Introduction  介绍

Metal is very attractive for decorative, representational, instrumental, offensive or ritual purposes in both modern and ancient time. Moreover, metal is relatively easy to process into some small accessories. Among them, the bracelet is one of the most common ornaments on the wrist or ankle, which is circular or oval in shape. Technically, the bracelet was first made of bone or stone [1]. After entering Bronze Age, the ancestors began to cast or make bracelets from metal. Specifically, the ratio between the diameter of the bracelet and the diameter of its metal wire or the width of its metal sheet is usually greater than 5, which is the main feature that can be roughly distinguished from the armband [2]. In addition, bracelets differ in many aspects such as pattern, size, gap, cross-sectional shape, etc.
金属无论在现代还是古代，都具有极强的装饰性、表象性、工具性、攻击性或礼仪性等作用，而且金属比较容易加工成一些小的饰物，其中手镯是最常见的腕上或脚踝上的饰物之一，其形状为圆形或椭圆形。从技术上讲，手镯最早是由骨头或石头制成的[ 1 ]，进入青铜时代后，先民们开始用金属铸造或制作手镯。具体来说，手镯的直径与其金属丝的直径或金属片的宽度之比通常大于5，这是可以粗略地与臂章区别开来的主要特征[ 2 ]。此外，手镯在式样、大小、间隙、横截面形状等许多方面都有所不同。

In China, the earliest known metal bracelets appeared in some ruins of Qijia culture (around 420–3600 BC) and Siba culture (around 3900–3400 BC) in the northwest [3, 4]. After preliminary analysis, it was found that most of them were cast, and the raw materials were red copper or arsenic copper. In the ensuing time, the metal bracelets only spread southward along the crescent-shaped cultural-communication belt, and there was almost no trace in the Central Plains before Han Dynasty [5]. Especially when Southwest China entered bronze age, bronze bracelets began to appear on a large scale and became a prominent feature of this region [6].
在中国，已知最早的金属手镯出现在西北地区的齐家文化（约公元前420～3600年）和四巴文化（约公元前3900～3400年）的一些遗址中[ 3,4 ] 。初步分析发现，它们大多是铸造的，原料是红铜或砷铜。此后，金属手镯只是沿着月牙形文化交流带向南传播，在汉代以前的中原地区几乎没有踪迹[ 5 ]。特别是当西南地区进入青铜时代后，青铜手镯开始大规模出现，并成为这一地区的一个显著特征[ 6 ]。

Archaeological research in Southwest China in the pre-Han period is currently receiving increasing attention. Some of the results achieved in several relatively large areas have expanded our knowledge of various aspects of southwestern societies during the first millennium BC, as well as the prehistoric communications between this region and areas outside its borders [7,8,9]. Meanwhile, the information provided by metallurgical archaeology has also played a pivotal role. However, most previous research has focused on larger bronzes such as drums and weapons, with little attention paid to smaller objects involving accessories and ornaments [10]. As a portable item, the bracelet is more personalized and more capable of embodying cultural and technical exchange [11].
当前，西南地区汉前考古研究正受到越来越多的重视，在几个较大区域取得的一些成果拓展了我们对公元前一千年西南社会各方面的认识，以及这一地区与境外的史前交流[ 7，8，9 ] 。同时，冶金考古提供的信息也发挥了举足轻重的作用。但前人研究大多集中在鼓、兵器等体型较大的青铜器上，对配饰、饰物等较小器物的关注较少[ 10 ]。手镯作为随身物品，更具个性化，更能体现文化和技术交流[ 11 ]。

Therefore, this paper focuses on the technical status of the bronze bracelets in the southwest before the Han Dynasty. At that time, this region was quite closed and less subject to the Central Plains influence. The analysis of a batch of bracelets from the Fenjiwan site in Huili County can help us to better understand the local bronze culture. By revealing alloy characteristics and raw material sources, and combining relevant scientific data, some issues about the production status, technical level, use of raw material, workmanship exchange and even external interactions in Southwestern China can be discussed in more depth.
因此，本文着重研究汉代以前西南地区青铜手镯的技术状况。当时，这一地区相对封闭，受中原影响较小。对会理县粉鸡湾遗址一批手镯的分析，有助于我们更好地了解当地的青铜文化。通过揭示合金特性和原料来源，结合相关科学资料，可以更深入地探讨西南地区的生产状况、技术水平、原料使用、工艺交流乃至对外互动等问题。

