UDK 903.2(497.4)“633/634”:738.8; 903.2(497.4)“633/634”:552.1 Documenta Praehistorica XXXI Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia Andreja ibrat Gapari Department of Archaeology, University of Ljubljana, SI andreja.gasparic@ff.uni-lj.si ABSTRACT – The results of the mineralogical and chemical analyses of pottery from the Neolithic period from the Divaa region are presented. Pottery samples from two rock shelters, i.e. Mala Triglavca and Trhlovca, were included in the analyses, as well as sediment samples from other rock shelters, caves and rivers around this area. The mineralogical and chemical composition of the ceramic is uniform in most of the samples; the differences between the clay pastes of the vessels are in the use of a tempering material, mostly calcite grains. The sediment samples, especially from the cave deposits, point to a local production of the Neolithic pottery on the Karst plateau. IZVLEEK – Arheometrine analize neolitske keramike iz podroja Divae. Predstavljeni so rezultati mineralokih in kemijskih analiz neolitske keramike iz okolice Divae na Krasu. V analizo so bili vkljueni vzorci keramike iz dveh spodmolov, Male Triglavce in Trhlovce, pa tudi vzorci jamskih in renih sedimentov iz blinje okolice arheolokih najdi. Mineraloka in kemijska sestava keramike je enotna v skoraj vseh vzorcih; glavne razlike pa se pojavijo v tehnologiji neolitskih lonarjev, ki so naravni glineni masi dodajali mineralna zrna, predvsem kalcitna zrna ali sigo. Analiza sedimentov iz jamskih najdi pa kae na lokalen izvor naravne glinene mase na Krasu. KEY WORDS – mineralogy; Neolithic; pottery; provenience studies; transhumance INTRODUCTION Although investigations of the mineralogy and che-sent paper. The pottery of the Neolithic period was mistry of pottery and other materials in archaeology attributed to the Danilo culture, originating in the are widely used (Rice 1987; Shepard 1965; Spataro eastern Adriatic, according to J. Koroec (1960), . 1999; 2002), they have not been extensively ap-Batovi (1973; 1979) and F. Leben (1973). Similar plied to the study of pottery technology in Slovenia assemblages in the Triestine Karst in Italy, on the (Osterc 1975; 1986; Zupani and Bole 1997). These other hand, are usually ascribed to the so-called Vla studies, although also carried out on samples of Neo-ka group, which was first described by L. H. Barfield lithic pottery, have never reached beyond the mere (1972; 1999), and is still used as a description for technological aspect of the results. This paper will certain pottery types in Italian archaeology (Gilli show that such applications are not just essential and Montagnari Koklej 1993; 1994; Montagnari for the study of technology, but useful also in the Kokelj 2001). These vessels are predominantly study of social structures in the Neolithic period and found in caves all along the Karst plateau, mostly on the palae-oeconomy in the region. the Italian side of the border, in the Triestine Karst region. Nevertheless, this group has many similari- Pottery samples of the Neolithic and Eneolithic pe-ties with the middle Neolithic Danilo culture on the riod from two cave sites near Divaa in the south Dalmatian coast. The samples from the Eneolithic west part of Slovenia, located on the Karst plateau, period were selected for comparison with the Neowere used for the archaeometrical studies in the pre-lithic pottery. Andreja ibrat Gapari As a first, step all the pottery fragments were analysed on a macroscopic level in order to describe the potsherds according to their technological and typological properties. Within this range of information, samples for a detailed mineralogical description and analysis were selected. The samples were chosen according to their technological groups, their stratigraphic position, and their typological and cultural properties. For a first provenience study, some samples of clays and sediments around the two archaeological sites were also sampled and analysed. THE ARCHAEOLOGICAL SITES The rock shelters at Mala Triglavca and Trhlovca, sited only one kilometre apart, are located on the Karst plateau near the town Divaa (Fig. 1). The caves were excavated from the mid-1970’s to the mid-1980’s by the Ljubljana Institute for Archaeology under the supervision of Dr. France Leben. In both caves there was a long stratigraphic sequence of layers dating from the Neolithic to the Modern Era; in Mala Triglavca there were also layers containing archaeological finds dating to the Mesolithic (Leben 1988). The principles of arbitrary excavation were employed, and all the material remains were documented in this context; in Mala Triglavca the finds are attributed to horizontal sections1 and in Trhlovca as layers. The Neolithic layers2 included pottery, various bone and stone artefacts, and numerous animal bones, both wild and domestic species. Wild animals yielded the majority of bones: stag, wild boar, and brown bear; among the domestic species were many sheep, goat, cattle and dog bones (Budja 1995; 1996; 2001; Leben 1967; 1988; Petru 1997; Pohar 1990). The pottery assemblage from the Neolithic and Eneolithic layers is quite modest, since only 690 fragments were found in Mala Triglavca and 785 potsherds from Trhlovca3. The assemblage included some typical ceramic vessels of the Neolithic period in this region, including bowls ornamented with triangles and tulip shaped cups (Dacar 1999; Toma 1999.18–57; ibrat 2002). A rhyton fragment has been excavated at Mala Triglavca that has similarities with the rhyta of the Danilo culture on the Dalmatian Coast (ibrat 2002.t.19:8), as well as a potsherd ornamented with barbotine, a decorating technique traditionally attributed to the Starevo culture in the central Balkans (Dacar 1999. t.18:1; ibrat 2002.t.6:7, t.12:4). METHODOLOGY The macroscopic observation We observed three main groups of ceramic matrixes on the macroscopic level (as described by Horvat 1999)4 at Mala Triglavca. The group with calcium carbonate is by far the most abundant, since 78.3% of all the samples from the Neolithic and Eneolithic period belong to this group. The group with calcium carbonate and quartz was 18.9%, and the group with quartz only 2.7% of the total assemblage that is of the 690 potsherds analysed. In the oldest Neolithic layer all the samples (i.e. 215 fragments) belong to the group with calcium carbonate. In the second Neolithic layer the group with calcium carbonate comprised 97.6%, and the group with calcium carbonate and quartz 2.4% (of 329 potsherds from this layer). The group with quartz, but no calcium carbonate, comprised 0.7%, and for the first time appears in the Eneolithic layer; the group with calcium carbonate nevertheless still predominates, with 94.5% from a total of 146 potsherds (ibrat 2002.sl. 8–9, 14–17, 60). At Trhlovca most of the potsherd from Neolithic layers H and G also belong to the group with calcium carbonate (98.5% from 68 fragments from layer H, and 98.5% from 212 fragments from layer G). The group with quartz and calcium carbonate is represented by 1.5% each in layer H and G. The vessels from the younger layer, F, also contain calcium carbonate, but in a smaller part of the assemblage 1 The assemblage from Mala Triglavca is especially problematic since the material from some of the horizontal planums is evidently mixed (ibrat 2002.68–73). 