The Ancient Molinete, Plants, and Mercury Efficiency in Present-Day Artisanal Gold Mining, Northern Perú

Abstract

The molinete, a smaller version of the quimbalete or Inka mill, was also used to crush gold ore in the ancient Andes and its present-day use, though limited, continues. It consists of a large, movable upper stone, or chungo, and a lower, stationary stone, or muscha, that has a depression to seat the chungo. Each stone may weigh approximately one ton. The worker is seated and rocks the chungo, by foot on the lower, water-lubricated mushca. The weight of the chungo and back and forth motion crushes the ore thereby releasing the gold grains into the muddy slurry to which mercury is added to form a gold-mercury amalgam. After the amalgam is recovered, it is squeezed in a cloth to recover excess mercury, and then it is burned to volatilize the remaining mercury leaving an anthropogenic gold nugget. Spot geochemical sampling for Inductively Coupled Plasma (ICP) analysis of the molinete gold ore indicated average ~43 ppm Au and post-amalgamation mud contained average ~33 ppm Au which indicates an efficiency of ~23%. Local plants such as pegorondo or murmuncho were added to the muddy slurry in the muscha depression to aid amalgam recovery and help lubricate molinete movement; however, the use of these plants has been discontinued since the 1940s. Most importantly, since mercury is used with molinetes and quimbaletes to recover gold at artisanal gold processing sites today, then, the documented use of these artisanal ore-crushing mills in the past is consistent with mercury use for gold amalgamation in pre-contact Perú.

Share and Cite:

Brooks, W E., and Cano, A. (2026) The Ancient Molinete, Plants, and Mercury Efficiency in Present-Day Artisanal Gold Mining, Northern Perú. Archaeological Discovery, 14, 75-84. doi: 10.4236/ad.2026.141003.

1. Introduction

The molinete is an ancient artisanal stone crushing mill (Larco Hoyle, 1945, 2001) and it is a smaller version of the ancient quimbalete or Inka mill (Petersen, 2010; Brooks, 2025). The purpose of the molinete is comminution, or crushing and pulverizing the ore, thereby releasing the gold for separation, chemical treatment, and recovery (Thrush, 1968). Synonymous regional terms for these artisanal ore-crushing mills include: bimbalete or bambalete, which comes from an indigenous word meaning shake or move from one side to the other while staying in the same place (Petersen, 2010); piruro or drum-wheel (Petersen, 2010); chancadora (Larco Hoyle, 1945, 2001); and maray (de Nigris & Riart, 2013). Jaw crushers, also called chancadoras, have a similar role in initial ore processing in Perú’s present-day open-pit copper-gold mines.

Artisanal mills were widely used in the past and descriptions are numerous, for example: Agricola (1912); Atlas (2000); Barba (1923); Bargalló (1955, 1969); Brooks (2025); Costa et al. (2009); Florsch et al. (2015); Kaptan (1982); Olaechea (1901); Petersen (2010); Rivero y Ustariz (1857); and Romaña (1908). These mills have been found at pre-contact mining sites in Argentina, Bolivia, Chile, and Perú (Ahlfeld & Schneider-Scherbina, 1964; Larco Hoyle, 1945, 2001; Petersen, 2010). A review of Andean crushing mills and sites is provided in de Nigris & Riart (2013) and mechanics of these mills is discussed in Florsch et al. (2015); however, neither study acknowledges the essential use of mercury (azogue), without which gold could not be recovered from the muddy slurry in the muscha, for example, in La Rinconada, Perú (Cuentas & Velarde, 2025).

Given that there are only two ways to produce industrial amounts of gold—the oldest is amalgamation and the other is cyanide, which only dates to the 1880s in the US (Craig et al., 2001), then mercury amalgamation must be considered as key to archeological studies of ancient gold production. However, most archaeologists resist this logic and provide no evidence of alternative methods. Hard geochemical evidence for pre-contact use of mercury for gold amalgamation is based on comparison of the high mercury content of pre-cursor alluvial gold (>5000 ppm Hg) and the low mercury content of artifact gold (<20 ppm Hg) resulting from burning (refogado) the gold-mercury amalgam to volatilize the mercury (Petersen, 2010; Brooks et al., 2013).

Approximately 1.5 tons of gold per month are produced from Perú’s small-scale gold mines that use artisanal methods that include the ages-old technique of gravity separation and mercury amalgamation (Ahern, 2016; Al-Hassan & Hill, 1986; Brooks et al., 2007; Cánepa, 2005; Chauvin, 2018; Larco Hoyle, 1945, 2001; Soto-Viruet, 2018). The abundant alluvial gold sources (Noble & Vidal, 1994; Atlas, 1999; Atlas, 2000) in Perú likely provided the tons of gold used by Atahualpa as ransom for his release from the Spanish before his execution in 1533.