The archaeological context and samples
考古背景和样本

The background of Huili  惠立的背景

Huili has convenient water and land transportation, and has been a turning point in the trade and commerce between southwest Sichuan, western Yunnan, and South Asia since ancient times. Known as the the ‘lock of Sichuan and Yunnan’, it is across the Jinsha River from Yunnan Province and is the gateway and fortress of the Southern Silk Road. In addition, Huili could also reach eastward to Chongqing by the Jinsha River. Previous archaeological investigations have found stone models for casting copper spears and dagger-axe in Huili, which indicated that local casting activities existed as early as the Warring States Period [12, 13].
会理水陆交通便利，自古以来就是川西南、滇西与南亚之间商贸往来的枢纽，素有“川滇锁关”之称，隔金沙江与云南相望，是南方丝绸之路的门户和要塞。此外，会理还可经金沙江向东直达重庆。前期考古发现，会理曾出土铸造铜矛、戈等石模，表明当地早在战国时期就有铸造活动[ 12,13 ]。

As recorded in Historical Records (史记), ‘Of the dozens of chiefs north of Dian (滇), the most important is Qiongdu (邛都)’. Book of the Later Han (后汉书) depicted that Qiongdu was conquered by Emperor Wu of Han, and the land became an administrative unit of the Han Dynasty. Historians have concluded that the Qiongdu people’s scope of activity was centered on today’s Xichang City (西昌), Liangshan Yi Autonomous Prefecture (凉山彝族自治州), Sichuan Province and Huili County was on the southern border [6].
《史记》载：“滇北数十酋，首以邛都。” 《后汉书》记载，邛都为汉武帝所灭，成为汉朝的一个行政区。史学家认为，邛都人的活动范围以今四川省凉山彝族自治州西昌市为中心，南部边境为会理县 [ 6 ]。

The archaeological site  考古遗址

The Fenjiwan site is located in Lixi Town, Huili County, within 5 km from the Jinsha River (Fig. 1). Huili County Office for the Preservation of Ancient Monuments revealed 150 densely arranged earthen-pit burials. The walls of these tombs are straight and neat. Generally, the tombs are elongated rectangular in shape, 2 to 4 m long and 0.4 to 0.6 m wide, and up to 2 m deep. Judging from the remaims, the burial style is mainly extended supine, and no funeral implements are found. Pottery occupies a large proportion in funerary objects, accounting for about 90%. Each tomb is usually buried with 1 to 4 pieces of pottery, a few tombs up to 10 pieces. Most of the funerary objects are placed at the head of the tomb, occasionally in the middle of the tomb. Black smoke marks on most potteries indicate that they were practical utensils in life. The yellow color characteristic of the local clay suggests that the ceramic vessels were locally produced. Besides, the number of bronze artifacts is small and most of them are crudely made. The main types of bronze are spears, knives, bracelets and ornaments [12, 14]. According to typology analysis, these tombs should be in the Warring States period, and most of the objects seem to come from local or nearby areas [14, 15].
分鸡湾遗址位于会理县礼溪镇，距金沙江5公里（图1 ）。会理县古迹保护所发现150座密集排列的土坑墓。这些墓葬墓壁挺拔整齐，墓葬一般呈长方形，长2～4米，宽0.4～0.6米，深可达2米。从遗迹看，墓葬形式以仰卧长葬为主，未发现陪葬品。随葬品中陶器所占比例较大，约占90%。通常每个墓葬随葬陶器1～4件，少数墓葬随葬陶器达10余件。随葬品大多置于墓首，偶有置于中部。多数陶器上有黑色烟痕，表明为生活用具。当地黏土的黄色特征表明这些陶器是当地生产的。此外，青铜器数量较少，且大多制作粗糙。青铜器的主要类型是矛、刀、镯和装饰品[ 12,14 ]。根据类型学分析，这些墓葬应为战国时期，大多数器物似乎来自当地或附近地区[ 14,15 ] 。

Fig. 1  图 1
figure 1

The geographical location of Qiongdu (today’s Huili) and other find spots of bracelets discussed in this paper
本文讨论的邛都（今会理县）的地理位置及其他手镯发现地

Full size image

It is worth noting that gray-green flat-round pebbles were found in almost all tombs. There are only one pebble object in most tombs, while two or three in a few tombs. This funeral custom is very rare in the tombs of the same period in Southwest China. This discovery coincides with the record of Huayangguo Zhi (华阳国志), which is a local chronicle book completed around 350 AD, specialized in ancient China’s southwestern region. It should be that the recorder was also impressed by this unique burial custom and felt the need to take notes [16].
值得注意的是，几乎所有墓葬中都发现了灰绿色扁圆形卵石，大多数墓葬中只有一枚卵石，少数墓葬中有两枚或三枚卵石。这种丧葬习俗在西南地区同时期墓葬中十分少见。这一发现与《华阳国志》的记载相吻合。《华阳国志》是一部成书于公元350年前后，专门记载中国古代西南地区的地方志。应该记录者也对这种独特的丧葬习俗印象深刻，并觉得有必要做些记录[ 16 ]。

