 751).
897; 127: 1404,1409) imajo jezičast Izvor južnoitalsko-picenske variante kratkih mahair z ročajno ploščo, na kateri so v obliki trikotnika razmeščene tri zakovice, lahko najverjetneje iščemo že v 8. stoletju (Tomedi 2002, 113; npr. Beinhauer 1985, t. 66: 781; 68: 806; 86: 973). V vzhodnopredalpskem prostoru kaže te značilnosti meč iz Črnomlja (Gabrovec 1992, 250, sl. 2: 18).
34	Mihovilic 1991, t. 1: 8; 3: 7-10; ead. 1995, 284-285, sl. 14; t. 20: 1-3; Moretti 1983, 155, t. 43: 5.
35	Guštin 1974, 87.
Doline in Sanskega Mosta datirane v 6. stoletje pr. n. št. (stopnji 2b in 2c).36
V mlajših fazah železne dobe ostane uporaba mečev z odklonjenim ročajem značilna za plemena severnojadranskega zaledja, medtem ko se v osrednjih delih Balkana pričnejo pojavljati druge oblike ukrivljenih enoreznih orožij.37
Prehodno obliko k najmlajšim izvedbam mahair predstavljata meč iz poznohalštatskega groba 135 na šmihelski nekro-poli Za Polšno (sl. 5: 3), v katerega so bile pridane še uhata sekira in dve sulični osti s klasično oblikovanim listom,38 ter še ena najdba brez znane celote iz istega grobišča (sl. 5: 4).39 Nekaj nad 50 cm dolgi mahairi variante Šmihel imata čokato, do 3,6 cm široko rezilo s skoraj ravnim hrbtom, izrazito izbočen osrednji del ostrine in stanjšano konico ter izbočen prehod med hrbtom in jezičastim ročajem.40 Meč brez ohranjene grobne celote je imel ohranjen kratek koničnik iz železne pločevine ter ustje s polkrožno zanko za obešanje, od drugega meča pa se razlikuje tudi po sploščeno izkovanem hrbtu v obliki črke T.41 Podobne oblike je tudi nekoliko širša mahaira iz japodske nekropole Prozor pri Otočcu v Liki, ki prav tako nima sporočenih natančnejših najdiščih podatkov (sl. 5: 2).42 Krajši od opisanih je 41,2 cm dolg primerek z manjkajočim ročajnim trnom in poškodovano konico rezila iz groba 10 na nekropoli sveto-lucijske skupine na Jelenšku nad Godovičem, kjer se je pokopavalo izključno v stopnji IIc.43
Mahaire variante Ljubljanica, ki smo jo poimenovali po najdišču primerka z okrašenim ročajem, so poleg najdbe z Repelca znane še s štirimi primerki, in sicer z enim iz Šmihela, enim iz neznanega najdišča "na Kranjskem" ter dvema iz Ljubljanskega barja oz. reke Ljubljanice. Od prej naštetih oblik se razlikujejo po daljšem rezilu, saj merijo med 65 in 80 cm, skupna pa so jim tudi razmerja posameznih delov ter usmeritev rezila oz. kot med hrbtom in ročajem (sl. 6). Označuje jih vitko usloče-no rezilo klinastega preseka ter oster prehod v ročajni del z jezičastim branikom. Ostrina je bolj ali manj vzporedna s hr-
btom, z največjo širino na sredini rezila. Zaobljen del železnega dela branika je obrnjen proti ostrini, z navpične ročajne strani pa je nanj prvotno nasedala organska ali kovinska obloga. To velja tudi za zaključek ročajne plošče, medtem ko je bil sam držaj pravokotnega preseka izdelan iz organske obloge ali nanizanih diskov.
Geografsko najbližja primerjava za mahairo z Repelca izvira iz grobišča Za Polšno pri Šmihelu (sl. 6: 1), kjer je A. Burger je pod Szombathyjevim oz. Hoernesovim vodstvom v letih 18851886 izkopal 96 grobov.44 Meč s poškodovano konico nima ohranjene grobne celote, vendar so iz te nekropole sicer znani tako poznohalštatski kot srednjelatenski pokopi. Poznolaten-skih in zgodnjerimskih grobov, o katerih bi smeli domnevati na podlagi gradiva iz območja naselbine, ne poznamo.
