Next Article in Journal
Reframing Place Identity for Traditional Village Conservation: A Theoretical Model with Evidence from Dali Dong Village
Previous Article in Journal
Analysis of the Experiences of Visitors the Museum Offer of Tourist Destination
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Heritage
Volume 8
Issue 10
10.3390/heritage8100426
heritage-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Preserved 800-Year-Old Liquid Beer in a Jin Dynasty Vase: Evidence of Malted Sorghum–Wheat Fermentation in Xi’an, China

by
Li Liu
1,*,
Xinyi Liu
2,
Chunlei Yu
3 and
Yifei Miao
3
1
Department of East Asian Languages and Cultures, Stanford University, Stanford, CA 94305, USA
2
Department of Anthropology, Washington University in St. Louis, St. Louis, MO 63130, USA
3
Shaanxi Provincial Academy of Archaeology, Xi’an 710054, China
*
Author to whom correspondence should be addressed.
Heritage 2025, 8(10), 426; https://doi.org/10.3390/heritage8100426
Submission received: 6 June 2025 / Revised: 28 September 2025 / Accepted: 30 September 2025 / Published: 9 October 2025
Browse Figures
Versions Notes

Abstract

This study investigates a rare case of liquid alcohol preserved in a glazed ceramic vase from the tomb of Li Jurou (AD 1226), Jin dynasty, Xi’an, China, to provide new insights into medieval brewing traditions. We employed a multi-proxy approach combining microfossil and isotopic analysis, experimental brewing with sorghum, and incorporated previously published proteomic data to illuminate its origin. Microfossil analysis revealed yeast cells and starch granules with damage patterns diagnostic of enzymatic saccharification and mashing, indicating the use of malted sorghum and wheat, alongside cooked rice and foxnut. The starch damage features observed in the archaeological sample are consistent with patterns documented in experimental beer brewing with sorghum and wheat/barley. Stable isotope analysis yielded a δ13C value of –18.5‰, consistent with mixed C3 and C4 inputs. Two-component isotopic modeling revealed that C4 plant (likely sorghum) contributed 40–50% of the ingredients, with C3 plants such as wheat, rice, and foxnut making up the remainder. These findings align with proteomic results identifying sorghum proteins in the liquid. The combined evidence distinguishes this beverage from qu-based fermentation and links it instead to li-type brewing, rooted in malted cereals and associated with ritual practices. This represents the earliest direct archaeological evidence of sorghum beer in China, highlighting both technological innovation and cultural adaptation in historical alcohol production.

1. Introduction

In recent years, advances in archaeological science have greatly accelerated research on ancient fermented alcoholic beverages in China, revealing the origin of this technique to be from 10,000–9000 years ago, at the onset of the Neolithic era in the Lower Yangzi River region [1,2,3]. By 9000–7000 cal. BP, alcoholic fermentation techniques spread rapidly northward to the Yellow River region [4,5,6], and they continued to flourish throughout the Middle and Late Neolithic periods (ca. 7000–3800 cal. BP) across a very broad area extending from the east coast to the northeast margin of the Tibetan Plateau [7,8,9,10,11,12,13,14,15]. This tradition reached a new height during the early dynastic period (ca. 3800 BP—AD 200), when bronze vessels came into frequent use for serving and storing alcoholic beverages [4,16,17,18,19]. The dominant brewing methods involved producing cereal-based beer, using either malted cereals or a qu-starter containing filamentous fungi that generate saccharifying enzymes. The main ingredients included rice and millets, along with other cereals and tubers summarized in Ref. [20].
Previous studies on ancient Chinese alcoholic beverages have concentrated primarily on solid residues adhering to pottery surfaces and liquid remains preserved in bronze vessels. With the rise in porcelain technologies—particularly during the Tang and Song dynasties (AD 618–1279)—glazed ceramics gradually became the predominant form of drinking vessel [21]. This technological shift not only reduced the likelihood of preserving liquid alcohol but also led to limited scholarly attention to the solid residues found in glazed vessels. As a result, the long-term development of brewing practices remains insufficiently understood. To help fill this gap, the present study investigates a rare case of liquid preserved in a glazed ceramic vase from the Jin dynasty tomb of Li Jurou, offering fresh insights into medieval brewing traditions.

2. Archaeological Background and Previous Studies of the Residue

2.1. Archaeological Background

The Jin dynasty (1115–1234) was established by the Jurchens, a people from the northeast. It was founded by Wanyan Aguda after overthrowing the Liao dynasty, with the dynastic name “Jin” (Gold) chosen to contrast with the Khitan’s “Liao” (Iron). The Jin rapidly expanded its territory and, in 1127, launched the Jingkang Incident, capturing the Northern Song emperors Huizong and Qinzong and bringing the Northern Song to an end. The Jin rulers gradually adopted policies of Sinicization, implementing Confucian education, instituting the civil service examination, and establishing their capital in the Central Plains, thereby absorbing many aspects of Chinese institutions and culture. In the early 13th century, the Mongols rose to power; under Genghis Khan, they launched attacks on the Jin. The dynasty eventually fell in 1234 when it was defeated by a joint Mongol–Southern Song force [22].
The Li Jurou tomb examined in this study is located in Guanyinmiao Village, Yanta District, Xi’an. It is a small brick chamber tomb from the Jin dynasty, found undisturbed and excavated in 2014 by archeologists from Shaanxi Provincial Institute of Archaeology. The tomb features a vertical shaft leading to a brick-vaulted chamber in a T-shaped layout, comprising a tomb passage, a sealed doorway, and the main chamber. Within the chamber, a coffin platform is accompanied by several ceramic lamps and ritual vessels, all arranged in a compact, modest configuration (Figure 1A,B) [23].
There is no evidence of looting or human disturbance of the artifacts prior to excavation. However, some natural movements appear to have displaced certain artifacts from their original positions, likely as a result of the 1556 earthquake centered in Huaxian County, approximately 75 km northeast of the Li Jurou tomb [23].
The burial yielded a complete Yaozhou kiln liquor set, including a Yuhuchun vase, two small cups, and a rectangular celadon tray. In addition, a pair of Yaozhou kiln meiping (plum vases) was unearthed, each measuring 32.5 cm in height, 10.6 cm in body diameter, 4 cm at the rim, and 8 cm at the base. One vase (M1:16) was found leaning against the chamber wall in a corner, likely having been placed upright originally but displaced by the earthquake. Its clay seal on the mouth remained intact, and liquid was still preserved inside (Figure 1C). This assemblage of drinking wares offers valuable insights into mid-level elite drinking practices during the Jin dynasty [23].
Upon discovery, the vase mouth, still bearing its original clay seal, was immediately enclosed with newly purchased clean plastic sheeting, secured with string, placed in a box, and transported to the Jingwei Archaeological Station of the Shaanxi Provincial Institute of Archaeology for storage.
A brick tomb deed (maidiquan) bearing a 226-character inscription was recovered from the burial. The text outlines the personal history of Li Jurou and records that his interment took place in the third year of the Zhengda reign (AD 1226). According to this record, he had served as Transport Commissioner of the Eastern Route of Shaanxi and acting Minister of the Six Boards, holding a senior third-rank official title. Despite his former status, Li Jurou was hastily buried [23]. The tomb was built during the late Jin dynasty, a time of political crisis: by 1226, the Mongols had already occupied much of the north [22]. The burial practice thus reflects the upheavals of the era.