The bronze bracelet and slag samples
青铜手镯和炉渣样品

The bracelets from the Fenjiwan site are made of plain sheet metal without decoration (Fig. 2). They are relatively uniform in size, leaving a certain gap, and are around 6 cm in diameter. The width and thickness of these bracelets range from 0.1 cm to 0.3 cm, and no more tham 0.5 cm. In addition, the cross sectional shape of all the bracelets is rectangular rather than circular or elliptical. This very popular type in Southwest China makes it difficult to determine whether they are influenced by any particular cultural factors [11, 15].
分鸡湾遗址出土的手镯均为素板状，无任何纹饰（图2 ）。手镯大小比较均匀，留有一定间隙，直径约6厘米。手镯的宽度和厚度在0.1厘米至0.3厘米之间，最厚的不超过0.5厘米。此外，手镯的横截面形状均为矩形，而非圆形或椭圆形。这种类型在西南地区十分流行，很难判断它们是否受到某种特定文化因素的影响[ 11,15 ]。

Fig. 2  图 2
figure 2

The typical pattern of bracelet in this study
本研究中手镯的典型图案

Full size image

In addition, for data comparison, Huili County Office for the Preservation of Ancient Monuments provides nine ancient slags (around the 15th century) from Raojiadi (饶家地). It is also located in Lixi Town, Huili County, less than 2 km from the Fenjiwan site and therefore these slags can adequately reflect the information of local deposits and geochemical background.
另外，为便于数据比对，会理县古迹保护办公室提供了饶家地遗址的9件古代炉渣（年代大约为15世纪），该遗址也位于会理县里溪镇，距离粉鸡湾遗址不到2公里，因此这些炉渣可以充分反映当地矿床和地球化学背景的信息。

Analytical methods and results
分析方法和结果

Component analyses  成分分析

Since these bronze bracelet are intact and pretty precious, pieces removing or sampling is not encouraged. By the chance of cultural relic restoration, a small part of the surface area of the bracelets was polished until the metal base was exposed. Using a hand-held XRF (Niton XL3t 900He by Thermo Fisher Scientific, Billerica, USA) to obtain the alloy compositions instantly. Slags data were acquired in a similar way. The main filter operates at voltage of 40 kV and current of 100 μA. The alloy mode was selected for bronzes and mining mode for slags. The X-ray beam spot on the sample was about 3 mm in diameter. Elemental data were collected with acquisition time set to 120 s. Final results was obtained by averaging three random analyses.
由于这些青铜手镯保存完整且十分珍贵，因此不鼓励将其拆下或取样。借着文物修复的机会，对手镯的一小部分表面进行了抛光，直至露出金属底部。使用手持式XRF（美国比勒里卡赛默飞世尔科技公司的Niton XL3t 900He）即时获得合金成分。炉渣数据以类似方式获取。主滤波器在40 kV的电压和100 μA的电流下工作。青铜器选择合金模式，炉渣选择采矿模式。样品上的X射线束斑直径约为3 mm。元素数据采集时间设置为120 s。通过对三次随机分析进行平均获得最终结果。

