Socketed axes are a well-documented category of finds in the Balt territory during the Roman Period, the Migration Period, and the Early Middle Ages. They have their direct prototypes in the Early Iron Age in the West Balt Barrow Culture, as indicated by their substantial size and morphology, especially the bulge of the socket and the way the edge tapers to the axe base (Fig. 1). Socketed axes recorded in the Bogaczewo and Sudovian Cultures (List 1) have sockets of different depth, usually reaching the base of the edge; the socket end is usually wedge-shaped or flat. The sockets are round in cross-section and carefully finished, although sometimes a longitudinal seam is noticeable. None retain horizontal rivets or nails used to fasten the handle. The length of the axes ranges between 10 and 16 cm, socket diameter – between 2.3 and 4.4 cm. The blade is flared out on both sides, usually symmetrically (this does not apply to the majority of the later finds from Lithuania). Adzes Some finds are distinguished from socketed axes by the transverse asymmetry of the working edge, i.e., the fact that the tool is visibly flattened on its outer side (the edge flares on one side only), as well as the fact that the working part is bent outwards in its central part which identifies them as wood working tools – adzes (List 2). Also the fact that the socket is square in cross-section suggests an adze: this kind of hafting is more secure as it prevents the tool from revolving during work. It seems that this was much more probable for smaller tools which were more susceptible to tensions occurring during repeated revolving actions, which may take place when working wood, e.g., when gouging. This does not mean that a round socket prevented the tool from being used as an adze. Some adzes have unfinished sockets, or even ones incomplete in their bottom part: apparently, this was due to their careless manufacture. Quite often the edge is asymmetrical and hence the top part is shaped unevenly. The adzes made by the Balts were also quite small, not exceeding 10 cm in length. This is true and concerns also of the majority of known ‘barbarian’ adzes from the Roman Period. Longer specimens with symmetrical edges, and sometimes, round-sectioned sockets are very rare, e.g., the two finds from sacrificial lake site on Lake Lubanowo, Gryfino County in the territory of the Lubusz Group and the adze from the bog site at Żarnowiec, Puck County, attributed to the Wielbark Culture. Based on the described morphological attributes nine finds from the Bogaczewo Culture area have been identified as adzes (Fig. 2:a–c; List 2:1–6). They have quadrangular sockets, often unclosed (Fig. 2:c; List 2:4.1, 2:5). The asymmetry is sometimes so pronounced that the tool definitely cannot have been used for cutting but only for gouging (List 2:4.2). The Bogaczewo Culture adzes differ from socketed axes by their significantly smaller length, which is usually 7–8 cm. Another characteristic feature useful in woodworking is also the fan-shaped edge. The context of discovery of these tools does not discount their being adzes: they have been found together with weapons (Fig. 2:a.c; List 2:1, 2:3, 2:5, 2:6) but also in assemblages with weapons and a plane (List 2:4.3), or without weapons, but with a plane (List 2:4.2). Naturally, they may have been used for combat, but it is rather a secondary function. Adzes have been recorded also the Sudovian Culture at Szwajcaria (Fig. 2:d–f, 3). They are quite small (their length 8.5, 9.8 and 11.3 cm) and light (resp., 0.13, 0.09 and 0.12 kg, whereas most axes weigh more than 0.25 kg); with the asymmetry of the edge visible from the side and a polygonal socket. Socketed adzes continue practically in an unchanged form in, i.a., the La Tène, Pre-Roman and Roman Periods in different cultural units of West, Central and North Europe. They are documented for an exceptionally broad chronological and territorial spectrum and may be considered a very popular form of a woodworking tool in the Iron Age. At present their dating in the Bogaczewo and Sudovian Cultures takes in the mature phase of the Early Roman and the beginning of the Younger Roman Period but this may be changed by future research. Typology and chronology of socketed axes Socketed axes from the Bogaczewo Culture (Fig. 6) have sockets of a different depth, but usually reaching to the beginning of the working edge, which is either wedge-shaped or flat. The length of the axe heads ranges between 10 and 14 cm. They may be divided into three types. Type I includes four specimens with a massive socket, clearly thicker in its central part, with a waist between the socket and the edge, which is fan-shaped (Fig. 4; List 1:13.2). Type II includes eight forms similar to those of Group I but with an evenly tapering socket to a well-defined edge giving the artefact an hour-glass shape (Fig. 5:a–c; List 1:4, 1:5, 1:6.4, 1:9, 1:11.1, 1:16). Type III includes two axes with parallel or almost parallel walls of the socket, with the edge gently sloping from the socket; in both specimens the entrance of the socket was slanting, which can, however, be considered as a distinctive feature (Fig. 5:d.e). In some cases the identification as type (II/III) is only tentative (List 1:10, 1:11.2), in others – unfeasible due to the lack of sufficient data (List 1:2, 1:3, 1:8, 1:14, 1:15). Some of the axes retain traces of production in the form of a seam which proves that the iron was bent to form the socket (see Fig. 4:a, 5:d). Based on the collected material it may be assumed that the form of the socketed axes did not change much in time. So far they have been discovered in assemblages dated to phases B1b–B2a (List 1:11.1), B1c–B2a (Fig. 5:b), B2 (Fig. 5:e), B2b (List 1:13.2) and B2b–B2/C1 (List 1:15). At present it does not seem feasible to narrow down the dating of the respective groups of socketed axes, although it may be supposed that type I, massive specimens with sockets of a large diameter, derives directly from forms known in the West Balt Barrow Culture (Fig. 1:b.c) where similar axes have been found. In each case the chronology may be broader and the absence of respective forms in different periods may be due to the insufficient investigation