2.2. Previous Study of the Liquid

After excavation, vase M1:16 was stored at room temperatures in the facilities of the Jingwei Station of Shaanxi Provincial Institute of Archaeology. It was first unsealed in 2018 to extract a liquid sample for scientific analysis by researchers at the Cultural Heritage Protection Center of Northwestern Polytechnical University. This study used protein-based mass spectrometry to determine the contents of the liquid. The researchers applied SDS-PAGE followed by LC-ESI-MS/MS to identify peptides surviving in the ancient liquid. Three species-specific proteins were identified: a glycosyltransferase from Sorghum bicolor (suggesting sorghum as the raw material), a cytochrome b-c1 subunit from yeast Nadsonia fulvescens, and a calcium-dependent protein kinase from yeast Wickerhamomyces ciferrii—both known for alcoholic fermentation. These findings provide direct molecular evidence that the vase contained sorghum-based liquor produced through yeast fermentation, confirming the vessel’s historical use for storing alcoholic beverages. This discovery represented the first confirmation of sorghum-based brewing practices in ancient China [24]. However, the specific brewing method remained unidentified.

3. Materials and Methods

To investigate the brewing method of this beverage, we conducted additional analyses using different methodologies. In 2019, through the Shaanxi Provincial Institute of Archaeology, we obtained a small subsample of the liquid from the vase M1:16 (Figure 1D). The vase was unsealed, and a small amount of liquid was poured into a clean glass container and sealed with a plastic cap (Figure 1D). This sample has been stored in a refrigerator maintained at 1.7–3.3 °C, designated for fermentation-related materials at the Stanford Archaeology Center.
This study employed two analytical approaches: microfossil analysis and isotopic testing. The liquid sample is hereafter referred to as the LJR liquid.

3.1. Microfossil Analysis

The purpose of the microfossil analysis was to detect microbotanical remains (starch grains and phytoliths) and fungal remains (molds and yeast cells) potentially associated with brewing activities, thereby inferring the brewing methods and ingredients utilized. This approach has been successfully applied to recover evidence of alcohol brewing by analyzing solid residues adhering to fermentation-related pottery vessels [1,5,8], as well as liquid remains preserved in bronze containers [16,18]. This part of the research was conducted at the Stanford Archaeology Center.
Since the vase was sealed and placed on the brick floor of the tomb chamber rather than buried in soil, the LJR liquid was unlikely to have been contaminated within the tomb. Post-excavation handling was carried out with great care to prevent contamination. Consequently, it was unnecessary to treat the samples with a heavy liquid (e.g., sodium polytungstate), as is commonly performed to separate microfossils from sediments in sediment-containing dry residue samples. Instead, we employed a non-chemical recovery method previously used successfully to analyze liquid from a bronze bottle dating to ca. 2300 BP [16].
Three microscope slides were prepared. Approximately 50 µL of liquid was directly extracted from the bottom of the sample using a pipette and deposited onto a microscope slide. A coverslip was then applied, and the sample sealed with nail polish. Two slides were prepared using this method.
Given that the LJR liquid had been previously identified as a fermented cereal beverage, it was expected that fermentation would have caused substantial damage to the starch grains, including extensive gelatinization that could obscure morphological features. To address this issue, we employed trypan blue staining [25], a method that selectively stains the outer layers of damaged starch granules blue, while leaving intact granules unstained. This staining method was applied on the third slide as follows: 30 µL of the liquid was dropped onto a slide, allowed to dry slightly, stained with 30 µL of trypan blue, covered with a coverslip, sealed, and observed after 15 min.
Microscopic observation was conducted using a Zeiss Axio Scope A.1 optical microscope at magnifications of 200× and 400×. Images were captured with a Zeiss Axiocam HRc3 digital camera and processed with Zeiss AxioVision software (version 4.9.1).

3.2. Identification of Fermented Starches Through Experimental Brewing and Ethnographic Observations

3.2.1. Experimental Brewing for Starch Identification

For the current project, we conducted a brewing experiment with malted sorghum, involving 3 days of sprouting, 2 h of mashing at 65 °C, and 10 days of fermentation, to document the resulting damage patterns on starch granules. The results show a progressive increase in enzymatic digestion damage over the course of fermentation, providing a comparative reference for the LJR liquid (see Section 4.1.2).
Starch taxonomic identification in this study was conducted according to established protocols outlined in earlier published studies, with particular attention to morphological features observed under light microscopy. The diagnostic features include granule shape and size, as well as general characteristics such as the hilum, fissures, lamellae, and extinction cross [26] (pp. 34–42).
To identify starch morphological modifications resulting from fermentation and cooking, we have conducted a series of fermentation experiments at Stanford Archaeology Center, primarily focusing on barley, wheat, millet, and rice. In this paper, we draw on results from these experiments [27,28], along with other published experimental studies [29,30]. We also incorporate ethnographic observation of traditional malt-based fermentation practices preserved in remote rural areas in north Shaanxi, China [5].

3.2.2. Identifying Traditional Fermentation Methods in China

Fermented beverages in ancient China primarily fall into two categories based on their brewing methods. One method is qu-based fermentation, commonly referred to as jiu in ancient texts [31]. This technique involves two sets of raw materials: (1) a qu starter, composed of a mixture of filamentous fungi (molds), yeast, and uncooked or semi-cooked cereals; and (2) the main fermentation ingredients, typically consisting of steamed or boiled cereals combined with starch-rich tubers. These two components produce distinct starch damage patterns. The cooked ingredients are primarily represented by heavily gelatinized starches, while the uncooked components in the qu starter exhibit features indicative of enzymatic digestion [28,30].
The second method involves the use of malted cereals, referred to as li in ancient texts, where malting and mashing convert starch into sugars, followed by yeast fermentation to produce alcohol [31]. Based on experimental studies, during saccharification, starch granules undergo enzymatic digestion during malting and further degradation from mashing in low-temperature hot water (65–70 °C). These combined processes lead to distinctive morphological alterations observable under microscopy, including pitting, deep fissures, exposed lamellae, broken edges, loss of birefringent crosses, central depressions, mild gelatinization, and hollowed centers with relatively intact peripheries [27].
This controlled malting–mashing experiment, however, documented only the morphological changes in malted cereals, without considering the scenario of mixing malts with additional cooked ingredients—a traditional brewing method still practiced in northern Shaanxi, called hunjiu (turbid beer), which commonly employs crops such as wheat and maize grown in north China today. The brewing process can be divided into five main steps: (1) grinding malted wheat or maize (Zea mays) into flour to serve as the saccharification starter; (2) steaming broomcorn millet (Panicum miliaceum) flour to form a cake; (3) mixing the millet cake with the wheat or maize malt flour; (4) placing the mixture in a mashing jar, adding boiled water, and stirring it into a paste; and (5) covering the jar and keeping it in a warm place for 24 h to complete fermentation. Finally, water is added to the paste, which is then brought to a boil and served hot. The resulting beverage is yellowish, porridge-like, low in alcohol, and slightly sour and sweet in taste ([5]: Supporting Information, Section 2). Based on our analysis, starches in this beverage exhibit clear signs of enzymatic hydrolysis (originating from malted wheat or maize) and extensive gelatinization (resulting from steamed broomcorn millet), while showing no evidence of mold activity.
Identifying fermentation techniques from ancient residues poses significant analytical challenges. A key distinction between qu-based and malt-based fermentation lies in the types of damaged starches and microbial remains preserved. Qu-based fermentation typically produces microfossil evidence of fermentation-related molds—such as hyphae, fungal spores, sporangia, and cleistothecia—along with enzymatically damaged starches and heavily gelatinized starch aggregates. In previous experimental studies, the molds are primarily identified as Monascus and Rhizopus [28,30]. In contrast, malt-based fermented beverages are characterized by starch granules with hollowed centers, intact peripheries, and slight expansion from mild gelatinization—features indicative of malting and mashing processes [27]—but lack evidence of mold activity. Additionally, malt-based beverages may contain a higher concentration of husk phytoliths than qu-based ones, especially in early Neolithic contexts where dehusking may have been incomplete [5]. If cooked starchy foods were incorporated into malt-based fermentation, the presence of heavily gelatinized starches would also be expected.