Lepo ohranjen, 79 cm dolg meč iz neznanega najdišča "na Kranjskem" (sl. 6: 2), ki ga danes hrani Peabody Museum Har-vardske univerze, je v Združene države Amerike prispel z delom dediščine vojvodinje Mecklenburške.45 Avkcijski katalog Treasures of Carniola ga omenja kot tipičen meč ilirskih plemen, vendar ne navaja niti točnejših podatkov o najdišču niti načina pridobitve.46
"Čudno oblikovana železna sablja" (sl. 6: 4), ki jo je leta 1846 Deželnemu muzeju v Ljubljani predal posestnik Anton Breskvar, je bila odkrita pri zajemanju vode na barjanski ledini V Blekah med izlivom Iščice in jarkom Preprošica pri Volarju, okoli 1500 m južno od ljubljanske Špice. Dobro ohranjen meč, ki ima na zaključku ročaja nasajen mandljasto oblikovan glavič iz bronaste zlitine z grobo obdelano površino, je ležal na površini gline pod šoto.47 A. Müllner je "jatagan" na podlagi Hoernesovih izvajanj o mečih iz Šmihela pripisal halštatskim oblikam, najdiščne okoliščine pa je sprva enačil s stratigraf-skim položajem tovorne ladje iz bližnjih Lip.48
Novejšo najdbo v obravnavani skupini mahair predstavlja primerek z okrašenim ročajem iz struge Ljubljanice pri izlivu Zrnice, jugovzhodno od Blatne Brezovice. Meč (sl. 6: 5) je
36	Čovic 1987, 246, 250-251, sl. 15: 3,20.
37	Med 30 in 42 cm dolge nože iz najdišč v zahodni Bosni in Srbiji, datirane v konec 5. in 4. stoletje pr. n. št., označuje rahlo ukrivljen hrbet, simetrično izbočena ostrina in ročajna plošča s tremi do petimi zakovicami. Mlajše primerke znotraj omenjenega časovnega okvira predstavljajo primerki tipa Sanski Most z žlebom pod hrbtom rezila in sedlasto oblikovanim branikom (Parovic-Pešikan 1982, 35-39, t. 3). Po Čovicu naj bi bili značilni za fazo IIIb (350-300/275 pr. n. št.; Čovic 1987, 258, 262, sl. 16: 25). Drugo skupino predstavljajo falkate oz. meči z enakomerno ukrivljenim hrbtom in esasto ostrino s težiščem v zgornji tretjini ter specifično oblikovanim ročajem v obliki konjske ali ptičje glave. Na Balkanu se pojavljajo na prostoru med bosansko Posavino, Pomoravjem in Makedonijo, kjer so datirani med 6. in 3. stoletje pr. n. št. (Parovic-Pešikan 1982, 25-35, t. 2: 1-3; 5: 1-3; glej tudi novejšo najdbo iz okolice Prilepa - Staro oružje, 11, kat. br. 103). Sorodni so meči z ravnim hrbtom in rahlo izbočeno ostrino (npr. Parovic-Pešikan 1982, t. 1: 1,2).
38	Hoernes 1888, 230, t. 4: 4; Guštin 1974, 81, sl. 9: 1; id. 1979, 44, t. 56: 2.
39	Hoernes 1888, 230, t. 4: 5; Guštin 1979, 48, t. 85: 5.
40	Naštete značilnosti srečamo že pri 54,2 cm dolgem meču z odlomljenim ročajnim delom iz groba 126 na Sanskem Mostu (sl. 5: 1), ki ga je Čovic datiral v stopnjo Donja Dolina-Sanski Most II c oz. konec 6. stoletja. Grob je vseboval še dve lovorolistni sulični osti, del sklede in lonec (Fiala 1899, 99, sl. 140; Čovic 1987, 251-252, sl. 15: 21).
41	Način pripenjanja nožnice na pas neposredno na sedlastem prehodu branika v ostrino meča je razviden pri mahairah iz Novega mesta (Knez 1993, 23-24, t. 19) in Cveteža pri Vačah, kjer dva obročka visita iz zank na figuralno okrašeni bronasti prevleki (Stare 1955, 14, t. 1).