of settlements, but also to the ‘filtering’ by the burial rites. The latter is associated with the discontinuation of the practice of placing weapons in graves in the Bogaczewo Culture in the mature phase of the Younger Roman Period, probably under the influence of the Wielbark Culture burial rites. In comparison to the finds from Bogaczewo Culture, the axes recorded in the Sudovian Culture have a broader spectrum of dimensions: besides the smaller forms, ranging from 10 to 14 cm, there are substantially larger specimens: 15.9 cm (Fig. 7:d) and 17.5 cm (Fig. 7:e; also List 1:17.2). Also the morphology of the axes differs. So far no finds of type I axes have been made in the Sudovian Culture. The unthickened sockets and generally slimmer shape may have been due to the chronological, not only cultural, differences, however, due to the scarcity of the well-dated assemblages from the Bogaczewo Culture, this observation must remain hypothetical, and may be verified by future studies. The Sudovian Culture has yielded two specimens of type II (List 1:18, 1:21). The waist in the socket marking out the beginning of the edge is not as prominent in axes type II as in the specimens from the Bogaczewo Culture. For that reason three axes were attributed to type II/III (Fig. 7:a; also List 1:17.1 & 2, 1:19.3). It should be noted in one specimen has a socket with a polygonal cross-section (Fig. 7:a), which is untypical, but on the evidence of its other morphological features it has been determined as a socketed axe. In comparison to the Bogaczewo axes the sockets had smaller diameters and the edges were quite wide. Type III is the most numerous (Fig. 7:b–d; also List 1:19.5, 1:20.2); sockets of these specimens from Sudovian Culture were slimmer and had smaller diameters than the more squat artefacts from the Bogaczewo Culture. Two specimens from Szwajcaria retain seams and traces of hammering of the socket (Fig. 7:d; also List 1:19.2 & 5). One of the axes has a slightly asymmetrical edge with a marked beard (Fig. 7:e) and does not fit in the presented scheme, but fulfils the criteria for type 2 after A. Malonaitis (2008). This may be an indication of the influence from the territory of Lithuania justified by the geographical closeness. The Sudovian socketed axes are usually slimmer and often longer than their Bogaczewo Culture equivalents; besides, their sockets have smaller diameters (2.3–3.6 cm on the outside whereas those of the Bogaczewo Culture are 3.5–4.4 cm). This may be due to the fact that a different, more secure fastening was used, e.g., leather straps or wooden wedges. On the other hand it seems justified to assume that in the Sudovian Culture (in other other words from the beginning of the Younger Roman Period) hafts from some other more durable material were used, allowing sockets with smaller diameters. Axes in the Sudovian Culture are typical of the earlier phases of this cultural unit, i.e., the Younger Roman Period. The functions of socketed axes Based on the discoveries of socketed axes with preserved hafts made in at Danish bog sites from the Pre-Roman and Roman Periods one may assume that also the Balt socketed axes had knee-shaped hafts both unipartite and composite, i.e. two-piece ones (the latter are knowns from Vimose and Hjortspring – Fig. 9:6). The context of discovery of the Scandinavian socketed axes may raise some doubts as to their function (although they may have been used in battle, they were more likely designed for camp work, grubbing, or other purposes which arose during the military expeditions; one should not neglect boat-building here) but the situation of the West Balt circle is more clear. The socketed axes frequently appear in burial assemblages here, including inhumations. The location of the socketed axes may suggest how they were hafted: they had knee-shaped hafts bent at an angle of ca 70–80˚ to the axis of the head. Also, bearded axes were hafted at a similar angle, which suggests the of function of socketed axes and bearded axes. The edge was placed vertically, which excludes the possibility of their being adzes. The case of grave 312 from Marvelė, raj. Kaunas in the Central Lithuanian Group is particularly interesting (Fig. 10:a): the edge of the axe rested next to the shoulder belt type Vidgirdiai, taken off and placed above the head of the deceased. In the Lithuanian material there is evidently a greater correlation between shoulder belts type Vidgirdiai and socketed axes than for battle knives and shafted weapons. Thus it may seem that these belts were used for suspending weapons, including the socketed axes, and the connection with a warrior’s belt would confirm the military use of these axes. Another important premise as to the function comes from the Sudovian cemetery at Szwajcaria, grave S.12 (List 1:19.4). It is important to note the clear spatial division of the grave goods at this site (Fig. 10:b). Household tools (a spoon-bit auger and a sickle-shaped knife) were placed next to the right leg of the skeleton resting supine. The weapons, on the other hand (the head of a shafted weapon, the possible arrowhead and the socketed axe) and tools usually suspended from the belt of the warrior (a knife, a bar-shaped fire steel) had been deposited next to and above the shoulder of the deceased. The location of the socketed axe suggests its symbolic connection with a military activity rather than an agricultural or crafts activity. The West Balt circle is not uniform as regards the contexts in which the socketed axes were found. They have been recorded almost without exception in weapon graves. Whereas in the Bogaczewo and Sudovian cultures they are not accompanied by other tools for woodworking, in the Dollkeim-Kovrovo Culture this happens quite often which was tentatively explained by K. Raddatz by the exceptional role played in the region by woodworking, comparable to the status of smiths in the Celtic and La Tène worlds; in his opinion the Balts from Sambia, Natangia, and Nadrovia would have represents a specific form of Holzkultur. Although it is impossible to establish the purpose of the socketed axes, ultimately, the context of their discovery strongly suggests that indeed they were used as weapons, like the bearded axes.