3.3. Identification of Yeast Cells in Fermented Beverages

Regardless of the brewing method, yeast is essential for converting sugars into alcohol; its presence is therefore a key indicator of alcoholic fermentation. Yeast cells can survive and be preserved in the residues of fermented beverages, and their morphology can be directly examined under a light microscope.
Traditional brewing utilizes various yeast species, with Saccharomyces cerevisiae being the most common. These cells are typically round to oval, measuring approximately 5–10 µm in length. Yeast primarily reproduces by budding, whereby a small protrusion develops on the cell surface, enlarges to the size of a mature cell, and then separates from the parent to form an independent cell [32,33] (pp. 143–149) (Figure 2(A1)).
Our brewing experiments show that yeast cells typically retain a clearly defined cell wall, and although some internal organelles may still be visible, they cannot be readily identified. The presence of cells in the budding stage is the most reliable criterion for identifying yeast; however, these are infrequently preserved in archaeological residues. Accordingly, the observation of round or oval cells—particularly when two or more remain connected—provides strong evidence for yeast. We also observed that, after fermentation is complete, yeast cells often occur in clusters (Figure 2(A2,A3)).

3.4. Isotopic Analysis

We conducted isotopic analysis at Washington University in St. Louis to determine the carbon and nitrogen isotope compositions of the substance and to infer its original source. For pretreatment, we followed a protocol commonly used in isotopic analysis within the beer industry to enable comparison with published isotopic values of modern beers [34]. Ten mL of the ancient beverage was placed into a glass Petri dish and dried at 60 °C until no liquid remained. The resulting residue was then ground into a fine powder using a mortar and pestle. Approximately 2 mg of the powdered sample was weighed into a tin capsule for analysis.
Combustion was carried out using an Elemental Analyzer coupled with a Thermo Delta V Advantage Continuous Flow Isotope Ratio Mass Spectrometer (EA-IRMS), housed in the Department of Earth, Environmental, and Planetary Sciences at Washington University. Raw δ13C and δ15N values were normalized to VPDB and AIR standards, respectively, using international reference materials USGS 40 (δ13C = –26.4‰, δ15N = –4.5‰) and USGS 41a (δ13C = 36.6‰, δ15N = 47.6‰). Two in-house standards were used for linearity correction and as check standards: acetanilide (δ13C = –29.5‰, δ15N = 47.6‰) and a laboratory protein standard (δ13C = –27.0‰, δ15N = 5.9‰). No replicates were conducted due to the limited mass after pre-treatment.

4. Results

4.1. Results of Microfossil Analysis

The residue sample contained abundant microfossils, primarily yeast cells and starch granules, with no phytolith or mold present. This association of microorganisms, coupled with the absence of mold activity, strongly indicates that the liquid was malt-based rather than qu-based.

4.1.1. Yeast Remains

A total of 66 yeast cells were identified, representing two morphological types: spherical or oval forms (N = 56; 4.22–11.5 μm in length) and elongated oval forms (N = 10; 7.48–12.68 μm in length) (Table 1). The cells occur either as single units or in clusters. Several exhibit budding characteristics, indicated by small protrusions emerging from the parent cell or by pairs of cells remaining connected, suggesting an active stage of reproduction comparable to modern reference examples (Figure 2(4–9); cf. 2 and 3).
According to the aforementioned mass spectrometry analysis, the LJR liquid contained two yeast species: W. ciferrii and N. fulvescens [24]. The former is typically characterized by spherical or oval cells [35], whereas the latter displays elongated oval forms [36]. Although these morphotypes correspond to those observed in our microscopic analysis, the taxonomical identities of the yeasts cannot be determined based on morphology alone.

4.1.2. Starch Remains

A total of 160 starch granules were recorded, most of which exhibited damage characteristic of enzymatic hydrolysis and gelatinization, including central cavities, broken edges, loss of extinction crosses, mild gelatinous deformation, and aggregation into gelatinized masses. On the trypan blue-stained slide, the starch granules displayed clearer outlines, enabling more confident species-level identification. Relatively intact starch grains were classified into four types, morphologically corresponding to sorghum, Triticeae trabe, rice, and foxnut (N = 118; 73.8%). Gelatinized starch masses (N = 15; 9.4%) and retrograded starch network structures (N = 2; 1.3%) were also recorded. Severely damaged starch granules or those lacking diagnostic features were classified as unidentified (UNID; N = 25; 15.6%) (Table 1; Figure 3). The morphological features of these starch types, along with corresponding modern reference features, are summarized in Table 2.
Type I (sorghum, Sorghum sp.): A total of 46 examples were identified, round or polygonal in shape, measuring 4.79–25.07 μm in length. Some appeared singly, while others were found in compound clusters. Most exhibited damage, such as deep fissures, central voids, blurred extinction cross, and mild gelatinization, consistent with fermentation-related enzymatic activity (Figure 3(1,2,6,9); cf. Figure 4(1–3)). Type I starch morphologies broadly resemble those of several species under Panicoideae subfamily, such as foxtail millet (Setaria italica), broomcorn millet, maize, and sorghum. However, maize and millets can be excluded: maize was not introduced to China until the 16th century [37], and millet starch granules are considerably smaller than the size range observed in Type I. In our reference database, broomcorn millet starches measure 4.81–10.86 µm; and foxtail millet starches measure 4.04–16.85 µm. By contrast, Type I starches most closely match modern sorghum in both morphology and size range of 5.90–26.81 µm (Table 2). This identification is further supported by the detection of sorghum components in the aforementioned mass spectrometry analysis.
The brewing experiment with malted sorghum revealed damage patterns characterized by deep fissures, pitting, and central depressions resulting from malting and mashing, along with hollowed centers and relatively well-preserved peripheries showing mild gelatinization during subsequent fermentation (Figure 4(1–3)), comparable to those observed in the LJR liquid (Figure 3(1,2,6)). These features—reflecting the combined effects of enzymatic digestion during malting, mashing, and fermentation—were likewise observed in our malt-based experiments with various cereals, as well as in an ethnographic reference sample of millet beer (hunjiu) from northern Shaanxi, which employed locally grown sprouted maize or wheat as saccharification agents [5].
Type II (Triticeae tribe, e.g., wheat): 62 examples. Triticeae starch granules exhibit a bimodal size distribution: large A-type (10–35 μm, lenticular) and small B-type (1–10 μm, spherical). We recorded 13 A-type (11.16–29.36 μm) and 49 B-type (3.3–9.84 μm) starches (Figure 3(3,8); cf. Figure 4(4); Table 2). B-type granules were relatively well preserved, whereas A-type granules exhibited severe fermentation-related damage, showing deep fissures, central depression with mild gelatinized expansion, and broken edges (Figure 3(4,5)). The damage patterns are consistent with those of mashed wheat or barley in our modern references from beer brewing (Figure 4(5,6)) [27]. The most likely source of these starches is wheat, widely cultivated in northern China after the Tang dynasty (AD 618–907) [38].
Type III (Rice, Oryza sativa): Two compound clusters consisting of small polygonal granules (3.28–6.13 μm in size) were observed exclusively on stained slides. These morphologies are consistent with those of rice (Table 2). These clusters exhibit varying degrees of gelatinization, ranging from the loss of birefringent crosses to partial disintegration of the granules. Their appearance closely resembles that of modern fermented rice starch aggregates (Figure 3(10,12); cf. Figure 4(7,8,10)).
Type IV (foxnut, Euryale ferox): Eight compound aggregates, overall round or oval in shape (13.06–46.27 μm), containing numerous small polygonal granules (size 1.43–3.88 μm). These characteristics are consistent with the starch morphology of foxnut (Figure 3(7); cf. Figure 4(9); Table 2). Many of the small granules within the aggregates appear blurry or partially degraded, and some aggregates are enlarged, likely due to cooking and fermentation.
Retrograded starches: Two examples of starch retrogradation networks were observed (Figure 3(11)). Starch retrogradation occurs when gelatinized starch molecules (amylose and amylopectin) realign and crystallize during cooling or storage, forming a three-dimensional network [39]. Retrograded starch networks have also been identified in modern brewing samples made from cooked millet (Figure 4(11)) [28].
The retrogradation networks in the LJR liquid appear damaged and fragmented, likely due to post-depositional degradation over approximately 800 years of burial. While the specific plant sources of the retrograded starches remain unidentified, the evidence clearly indicates that some of the ingredients underwent thermal processing during fermentation.
In summary, the starch assemblage in the fermented beverage comprises multiple plant sources, including malted sorghum and wheat, as well as cooked rice and foxnut.