2 In Mala Triglavca the oldest Neolithic layer includes the horizontal planums 3,05–3,25m/2,90–3,25m; the second Neolithic layer includes planums 2,70–3,00m/2,70–2,90m/2,75–2,90m and the Eneolithic layer includes planums 2,70m/2,60–2,75m (ibrat 2002.60–61). In Trhlovca the layers H, G and F are all Neolithic layers, only layer F has material with similarities to the Eneolithic layer E at this site (Toma 1999.47–50) (Tab. 1). 3 The pottery assemblage from both cave sites has been already sorted by the excavation team in the 1980’s, mostly according to known typological finds. Therefore we must stress that all the later analyses on pottery were done on a smaller sample than originally excavated. 4 The technological description on the macroscopic level included the presence of mineral and organic inclusions, their size and frequency in the vessels, but also the surface treatment, hardness, colour and the firing methods and atmosphere (Horvat 1999. 159–161). Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia (79.6% from a total of 505 fragments). A higher percentage of potsherds, i.e. 19.2%, were attributed to the group with quartz and calcium carbonate on the macroscopic level in this layer. The group with quartz inclusions was 1.2%, and appears for the first time in layer F (Toma 1999. 26, sl. 9). In the pottery assemblages from both rock shelters the group with calcium carbonate inclusions predominates according to results of macroscopic observation. Only in layer F from Trhlovca does the percentage of such potsherds drop to below 90%. The group with quartz and calcium carbonate was approximately 2% at both sites, but increases in layer F in Trhlovca to 20%. This group appears only in the second Neolithic layer in Mala Triglavca, but is not present in the oldest layer. The group with quartz inclusions appears in the Eneolithic layer at Mala Triglavca and in layer F at Trhlovca for the first time, but forms only some 1% of the total assemblage in these layers. Sampling We analysed 43 pottery samples from both rock shelters (24 from Mala Triglavca and 19 from Trhlovca cave) (Tab. 1) and 6 sediment samples from different locations in the microregion (the site catchment analysis was carried out within a radius of 5 km) (Fig. 1). One of the sediment samples was taken directly from the archaeological layer at Trhlovca, the other from Divaka jama and two samples were taken from denuded caves5 called Radvanj, near the Mala Triglavca rock shelter, and Lipove doline. Also, two alluvial samples were taken from a stream, Globoki potok near the village of Dane and another from the River Reka near koflje (Tab. 4). Pottery samples were chosen from the macroscopic observations, on the basis of their stratigraphic position, the typology of the vessels, and their cultural relevance (Fig. 2). The sediments were gathered according to their proximity to the archaeological sites in question, their workability (high clay content) and origin. Only the alluvial sediments failed to meet the workability criterion, since they were mainly composed of quartz sand. The analyses Various methods of analysis were used for the determination of minerals in the pottery and sediment samples. The mineralogical composition was determined by means of optical microscopy, X-ray powder diffraction, and scanning electron microscopy. The pottery and sediment samples were analysed for their chemical composition with the inductively coupled plasma optical emission spectrometry (ICPOES) method (Bishop et al. 1982; Nφlte 2003) in the ACME Laboratory in Vancouver, Canada. The optical mineralogy and X-ray powder diffraction techniques applied at the Department of Geology in Ljubljana. For the optical microscopy we used samples, prepared as polished thin sections, which are useful for the identification of different kinds of minerals and other grains in pottery, their abundance and associations, particle orientation, void size, shapes and locations, surface treatments, and alterations due to firing or post-depositional factors. One of the more useful characterizations was granulometry and heavy mineral analysis (Bari and Tajder 1967; Grimshaw 1971; Rice 1987.348–350; Whitbread Fig. 1. Locations of the archaeological sites and the locations of the sedi- ment samples: 1 – Trhlovca; 2 – Divaka jama; 3 – dolina Radvanj; 4 – Ma- la Triglavca; 5–Lipove doline; 6–Pared near Dane; 7–koflje (river Reka). 5 Caves, in which denudation had removed their upper parts, yet are recognized as caves due to the typical sediments they contained and other features. There are three types of relief features controlled by denudation of rocks above the caves. One of the types are roofless caves, that present a longer section of passages and have been filled by flowstone and allochthonous fluvial sediments that have been deposited in a cave environment (Mihevc, Slabe and ebela 1998.167–170; Mihevc 2001.15–41). Andreja ibrat Gapari 1986). With this method we could determine the mi-under the microscope. Smaller particles of the sedneralogical composition of pottery, distinguish bet-iment samples were sieved out (i.e. particles small- ween clay pastes and temper, discover secondary er than 0.063mm or 630΅m). minerals, and estimate the firing temperature. We also analysed the sediments with optical mineralo-With the X-ray powder diffraction technique we analgy, although we could only observe individual grains ysed the bulk mineralogical composition of the pot- SAMPLES YEAR OF SAMPLING THIN SECTION NUM. SITE LAYERS DESCRIPTION 1 2003 1 Mala Triglavca 3,05-3,25m the wall of the vessel 2 2003 2 Mala Triglavca 3,05-3,25m the wall of the vessel 3 2003 3 Mala Triglavca 2,70-2,90m the wall of the vessel 4 2003 4 Mala Triglavca 2,70-3,00m the wall of the vessel 5 2003 5 Mala Triglavca 2,75-2,90m piece of plain rim with wall 6 2003 6 Mala Triglavca 2,70m piece of rim with wall 13 2000 15 Mala Triglavca 3,05-3,25m the wall of the vessel 14 2000 13 Mala Triglavca 2,70-3,00m the wall of the vessel 15 2000 3 Mala Triglavca 2,70-3,00m handle 16 2000 5 Mala Triglavca 2,70-3,00m the wall of the vessel 17 2000 10 Mala Triglavca 2,75-2,90m the wall of the vessel 18 2000 14 Mala Triglavca 2,70-2,90m the wall of the vessel 19 2000 11 Mala Triglavca 2,70-2,90m the wall of the vessel 20 2000 4 Mala Triglavca 2,90-3,05m the wall of the vessel 20 2003 144 Mala Triglavca 2,90m the wall of the vessel 21 2000 16 Mala Triglavca 2,75-2,90m the wall of the vessel 22 2000 9 Mala Triglavca 2,90-3,05m bowl 22 2003 119 Mala Triglavca 2,90-3,05m the wall of the vessel 23 2000 7 Mala Triglavca 2,70m piece of rim with wall 23 2003 68 Mala Triglavca 2,90-3,05m plate 24 2000 1 Mala Triglavca 2,90-3,05m the wall of the vessel 24 2003 148 Mala Triglavca 2,70-3,00m handle 25 2003 153 Mala Triglavca 2,70-3,00m piece of plain rim with wall 29 2003 374 Mala Triglavca 2,60-2,75m piece of base with wall 1 2000 28 Trhlovca H the wall of the vessel 2 2000 26 Trhlovca H the wall of the vessel 3 2000 27 Trhlovca H the wall of the vessel 4 2000 29 Trhlovca F,3,4 footed bowl 5 2000 30 Trhlovca G the wall of the vessel 6 2000 18 Trhlovca G the wall of the vessel 7 2003 7 Trhlovca G the wall of the vessel 8 2000 24 Trhlovca G the wall of the vessel 8 2003 8 Trhlovca G pot 9 2000 25 Trhlovca H the wall of the vessel 9 2003 9 Trhlovca G the wall of the vessel 10 2000 2 Trhlovca F piece of plain rim with wall 10 2003 10 Trhlovca G the wall of the vessel 11 2000 21 Trhlovca G bowl 11 2003 10109 Trhlovca F the wall of the vessel 12 2003 10115 Trhlovca F, E piece of plain rim with wall 13 2003 10123 Trhlovca F,D dish 14 2003 10101 Trhlovca F dish 15 2003 10066 Trhlovca G bowl Tab.1. Pottery samples for optical and x-ray analysis from Mala Triglavca and Trhlovca, the description of pottery types and their stratigraphic context. Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia tery and sediment samples (Grimshaw 1971; Klein and Hurlbut 1993). Approximately 2g of a sample was ground into fine powder for this method. All the results were presented on the computer as diagrams and detailed data. Since this technique gives the results of the bulk composition of a sample, we could not distinguish between natural and added inclusions. Nevertheless, we were able to identify most of the minerals present in the samples. The analysis with the scanning electron microscope (Goldstein et al. 2003; Reed 1996) was provided by the Institute Joef Stefan in Ljubljana at the centre for electron microscopy. Six samples of pottery from both sites have been analysed with this technique so far. We also performed a point chemical analysis using an energy dispersive spectrometer (EDS) for a detailed study, in which we were interested not only in the chemical composition of calcite and quartz grains, but also in limestone and chert grains in the samples. We tried to establish whether these grains are of a uniform composition in the pottery. We will present only some of the preliminary results obtained by electron microscopy, since only a fraction of the samples have been analysed (i.e. 5 samples). Some of the pottery samples from both archaeological sites (i.e. 10 samples from Mala Triglavca and 10 samples from Trhlovca) and all the sediment samples were sent to ACME Laboratory in Vancouver, Canada for a chemical analysis. All the samples were crushed into powder and some 5g of each sample were sent for analysis by inductively-coupled plasma emission spectrometer for major, minor, and trace elements6. In this paper we present only some preliminary results of this analysis (Tab. 5). THE RESULTS Pottery The mineralogical composition of the pottery samples contained quartz, mica (i.e. muscovite) and feldspar in all of the analysed samples (Tab. 2; 3). Hematite was determined in 6 out of 24 samples from Mala Triglavca and 7 out of 19 samples from Trhlovca. We also found grains of clay pellets, argillaceous rock fragments, limestone, chert, and quartz sandstone in various quantities by optical mineralogy (Tab. 2). Argillaceous rock fragments were present in 8 samples from Mala Triglavca and 10 samples from Trhlovca; clay pellets were present in all of the analysed samples. Limestone was present in 11 samples from Mala Triglavca, and 9 samples from Trhlovca. Chert grains were present as individual grains or as part of quartz sandstone; these grains were discovered in 18 samples from Mala Triglavca and 16 samples from Trhlovca. In most of the samples (i. e. in 22 samples from Mala Triglavca and 14 samples from Trhlovca) calcite grains7 were found in various quantities (Fig. 3), but it was most probably added as temper; only on rare occasions were calcite grains naturally included in the clay (for example, in sample 6/2003 from Mala Triglavca). In a smaller percentage of samples, secondary calcite was observed inside pores or on the surface of the vessels, i.e. in 3 samples from Mala Triglavca and in 6 samples from Trhlovca. Calcite grains were present abundantly or very abundantly in 75% of all samples from Mala Triglavca and in 42.1% of all samples from Trhlovca (Tab. 2). The grains are of angular shape, poorly sorted, and of an average size of around 0.35mm (350΅m), but they can be a few millimetres in some samples; these criteria are usually attributed to tempering materials (Rice 1987.406–411). In our opinion most of the grains, especially in the sand and pebble grain range, can be attributed to human activity, and that calcite was indeed used as a tempering material in the manufacture of the Neolithic pottery in this area. Calcite grains can be a problematic material in pottery making, because of its decomposition into quicklime in firing, and the effect of “lime popping” when the vessel is being cooled. Many solutions have been proposed for this problem, from wetting the vessels after heating, to adding salt to the paste and firing the vessels below the decomposition temperature, which means less than 850°C (Grimshaw 1971. 280; Rice 1987.97–98). This last solution applies to the pottery from Mala Triglavca and Trhlovca, since we already proved that the firing temperature for vessels with calcite temper rarely exceeded this level. Again, the question remains whether this was done deliberately, or the potters were not able to reach a higher firing temperature with their firing technology. Nevertheless, calcite can be very useful in cooking pots, as assumed by other researchers (Rice 1987.410; Rye 1981.33). 6 A major or main component means that the concentration of a given element in the sample is at least 10%; the minor component means that the concentration is between 10 and 0,01%; and a trace element means that the component has a lower concentration, less than 0,01%, therefore usually expressed in part per million (ppm) (Nφlte 2003.8). 7 The calcite grains in these samples are in fact composed of calc-sinter. Andreja ibrat Gapari Fig. 2. Neolithic pottery from Mala Triglavca (1 – sample 22/2000; 2 – sample 24/2000; 3 – sample 24/2003; ibrat 2002.t.7:4, 12:4, 19:8) and Trhlovca (4 – sample 4/2000; 5 – sample 11/2000; 6 – sample 9/2000; 7 – sample 11/2003; Dacar 1999.t.3:2, 18:1-2, 34:2). (1:2) Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia The vessels from the Neolithic and Eneolithic period have great strength8 and this may be due to two reasons: the formation of secondary calcite on the surface and inside the walls of the potsherds in post- deposition; and the relatively high content of mica, i.e. muscovite flakes in the natural pastes of these vessels. Strength is associated with the ability of a material to withstand pressure being applied to it without rupturing or deforming (Grimshaw 1971. 