42	Ljubic 1889, 147, t. 29: 215; Drechsler-Bižic 1956, t. 1: 2.
43	Bratina 1997; za podatek o najdbi se zahvaljujeva izkopavalcema grobišča Patriciji Bratina (ZVKDS OE Nova Gorica) in Dragu Svoljšaku (NMS).
44	Hoernes 1888, 230, t. 4: 6; Guštin 1974, sl. 9: 6; id. 1979, 48, t. 85: 1.
45	Guštin 1974, 81; za dovoljenje za objavo se zahvaljujeva dr. Mitji Guštinu.
46	Treasures of Carniola, 81.
47	Narodni muzej Slovenije - inv. št. B 5051; Verzeichniß der im Jahre 1946 dem Museum in Laibach verehrten Geschenke. - Illyrisches Blatt 1847, 188, št. 38: "Vom Herrn Anton Einsiedler Breskquar, Realitäten-Besitzer in Laibach: - ein alter, eigenthüm-lich geformter Eisensäbel, stark vom Roste zersressen, mit einem messingenen Griffe, im Moorgrunde am Volar zwischen dem Ischzaflusse und dem Perproshza-Graben v' Blekah unter dem Torfe am Lehmboden beim Wasserschöpfen aufgefunden."; Desch-mann 1888, 142, št. 9; Müllner 1894, 149, t. 12: 18; id. 1899, 151-152, sl. 3; Stare 1952-1953, 72, risba 1; Gabrovec 1966a, 251, t. 8: 1; za izdelavo risbe in dovoljenje za objavo se zahvaljujeva Arheološkemu oddelku Narodnega muzeja Slovenije.
48	Müllner 1894, 149.
našel potapljač Miro Potočnik leta 1992. Čeprav so podatki o neposrednih najdiščnih okoliščinah meča omejeni na dejstvo, da je bil najden na glini v plasti finozrnatega peska ob levem bregu, je treba poudariti, da je odsek struge in obrežja med izlivom Bistre in Kaminom, še posebej območje samega izliva Zrnice, izjemno zanimivo arheološko najdišče. Med dokumentiranimi najdbami iz vode izstopajo skupina srednje- in pozno-bronastodobnih kovinskih predmetov,49 poznohalštatska nare-brena zapestnica in železna uhata sekira50 ter najdbe iz poz-nolatenskega obdobja, ki nakazujejo obstoj odlagališča dolgega trajanja. Poleg fibule tipa Posočje, ročke tipa Gallarate51 ter dveh le nekaj deset metrov narazen odkritih bronastih plastik votantov52 izvira iz okoli 50 m dolgega odseka tudi zdaj že znamenita skupna najdba keltskih in republikanskih srebrnikov iz časa kmalu po 147 pr. n. št., za katero P. Kos in A. Šemrov domnevata, da predstavlja slučajno izgubljen denar na pomembni transportni komunikaciji.53
Opis meča podajamo po fotografiji, na podlagi katere smo izdelali risbo. Meč z ohranjeno dolžino 72 cm ima odlomljeno zgornjo tretjino rezila. Močno odklonjen 16 cm dolg ročajni trn pravokotnega preseka ima na prehodu v rezilo širok trn, na katerega je pritrjen branik iz bronaste zlitine. Za razliko od ostalih mahair obravnavane variante je na strani proti rezilu branik raven, proti ročaju pa zaobljeno usločen. Branik je na trn pritrjen s sedmimi zakovicami okroglega preseka, od katerih so štiri razmeščene ob sprednjem robu branika, dve na sredini in ena proti ročaju. Zaključek ročaja je opremljen z gobasto oblikovanim glavičem s tremi polkrožnimi roglji, višjima stranskima in nižjim srednjim, pod katerimi so vidne štiri okrasne zakovice. Pritrditev je zagotavljala dodatna zakovica na začetku glaviča. Med glavičem in branikom so na ročaj v enakomernem razmiku nasajeni trije razmeroma široki ovalni kovinski obročki s poševno narezanimi robovi. Ti kažejo, da je bila obloga prvotno verjetno sestavljena iz izmenično nameščenih kovinskih in organskih elementov. Vsak