4.2. Isotopic Results

The carbon isotope ratio (δ13C) and nitrogen isotope ratio (δ15N) are reported using the standard notation: δ = (Rsample/Rstandard − 1) × 1000, where Rsample and Rstandard represent the 13C/12C and 15N/14N ratios of the sample and standard, respectively. Pee Dee Belemnite (VPDB) is used as the standard for carbon, and atmospheric nitrogen (AIR) for nitrogen.
The measured δ13C value of the LJR liquid sample is –18.5‰, and the δ15N value is 15.5‰. The carbon isotope ratio is particularly informative, as it is indicative of the botanical source in fermentation. Terrestrial plants can be broadly classified into two major categories—C3 and C4—based on their photosynthetic pathways (a third group, CAM, is negligible in the East Asia diet). In the C3 pathway, CO2 is fixed by the enzyme Rubisco into a 3-carbon compound within mesophyll cells. In contrast, the C4 pathway begins with CO2 fixeation by PEP carboxylase into a 4-carbon compound in mesophyll cells, which is then transported to bundle-sheath cells where CO2 is concentrated for Rubisco. These distinctions result in different degrees of isotopic fractionation: C4 plants discriminate less against 13C compared to C3 plants. Consequently, δ13C value of C3 plants typically range between −24 to −35‰, while those of C4 plants fall between −10 and −14‰, with no overlap between the two groups. The measured δ13C value of –18.5‰ thus suggests a mixed contribution from both C3 and C4 plants. Importantly, previous studies have shown that fermentation itself does not cause significant carbon isotope fraction (less than 1‰) [40,41].
C3 plants adapt to moderate environments with relatively low water and nitrogen stress. They constitute the majority of human plant foods, including most nuts, fruits, tubers, pulses, and cereals such as wheat, rice, barley, rye, and oats. In contrast, C4 plants are better adapted to hotter, drier and saline environments. Although fewer in number, they include several nutritionally important cereals such as millets, maize, sorghum and sugarcane.
While a range of C3 plants were available in 13th -century China as potential fermentation sources (e.g., rice, wheat, barley, and variety of fruits), the options for C4 plants were relatively limited, since the Li Jurou tomb predates the Columbian Exchange. We will return to this point in a later section.
It is noteworthy that the δ13C value measured from the LJR liquid specimen closely aligns with carbon isotopic compositions reported for modern beer. Brooks and colleagues [34], for example, analyzed 160 beers worldwide and found δ13C values ranging from −27.3‰ to −14.9‰, reflecting differences in the proportion of C3 versus C4 plant carbon in the final product. Of these, 31% of beers exhibited purely C3 signatures, while the remaining 69% contained a mixture of C3 and C4 sources. Notably, the distribution of δ13C ratios in beers brewed with both C3 and C4 ingredients peaks at −19‰, a value comparable to that of our sample (−18.5‰).

5. Discussion

Drawing on evidence from microfossil analysis and isotopic testing, the following discussion addresses the fermentation method, brewing ingredients, and the broader implications of the LJR liquid for the historical development of fermented alcoholic beverages in early China.

5.1. The Fermentation Method and Ingredients

Microfossil analysis indicates that the LJR liquid contained two yeast species and starch remains from at least four plant sources: sorghum, wheat, rice, and foxnut. The majority of starch grains displayed fermentation-related damage, confirming that the liquid was a fermented alcoholic beverage, consistent with previous mass spectrometry findings. Importantly, no filamentary fungi associated with qu fermentation starters were observed, suggesting that the beverage was not produced using qu.
Many starches from sorghum and wheat exhibited damage consistent with enzymatic digestion and mild gelatinization caused by malting and mashing, suggesting that malted sorghum and wheat may have been used as saccharification agents for making beer.
The presence of heavily gelatinized starch granules and retrograded starch networks suggests that some of the brewing ingredients were cooked. The gelatinization observed in rice starch indicates that steamed rice was likely used as a fermentable substrate. Foxnut, which is rich in starch, probably served as an additional ingredient. Notably, foxnut starch granules have been identified in fermentation vessels at several Neolithic sites, dating back as early as 8000 years ago [20]. In modern times, some breweries in China have used wheat malt in combination with rice and foxnut to produce foxnut beer [42]. These findings point to a long history of foxnut use in alcoholic fermentation in China.
Based on these observations, we suggest that the LJR liquid derives from a malted grain beer, brewed using sprouted sorghum and wheat as saccharification agents, along with cooked rice and foxnut as the primary fermentable ingredients. This brewing method—combining both malted and cooked cereals—differs from typical beer brewing practices in the West, where all cereal ingredients (mainly barley, wheat, and rye) are usually subjected to malting, mashing, and fermentation [43].
When examining the identified starch assemblage (N = 118), sorghum accounts for 39.0%, wheat 52.5%, rice 1.7%, and foxnut 6.8%. However, starch residue analysis alone cannot reliably determine the relative proportions of plant ingredients in the original fermentation mixture, as preservation processes may have altered the representation of starches. This issue is further explored through isotope analysis in the following section.

5.2. C4 vs. C3 Plant Source

Stable isotope analysis provides insights into the relative contributions of C3 and C4 plants in the beverage ingredients. While isotopic evidence alone does not allow taxonomically specific identification, integrating it with starch residue data enhances the resolution of ingredient identification.
One inference from the isotopic results is that the ancient beverage was likely made using a combination of both C3 and C4 plants. This interpretation aligns with the microfossil analysis, which also indicates taxa representing both photosynthetic pathways. The pre-industrial average δ13C value of C3 plants is approximately −25.5‰ and that of C4 plants −11.5‰ (both adjusted for the Suess effect with a +1.5‰ offset). Allowing for typical pre-industrial variation (C3: −24‰ to −27‰; C4: −9‰ to −12‰), a two-component isotopic mixing model [40] suggests that, in the LJR liquid, C3 plants (including wheat, rice, and foxnut) likely made up approximately 43–63% of the ingredients, while C4 plant (including sorghum) contributing around 37–57%. For comparison, modern beers brewed with both C3 and C4 sources contain 28–55% C4 carbon at a global level [34].
% C4 carbon = (δ13Csample − δ13CC3)/(δ13CC4 − δ13CC3) × 100
However, it should be noted that the δ13C value measured in the residue is likely enriched in 13C relative to the original grain carbohydrates. Such enrichment occurs because fermentation preferentially volatilizes 13C-depleted ethanol (accounting for roughly two-thirds of the sugar carbon) and, to a lesser extent, 13C-enriched CO2, leaving behind a residual mash typically enriched by less than 1‰ [44]. When this enrichment is taken into account, the actual contribution of C4 carbon (likely sorghum) may be slightly lower than the initial estimate, though it would still plausibly constitute 40–50% of the plant input. This interpretation is further supported by previous LC-MS/MS analyses conducted on the same sample, which detected proteins most consistent with sorghum [24]. In comparison with the starch analysis results, the estimated 40–50% contribution from C4 carbon is comparable to the microbotanical evidence, where sorghum starch constitutes about 39% of the identified granules.
As noted, the measured δ15N value lies at the higher end of the range for plant tissues grown in temperate environments. Grain-based fermentation itself does not typically cause a significant shift in δ15N, since the primary fermentation produces (CO2 and ethanol) are not nitrogen-bearing volatiles. Although post-fermentation heating could, in principle, drive off 14N-rich volatiles and thereby enrich the residue in 15N, this effect occurs only at elevated temperatures. The elevated δ15N value therefore most likely reflects either the cultivation of grains in 15N-enriched soils—through natural processes such as denitrification or anthropogenic inputs like manuring [45]—or, less plausibly, the inclusion of nitrogen-rich additives during the fermentation (e.g., fish sauce, dairy products, or ash lye). While the latter possibility cannot be completely excluded, it is not characteristic of East Asian fermentation traditions and has been documented in ancient Chinese sources. At present, the available data do not allow us to distinguish between these scenarios, leaving this question open for future research.