832). The pottery from both rock shelters shows many similarities in terms of technological properties and a more uniform mineralogical composition than previously hypothesised on the basis of macroscopic observation. The basic mineralogical composition is essentially the same in most of the vessels analysed, in those with calcite temper as well as in vessels without any tempering materials (Figs. 3, 4). Even in terms of chronological differences, we could establish almost no variation in the mineralogical composition of the ceramic pastes; the use of calcite as a temper is somewhat less popular in the Eneolithic than in the Neolithic period, but this was already determined on the macroscopic level of observation (ibrat 2002.52–58). When we consider the stylistic and typological data and the composition of such pottery, we can establish that all the typical Neolithic vessels from this region that have similarities with the Danilo culture are of a similar mineralogical and chemical composition (Fig. 2:1, 3–5). In terms of production centres and trade we can say that the Neolithic pottery from the Karst plateau was manufactured in this area by using locally available mate- Fig. 3. Thin section photomicrograph of sample 14 (2000) from Mala Triglavca. The clay matrix con- tains calcite grains (x40; || N; the black line is 250 ΅m wide). Fig. 4. Thin section photomicrograph of sample 9 (2000) from Trhlovca. The potsherd is decorated with barbotine and the clay matrix includes quartz grains and clay pellets (x40; || N; the black line is 250΅m wide). rials for all kinds of pottery. This is proved by a certain type of pottery, namely that of the barbotine decorated potsherds (Fig. 2:2, 6–7), which can be best linked to the Starevo and Starevo-Impresso culture in the eastern and central Balkans. These potsherds from Mala Triglavca and Trhlovca do not differ in any way from the rest of the ceramic assemblage and were as such also manufactured on the Karst (Tab. 2; 3). On the other hand, there is one pottery sample from Trhlovca (i.e. sample 10/2000; Tab. 1–3) – dating to the Eneolithic period – that has a somewhat different chemical composition than all the other pottery and sediment samples analysed with the inductively coupled plasma optical emission spectrometer (ICPOES). The composition of major and minor elements is similar to the other samples, but there is a significant change in the quantity of some trace elements (manganese, molybdenum and lead among others). These results could point to a different natural clay source having been used for this vessel, since none of the sampled sediments contained a similar quantity of these elements. There was a slight difference in the amount and variety of clay pellets and argillaceous rock fragments than was observed on the macroscopic and microscopic levels in this sample, but only the chemical analysis pointed to such diversity in composition. One aspect that is common in pottery analysis is archaeothermometry, i.e. the determination of the 8 Strength is a measure of the response to stresses involving the entire ceramic body, while hardness denotes deformations affecting the surface of the vessel (Rice 1987.354). Andreja ibrat Gapari temperature at which the pottery was initially fired. working firing temperature for pottery from the For the estimation of the initial firing temperature physical properties and mineralogy of the samples. we can use different techniques (Rice 1987.426– Our results show that the temperature was below 435). In our analysis we also tried to establish the 700°C for most of the pottery, because of the very SAMPLE YEAR OF SAMPLING SITE quartzquartzsandstonechertcalcitelimestonesecondaycalcitemuscoviehematitefeldsparschloriteanhydritezirconclaypelletsARF1 1 2003 Mala Triglavca A R R VA R R R P 2 2003 Mala Triglavca VA R R VA P R P R 3 2003 Mala Triglavca A A R P R R P 4 2003 Mala Triglavca VA R R A R R P R 5 2003 Mala Triglavca A R A P A 6 2003 Mala Triglavca A R R R R P R R A R 13 2000 Mala Triglavca A R A P A 14 2000 Mala Triglavca P VA R R A 15 2000 Mala Triglavca VA R R P R P R A 16 2000 Mala Triglavca P R VA R A A 17 2000 Mala Triglavca A R A P A R 18 2000 Mala Triglavca VA R P R A 19 2000 Mala Triglavca A R R A P P R A 20 2000 Mala Triglavca P A P A 20 2003 Mala Triglavca P P R R P R 21 2000 Mala Triglavca A R A R A 22 2000 Mala Triglavca P A P A 22 2003 Mala Triglavca VA R VA R P A 23 2000 Mala Triglavca A R R R R ZO R 23 2003 Mala Triglavca P R R A R R P 24 2000 Mala Triglavca R A R R A 24 2003 Mala Triglavca A R A R P 25 2003 Mala Triglavca P R VA R R P 29 2003 Mala Triglavca VA R P A R P R 1 2000 Trhlovca VA R P R R R A 2 2000 Trhlovca P P R A R R R A 3 2000 Trhlovca A R VA P P 4 2000 Trhlovca P R R P R A 5 2000 Trhlovca A R R R VA R R P R 6 2000 Trhlovca A A R P 7 2003 Trhlovca VA R P R P R 8 2000 Trhlovca R A R R A R 8 2003 Trhlovca A R R A R P P R R 9 2000 Trhlovca VA R P R A 9 2003 Trhlovca A R R VA P R R R R P 10 2000 Trhlovca VA R R P R VA P 10 2003 Trhlovca A P P P R R P P R R R A R 11 2000 Trhlovca R A R R 11 2003 Trhlovca VA R A A R 12 2003 Trhlovca VA P R P R A P 13 2003 Trhlovca A R R R P R P P R 14 2003 Trhlovca P R VA R A R P R 15 2003 Trhlovca A R P R P R P R Tab. 2. The results of the optical microscopy analysis for the pottery samples (VA – very abundant; A – abundant; P – present; R – rare; 1 – argillaceous rock fragments). Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia good state of preservation of calcite grains, which did not decompose9, as observed under the microscope. In addition, most quartz grains in our samples were cracked because of the volume change in quartz inversion around 573°C 10. Therefore, we conclude that most of the pottery was fired at low temperature, only around 600 to 700°C. Only a small number of potsherds contained calcite grains that were partly decomposed, and those vessels could have been fired to a higher temperature, that is, up to 800°C, but not higher, since the grains did not decompose completely. These firing temperatures could only be estimated for the potsherds containing calcite grains – for the rest of the pottery we do not have enough information to ascertain the firing temperature. The only guideline in the samples without calcite is quartz, and its inversions to – quartz around 573°C and tridymite around 870°C (Grimshaw 1971.158). Since tridymite was not discovered in the samples we can estimate that most of the samples were fired at around 600 to 800°C. Sediments All of the sediment samples were composed of quartz, feldspar, mica (i.e. muscovite, biotite) and chlorite, according to the mineralogical analyses (Tab. 4). Ferrous oxides, such as hematite and goethite, were discovered only in the cave sediments. A similar mineralogical composition has been attested for sediments from doline on the Karst (Zupan Hajna 1998. 