Lead isotope analyses  铅同位素分析

Because the analysis of lead isotopes requires very small amounts of samples and does not need to touch the metallic body of the samples, some metal powders are collected for analysis, which are mainly scraped off after polishing to the metal base. Lead isotope analyses were carried out at School of Archaeology and Museology, Peking University in Beijing, China and by a MC-ICP-MS (VG AXIOM, Thermo-Elemental Inc., Winsford, England). Firstly, about 2 mg of bronze powders needed to be totally dissolved in mixed solution of 3 ml of HCl and 1 ml of HNO3. Later, the clear solution was leached and diluted with deionized water to 10 ml. The solutions were then measured to detect the lead contents by ICP-AES (PHD, Leeman Labs Inc., California, USA). According to the results representing the lead contents, the solutions were diluted to 1000 ppb. The thallium (T1) standard SRM997 was added in the solutions. Lead isotope analyses were carried out by a MC-ICP-MS (VG AXIOM, Thermo-Elemental Inc., Winsford, England). The spectrometer is a double focusing magnetic sector instrument equipped with an array of 10 variable Faraday collectors. And it has a further fixed Faraday and an electron multiplier detector. Based on repeated analysis of SRM981, the overall analytical 2σ error for all lead isotope ratios was less than 0.106% (Table 1). The results are in good agreement with published values.
由于铅同位素分析所需样品量极小且不需要接触样品的金属本体，因此会收集一些金属粉末进行分析，这些金属粉末主要在抛光至金属基体后刮掉。铅同位素分析在北京大学考古文博学院进行，采用 MC-ICP-MS（VG AXIOM，Thermo-Elemental Inc.，Winsford，英国）。首先，将约 2 mg 青铜粉末完全溶解在 3 ml HCl 和 1 ml HNO 3的混合溶液中。然后，将澄清溶液浸出并用去离子水稀释至 10 ml。然后用 ICP-AES（PHD，Leeman Labs Inc.，加利福尼亚州，美国）测量溶液以检测铅含量。根据代表铅含量的结果，将溶液稀释至 1000 ppb。在溶液中加入铊（T1）标准品 SRM997。铅同位素分析采用 MC-ICP-MS（VG AXIOM，Thermo-Elemental Inc.，Winsford，英国）进行。该光谱仪为双聚焦扇形磁场仪，配备 10 个可变法拉第收集器阵列。它还配有一个固定法拉第和电子倍增器检测器。基于对 SRM981 的重复分析，所有铅同位素比值的总体分析 2σ 误差小于 0.106%（表 1）。结果与已发表的值一致。

Table 1 SRM981 solution was used as standard reference material in MC-ICP-MS analysis. The results of three runs for the SRM981 determination, analytical error and published values from [17] are shown
表 1 SRM981 溶液在 MC-ICP-MS 分析中用作标准参考材料。SRM981 测定的三次运行结果、分析误差和来自 [ 17 ] 的公布值如下所示
Full size table

Result and discussion  结果与讨论

Chemical compositions in bracelets
手链中的化学成分

The pXRF analysis indicates that the bracelets have a concentrated chemical composition with a copper content of about 80% and a tin content of 15%. Based on lead content, they can be divided into two categories (Table 2). Six of them belong to the first category (HL01 ~ 06), with lead content roughly between 2 to 3%. While the lead content of the other category (HL07 ~ 10) do not exceed 0.6%. Some scholars propose that when the lead content is greater than 2%, it should be regarded as an intentional alloy composition [18]. Conversely, lead with low amount indicates that it comes from the original copper ore. According to such criterion, these bracelets have completely different alloy systems. One is Cu-Sn–Pb ternary alloy, and the other is Cu-Sn binary metal.
pXRF分析表明该批手镯化学成分集中，铜含量约为80%，锡含量为15%。根据铅含量可分为两类（表2 ）。其中6件属于第一类（HL01～06），铅含量大致在2～3%之间，而另一类（HL07～10）的铅含量不超过0.6%。有学者认为，当铅含量大于2%时，应视为故意加入的合金成分[ 18 ]，反之，铅含量较低则表明来自原始铜矿石。按照这样的标准，这批手镯属于完全不同的合金体系，一类为Cu-Sn-Pb三元合金，一类为Cu-Sn二元金属。

Table 2 The alloy compositions of bronze bracelets in the pre-Han period excavated from Southwest China (wt%, Blank indicates under detection limit)
表2 西南地区出土汉代前期青铜手镯的合金成分（wt%，空白表示低于检测限）
Full size table