obroček je bil na ročajni trn pritrjen z dvema zakovicama. Takšna konstrukcija držajnega dela je znana že pri mečih iz pozne bronaste dobe,54 pri halštatskih kratkih mečih in bodalih iz stopenj Ha C2 in Ha D,55 srečamo pa jo tudi pri mečih srednje-56 in poznolatenske sheme.57 V tem oziru je zanimiv tudi kratek enorezen meč s klekastim hrbtom in dvojno ukrivljeno ostrino iz zgodnjecesarskega grobišča v Sta-rigradu (Argyruntum). Na ročajnem trnu okroglega preseka ima nameščen trakast bronast branik, za njim pa ornamentiran
cilindričen okov, dve okrogli ploščici in masiven bronast gumb bikonične oblike, ki se je prvotno verjetno nahajal na koncu trna.58 Analogij za glavič ročaja mahaire iz izliva Zrnice, razen preprostejše izvedbe na drugem primerku iz Ljubljanskega barja, ne poznamo, branik pa ima dobro primerjavo na meču iz Gorenjega Vrhpolja pri Šentjerneju na Dolenjskem.59 Meč brez natančnejših najdiščnih podatkov, ki ga je M. Guštin pripisal varianti Tržišče-Donja Dolina, ima odlomljen večji del rezila klinastega preseka z največjo širino 3,6 cm, na ostanku pa so ohranjeni sledovi lesenih platic nožnice. Bronast ročajni trn pravokotnega preseka je nasajen na podaljšek rezila, trdnost stika pa je zagotovljal bronast branik z grobo obdelanimi robovi. Na stanjšan zaključek trna je prvotno najverjetneje nasedala kovinska prečka.
SKLEP
Meč iz groba 25 z Repelca sodi med dolge mahaire variante Ljubljanica, za katere so predhodni raziskovalci na podlagi tipološkega razvoja oz. težnje k podaljševanju rezila domnevali latensko starost.60 Datacija mečev te variante ostaja odprta, ker sestava groba z Repelca ni povsem zanesljiva, ostalih štirih mečev pa ni mogoče časovno opredeliti. Ost z deltoidnim listom in dolgim tulom, ki je v grobu ležala tik ob mahairi, bi lahko sodila tako v pozni halštat kot v laten. Črepinja posode, okrašene z glavničenjem, govori za zadnji stoletji pr. n. št. ali zgodnjerimsko obdobje, ta datacija pa velja tudi za deformirano bronasto fibulo srednjelatenske sheme iz vrha zasutja grobne jame. Popolno presenečenje so rezultati radiokarbonske analize, ki postavljajo oglje z dna jame groba 25 v starejši halštat.
V prid domneve o pripadnosti groba 25 z Repelca latenske-mu obdobju, ko je bilo med posoškimi bojevniki sicer razširjeno orožje in ostala vojaška oprema keltskega tipa,61 govori tudi že večkrat obravnavan privzem poznohalštatskih oblik v latensko oborožitev. V poznolatenskem grobu 1 v Idriji pri Bači sta bila poleg delov noše, sulične osti in pretežno poljedelskega orodja najdena tudi dva večja noža.62 Eden po obliki močno spominja na mahairo, saj ima usločeno ozko rezilo in odklonjen ročajni trn, na kar je opozoril že M. Hoernes. 63
T. i. "halštatska renesansa" je v Posočju in na Notranjskem izpričana še z idrijsko varianto negovskih čelad iz 1. stoletja pr. n. št.64 in s pilumom podobnimi suličnimi ostmi, na ob-
49	Gaspari 2004.
50	Hrani Narodni muzej Slovenije; neobjavljeno.
51	Potočnik 1987, sl. 4; Breščak 1995, sl. 4: 20.
52	Istenič 2003; Gaspari, Krempuš 2003.
53	Kos, Šemrov 2003.
54	Reinecke 1965, t. 2: 642.
55	Glej npr. železne ali bronaste diske na kratkih mečih z antenskim ročajem in bodalih tipa Hallstatt (Sievers 1982, 18-24, t. 5-11).