5.3. The Tradition of Making Li Beer

The fermented beverage represented by the LJR liquid corresponds to li, a term that first appears in oracle-bone inscriptions from the late Shang Dynasty (ca. 1250–1026 BC). In later texts, li is described in greater detail as a beverage made from malted cereals and brewed for a single day. It was unfiltered, slightly sweet, low in alcohol content, typically prepared from broomcorn millet, and commonly used as an offering in ancestral rituals [46,47]. This description of li alcohol broadly aligns with the LJR liquid, both in its characteristics and ritual function.
The li beverage was believed to have gradually declined. The Tiangong Kaiwu (The Exploitation of Heavenly Treasures; published in 1637) records: “In ancient times, qu was used to make jiu (strong fermented beverage), and nie (malted cereal) was used to make li (sweet fermented beverage). In later generations, as people came to dislike the bland taste of li, its method was lost, and with it, the technique of making nie also disappeared.” This statement has often been regarded as the reasons for the decline of li in China. However, this brewing method has actually persisted in some remote regions, exemplified by the traditional millet-based turbid beer (hunjiu), as abovementioned (Section 3.2.2), still produced in northern Shaanxi on the Loess Plateau ([5]: Supplementary Information Text). This unfiltered, porridge-like beverage with low alcohol content bears close similarity to the LJR liquid from the Jin Dynasty, particularly in its brewing technique, suggesting continuity of this ancient method.

5.4. The Introduction of Sorghum and the Production of Sorghum Beer

Sorghum, originally domesticated in Sub-Saharan Africa, was introduced into East Asia during the historical period. The timing of its arrival in China has long been subject to speculation and debate, largely due to the scarcity of definitive archaeological evidence [48,49]. It is broadly believed to have been associated with the Arab Agricultural Revolution (8th–13th centuries), which facilitated extensive crop exchanges between Africa and Asia, with sorghum being a prominent example [50]. Sorghum subsequently became widely cultivated in China [51].
Archaeologically, the earliest carbonized sorghum remains in China have been identified in Beijing, dating to the early to middle Jin Dynasty [52], predating the sorghum microremains recovered from the LJR liquid, which dates to the late Jin Dynasty, just prior to the Yuan period.
This finding is especially significant in light of recent genetic studies that have identified unique selection signatures in Chinese sorghum varieties associated with brewing efficiency [53]. Interestingly, textual references to sorghum—frequently in the context of alcohol brewing—increased markedly during the Yuan Dynasty (AD 1271—1368) [54] (p. 251).
This study also provides the first direct identification of sorghum as an ingredient in beer in China, setting this beverage apart from earlier forms of li beer, which were primarily brewed with millet, barley, or wheat. Given sorghum’s dual role in Africa as both a staple food and a key ingredient in beer production [43,55], the discovery of sorghum-based beer in China offers new insights into the culinary and ritual functions that sorghum may have assumed during its early introduction and adoption.

6. Conclusions

The analysis of the Jin dynasty ceramic vase from the tomb of Li Jurou provides compelling evidence for the use of malted grains—specifically sprouted sorghum and wheat—in brewing a fermented alcoholic beverage intended for mortuary ritual purposes. By integrating microfossil and isotopic analyses with prior proteomic results, we were able to reconstruct both the composition and the brewing process of the liquid residue with a high degree of specificity. The convergence of evidence from these complementary approaches distinguishes this beverage from qu-based fermentation traditions and instead links it to li-type brewing, rooted in the use of malted cereals.
This discovery is especially significant as it offers a scientific confirmation of multi-cereal beer production in ancient China, incorporating both sorghum and wheat—two of the most important sources for beer production historically and today. The inclusion of sorghum, likely a relatively recent introduction during the historic period, into ritual alcoholic production reflects both technological innovation and cultural adaptation. This finding also resonates with recent genetic research identifying unique liquor-making properties in Chinese sorghum, thereby situating such selection within a deeper historical context [53]. Moreover, it suggests that sorghum, long associated with beer brewing in Africa, rapidly assumed a comparable role in Chinese mortuary and ceremonial traditions. Both starch and isotopic analyses indicate that sorghum may have constituted a substantial proportion of the fermentation ingredients.
More broadly, this study demonstrates how a multi-proxy analytical framework can shed light on the social and technological dimensions of ancient fermentation practices. The incorporation of new crops into established brewing traditions highlights the dynamic interplay between subsistence strategies and symbolic practice in China’s historical past.

Author Contributions

Conceptualization, L.L. and X.L.; methodology, L.L. and X.L.; formal analysis, L.L. and X.L.; resources, C.Y. and Y.M.; writing—original draft preparation, L.L. and X.L.; writing—review and editing, L.L. and X.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The original contributions presented in this study are included in the article material. Further inquiries can be directed to the corresponding author.