Moreover, we have also collected all published chemical composition data of bronze bracelets unearthed in the Southwest China before the Han Dynasty (Table 2, Fig. 3). The results shows that the alloy types of these bracelets are complex and the bronze techniques are diverse. Some bracelets even applied special craftsmanship. For instance, tin plating was found in a bracelet from Eryuan (data in the lower right corner of Fig. 3) [20]. Unlike the early copper bracelets in the Northwest China, the bracelets in the southwest are more often added with tin. However, some bracelets from Deqin and Yanyuan are made of copper, which reflect the primitive nature of the material processing [19]. The geographical location of the two places is also relatively close to the north, supporting the idea that metal bracelets were transported from the northwest to the southwest [11]. The emergence and popularity of Cu-Sn alloys in small objects shows that it was known that adding tin could lower the melting point and obtain better properties of the bronze, such as improving the hardness [24]. Furthermore, certain other metallic minerals like lead and arsenic (e.g. two bracelets from Yanyuan) were probably added intentionally or the raw metal ores rich in these elements were deliberately selected to improve the malleability or to enhance the color [23]. However, even with intentional addition of tin and lead, the proportions vary considerably.
此外，我们还收集了所有已发表的西南地区汉代以前出土青铜手镯的化学成分数据（表2 、图3 ），结果显示这些手镯的合金类型复杂，青铜工艺多样，有的甚至采用了特殊的工艺，如洱源出土的一件手镯就镀了锡（数据见图3右下角）[ 20 ]。与西北地区早期的铜手镯不同，西南地区的手镯更多时候添加锡，但德钦、盐源出土的一些手镯以铜为主，体现了材料加工的原始性质[ 19 ]。两地的地理位置也比较靠北，支持金属手镯是从西北向西南运输的观点[ 11 ]。铜锡合金在小器物中的出现和流行表明，添加锡可以降低青铜的熔点并获得更好的性能，例如提高硬度 [ 24 ]。此外，某些其他金属矿物（如铅和砷）（例如来自燕源的两个手镯）可能是故意添加的，或者故意选择富含这些元素的金属矿石原料来提高延展性或增强颜色 [ 23 ]。然而，即使是有意添加锡和铅，其比例也会有很大差异。

Fig. 3  图 3
figure 3

The Sn–Pb contents of bronze bracelets before Han Dynasty unearthed in Southwest China
西南地区出土汉代以前青铜手镯的锡铅含量

Full size image

In general, metal bracelets do not contain more than 16% tin (the outlier in Fig. 3 is affected by tin plating). After simulating the forging process of bronze with 5–30% tin content, it was noted that copper-tin binary alloys have two ductile forging zones: bronze with tin content less than 18% at 200–300 °C and 20–30% tin at 500–700 °C. The former is more suitable for some degree of thermal processing [25]. From this point of view, the ancestors in Southwest China may have recognized this rule of thumb. Lower machining temperatures are not only easier to achieve and control, but also allow sufficient time for the forging process to further ensure a good structure and performance [26]. Moreover, the tin content of the bracelets unearthed from different places is relatively concentrated and has its own range. The bracelets from Huili, Luliang and Eryuan (only partially) contain relatively similar tin content. And the Huili bracelets with low lead content have almostly the same alloy ratio as the Luliang bracelet and two Yanyuan bracelets, suggesting that potential technical exchanges have taken in these places.
一般而言，金属手镯的含锡量不超过16%（图3中的异常值受镀锡影响）。模拟含锡量5%~30%青铜的锻造过程后发现，铜锡二元合金存在两个延性锻造区：含锡量小于18%的青铜在200~300 ℃，含锡量20%~30%的青铜在500~700 ℃。前者更适合某种程度的热加工[ 25 ]。从这个角度来看，中国西南地区的先民可能已经认识到这个经验法则，较低的加工温度不仅更容易实现和控制，还能为锻造过程提供充足的时间，进一步保证良好的组织和性能[ 26 ]。而且，不同地点出土手镯的锡含量相对集中，有各自的范围。会理、陆良和洱源（仅部分）出土的手镯含锡量相对相似。而铅含量较低的回黎手镯，其合金比例与吕梁手镯及两件盐源手镯基本相同，暗示这些地方曾经发生过潜在的技术交流。

However, the alloy composition of Huili bracelets is more regular and concentrated than that of bracelets from other regions, indicating a stable craft characteristic and good bronze-making skill. It is noteworthy that the difference in lead content is not only found on Huili bracelets. Similar situations arise in Chuxiong Wanjiaba and Jianchuan Haimenkou, where the fluctuations in lead content are more dramatic, which may reflect the experimental stage of adding lead material into bronzes at that time [21, 24]. According to the statistics of the alloy composition analysis of ancient Chinese bronzes by previous scholars, the biggest feature of the alloy technology of bronzes in the Central Plains after Shang Dynasty is that a large amount of lead was added to the bronze, and the lead content is usually above 5%. While most of the bronzes in Southwest China are of little lead, mainly tin bronzes [10]. This shows that the bronze technology route in Southwest China at that time was from Cu–Sn binary alloy to Cu–Sn–Pb ternary alloy.
但会理手镯的合金成分相较于其他地区手镯更加规则、集中，体现出稳定的工艺特征和良好的青铜制作技术。值得注意的是，铅含量的差异不仅存在于会理手镯上，楚雄万家坝和剑川海门口也有类似的情况，铅含量的波动更为剧烈，可能反映了当时青铜器中添加铅材料的实验阶段[ 21,24 ]。从前辈学者对中国古代青铜器合金成分分析的统计来看，商代以后中原青铜器的合金技术最大的特点就是在青铜器中加入大量的铅，铅含量通常在5%以上，而西南地区青铜器大多含铅量较少，以锡青铜为主[ 10 ]。这说明当时西南地区的青铜技术路线是从Cu-Sn二元合金到Cu-Sn-Pb三元合金。