56	Običajno imajo ročajni trni nameščene po dve oz. tri okrogle železne ali bronaste ploščice z narezanim, navadno pa tudi nekoliko stanjšanim robom (Keltoi, 78; Lejars 1994, 48).
57	V stopnjo Lt D1 sodi npr. meč iz groba 225 v Poveglianu pri Veroni, pri katerem je na ročaju ohranjenih 8 tankih bronastih ploščic in dva koščena obročka (Restituzioni 1993, 11, 15). Gosto nasajene okrogle ali ovalne kovinske ploščice so značilne tudi za ročaje mečev iz severovzhodne Galije in germanskega območja, datirane v mlajši del stopnje Lt D2 (Verwers, Ypey 1975; Lejars 1996, 90).
58	Abramic, Colnago 1909, 110, sl. 82; Stipčevic 1960, 89, sl. 1: 3.
59	Guštin 1974, 81, sl. 3; Knez 1993, 24; Dular 2003, 178, t. 13: 5. Za podatek o najdišču meča se zahvaljujeva dr. Draganu Božiču.
60	Gabrovec 1966a, 255; Guštin 1974, 81, op. 16.
61	Med neobjavljenimi najdbami omenimo fragmente ščitnih grb, nožnic in mečev z Repelca (Tolminski muzej), sulično ost z bajonetnim listom iz Žage pri Bovcu (neobj., hrani Gorenjski muzej, Kranj; za informacijo se zahvaljujeva Miji Ogrin) in del nožnice poznolatenske sheme z Gradu pri Reki (neobjavljeno; podatek je posredoval dr. Dragan Božič).
62	Guštin 1991, 13-14, t. 1: 9.
63	Hoernes 1888, 232-233.
64	Guštin 1990, 128-129; id. 1991, 52.
močju mokronoške skupine pa z uhatimi sekirami65 in bronasto čelado vrste Filottrano iz Bele Cerkve.66 Uporaba enorez-nih mečev oz. dolgih nožev je značilna za Vinico in japodski prostor,67 ob koncu latenskega obdobja pa tudi za mokrono-ško skupino, ki vsaj pri orožju velja za skoraj docela keltizira-no. Zgodnjerimske izvedbe ukrivljenih nožev "japodske vrste",68 ki predstavljajo izvedenko enoreznih mečev tipa Sanski Most, sežejo na zahodu prek grobišča Strmec pri Beli Cerkvi vse do Ljubljanskega barja, kjer so izpričane vsaj s 5 najdbami iz Ljubljanice. Dokaz za tesne povezave prebivalstva južnega dela Ljubljanske kotline, ki velja za zahodni rob s Tavriski poseljenega območja, z nekeltskimi sosedi v zaledju Caput Adriae je tudi precejšnje število najdb najmlajših izvedb certoških fibul (vrste XIIb in notranjske različice vrste VIIf) ter fibul tipa Picugi in Kastav, najdenih v strugi Ljubljanice.69
Andrej Gaspari Univerza na Primorskem Znanstveno-raziskovalno središče Inštitut za dediščino Sredozemlja Garibaldijeva 18 SI-6000 Koper andrej.gaspari@siol.net
Miha Mlinar Tolminski muzej Mestni trg 4 SI-5220 Tolmin miha.mlinar@tol-muzej.si
65	Gabrovec 1966, 183, t. 8: 2; Windl 1975, sl. 2; Križ 2001, 83, sl. 63.
66	Guštin 1990, 128; Božič 1987, 870.
67	Na temu mestu izpostavljamo okoli 50 cm dolge meče z ravnim ali ukrivljenim hrbtom in izbočenim rezilom, ki prehaja v ročajno ploščo z več zakovicami (Hencken 1978, sl. 99: a,f; 365: i). Ročaji nekaterih bolje ohranjenih mečev iz Vinice se končajo s trapezasto zoženim delom, na katerega je nasajena ovalna ploščica (gr. 59/48, 96/239, 110/311, 110/312). Za pojasnitev izvora mečev v Mecklenburški zbirki se lepo zahvaljujeva dr. Draganu Božiču.
68	Božič 1999, 199.
69	Stare 1952-1953, 73, t. 1: 1,2.