Acknowledgments

We are grateful to Anding Shao of the Shaanxi Provincial Institute for the Preservation of Cultural Heritage for his assistance in acquiring the archaeological samples. We acknowledge Petra Vaiglova in assisting lab analysis for isotopic measurements. This research was partially supported by the Min Kwaan Chinese Archeology Project at the Stanford Archeology Center, Stanford University.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Liu, L.; Zhang, J.; Li, J.; He, Y.; Gao, Z.; Jiang, L. Identification of 10,000-year-old rice beer at Shangshan in the Lower Yangzi River valley of China. Proc. Natl. Acad. Sci. USA 2024, 121, e2412274121. [Google Scholar] [CrossRef] [PubMed]
  2. Liu, L.; Wang, H.; Sun, H.; Chen, X. Serving red rice beer to the ancestors ca. 9000 years ago at Xiaohuangshan early Neolithic site in south China. Holocene 2023, 33, 1012–1020. [Google Scholar] [CrossRef]
  3. Wang, J.; Jiang, L.; Sun, H. Early evidence for beer drinking in a 9000-year-old platform mound in southern China. PLoS ONE 2021, 16, e0255833. [Google Scholar] [CrossRef] [PubMed]
  4. McGovern, P.E.; Zhang, J.; Tang, J.; Zhang, Z.; Hall, G.R.; Moreau, R.A.; Nunez, A.; Butrym, E.D.; Richards, M.R.; Wang, C.-S.; et al. Fermented beverages of pre- and proto-historic China. Proc. Natl. Acad. Sci. USA 2004, 101, 17593–17598. [Google Scholar] [CrossRef]
  5. Liu, L.; Wang, J.; Levin, M.J.; Sinnott-Armstrong, N.; Zhao, H.; Zhao, Y.; Shao, J.; Di, N.; Zhang, T. The origins of specialized pottery and diverse alcohol fermentation techniques in Early Neolithic China. Proc. Natl. Acad. Sci. USA 2019, 116, 12767–12774. [Google Scholar] [CrossRef]
  6. Liu, L.; Li, Y.; Zhao, Y.; Chen, X.; Gu, W. Beyond subsistence: Evidence for red rice beer in 8000-year old Neolithic burials, north China. J. Archaeol. Sci. Rep. 2023, 51, 104168. [Google Scholar] [CrossRef]
  7. McGovern, P.E.; Underhill, A.; Fang, H.; Luan, F.; Hall, G.; Yu, H.; Wang, C.; Cai, F.; Zhao, Z.; Feinman, G. Chemical identification and cultural implications of a mixed fermented beverage from late prehistoric China. Asian Perspect. 2005, 44, 249–275. [Google Scholar] [CrossRef]
  8. Wang, J.; Liu, L.; Ball, T.; Yu, L.; Li, Y.; Xing, F. Revealing a 5,000-y-old beer recipe in China. Proc. Natl. Acad. Sci. USA 2016, 113, 6444–6448. [Google Scholar] [CrossRef]
  9. Liu, L.; Wang, J.; Liu, H. The brewing function of the first amphorae in the Neolithic Yangshao culture, North China. J. Anthropol. Archaeol. Sci. 2020, 12, 118. [Google Scholar] [CrossRef]
  10. Feng, S.; Liu, L.; Wang, J.; Levin, M.J.; Li, X.; Ma, X. Red beer consumption and elite utensils: The emergence of competitive feasting in the Yangshao culture, North China. J. Anthropol. Archaeol. 2021, 64, 101365. [Google Scholar] [CrossRef]
  11. Liu, L.; Wang, J.; Chen, X.; Liang, Z. The quest for red rice beer: Transregional interactions and development of competitive feasting in Neolithic China. Archaeol. Anthropol. Sci. 2022, 14, 78. [Google Scholar] [CrossRef]
  12. Liu, L.; Chen, J.; Wang, J.; Zhao, Y.; Chen, X. Archaeological evidence for initial migration of Neolithic Proto Sino-Tibetan speakers from Yellow River valley to Tibetan Plateau. Proc. Natl. Acad. Sci. USA 2022, 119, e2212006119. [Google Scholar] [CrossRef] [PubMed]
  13. He, Y.; Liu, L.; Sun, Z.; Shao, J.; Di, N. “Proposing a toast” from the first urban center in the north Loess Plateau, China: Alcoholic beverages at Shimao. J. Anthropol. Archaeol. 2021, 64, 101352. [Google Scholar] [CrossRef]
  14. Liao, J.; Yang, Y.; Gu, W.; Yao, L.; Wei, Q.; Luo, W.; Gong, Y.; Ding, L.; Gu, C.; Zhang, J. A New Filtered Alcoholic Beverage: Residues Evidence From the Qingtai Site (ca. 5,500–4,750 cal. BP) in Henan Province, Central China. Front. Earth Sci. 2022, 10, 884630. [Google Scholar] [CrossRef]
  15. Liao, J.; Yang, Y.; Gu, W.; Gong, Y.; Sun, B.; Cui, Q.; Zhou, X.; Zhang, J. Pottery vessels and alcohol consumption at the late Neolithic Shuanghuaishu site in Central China. npj Herit. Sci. 2025, 13, 348. [Google Scholar] [CrossRef]
  16. Liu, L.; Zhang, Y.; Ge, W.; Lin, Z.; Sinnott-Armstrong, N.; Yang, L. Revealing the 2300-Year-Old Fermented Beverage in a Bronze Bottle from Shaanxi, China. Fermentation 2024, 10, 365. [Google Scholar] [CrossRef]
  17. Li, J.; Yang, J.; Cao, J.; Nan, P.; Gao, J.; Shi, D.; Han, B.; Yang, Y. Characterization of liquor remains in Beibaie site, central China during the 8th century BCE. Microchem. J. 2022, 177, 107293. [Google Scholar] [CrossRef]
  18. Li, Y.; Zhang, G.; Nan, P.; Yang, J.; Cao, J.; Ma, Z.; Ge, W.; Wen, R. Wine or Beer? A reinvestigation of residues from bronze vessels from the Beibai’e cemetery, Shanxi China. Herit. Sci. 2023, 11, 184. [Google Scholar] [CrossRef]
  19. He, Y.; Zhao, H.; Liu, L.; Xu, H. Brewing and Serving Alcoholic Beverages to Erlitou Elites of Prehistoric China: Residue Analysis of Ceramic Vessels. Front. Ecol. Evol. 2022, 10, 845065. [Google Scholar] [CrossRef]
  20. Liu, L. Archaeological evidence for fermented alcoholic beverages in ritual feasts. In Microbial Fermentations in Nature and As Designed Processes; Hurst, C.J., Ed.; Wiley: Hoboken, NJ, USA, 2023; pp. 209–224. [Google Scholar]
  21. Sun, J. Tang Song shiqi de chaju yu jiuju [The tea and wine vessels of the Tang and Song Dynasties]. Zhongguo Lishi Bowuguan Guankan 1982, 00, 113–123. [Google Scholar]
  22. Franke, H.; Twitchett, D.C. The Cambridge History of China: Alien Regimes and Border States, 907–1368; Cambridge University Press: Cambridge, UK, 1978; Volume 6. [Google Scholar]
  23. Yu, C.; Miao, Y. Shaanxi Xi’an Jindai Li Jurou mu fajue jianbao [Preliminary excavation report of the Li Jurou tomb of Jin dynasty in Xi’an, Shaanxi]. Kaogu Yu Wenwu 2017, 2, 40–49. [Google Scholar]
  24. Zhu, Z.; Yu, C.; Miao, Y.; Lu, Z.; Yang, J. Mass spectrometry identification of the liquor contained in the plum vase excavated from Jurou Li’s Grave of the Jin Dynasty (1115–1234 CE) in Xi’an, Shannxi, China. Herit. Sci. 2018, 6, 45. [Google Scholar] [CrossRef]
  25. Barton, H. Starch residues on museum artefacts: Implications for determining tool use. J. Archaeol. Sci. 2007, 34, 1752–1762. [Google Scholar] [CrossRef]
  26. Torrence, R.; Barton, H. (Eds.) Ancient Starch Research; Left Coast Press: Walnut Creek, CA, USA, 2006. [Google Scholar]
  27. Wang, J.; Liu, L.; Georgescu, A.; Le, V.V.; Ota, M.H.; Tang, S.; Vanderbilt, M. Identifying ancient beer brewing through starch analysis: A methodology. J. Archaeol. Sci. Rep. 2017, 15, 150–160. [Google Scholar] [CrossRef]
  28. Liu, L.; Liu, Z. Millet Beer Brewing in North China: Exploring Traditional Methods and their Significance in Archaeological Research. Ethnoarchaeology 2023, 15, 138–152. [Google Scholar] [CrossRef]
  29. Henry, A.G.; Hudson, H.F.; Piperno, D.R. Changes in starch grain morphologies from cooking. J. Archaeol. Sci. 2009, 36, 915–922. [Google Scholar] [CrossRef]
  30. Chen, R. Red Rice Alcohol in Southern China: Two Ways of Traditional Brewing and Their Implications for Archaeological Research. Ethnoarchaeology 2024, 16, 281–302. [Google Scholar] [CrossRef]