2. Using the Molinete

The molinete consists of a movable, upper stone, or chungo and a lower, stationary stone base, or muscha, that has a depression to seat the chungo. These stones may weigh one ton, or more, each. The worker, or moledor, is seated (Figure 1) and rocks the chungo with his feet (Figure 2) or may be aided by ropes attached to the chungo. In the 1940s smaller rocks encircled the muscha in order to help contain the muddy mixture (Larco Hoyle, 1945, 2001). Water is added to the muscha which helps the initial gravity separation of the heavier gold from the ore. Mercury is then added to the muscha and amalgamates the millimeter-sized or smaller gold particles. To recover the amalgam, a rock is propped under the chungo for safety. The amalgam is squeezed in a cloth to recover excess mercury and then burned to volatilize the mercury leaving an anthropogenic gold nugget.

Mercury is commercially available in Perú (Brooks et al., 2007) (Figure 3). And in the past, mercury was available from mercury occurrences in Perú that include mines such as Chonta and Huancavelica (Arana, 1901; Petersen, 2010; Giles, 1990; Brooks, 2020).

Figure 1. Molinete workplace, Pataz, northern Perú.

Figure 2. Worker seated at molinete, Pataz, northern Perú.

Figure 3. Commercial mercury used at molinete work site, Pataz, northern Perú.

3. Plants and Gold Recovery

Artisanal gold that is recovered today using plants or other non-mercury methods is referred to as “green gold” however, the use of plants as a part of gold processing dates to Roman time at Las Médulas, Spain. The gold washing tables, called agogae, were lined with moss or heather, locally called brezo [Ericaceae], which helped trap the fine-grained gold particles (Fernández-Lozano et al., 2021). The plants were removed, cleaned, and mercury was added to amalgamate the trapped fine-grained alluvial gold particles. This is mechanically comparable to the legendary Golden Fleece and the use of animal skins or specialized carpets and mats that are similarly used today to trap the fine-grained gold in sluice boxes.

In Chocó, western Colombia, artisanal miners use plant leaves, commonly called cedro playero to aid the final separation of the fine-grained gold and platinum from the lighter waste material (Castillo, 2007) and mercury is not used. The plant leaves are crushed by hand and the frothy liquid is mixed in water to make a flotation foam that is added to the gold pan. The heavier gold sinks as a heavy-mineral separate from the lighter minerals that cling to the foam. The plants were identified as Balso [Ochroma pyramidale] and Malva [Hibiscus furcellatus] (Brooks et al., 2015). Chocó gold miners were awarded a United Nations environmental award for producing green gold (Silva Herrera, 2010; Brooks, 2014).

Near Tulpo, northern Perú, the use of plants in gold-processing was documented in the 1940s by Larco Hoyle (1945, 2001) and the local plant names include: 1) pegorondo, which helps clean the mercury of impurities, and 2) murmuncho, which forms a viscous mass to capture the gold and also help lubricate the movement of the molinete. Other plants listed include: el shinac, shirac [Iochroma umbellata] (Figure 4), la verbena [Verbena litoralis] (Figure 5), el cuiguyum [Solanum glutinosum] (Figure 6), and el negush negush. Only some of the listed plants were found in the field and the other local plant names given in Larco Hoyle (1945, 2001) were not cross-referenced in Flora of Perú (MacBride, 1936a, 1936b). However, the use of plants with molinetes has been discontinued since the 1940s.

Figure 4. Shirac [Iochroma umbellata], northern Perú.

Figure 5. La verbena [Verbena litoralis], northern Perú.

Figure 6. El cuiguyum [Solanum glutinosum], northern Perú.

Agricola (1912) describes a unique mercury retorting process in which mercury is retorted from cinnabar in a closed work area or hut that contains a shrub or small tree. The mercury is retorted from cinnabar, the common ore of mercury, and the vapor condenses on the cooler plant leaves and can then be collected.