279–290). Sediments from Trhlovca, Divaka jama and both of the alluvial samples showed traces of calcite or calcium oxides according to the X-ray diffraction and chemical analyses (Tab. 4; 5). The alluvial sediments contained some minor quantities of calcite, but this can be explained by the presence of mollusc shells in these samples, as was observed on the macroscopic level. Mollusc shells were not discovered in the cave sediments or the pottery sample – therefore, we conclude that these alluvial sediments were not used for pottery production at Mala Triglavca and Trhlovca. Nevertheless, all the sediments show a similar mineralogical composition, which is not surprising since those sediments all originate in the same type of rock, flysch11. Flysch rocks are common around the Karst plateau, as they are present along the Slovene coastal region, in the Vipava valley to the northeast and in the Brkini and the Reka valleys to the southeast (Gams 1988, 81–83). According to our results, the mineralogical composition of pottery from Mala Triglavca and Trhlovca can be better compared with the composition of the analysed cave sediments than with the alluvial samples (Tabs. 2–5). Our main arguments for this are the much higher quartz sand content, the lack of hematite or other ferrous minerals, and the presence of mollusc shells in the alluvial sediments, as opposed to the cave sediments and the pottery. In contrast, we have to stress that the alluvial sediments we sampled did not contain enough clay for pottery manufacture to begin with. Discussion Our results can be best compared to archaeometric studies of pottery from Edera Cave/Stenaca in the Triestine Karst (Spataro 1999) and from various open air and cave sites along the eastern Adriatic Coast (Spataro 2002). Most of the pottery analysed from these sites are Neolithic and Eneolithic. In Edera, which is located some 25 km away from our two archaeological sites, the potsherds could be divided into two main groups according to thin section and X-ray diffraction analyses. The best represented group includes vessels made from local materials, as demonstrated by the inclusion of many calcite grains; and the second group is that of imported ware. In the fabrics of the second group several chert, granite and quartz sandstone grains were detected. The granite grains, part of an igneous rock, contained minerals K-feldspar, quartz and biotite. Some traces of chlorite were also discovered (Spataro 1999.70–72). The main problem with this analysis is the fact that no real provenience studies took place, since no samples of locally available sediments, clays or rocks were analysed. The author concluded that one group was comprised of locally-made vessels only on the present of calcite in the samples. No real comparison was made between both groups on the basis of their pastes without considering calcite grains, which were probably added as temper. The pottery from layer 2a at Edera/Stenaca that was attributed to the Vlaka group (Biagi et al. 1993. 49; Spataro 1999.70–72) is typologically and stylistically similar to the vessels from Mala Triglavca and Trhlovca. The local group of clays as described at Edera/Stenaca, which contained calcite grains, and the clay pastes from the rock shelters Mala Tri 9 The calcite decomposition happens in the temperature range from 700 to 900°C (Rice 1987.98) . 10 The first inversion of quartz occurs rather rapidly around this temperature and is a chamge from low to high quartz (Rice 1987.95) . 11 Flysch rocks of this region are composed of layers of sandstone and carbonate marl. Andreja ibrat Gapari glavca and Trhlovca, also have a similar mineralogi-yer 3a in Edera/Stenaca (Biagi et al. 1993.47–49; cal composition. The presence of the non-local group Biagi and Spataro 2001.32–35; Boschin and Riewith granite grains as described in Edera/Stenaca del 2000). Whether granite grains in the clay pastes was discovered in the pottery samples from the Slo-are in fact a non-local material should have been tesvene Karst region. Some of the vessels with this non-ted with an appropriate sampling of sediments and local composition are from the late Castelnovien la-rocks in this region. SAMPLES YEAR OF SAMPLING SITE quartz calcite muscovite plagioclase K-feldspars chlorite hematite 1 2003 Mala Triglavca xx xx x x 2 2003 Mala Triglavca xx xx x x 3 2003 Mala Triglavca xx xx x x * x 4 2003 Mala Triglavca xx x x x * 5 2003 Mala Triglavca xx xx x x x 6 2003 Mala Triglavca xx x x x x x 13 2000 Mala Triglavca xx xx x x 14 2000 Mala Triglavca xx xx x x 15 2000 Mala Triglavca xx xx x x * 16 2000 Mala Triglavca xx xx x x 17 2000 Mala Triglavca xx xx x x * * 18 2000 Mala Triglavca xx xx x x * * 19 2000 Mala Triglavca xx xx x x * 20 2003 Mala Triglavca xx xx x x * 21 2000 Mala Triglavca xx xx x x * 22 2000 Mala Triglavca xx xx x x 22 2003 Mala Triglavca xx xx x x * 23 2000 Mala Triglavca xx x x x * 23 2003 Mala Triglavca xx xx x x x 24 2000 Mala Triglavca xx xx x x 24 2003 Mala Triglavca xx xx x x 25 2003 Mala Triglavca xx xx x x 29 2003 Mala Triglavca xx xx x x * x 1 2000 Trhlovca xx xx x x * 2 2000 Trhlovca xx xx x x 3 2000 Trhlovca xx xx x x 4 2000 Trhlovca xx xx x x 5 2000 Trhlovca xx xx x x * * 6 2000 Trhlovca xx xx x x * 7 2003 Trhlovca xx x x x * * 8 2000 Trhlovca xx xx x * * 8 2003 Trhlovca xx xx x * 9 2000 Trhlovca xx * x x x 9 2003 Trhlovca xx xx x x 10 2000 Trhlovca xx x x * * 10 2003 Trhlovca xx xx x x x * 11 2000 Trhlovca xx xx x x * 11 2003 Trhlovca xx x x x 12 2003 Trhlovca xx x x x * * * 13 2003 Trhlovca xx x x x * 14 2003 Trhlovca xx xx x * x 15 2003 Trhlovca xx xx x x Tab. 3. The results of the X-ray diffraction analysis (xx – major quantity; x – minor quantity; * – trace) . Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia SAMPLES SITE DESCRIPTION quartzcalcitemuscovitebiotitefeldsparsorthoclasechloriteclay mineralsgibbsitehematitegoethitepyrithe MACROSCOPIC DESCRIPTION 1 Trhlovca cave sediment brown colour; calcite; bones X X X X X X X X X 2 Divaka jama cave sediment yellow colour X X X * X * X * * 3 Dolina Radvanj cave sediment brown red colour X X X X X X 4 Lipove doline cave sediment brown red colour X X X X X X 5 Dane (Globoki potok) river sediment grey brown colour; shells X X X X X 6 Reka river sediment grey colour; X X X X X X quartz sand Tab. 4. The mineralogical composition of the sediment samples according to the optical microscopy and X-ray diffraction method (X – major component; * – trace quantity). The analysis of the rhyton from Edera/Stenaca has demonstrated its local production as well as the analysis of the rhyton from Mala Triglavca. These vessels are found in numerous caves and rock shelters on the Triestine Karst and have typological similarities with the Danilo Culture in Dalmatia and the Kakanj Culture in central Bosnia (Montagnari Kokelj and Crismani 1993). Chapman has proposed that these were salt containers and argued that their symbolic meaning could be transferred between different groups not as whole pots, but only as fragments (Chapman 2000.65–68). Nevertheless, the rhyta were probably transported as ideas, not as real vessels or parts of vessels to the northern parts of the Adriatic, the Caput Adriae, since most of the rhyta found in this region were produced locally (Spataro 2002.199). Spataro analysed the mineralogical composition of potsherds from Neolithic and Eneolithic sites on both sites of the Adriatic coast. Samples were taken from 13 sites according to their stratigraphic sequence and typology, but no preliminary macroscopic studies of all the