  31. Ling, C. Zhongguo jiu zhi qiyuan [The origins of alcohol in China]. Bull. Inst. Hist. Philol. 1958, 29, 883–901. [Google Scholar]
  32. St-Germain, G.; Summerbell, R. Identifying Fungi: A Clinical Laboratory Handbook; Star Publishing Company: Belmont, CA, USA, 2011. [Google Scholar]
  33. Boulton, C.; Quain, D. Brewing Yeast and Fermentation; Blackwell Science Ltd.: Oxford, UK, 2001. [Google Scholar]
  34. Brooks, J.R.; Buchmann, N.; Phillips, S.; Ehleringger, B.; Evans, R.D.; Lott, M.; Martinelli, L.A.; Pockman, W.T.; Sandquist, D.; Sparks, J.P.; et al. Heavy and Light Beer: A Carbon Isotope Approach To Detect C4 Carbon in Beers of Different Origins, Styles, and Prices. J. Agric. Food Chem. 2002, 50, 6413–6418. [Google Scholar] [CrossRef]
  35. Choi, J.Y.; Hwang, H.J.; Cho, W.Y.; Choi, J.I.; Lee, P.C. Differences in the Fatty Acid Profile, Morphology, and Tetraacetylphytosphingosine-Forming Capability Between Wild-Type and Mutant Wickerhamomyces ciferrii. Front. Bioeng. Biotechnol. 2021, 9, 662979. [Google Scholar] [CrossRef]
  36. Bioresource Collection and Research Center. Nadsonia fulvescens. 1989. Available online: https://mcoddb.bcrc.firdi.org.tw/detail.jsp?bcrc_id=22844 (accessed on 8 April 2025).
  37. Ho, P.-T. The introduction of American food plants into China. Am. Anthropol. 1955, 57, 191–201. [Google Scholar] [CrossRef]
  38. Zeng, X. Lun xiaomai zai gudai zhongguo zhi kuozhang [Expansion of wheat in ancient China]. Zhongguo Yinshi Wenhua 2005, 1, 99–133. [Google Scholar]
  39. Wang, S.; Li, C.; Copeland, L.; Niu, Q.; Wang, S. Starch Retrogradation: A Comprehensive Review. Compr. Rev. Food Sci. Food Saf. 2015, 14, 568–585. [Google Scholar] [CrossRef]
  40. Stern, L.A.; Chamberlain, C.P.; Blum, J.D.; Fogel, M.L. Isotopic Lessons in a Beer Bottle. J. Geosci. Educ. 1997, 45, 157–161. [Google Scholar] [CrossRef]
  41. DeNiro, M.J.; Epstein, S. Mechanism of Carbon Isotope FractionationAssociated with Lipid Synthesis. Science 1977, 197, 261–263. [Google Scholar] [CrossRef] [PubMed]
  42. Liu, Y.; Li, Y.; Chen, Z.; Zhang, C. Study on Fermentation Technology of Beer Brewing from Euryale Ferox. Bioprocess 2022, 12, 248–256. [Google Scholar] [CrossRef]
  43. Arthur, J.W. Beer: A Global Journey Through the Past and Present; Oxford University Press: Oxford, UK, 2022. [Google Scholar]
  44. Hobbie, E.A.; Werner, R.A. Intramolecular, compound-specific, and bulk carbon isotope patterns in C(3) and C(4) plants: A review and synthesis. New Phytol. 2004, 161, 371–385. [Google Scholar] [CrossRef]
  45. Bogaard, A.; Heaton, T.H.E.; Poulton, P.; Merbach, I. The impact of manuring on nitrogen isotope ratios in cereals: Archaeological implications for reconstruction of diet and crop management practices. J. Archaeol. Sci. 2007, 34, 335–343. [Google Scholar] [CrossRef]
  46. Wen, S.; Yuan, T. Yinxu Buci Yanjiu: Kexue Jishu Pian [A Study of Oracle Bones from Yinxu: Science and Technolgy]; Sichuansheng Shehui Kexueyuan Chubanshe: Chengdu, China, 1983. [Google Scholar]
  47. Fan, S. “Shijing” yu jiu wenhua [“Classic of Poetry” and alcohol culture]. Shijing Yanjiu Congkan 2004, 6, 201–214. [Google Scholar]
  48. Liu, L.; Crawford, G.; Lee, G.-A.; Chen, X.; Ma, X.; Li, J.; Zhang, J. Zhengzhou Dahecun Yangshao Wenhua “gaoliang” yicun de zai yanjiu [Re-analyse of the Yangshao “sorghum” remains at Dahecun in Zhengzhou]. Kaogu 2012, 1, 91–96. [Google Scholar]
  49. Zhao, L. Shilun Gaoliang Chuanru Zhongguo de Shijian, Lujing Ji Chubu Tuiguang [Discuss on the Time, Path of Sorghum’s Introduction into China and its Preliminary Popularization]. Zhongguo Nongshi 2019, 1, 3–14. [Google Scholar]
  50. Watson, A.M. Agricultural Innovation in the Early Islamic World: The Diffusion of Crops and Farming Techniques, 700–1100; Cambridge University Press: Cambridge, UK, 2008. [Google Scholar]
  51. Ge, F.; Xie, P.; Wu, Y.; Xie, Q. Genetic architecture and molecular regulation of sorghum domestication. aBIOTECH 2023, 4, 57–71. [Google Scholar] [CrossRef]
  52. Beijing Institute of Archaeology. Xinshidai Beijing kaogu gongzuo huigu yu zhanwang [A review and outlook of Archaeological work in Beijing in the New Era]. Zhongguo Wenwubao. 23 September 2022, p. 5. Available online: https://mp.weixin.qq.com/s/H4106ldTB2GIAy6vK9GTZw (accessed on 6 October 2025).
  53. Zhang, L.; Ding, Y.; Xu, J.; Gao, X.; Cao, N.; Li, K.; Feng, Z.; Cheng, B.; Zhou, L.; Ren, M.; et al. Selection Signatures in Chinese Sorghum Reveals Its Unique Liquor-Making Properties. Front. Plant Sci. 2022, 13, 923734. [Google Scholar] [CrossRef]
  54. Han, M.L. Historical Agricultural Geography of China; Peking University Press: Beijing, China, 2012. [Google Scholar]
  55. Arthur, J.W. Brewing beer: Status, wealth and ceramic use alteration among the Gamo of South-Western Ethiopia. World Archaeol. 2003, 34, 516–528. [Google Scholar] [CrossRef]
Heritage 08 00426 g001
Figure 1. Jin Dynasty plum vase and liquid sample analyzed in this study. (A) Geographic location of the Li Jurou tomb. (B) Structure of the tomb chamber, showing the vase containing liquid in the northwest corner (arrow). (C) Assemblage of drinking wares: (1) meiping vase; (2) meiping vase with liquid contents (M1:16) analyzed in this study; (3) Yuhuchun vase; (4) two small cups and a rectangular celadon tray. (D) Liquid sample extracted for analysis (vial capacity: 15 mL).
Figure 1. Jin Dynasty plum vase and liquid sample analyzed in this study. (A) Geographic location of the Li Jurou tomb. (B) Structure of the tomb chamber, showing the vase containing liquid in the northwest corner (arrow). (C) Assemblage of drinking wares: (1) meiping vase; (2) meiping vase with liquid contents (M1:16) analyzed in this study; (3) Yuhuchun vase; (4) two small cups and a rectangular celadon tray. (D) Liquid sample extracted for analysis (vial capacity: 15 mL).
Heritage 08 00426 g001
Heritage 08 00426 g002
Figure 2. Yeast cell morphology (Saccharomyces cerevisiae) compared with cells from the LJR liquid. (A) Modern yeast: (1) Diagrammatic cross-section of a yeast cell after [33] (Figure 4(1)); (2) yeast cells during beer fermentation, some showing the budding stage (courtesy of Jing-A Brewery); (3) clustered yeast cells after fermentation (from a mead-brewing experiment at Stanford). (B) Ancient yeast from the LJR liquid: (4) rounded cell; (5) oval cell with a developing bud on top; (6) round cell with a developing bud on top; (7) two connected round cells; (8) cluster of round/oval cells; (9) two connected elongate oval cells.
Figure 2. Yeast cell morphology (Saccharomyces cerevisiae) compared with cells from the LJR liquid. (A) Modern yeast: (1) Diagrammatic cross-section of a yeast cell after [33] (Figure 4(1)); (2) yeast cells during beer fermentation, some showing the budding stage (courtesy of Jing-A Brewery); (3) clustered yeast cells after fermentation (from a mead-brewing experiment at Stanford). (B) Ancient yeast from the LJR liquid: (4) rounded cell; (5) oval cell with a developing bud on top; (6) round cell with a developing bud on top; (7) two connected round cells; (8) cluster of round/oval cells; (9) two connected elongate oval cells.