As for slags, the main chemical element results reveal generally high iron content (Table 3). Regarding to ancient metallurgical industry of China, the iron content of copper slag is relatively high, which is more than 20%, while the iron content of iron slag is generally less than 10% [27]. Accordingly, it can be considered that the slags may be related to copper smelting. The lead content of all slag samples from Raojiadi is below detection limit which demonstrates that the raw metal ores in Huili should not contain high levels of lead. The lead isotopes of these slags reflect the characteristics of the copper deposits. In fact, even if these slags are independent with copper smelting, they can also reflect the geochemical characteristics of the local metallic minerals to a certain extent.
炉渣主要化学元素分析结果普遍显示铁含量较高(表3 )。相对于中国古代冶金工业，铜渣含铁量较高，可达20%以上，而铁渣含铁量一般在10%以下[ 27 ]，因此可以认为该类炉渣可能与铜冶炼有关。绕家地所有炉渣样品的铅含量均低于检出限，说明会理地区原矿不应含有高浓度的铅。该类炉渣的铅同位素组成反映了铜矿床的特征。事实上，即使该类炉渣与铜冶炼无关，也能在一定程度上反映当地金属矿物的地球化学特征。

Table 3 Main chemical compositions of slags collected from Raojiadi, Huili County (wt %) (Blank indicates under detection limit)
表3 会理县绕家底矿渣主要化学成分（wt%）（空白表示低于检测限）
Full size table

Lead isotopic result  铅同位素结果

The results of lead isotope analysis of Huili bronze bracelets are given in Table 4. Comparing these LIA data with those of the late Shang bronzes in ancient China, they do not indicate highly radiogenic lead [28]. These lead isotope ratios need to be treated in two parts. One part reflects the characteristics of the lead ore, and the other part reveals the characteristics of the copper ore. However, the isotope data is also divided into two groups without overlap, and is completely consistent with the classification based on chemical composition. The first group (HL01-06) ranges from 17.5 to 17.7 for 206Pb/204Pb, 0.88 to 0.89 for 207Pb/206Pb and 2.16 to 2.19 for 208Pb/206Pb. And the second group (HL07-10) fluctuates from 18.2 to 18.6 for 206Pb/204Pb, 0.84 to 0.86 for 207Pb/206Pb and 2.06 to 2.10 for 208Pb/206Pb.
会理青铜手镯的铅同位素分析结果见表4。与中国商代晚期青铜器的铅同位素分析数据相比，并不代表具有高度的放射性成因铅[28]。这些铅同位素比值需要分为两部分处理，一部分反映铅矿石的特征，另一部分反映铜矿石的特征。但同位素数据也分为两组，互不重叠，与按化学成分分类完全一致。第一组（HL01-06）的206 Pb/ 204 Pb为17.5～17.7，207 Pb/ 206 Pb为0.88～0.89，208 Pb/ 206 Pb为2.16～2.19 。第二组（HL07-10） 206 Pb/ 204 Pb 比值波动于 18.2 至 18.6 之间， 207 Pb/ 206 Pb 比值波动于 0.84 至 0.86 之间， 208 Pb/ 206 Pb比值波动于 2.06 至 2.10 之间。

Table 4 Lead isotope ratios of the bronze bracelets and slags from Huili County
表4 会理县青铜手镯及炉渣的铅同位素比值
Full size table

Most of the lead isotope data collected for lead-bearing minerals in Yunnan are concentrated in the following ranges: 18.0 to 18.5 for 206Pb/204Pb, 0.84 to 0.86 for 207Pb/206Pb, and 2.05 to 2.15 for 208Pb/206Pb (region A in Fig. 4), which indicates that these deposits are very similar in terms of metallogenic history and ore evolution [10]. Based on geochemical provinces theory and the latest summary of lead isotope distribution in China, the first group (HL01-06) is definitely not the lead material produced in the Southwestern China, but from the Yangtze river basin. While the second group (HL07-10) belongs to South China, similar to copper in Yunnan and southern Sichuan [29, 30].
云南地区含铅矿物的铅同位素数据大多集中在以下几个范围： 206 Pb/ 204 Pb为18.0～18.5，207 Pb / 206 Pb为0.84～0.86，208 Pb/ 206 Pb为2.05～2.15 （图4中区域A），表明这些矿床在成矿历史和成矿演化方面非常相似[ 10 ]。根据地球化学省理论和中国铅同位素分布的最新总结，第一类（HL01—06）肯定不是产自中国西南地区的铅物质，而是来自长江流域；而第二类（HL07—10）则属于华南地区，与云南、川南地区的铜矿类似[ 29,30 ]。