4. Sampling

A molinete site near Pataz was sampled to determine the efficiency of mercury amalgamation. Samples were taken of the gold ore and the mud resulting from molinete processing with water and mercury. It is important to indicate that the molinete process is ongoing with little clean-up other than perhaps a spray with a hose after the amalgam is removed and therefore, the mud from the first step mixes with ongoing mud output from the molinete process. Spot samples obtained from each step were analyzed by ICP (Inductively Coupled Plasma) and fire-assay for gold content and results from spot sampling at the Pataz site are given on Table 1. The samples were taken as available and there is no continuity between samples of: 1) ore, and 2) mercury-bearing mud on Table 1. Spot samples (~400 g each) were taken at each step given below:

Step 1: gold ore is crushed by the molinete, ~23% of the gold, as a gold-mercury amalgam is removed from the mud during this step

Step 2: remaining gold-mercury containing mud flows out of the molinete, there is no further treatment to recover the remaining gold or mercury.

Table 1. Molinete geochemical sampling, Pataz, northern Perú.

PE251

Au ore

PE252

Au ore

PE253

after Hg

PE254

after Hg

PE255

after Hg

Au (0.003)

6.46

79.5

55.7

31.4

12.0

Ag (0.3)

0.3

7.0

11.7

7.1

6.5

Al (300)

44,004

34,941

21,788

32,669

20,137

As (2.0)

4028

294

2142

918

376

Bi (5.0)

<5

10

41

25

<5

Ca (300.0)

2013

3562

5186

9404

3252

Ce (1.0)

31

10

9

12

13

Co (1.0)

14

53

143

81

21

Cu (1.0)

34

114

1302

915

498

Fe (300)

32,653

158,228

>250,000

167,557

67,956

Hg (0.5)

3.8

6.7

>100

77.9

>100

K (300)

35809

3878

7036

10715

3722

La (1.0)

16

7

7

8

8

Li (2.0)

<5

15

<5

6

12

Mg (100.0)

3787

25,731

8760

11,746

9323

Mn (5.0)

248

753

216

287

710

Na (100)

5214

140

601

970

326

Ni (1.0)

9

15

11

8

20

Pb (3.0)

19

1056

4338

3409

1934

S (30)

13,719

91,121

175,367

102,647

19,043

Sb (2.0)

7

7

7

3

3

Sc (1.0)

8

16

6

9

9

V (3.0)

29

115

31

43

115

Zn (3.0)

22

687

2364

1736

1166

Multi-element ICP analyses in parts per million (ppm) (detection limit given to right of element, in parentheses); American Assay, Sparks, NV [ICP-I04AB28, Au-fire assay]. Sample Descriptions: PE251 Au ore, rusty, quartz vein breccia, abundant pyrite and sulfides; PE252 Au ore, rusty, mafic rock, with quartz breccia, abundant pyrite; PE243 mud after molinete crushing and addition of Hg; PE254 mud after molinete crushing and addition of Hg; PE255 mud after molinete crushing and addition of Hg.

5. Elements of Interest

Gold—Two spot samples of the gold-bearing ore were sampled and the gold content of the two samples was 6.46 ppm and 79.5 ppm for an average gold content of 43 ppm gold (Table 1). After amalgamation, spot samples of the outgoing molinete mud ranged from 12 ppm to 55.7 ppm gold for an average gold content of 33 ppm gold indicating that mercury removed ~23% of the gold during amalgamation. In Perú, a study of artisanal gold recovery using the quimbalete indicated that ~20% of the gold was recovered by amalgamation (Brooks, 2025) and in Colombia, a similar field study showed that <19% of the gold was recovered by amalgamation (Torkaman & Viega, 2023).

Silver—Two spot samples of gold-bearing ore indicated very low silver content of the ore.

Mercury—Two spot samples of the gold-bearing ore contained 3.8 ppm and 6.7 ppm mercury indicating the base level of mercury in the primary gold deposit. After amalgamation the mercury content of the outgoing molinete mud samples was predictably higher, from 77.9 ppm to >100 ppm mercury. The gold-mercury-containing mud from the molinete process goes untreated and may create an environmental risk, whereas the mud from the quimbalete process is treated with cyanide to recover the remaining gold (Brooks, 2025).

6. Conclusion

Perú is the leading gold producer in South America and ~1.5 tons of gold per month are produced from Perú’s numerous small-scale gold mines that use ages-old indigenous methods that include the use of molinetes, gravity separation in water, and mercury amalgamation. However, mercury is lost to the environment from amalgam burning as a vapor as well as residual mercury in the mud resulting from the molinete process. Amalgamation with a molinete removes ~23% of the gold and the gold remaining in the mud is not recovered. Present-day artisanal molinete technology and the use of mercury to produce gold is key to understanding past gold production in the pre-contact Andes.