Heritage 08 00426 g002
Heritage 08 00426 g003
Figure 3. Microfossil types identified in the LJR liquid. (1,2) Type I sorghum starch granules showing deep fissures and central depressions; (3) Type II Triticeae A-type starch; (4) Type II Triticeae A-type starch with collapsed center, damaged edges, and slight gelatinization and swelling; (5) Type II Triticeae A-type starch with hollowed center and preserved periphery; (6) Type I sorghum starch with missing center, preserved periphery, and slight gelatinization; (7) Type IV foxnut starch; (8) Type II Triticeae B-type starch granules; (9) aggregated cluster of sorghum starch granules; (10) aggregated cluster of rice starch with extinction crosses absent; (11) retrograded starch network structure; (12) gelatinized starch cluster, likely rice. (6,7,9–12) stained with trypan blue; (1–5,9) include DIC and polarized light images).
Figure 3. Microfossil types identified in the LJR liquid. (1,2) Type I sorghum starch granules showing deep fissures and central depressions; (3) Type II Triticeae A-type starch; (4) Type II Triticeae A-type starch with collapsed center, damaged edges, and slight gelatinization and swelling; (5) Type II Triticeae A-type starch with hollowed center and preserved periphery; (6) Type I sorghum starch with missing center, preserved periphery, and slight gelatinization; (7) Type IV foxnut starch; (8) Type II Triticeae B-type starch granules; (9) aggregated cluster of sorghum starch granules; (10) aggregated cluster of rice starch with extinction crosses absent; (11) retrograded starch network structure; (12) gelatinized starch cluster, likely rice. (6,7,9–12) stained with trypan blue; (1–5,9) include DIC and polarized light images).
Heritage 08 00426 g003
Heritage 08 00426 g004
Figure 4. Modern reference samples of starch. (1) Native sorghum; (2) sorghum germinated for 3 days and mashed for 2 h, with most granules showing enzymatic digestion and a few intact; (3) sorghum fermented for 10 days after mashing, showing hollowed centers with mild gelatinization; (4) wheat starch with large A-type and small B-type granules; (5) barley A-type starch after germination and mashing, displaying enzymatic damage with deep fissures and channels; (6) barley A-type starch after fermentation, showing hollowed centers with residual birefringence at the periphery; (7) rice fermented for 5 days, showing disappearance of the extinction cross; (8) fermented rice lacking birefringence; (9) foxnut starch; (10) gelatinized mass of rice starch; (11) retrograded starch network from cooked and fermented millet in huangjiu (yellow beer). (1–9) include DIC/brightfield and polarized light images).
Figure 4. Modern reference samples of starch. (1) Native sorghum; (2) sorghum germinated for 3 days and mashed for 2 h, with most granules showing enzymatic digestion and a few intact; (3) sorghum fermented for 10 days after mashing, showing hollowed centers with mild gelatinization; (4) wheat starch with large A-type and small B-type granules; (5) barley A-type starch after germination and mashing, displaying enzymatic damage with deep fissures and channels; (6) barley A-type starch after fermentation, showing hollowed centers with residual birefringence at the periphery; (7) rice fermented for 5 days, showing disappearance of the extinction cross; (8) fermented rice lacking birefringence; (9) foxnut starch; (10) gelatinized mass of rice starch; (11) retrograded starch network from cooked and fermented millet in huangjiu (yellow beer). (1–9) include DIC/brightfield and polarized light images).
Heritage 08 00426 g004
Table 1. Records of microfossil types in the LJR liquid (measured length: µm).
Table 1. Records of microfossil types in the LJR liquid (measured length: µm).
Stach Type IIIIIIIV
TaxonYeast (round, oval)Yeast (elonga-ted)Yeast total SorghumWheat A-typeWheat B-typeRice (cluster)Rice
(individual granules)		Foxnut (aggre-
gates)		Foxnut
(individual granules)		UNIDGelati-
nized
mass		Retro-
graded starch		Starch total
Total N5610664613492 8 25152160
Total % 84.815.2100.028.88.130.61.3 5.0 15.69.41.3100.0
Identified starch N 46622 8 118
Identified starch % 39.052.51.7 6.8 100.0
Measured N5610 461349 12629
Min length4.227.48 4.7911.163.3 3.2813.061.43
Max length11.512.68 25.0729.369.84 6.1346.273.88
Average length7.6810.03 12.1421.825.84 4.4825.192.05
Table 2. Comparison of morphological features of starch types in the LJR liquid and modern reference.
Table 2. Comparison of morphological features of starch types in the LJR liquid and modern reference.
Starch Type/TaxonShapeSize Range (μm)HilumExtinction CrossDamaged Characteristics
Ancient Type I. SorghumPolygonal and circular4.79–25.07Mostly centricMostly straight armsDeep fissures; blurry or missing extinction cross; hollowed center
Modern sorghumPolygonal and circular5.90–26.81Mostly centric; some with fissureMostly straight arms
Ancient Type II. Triticeae, A-typeLenticular11.16–29.36CentricStraight armsDeep fissures; missing parts; central depression; hollowed center
Modern wheat,
A-type		Lenticular10.00–35.00CentricStraight arms
Ancient Type II. Triticeae, B-typeCircular3.30–9.84Not observedNot observedNo birefringence observed
Modern wheat,
B-type		Circular1.00–10.00CentricStraight arms
Ancient Type III. riceCompound grains composed of many small polygonal granules3.28–6.13Not observedSome birefringent; no extinction cross observedOnly detected by staining with trypan blue
Modern riceIsolated single granules polygonal; compound grains composed of many small polygonal granules3.19–9.63CentricStraight arms
Ancient Type VI. foxnutRound/oval compound aggregates containing small polygonal granulesCompound: 13.06–46.27; granule: 1.43–3.88Not observedNot observedMost granules blurry; clearer after trypan blue staining; gelatinized compound larger than normal size
Modern foxnutRound/oval compound aggregates containing small polygonal granulesCompound: 9.51–32.33; granule: 1.71–3.96CentricStraight arms
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Liu, L.; Liu, X.; Yu, C.; Miao, Y. Preserved 800-Year-Old Liquid Beer in a Jin Dynasty Vase: Evidence of Malted Sorghum–Wheat Fermentation in Xi’an, China. Heritage 2025, 8, 426. https://doi.org/10.3390/heritage8100426

AMA Style

Liu L, Liu X, Yu C, Miao Y. Preserved 800-Year-Old Liquid Beer in a Jin Dynasty Vase: Evidence of Malted Sorghum–Wheat Fermentation in Xi’an, China. Heritage. 2025; 8(10):426. https://doi.org/10.3390/heritage8100426

Chicago/Turabian Style

Liu, Li, Xinyi Liu, Chunlei Yu, and Yifei Miao. 2025. "Preserved 800-Year-Old Liquid Beer in a Jin Dynasty Vase: Evidence of Malted Sorghum–Wheat Fermentation in Xi’an, China" Heritage 8, no. 10: 426. https://doi.org/10.3390/heritage8100426

APA Style

Liu, L., Liu, X., Yu, C., & Miao, Y. (2025). Preserved 800-Year-Old Liquid Beer in a Jin Dynasty Vase: Evidence of Malted Sorghum–Wheat Fermentation in Xi’an, China. Heritage, 8(10), 426. https://doi.org/10.3390/heritage8100426

Article Metrics

Back to TopTop