Fig. 4  图 4
figure 4

The lead isotope ratios of bronze bracelets and slags as well as copper ores from different localities in Huili County
会理县不同产地青铜手镯、炉渣及铜矿石的铅同位素比值

Full size image

As recorded in Huayangguo Zhi, Qiongdu was famous for producing copper ores in Han Dynasty. It is likely to be developed and utilized earlier than that time. More importantly, there are many copper mines within Huili County, such as the Lala deposit, the Datongkuang deposit, the Xiaoqingshan deposit, and the Tianbaoshan deposit etc. [31,32,33,34]. Thereinto, the Lala deposit produces copper ores with a wide range of lead isotopes, which are usually highly radiogenic. The same situation occurs only occasionally in the other Huili copper deposits. In Fig. 4, we plot only the common lead data for the above four copper deposits and the slags collected near the Fenjiwan site. The ellipse region C represents the extent of lowly radiogenic galena in Southwest China (according to [10, 30]). Since the distribution of lead isotopes in copper is always wider than that in lead, it is easy to understand that region C is covered by region A.
《华阳国志》记载，邛都早在汉代就以盛产铜矿而闻名，其开发利用的时间可能更早。更重要的是，会理县境内分布着拉拉矿床、大通矿床、小青山矿床、天宝山矿床等多处铜矿[ 31，32，33，34 ] 。其中，拉拉矿床的铜矿石铅同位素组成范围广泛，且通常具有强放射性，而同样的情况在其他会理铜矿床中只是偶尔出现。在图4中，我们只绘制了上述四个铜矿床常见的铅数据以及粉鸡湾遗址附近采集的炉渣。椭圆区域C表示中国西南地区低放射性方铅矿的范围（据[ 10，30 ]）。由于铜中铅同位素的分布始终比铅中的分布宽，因此很容易理解区域C被区域A覆盖。

The low-lead group of the bracelets falls into the region A rather than region C, again suggesting that the lead content is not from galena, but from copper ores. Although the data of these bracelets do not exaty match the lead isotope data for the entire southwest China (mainly Yunnan and southern Sichuan), but within the encirclement of the data of slags and ores from Huili (the isotopic ratios of the four bracelets can be derived from a linear combination of the data of slags and ores). Given the particularly large fluctuations in lead isotope of copper ores in this area, it is likely that the data of some small deposits in Huili may be consistent with that of the copper bracelets. Thus, these bracelets should be made of local copper.
低铅组手镯属于A区而非C区，也说明这些铅不是来自方铅矿，而是来自铜矿石。虽然这些手镯的数据与整个西南地区（主要是云南和川南地区）的铅同位素数据并不完全吻合，但都在会理矿渣和矿石数据的包围圈内（4个手镯的同位素比值可以通过矿渣和矿石数据的线性组合得到）。考虑到该地区铜矿石的铅同位素波动特别大，会理一些小矿床的数据可能与铜手镯的数据相一致。因此，这些手镯应该是当地的铜制品。

On the other hand, region B in Fig. 4 represents the predominated lead isotope ratios in bronzes in central China and middle reaches of the Yangtze River during the late Warring States period [35, 36]. Another group of bracelets completely falls within this range. Besides, we specially collected some LIA data of Yibi coins (蚁鼻钱) of ancient Chu State, as a typical example, for comparison with the high-lead group (Fig. 4). The Yibi coin, which is a unique currency with distinctive Chu style widely used during the Warring States period [37]. Considering that the coinage activities were strictly controlled by the central authority, the Yibi coins should be minted within the territory of Chu with domestic resources. The lead isotope ratios of the bracelets and Chu coins are similar. The three data points of the bracelets almost coincide with one data of the coin. Both the bracelets and the coins belong to region B, which indicates that the lead materials used for the high-lead group of Huili bracelets were probably from Chu at that time.
而图4中的区域 B 则代表了战国晚期华中及长江中游地区青铜器中铅同位素比值的优势区域 [ 35 , 36 ]，另一组手镯完全属于这个范围。此外，我们还专门收集了楚国蚁鼻钱的 LIA 数据，作为典型例子，与高铅组（图4 ）进行比较。蚁鼻钱是战国时期广泛使用的一种具有鲜明楚国风格的独特货币 [ 37 ]。考虑到铸币活动受到中央政府的严格控制，蚁鼻钱应是在楚国境内利用国内资源铸造的。手镯与楚钱的铅同位素比值相似，手镯的 3 个数据点与楚钱的一个数据几乎重合。镯子和钱币均属于B区域，这表明回黎镯子高铅组所用的铅料可能来自当时的楚国。