Acknowledgements

My gratitude and respect to Dr. Georg Petersen (deceased), University of Kiel, Germany and Lima, Perú for his insight and application of geology and geochemistry to archaeology in Perú by publication of Minería y Metalurgia en el Antiguo Perú [Mining and Metallurgy in Ancient Perú, 1970/2010]. Sincere thanks are expressed to Sr. Christian Ormeño for transport, security, and help in the field.

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

References

[1] Agricola (1912). De re metallica (Trans. by H. C. Hoover, & H. L. Hoover). Kessinger Publishing, 638.
[2] Ahern, N. (2016). Mercury in Gold Processing. In M. D. Adams (Ed.), Gold Ore Processing (2nd ed., pp. 753-766). Elsevier. [Google Scholar] [CrossRef
[3] Ahlfeld, F., & Schneider-Scherbina, A. (1964). Los yacimientos minerales y de hidrocarburos de Bolivia (p. 388). Ministerio de Minas y Petroleo, Departamento Nacional de Geología.
[4] Al-Hassan, A. Y., & Hill, D. R. (1986). Islamic Technology, an Illustrated History (p. 304). Cambridge University Press.
[5] Arana, P. P. (1901). Las minas de azogue del Perú. Imprenta El Lucero.
[6] Atlas (1999). Mineria y energía en el Perú. Ministerio de Energía y Minas, Instituto Geológico Minero Metalúrgico, Republica del Perú.
[7] Atlas (2000). Atlas de la pequeña minería en el Perú (Charmot Sarmiento, J., ed., p. 88). Ministerio de Energía y Minas, Dirección General de Minería, Republica del Perú.
[8] Barba, A. A. (1923). El arte de los metals (Trans. by R. E., & E. P. Mathewson). John Wiley & Sons.
[9] Bargalló, M. (1955). La mineria y metalurgía en la América española durante la época colonial. Compañía Fundidora de Fierro y Acero de Monterrey.
[10] Bargalló, M. (1969). La amalgamación de los minerales de plata en hispano colonial, con una ofrenda a Alvaro Alonso Barba (por Carlos Prieto). Compañía Fundidora de Fierro y Acero de Monterrey.
[11] Brooks, W. E. (2014). Colombia Mercury Inventory 2011. Geología Colombiana, 37, 15-50.
[12] Brooks, W. E. (2020). Locations of Cinnabar-Mercury Occurrences in Peru, Implications for Precontact Gold Production. Archaeological Discovery, 8, 245-259. [Google Scholar] [CrossRef
[13] Brooks, W. E. (2025). The Ancient Quimbalete and Mercury Efficiency in Present-Day Small-Scale Gold Processing, Perú. Archaeological Discovery, 13, 49-58. [Google Scholar] [CrossRef
[14] Brooks, W. E., Sandoval, E., Yepez, M., & Howard, H. (2007). Peru Mercury Inventory 2006. U.S. Geological Survey Open-File Report 2007-1252. [Google Scholar] [CrossRef
[15] Brooks, W. E., Schwörbel, G., & Castillo, L. E. (2013). Amalgamation and Small-Scale Gold Mining in the Ancient Andes. In N. Tripcevich, & K. J. Vaughn (Eds.), Mining and Quarrying in the Ancient Andes, Interdisciplinary Contributions to Archaeology (p. 213-229). Springer Publishing. [Google Scholar] [CrossRef
[16] Brooks, W. E., Sierra, J. A., & Palacios, F. (2015). Green Gold-Dirty Gold, Tadó, Dept. Chocó, Colombia. Natural Resources, 6, 534-542. [Google Scholar] [CrossRef
[17] Cánepa, C. (2005). Minería a pequeña escala en la Costa Sur Media del Perú (62 p). Instituto Geológico Minero y Metalúrgico, INGEMMET Boletín 3, Serie E: Minería.
[18] Castillo, N. (2007). Minería aurifera en el noroeste andino de Colombia, etnografía de la técnica. In R. Lleras (Ed.). Metalurgía en la América Antígua, Fundación de Investigaciones Arqueológicas Nacionales (p. 281-321). Instituto Francés de Estudios Andinos. [Google Scholar] [CrossRef
[19] Chauvin, L. (2018). Pope Brings Environmental Crusade to Peru (p. A8). Washington Post.
[20] Costa, M., Alfonso, P., & Palacios, S. (2009). Proceso de tratamiento para la recuperacion de oro en el asentiamiento minero artisanal de Misky, Perú. Segundo Congreso Internacional Sobre Geologia Y Minería En La Ordinación del Territorio y en el Desarrollo (pp. 231-242). Academia.