pottery assemblage were undertaken. As the results show, most of the pottery was made from locally available material. Calcite grains and rare organic material, were used as temper in the pottery from the eastern Adriatic coast sites, mostly in the middle Neolithic period; in one sample from Vela pilja on the island Korula grog temper was also found in the pottery. (Spataro 2002.193–199). In this study Spataro sampled local clays for her provenience study, but mostly used only one sample in the proximity of a given archaeological site. The potsherds were selected according to typology and ornamentation techniques characteristic of the Impresso and Danilo cultures on the eastern Adriatic coast. So our main criticism is of sampling techniques, which could not include all the possible clay matrixes within a site, since no technological observations were made beforehand. Some mineralogical and chemical analyses of pottery samples from this region were also carried out in Slovenia. The best example is the scientific study of Roman amphorae from Sermin near Koper on the Adriatic coast, beneath the Karst plateau. The samples were analysed for their chemical and mineralogical composition using similar techniques – inductively coupled plasma emission spectrometry, X-ray powder diffraction, and optical mineralogy. For the analysed amphorae the authors proposed an Adriatic origin on the basis of the mineralogical composition of the samples and their comparison to the geological structure in the eastern Adriatic, namely the composition of flysch rocks (Zupani, Bole 1997.98–99). The mineralogical composition of the pottery and flysch rocks is indeed similar on the general level, but in our opinion for a thorough provenience study, locally available clays should have been sampled. CONCLUSIONS Within the different pottery groups we were able to identify three ‘recipes’ employed by the Neolithic potters: one had no artificially added temper (Fig. 4); the predominant group had calcite grains added as temper (Fig. 3); and one possible group had grains of lime sandstone with micritic calcite cement added as temper (sample 5/2000 from Trhlovca; Fig. 5). Even potsherds that have a clear cultural reference to the Dalmatian coast, in the Danilo culture, such as the rython (Fig. 2:3; Fig. 7), or others with Andreja ibrat Gapari SAMPLE YEAR OF SITE SiO2 Al2O3 Fe2O3 MgO CaO Na2O K2O TiO2 P2O5 MnO Cr2O3 SAMPLING (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) 1 2000 Trhlovca 49,8 12,6 6,22 1,5 12,4 0,72 2,21 0,65 0,39 0,12 0,026 2 2000 Trhlovca 34,9 11,41 5,33 1,02 23,2 0,13 1,78 0,53 0,49 0,11 0,019 3 2000 Trhlovca 45,7 11,44 2,15 0,52 18,1 0,03 1,33 0,73 0,47 0,02 0,026 4 2000 Trhlovca 43,9 13,28 5,18 1,13 15,3 0,18 2,22 0,71 0,42 0,14 0,024 5 2000 Trhlovca 58,5 12,2 5,12 1,41 9,72 0,81 1,99 0,67 0,17 0,16 0,018 6 2000 Trhlovca 38,6 9,42 2,13 0,52 25,3 0,05 1,29 0,57 0,09 0,02 0,024 8 2000 Trhlovca 33 12,48 5,69 0,98 23,2 0,06 1,85 0,51 0,38 0,13 0,015 9 2000 Trhlovca 62,6 14,94 6,79 1,69 1,85 0,7 2,77 0,79 0,71 0,18 0,031 10 2000 Trhlovca 60 15,48 8,25 0,91 2,25 0,43 1,79 1,01 0,71 1,58 0,021 11 2000 Trhlovca 30,1 13,56 5,42 1,12 21,4 0,11 1,42 0,57 1,09 0,08 0,015 13 2000 Mala Triglavca 44,6 12,52 4,88 0,83 15,4 0,16 1,58 0,72 0,97 0,15 0,025 14 2000 Mala Triglavca 28,2 11,46 5,56 0,5 23,5 0,15 0,64 0,55 0,9 0,1 0,018 15 2000 Mala Triglavca 60,7 12,95 5,84 1,04 7,14 0,87 1,71 0,79 0,55 0,1 0,03 16 2000 Mala Triglavca 36,8 11,06 4,86 0,5 22,6 0,2 1,27 0,58 0,49 0,06 0,024 17 2000 Mala Triglavca 44,1 12,59 5,11 0,85 14,8 0,19 1,54 0,77 1,66 0,14 0,029 18 2000 Mala Triglavca 51,9 13,27 5,89 0,77 9,22 0,23 0,89 0,84 1,37 0,05 0,029 19 2000 Mala Triglavca 42,9 8,72 4,36 1,18 20,2 0,78 1,36 0,52 0,24 0,14 0,018 21 2000 Mala Triglavca 48,9 11,15 4,77 1,04 13,1 0,41 1,59 0,63 0,49 0,1 0,021 22 2000 Mala Triglavca 43,2 13,2 6,83 0,89 13,6 0,16 1,73 0,7 0,88 0,21 0,026 24 2000 Mala Triglavca 35,1 11,61 5,02 1,17 22,7 0,15 1,6 0,5 0,33 0,21 0,013 1 2004 Trhlovca 42,5 15,48 6,01 1,28 9,94 0,55 1,58 0,89 1,08 0,25 0,02 2 2004 Diva_ka jama 65,6 15,38 6,64 0,77 0,9 0,19 1,27 0,89 0,12 0,16 0,038 3 2004 Dolina Radvanj 56,9 18,28 7,81 1,3 0,85 0,52 1,64 1,02 0,05 0,12 0,03 4 2004 Lipove doline 76,4 9,93 4,18 0,67 0,44 0,17 1,12 0,67 0,04 0,06 0,036 5 2004 Pared 62,8 10,35 5,68 1,26 6,22 0,58 1,77 0,56 0,07 0,65 0,024 6 2004 Reka 82,6 6,29 2,97 0,82 1,03 0,79 0,94 0,41 0,05 0,06 0,024 Tab. 5. The results of the chemical analysis for the pottery and sediment samples. Only the major oxides are presented in form of percentage of mass. a reference to the Balkans, namely barbotine (Fig. 2:2,6; Fig. 6), were made of local clays and tempering materials. Some potsherds contained grains of lime sandstone with micritic calcite cement12 (Fig. 5) in the fabric, and this type of rocks can be found over a very limited area on the plateau. The nearest location of these lime sandstones is near Tomaj, a town located northwest ofthe Divaa region and only 15 km from our two archaeological sites (Otoniar 1999.32–33). These sandstone grains were probably naturally present in the clay matrix of the pottery from Mala Triglavca and Trhlovca, but could have been added as temper according to the size and angularity of these grains13. Nevertheless, these materials point to the possibility of transhumance on the Karst plateau and the gathering of material for the vessels in different places, not just around the two rock shelters. Mala Triglavca and Trhlovca are located some 15 km away from the coastline near Trieste/Trst in Italy. Hence the distance between the coast and the plateau and the distances around the plateau rarely reach more than 30 km. All major communications in this region can be carried out over short distances. Therefore, the lime sandstone temper could have been used in pottery manufacture near the geological area of these rocks, and the finished products transported by herders to the caves around modern-day Divaa, with other goods and the herds. Transhumance, i.e. the seasonal transfer of grazing animals to different pastures, often over substantial distances, can be an interpretative postulate for these groups as was shown by other studies (Sterud 1978. 381–384; Halstead 1996.21–26). By using the scanning electron microscope for some of our pottery samples we also found a phosphorite grain in one of the pottery samples from Trhlovca. Phosphorite is a sedimentary rock with a high con 12 Lime sandstone is composed of quartz, chert and limestone grains, which are combined with micritic calcite cement. 13 The angularity of the grains can be a result of the overall hardness of such grains due to quartz inclusions. Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia Fig. 5. Thin section photomicrograph of sample 5 (2000) from Trhlovca. In the middle is a grain of lime sandstone with micritic calcite cement (x40; + N; the black line is 250΅m wide). centration of phosphates in nodular or compact masses; one type of phosphorite is coprolites, fossilised animal or fish excrement. Phosphorite can be formed inside pottery in the sediment post-deposition, or may be naturally present in the clay that the potters used for their manufacture. In some potsherds from Mala Triglavca chemical analysis revealed a higher phosphorous content and similarly, in the clay sample from the deposit in Trhlovca. Phosphorus is closely associated with animal and human activity, because bones and teeth contain large amounts of this element. High concentrations of phosphorous in the soil often accumulate where humans have congregated and have discarded the bones of wild or domestic animals (Brady, Weil 1999.540). The presence of phosphorous minerals and the element itself in our samples links the potsherds to human and animal activities that took place after the deposition of the material in the caves and rock shelters, or indicates that the clay used for the manufacture of this potsherd was perhaps taken from a place in which these activities took place, for example from Trhlovca. New soil micromorphological data from four rock shelters in the Trieste Karst (Grotta Azzurra/Peina na Leskovcu, Edera/Stenaca, Caterina/Katrna pejca and Lonza) show that all post-Mesolithic deposits were coprolitic; that is, they are made up of thoroughly disaggregated and burned herbivore droppings, mostly of sheep and goat dung. The authors’ suggestion is that shepherds in the Karst would have used the rock shelters in a system of transhumance pasturage almost exclusively as stables. The low number of remains of material culture (mainly pottery) in the layers offers some evidence for such behaviour. Furthermore, if the specialized use of rock shelters is assumed, it might be reasonable to believe that people moved around (Boschian and Montagnari-Kokelj 2000.345–350). The presence of phosphorite, of which coprolites are one of the varieties, in one of the pottery samples links our results with those made in Italy. Although detailed soil micromorphological research has not as yet been carried out in the Slovene Karst region, we may assume that results similar to those from Mala Triglavca and Trhlovca can be expected, since they both contain many animal bones and fewer artefacts, of which mainly pottery was found. Also, many white to grey ash layers are found in the stratigraphic sequence of these sites, which could contain phytoliths and coprolitic aggregates (coprolitic deposits have been proven for Mala Triglavca according to the preliminary results of the soil micromorphological data; Budja, pers.comm.). The detailed study of the mineralogical and chemical composition of the Neolithic and Eneolithic pottery from two sites around Divaa on the Karst plateau has shown that we have very uniform vessel pastes used in the pottery technology. We could find almost no difference in pottery composition within one of the sites or between the two sites, by using tempering material, at least not in the Neolithic and Eneolithic pottery from this micro-region. The use of calcite grains as the predominant temper seems to be not only a technology typical of the Neolithic period in the Karst plateau, but also typical for a wider area in this period. Calcite grains are a common temper in pottery also on the Triestine Karst in layers with the so-called Vlaka group pottery (Spataro 1999.70–72) and along the Adriatic coast in the context of the Danilo culture (Spataro 2002.197). Fig. 6. Thin section photomicrograph of the sample 24 (2000) from Mala Triglavca. The potsherd is de- corated with barbotine, The clay matrix includes calcite and few quartz grains (x40; || N; the black line is 250΅m wide). Andreja ibrat Gapari Fig. 7. Thin section photomicrograph of sample 24 (2003) from Mala Triglavca. The rhyton includes calcite and quartz grains and clay pellets (x40; || N; the black line is 250΅m wide). On the other hand, some of the pots have lime sandstone with chert grains, which proves that people were mobile, and with them went materials around the Karst plateau. Clays from the region around Tomaj will have to be sampled in the future to establish the original location of this material. Since one of the potsherds from Trhlovca (i.e. sample 10/2000; Tab.1; 5) has a different chemical composition than the other analysed pottery and sediment samples, a different natural clay source has to be assumed. For this reason sediment samples from the Slovene coast region, which is not far the two archaeological sites, would also have too be sampled. New sediment samples and analyses will therefore be needed to produce a more accurate picture of Neolithic pottery technology, mobility, and transhumance in this area. Pottery from Mala Triglavca and Trhlovca from the ACKNOWLEDGEMENTS Neolithic period was made from local clays that This article is a part of the author’s postgraduate were taken from cave sediments, but not from the work on the structural and technological analysis of deposits near the river as shown by our investiga-Neolithic pottery at the Department of Archaeology, tion. The main tempering material was calcite, a mi-Faculty of Arts in Ljubljana. This work would not neral that is abundantly available on the Karst and have been possible without the assistance and guidcan be easily removed from cave walls and crushed. ance of certain people. I am grateful to my supervi- The clay already contained a lot of mica (muscovite) sor, Dr. Mihael Budja from the Department of Archaflakes that added significantly to the overall strength eology for supporting me in my studies; also to Dr. of the pots. The materials were locally available, cal-Breda Mirti, my co-supervisor, and Dr. Meta Dobnicite is abundant in caves, and the clay pastes were kar at the Department of Geology, who helped me also taken from cave deposits as shown by our pro-with optical mineralogy and the X-ray powder difvenience study. The content of phosphorous in some fraction technique. My thanks go also to Dr. Miran samples and a phosphorite grain in one of the sam-eh from the Institute Joef Stefan for his assistance ples from Trhlovca also show that the rock shelters in the use of the scanning electron microscope. My could have been used as stables for animals by Neo-deepest thanks go also to all those people, who assis lithic shepherds. Thus we conclude that most of the ted, guided or supported me in any way through all the work but are not being mentioned. All errors in Neolithic pottery was produced on a local scale and this article, however, are my own. from locally available materials on the Karst plateau.  REFERENCES BARFIELD L. H. 1972. The First Neolithic Cultures of North-eastern Italy. In Schwabedissen H. (ed.), Die Anfδnge des Neolithikums vom Orient bis Nordeuropa. Fundamenta A/3, Teil VII, Kφln: 182–216. 1999. The Moser collection in the Naturhistorisches Museum, Vienna. Atti Soc. Preist. Protost. XI, Trieste: 19–62. BARI L. and TAJDER M. 1967. Mikrofiziografija petrogenih minerala. Zagreb. BATOVI . 1973. Odnos jadranskog primorja prema jugoistonih Alpa u neolitu i eneolitu. Arheoloki vestnik 24: 62–122. 