Other archaeological and historical evidence can also support this judgment. Some traces of the suspected Chu culture have been found on certain types and decorative motifs of bronze wares unearthed in Yunnan [38], especially in some weapons (bronze axe and sword), clothing (hair ornament, hair style and feathered decoration) and musical instruments, etc. [39]. According to Historical Records, King Qingxiang of Chu sent a military force to the southwest in the 3rd century BC. A general of Chu, Zhuang Qiao (庄蹻), reached the Dian Lake as part of the Chu military campaign. When Chu’s homeland was invaded by Qin, Zhuang Qiao decided to stay in Yunnan and adopt indigenous ways to establish the Dian kingdom. In the process, it is likely that some artifacts and culture of Chu were brought into the southwest. However, since the exact date of the Fenjiwan sites could not be determined, the above statement is only a reasonable conjecture. Previous studies of bronzes unearthed in southern Sichuan have shown that there were multiple cultural interactions within Southwest China [40]. Nevertheless, from the perspective of lead isotopes, it can be confirmed that during the Warring States period, the southwest region had certain contacts and exchanges with the Chu state and even the Central Plains region.
其他考古学和历史证据也可以支持这一判断。在云南出土的某些青铜器类型和纹饰中，发现了一些疑似楚文化的痕迹[ 38 ]，特别是在一些武器（青铜斧和剑）、服饰（头饰、发髻和羽饰）和乐器等中[ 39 ]。据《史记》记载，公元前3世纪，楚庆襄王出兵西南，楚将庄蹻随楚军到了滇池。秦军入侵后，庄蹻决定留在云南，采用土著方式建立滇国。在这个过程中，很可能一些楚文物和文化被带入了西南地区。但由于分鸡湾遗址的具体年代无法确定，上述说法只是一种合理的推测。前人对川南地区出土青铜器的研究表明，西南地区内部存在多次文化互动[ 40 ]，但从铅同位素角度可以证实，战国时期西南地区与楚国乃至中原地区存在一定的接触与交流。

Conclusions  结论

In this paper, ten bronze bracelets and nine slags from Huili County have been analyzed. All bracelets contain about 80% copper and 15% tin, suitable for low temperature forging. It shows the stable process characteristics and production levels. According to the level of lead content, these bronze bracelets can be divided into two groups, which coincide with the classification by lead isotope. One group used local copper from Huili without adding lead; while the other group utilized lead material from the middle reaches of the Yangtze River.
本文对会理县出土的10件青铜手镯和9件炉渣进行了分析，所有手镯含铜量均在80%左右，含锡量在15%左右，适合低温锻造，工艺特点和生产水平稳定。根据含铅量的高低，可将这些青铜手镯分为两组，这与按铅同位素分类相一致，一组采用会理本地的铜，未添加铅；另一组采用长江中游的铅料。

Combined with other data of bracelets in Southwest China, it is clear that there are few bracelets directly made of copper, mostly tin bronze, revealing that at that time the ancestors may have recognized the experience of adding tin material to improve alloy properties. But the low lead content shows that the bronze technology route in Southwest China is from Cu-Sn binary alloys to Cu-Sn–Pb ternary alloys. Analysis of the sources of lead suggests that the southwest region had some contacts and exchanges with the the middle reaches of the Yangtze River and even the Central Plains region during the Warring States period. These scientific data provide new evidence for speculation beyond previous style analyses and typology comparisons.
结合西南地区其他手镯资料可知，直接用铜制成的手镯很少，多为锡青铜，揭示当时先民可能已经认识到添加锡材料改善合金性能的经验。但铅含量较低表明西南地区青铜技术路线是从Cu-Sn二元合金到Cu-Sn–Pb三元合金。铅的来源分析表明，战国时期西南地区与长江中游乃至中原地区存在一定的接触和交流。这些科学数据为以往风格分析和类型学比较以外的推测提供了新的证据。

Certainly, more research is needed to verify this conjecture. To further investigate, is is necessary to conduct more scientific analyses of other bronzes in Southwest China.
当然，这一推测还需要更多的研究来验证，为了进一步探究，还需要对西南地区其他青铜器进行更多的科学分析。