https://www.academia.edu/70426845/proceso_de_tratamento_para_la_recuperacion_de_oro_en_el_asentiamiento_minero_artesanal_de_Misky_Peru
[21] Craig, J. R., Vaughan, D. J., & Skinner, B. J. (2001). Resources of the Earth (520 p). Prentice-Hall.
[22] Cuentas, M., & Velarde, J. (2025). Evaluation of Mercury Use in the Amalgamation of Gold with Quimbaletes and Mills in La Rinconada, Puno. Encuentro Cientifico Internacional.
https://www.academia.edu/5497061/Evaluacion_del_uso_de_mercurio_en_los_processos_de_amalgamacion_de_oro_en_quimbaletes
[23] de Nigris, M. R., & Riart, O. P. (2013). Molinos mineros de tradición andina. De re Metallica, 20, 1-12.
[24] Fernández-Lozano, J., Nnandong, A. N., Rivera, S., Esbrí, J. M., Carrasco, R. M., Pedraza, J., Bernardo-Sánchez, A., & Higueras, P. (2021). Uso de mercurio para el beneficio de yacimientos secundarios de oro en epoca romana. Realidad o ficción? Geogaceta, 69, 1-4. https://www.researchgate.net/publication/349685762 [Google Scholar] [CrossRef
[25] Florsch, N., Téreygeol, P. J., & Cruz, C. (2015). The Ore-Dressing Grindstone Called a Quimbalete, a Mechanics-Based Approach. Archaeometry, 58, 881-891. [Google Scholar] [CrossRef
[26] Giles, C. F. (1990). Minas de mercurio conocidas en el Perú. INGEMMET Archivo Tecnico Codigo 9921.
[27] Kaptan, E. (1982). New Discoveries in the Mining History of Turkey in the Neighborhood of Gümüsköy and Kütahya (p. 1-10). Maden Tetkik ve Arama Enst.
https://dergi.mta.gov.tr
[28] Larco Hoyle, R. (1945). Los Mochicas (Pre-Chimú o Early Chimú) (325 p). Sociedad Gográfica Americana.
[29] Larco Hoyle, R. (2001). Los Mochicas, tomo II (350 p). Museo Arqueológico Rafael Larco Herrera, Pueblo Libre.
[30] MacBride, J. F. (1936a). Flora of Peru, Part I, no. 3 (356 p). Field Museum of Natural History.
[31] MacBride, J. F. (1936b). Flora of Peru, part II, no. 2 (410 p). Field Museum of Natural History. [Google Scholar] [CrossRef
[32] Noble, D. C., & Vidal, C. E. (1994). Gold in Peru. SEG Newsletter, No. 17, 1-13. [Google Scholar] [CrossRef
[33] Olaechea, T. (1901). Apuntes sobre el castillo y fundición de curamba. Anales de Ingenieros de Construcción Civil de Minas e Industria del Perú, 6, 1-21.
[34] Petersen, G. (2010). Mining and Metallurgy in Ancient Perú (Trans. by Brooks, W. E., 2010, of Minería y Metalurgia en el Antiguo Perú) (90 p). Museo Nacional de Antropología y Arqueología.
[35] Rivero y Ustariz, M. E. (1857). Exploitación del guano en tiempo prehistórico. Colección de Memorias Científicas, Agrícolas e Industriales, 1, 159.
[36] Romaña, E. (1908). Una inspección de los yacimientos de estaño de Bolivia y una exploración por el mismo metal en el Perú. Boletín del Cuerpo de Ingenieros del Perú.
[37] Silva Herrera, J. (2010). ONU premia chocoanos que se ganan la vida con el ‘oro verde’. El Tiempo.
[38] Soto-Viruet, Y. (2018). The Mineral Industry of Peru. U.S. Geological Survey 2017-2018 Minerals Yearbook.
https://pubs.usgs.gov/myb/vol3/2017-18/myb3-2017-18-peru.pdf
[39] Thrush, P. W. (1968). A Dictionary of Mining, Mineral, and Related Terms (1269 p). U.S. Department of the Interior.
[40] Torkaman, P., & Viega, M. M. (2023). Comparing Cyanidation with Amalgamation of a Colombian Artisanal Gold Mining Sample, Suggestion of a Simplified Zinc Precipitation Process. The Extractive Industries and Society, 13, Article 101208. [Google Scholar] [CrossRef

Copyright © 2026 by authors and Scientific Research Publishing Inc.

Creative Commons License

This work and the related PDF file are licensed under a Creative Commons Attribution 4.0 International License.