1979. Jadranska zona. In Praistorija Jugoslavenskih zemalja II, Neolitsko doba, Sarajevo: 473– 634. BIAGI P., STARNINI E. and VOYTEK B. 1993. The late Mesolithic and Early Neolithic Settlement of Northern Italy: Recent Considerations. Poroilo o raziskova Archaeometrical analysis of Neolithic pottery from the Divaa region, Slovenia nju paleolita, neolita in eneolita v Sloveniji XXI: 45–58. BIAGI P. and SPATARO M. 2001. Plotting the evidence: some aspects of the radiocarbon chronology of the Mesolithic-Neolithic transition in the Mediterranean basin. Atti della Societa per la Preistoria e Protostoria della regione Friuli – Venezia Giulia XII: 15–54. BISHOP R. L., RANDS R. L. and HOLLEY G. R. 1982. Ceramic Compositional Analysis in Archaeological Perspective. In M. B. Schiffer (ed.), Advances in Archaeological Method and Theory, Vol. 5. Academic Press, Inc. New York. BOSCHIAN G. and MONTAGNARI-KOKELJ E. 2000. Prehistoric Shepherds and Caves in the Trieste Karst (Northeastern Italy). Geoarchaeology: An International Journal 15/4: 331–371. BOSCHIN F. and RIEDEL A. 2000. The late Mesolithic and Neolithic of the Edera Cave (Aurisiana,Trieste Karst): a preliminary report. Quaderno 8, Societa per la preistoria e protostoria della regione Friuli – Venezia Giulia, Trieste: 73–90. BRADY N. C. and WEIL R. R. 1999. The Nature and Properties of Soils. New Jersey: Prentice-Hall, Inc. BUDJA M. 1995. Neolitizacija Evrope. Slovenska perspektiva. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji XXI, 1993(1994): 163–193. 1996. Neolithization in the Caput Adriae region: between Herodotus and Cavalli-Sforza. Slovenska perspektiva. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji XXXIII: 61–76. 2001. The transition to farming in Southeast Europe: perspectives from pottery. Documenta Praehistorica XXVIII: 27–47. CHAPMAN J. 2000. Fragmentation in Archaeology. People, places and broken objects in prehistory of south-eastern Europe. London: Routledge. DACAR M. 1999. Trhlovca. Ljubljana (Diplomsko de- lo, Filozofska fakulteta, Oddelek za arheologijo). GAMS I. 1998. Kras. In I. Gams, I. Vrier (eds.), Geografija Slovenije, Ljubljana: 55–90. GILLI E. and MONTAGNARI KOKELJ E. 1993. La Grotta dei Ciclami nel Carso Triestino (materiali degli scavi 1959–61). Atti della Societa per la Preistoria e Protostoria della regione Friuli – Venezia Giulia VII: 65–162. 1994. La Grotta delle Gallerie nel Carso Triestino. Atti della Societa per la Preistoria e Protostoria della regione Friuli – Venezia Giulia VIII: 121– 194. GOLDSTEIN J. I., NEWBURY D. E., ECHLIN P., JOY D. C., LYMAN C. E., LIFSHIN E., SAWYER L. and MICHAEL J. R. 2003. Scanning Electron Microscopy and X-Ray Microanalysis. Kluwer Academic/Plenum Publishers. New York. GRIMSHAW R. W. 1971. The Chemistry and Physics of Clays and Allied Ceramic Materials. John Wiley & Sons. London. HALSTEAD P. 1996. Pastoralism or household herding? Problems of scale and specialization in early Greek animal husbandry. World Archaeology 28(1): 20–38. HORVAT M. 1999. Keramika: tehnologija keramike, tipologija lonenine, keramini arhiv. Ljubljana: Znanstveni intitut Filozofske fakultete. KLEIN C. and HURLBUT C. S. jr. 1993. Manual of Mineralogy. (21st Edition). John Wiley and Sons Inc. New York. OTONIAR B. 1999. Kamnine na Krasu – Kreda. In A. Kranjc, V. Likar and M. alik Huzjan (eds.), KRAS: pokrajina, ivljenje, ljudje. Intitut za raziskovanje krasa ZRC SAZU, Zaloba ZRC, Ljubljana: 25–33. KOROEC J. 1960. Neolit na Krasu in v Slovenskem Primorju. Zgodovinski asopis 14: 5–33. LEBEN F. 1973. Opredelitev neolitske in eneolitske keramike iz jamskih najdi Jugovzhodno alpskega prostora. Arheoloki vestnik 24: 145–161. 1988. Novoodkrite prazgodovinske plasti v jamah na Krasu. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji XVI: 65–76. MIHEVC A. 2001. Speleogeneza Divakega krasa. Ljubljana: Zaloba ZRC, ZRC SAZU. MIHEVC A., SLABE T. and EBELA S. 1998. Denuded caves – an inherited element in the Karst morphology; the case from Kras. Acta Carsologica XXVII/1: 165–174. Andreja ibrat Gapari MONTAGNARI KOKELJ E. 2001. Some comments on the Vlaka Group or ‘Gruppo dei Vasi a Coppa’ of the Trieste Karst in the light of recent data. Preistoria Alpina 34: 245–251. MONTAGNARI KOKELJ E. and CRISMANI A. 1993. La presenza di »vasi a quattro gambe« nel neolitico del Carso Triestino. Aquileia Nostra LXIV: 9–66. NΦLTE J. 2003. ICP Emission Spectrometry. A Practical Guide. Weineim: Wiley-VCH Verlag. OSTERC V. 1975. Mineralna sestava in mikrostruktura keramike kolia ob Maharskem prekopu. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji IV: 123–135. 1986. Mineralogija keramike iz Ajdovske jame I. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji XIV: 97–111. PETRU S. 1997. Metode preiskav sledov uporabe na kamenih orodjih iz najdi Zakajeni spodmol in Mala Triglavca. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji XXIV: 79–98. POHAR V. 1990. Sesalska makrofavna v starejem holocenu. Poroilo o raziskovanju paleolita, neolita in eneolita v Sloveniji XVIII: 43–47. REED S. J. B. 1996. Electron Microprobe Analysis and Scanning Electron Microscopy in Geology. Cambridge: Cambridge University Press. RICE P. M. 1987. Pottery Analysis (A Sourcebook). The University of Chicago Press. Chicago. RYE O. S. 1981. Pottery Technology. Principles and Reconstruction. Manuals on Archaeology. Washington. SHEPARD A. O. 1965. Ceramics for the archaeologist. Washington D.C: Carnegie Institution of Washington. SPATARO M. 1999. La caverna dell’edera di Aurisina (TS): studio archeometrico delle ceramiche. Atti della Societa per la Preistoria e Protostoria della regione Friuli – Venezia Giulia XI: 63–89. 2002. The First Farming Communities of the Adriatic: Pottery Production and Circulation in the Early and Middle Neolithic. Quaderno 9. Trieste: Societa per la preistoria e protostoria della regione Friuli – Venezia Giulia. STERUD E. L. 1978. Prehistoric Populations of the Dinaric Alps: An Investigation of Interregional Interaction. In Redman C. L., Langhorne W. T., Berman M. J., Versaggi N. M., Curtin E. V. and Wanser J. C. (eds.), Social Archaeology. Beyond Subsistence and Dating, Academic Press, Inc., New York: 381–409. TOMA A. 1999. asovna in prostorska strukturiranost neolitskega lonarstva: Bela Krajina, Ljubljansko Barje, Dinarski Kras. Ljubljana (Magistrsko delo, Filozofska fakulteta, Oddelek za arheologijo). TURK I. 1980. Zaasno poroilo o najdbah favne v jami Triglavci pri Divai. Arhiv Intituta za arheologijo ZRC SAZU, Inv. t. E 340, Ljubljana. WHITBREAD I. K. 1986. The Characterisation of argillaceous inclusions in ceramic thin sections. Archaeometry 28/1: 79–88. ZUPANI N. and BOLE M. 1997. The Chemical and Mineralogical Analyses of the Amphorae. In J. Horvat (ed.), Sermin. Prazgodovinska in zgodnjerimska naselbina v severozahodni Istri, Ljubljana: Znastvenoraziskovalni center SAZU: 83–99. ZUPAN HAJNA N. 1998. Mineral composition of clastic sediments in some dolines along the new motorway Divaa-Kozina. Acta Carsologica XXVII/1: 277–296. IBRAT A. 2002. Neolitska in eneolitska lonenina iz najdia Mala Triglavca pri Divai. Ljubljana (Diplomsko delo, Filozofska fakulteta, Oddelek za arheologijo). cont ent s