Archaeologists at the Roman-era Red Sea port of Berenike in Egypt discovered a jar holding 7.55 kilograms of black peppercorns. The jar was found near a sanctuary of Serapis, thousands of kilometres from the wet forests of southern India where pepper is cultivated. It was one unusually concentrated find among many smaller deposits of pepper recovered at the port. Although the pepper travelled from far away, the jar cannot identify its growers or owners, or establish whether its contents were meant for food, commerce or a religious purpose.
Moving pepper from India to Egypt required a chain of growers, traders, ships, port labour and caravans operating with the seasonal monsoon winds. Other plants travelled through overlapping networks of exchange. Apples developed new genetic combinations as cultivated trees encountered wild relatives; rice reached western regions long before it became a major local crop; and some aromatics travelled as dried products without their living plants ever becoming established abroad. Over centuries, these exchanges linked Eurasia, Africa and the Indian Ocean through many intersecting routes.
Plants reveal a lasting consequence of ancient trade: once introduced, a crop could alter diets and farming practices long after the original shipment was forgotten. One peppercorn in a rubbish layer demonstrates that imported pepper was present at the site. A fruit pit at a mountain settlement might record a meal, a market purchase or local cultivation. An orchard of grafted trees, by contrast, required sustained investment and horticultural skill. Tracing a crop requires evidence for its transport, consumption and eventual adoption by local growers.
The exchange of crops began before the Silk Roads had a name
The term Silk Road became widely known through the nineteenth-century work of German geographer Ferdinand von Richthofen. Historians use the name for the routes connecting East Asia and Inner Asia with Iran and regions farther west, especially after the Han period. The routes changed with political control, warfare, river crossings, pasture, mountain snow and access to water. An oasis grower, a Sogdian merchant, a nomadic herder and an Indian Ocean sailor could all participate in that wider exchange without travelling the same route.
Exchange of useful plants through Inner Asia began centuries before the routes now called the Silk Roads acquired their familiar form. Wheat and barley originated as domesticated crops in Southwest Asia and moved east, while broomcorn millet travelled west from East Asia. The cereals spread in opposite directions, with different earliest secure dates in each region. Archaeobotanical work on northern Iran identified broomcorn millet there by approximately 2050 BCE, while evidence for rice at the investigated Caspian sites dates to around 120 BCE. Their remains record separate episodes of movement and cultivation over many centuries.
Mountain settlements were important participants in the spread of cereal crops. Archaeologists have found both wheat and millet at Bronze Age Central Asian sites where pastoral communities combined herding with some farming. Research on crop dispersal through the Inner Asian mountain corridors stresses that the securely dated evidence is uneven: an early grain found at one site cannot be taken as proof that every community between its place of domestication and its final destination was already growing it. A handful of grains might represent a meal or an offering; repeated concentrations of grain and processing debris give firmer grounds for identifying local farming.
Later Silk Road routes linked oases in the Tarim Basin and Tian Shan with communities in the Zarafshan valley and the Iranian plateau. Caravans could join other ancient trade networks, travelling south toward the Indus basin or west toward the Persian Gulf. In another direction, shipping across the Arabian Sea connected ports in India, Arabia and East Africa with Roman Egypt. Calling all of these connections a single ‘Silk Road’ can conceal the different environments, people and technologies that made them possible. For each crop, the useful questions concern its form, the communities through which it passed, and the date at which its presence or cultivation can be demonstrated.
How archaeologists identify a plant that travelled
Excavated seeds, fruit stones, nutshells and cereal grains give archaeologists direct evidence of plants used at ancient sites. Charring can preserve wheat kernels, while dry conditions may protect grape pips, date stones and other plant remains. Archaeobotanists recover tiny remains by flotation, passing excavated soil through water so that lighter charred fragments can be separated and examined. The species identification alone cannot show when the remains arrived or how they were used. Before treating a specimen as proof of early cultivation, researchers must assess its date, the possibility of intrusion from later layers, and whether it could have been imported as food.
Microscopic remains provide another way to identify plants at archaeological sites. Pollen is evidence of a plant’s presence, although wind and water can carry it far from the source and some grains cannot be assigned to a particular species. Phytoliths are silica structures formed within plant tissues; their shapes can help distinguish some crop groups, but not every plant can be identified reliably to species. Starch grains in residues or dental calculus may indicate food preparation or consumption. A chemical residue can support an identification, but compounds common to several species cannot establish the presence of one named imported spice without corroboration.
The dating and archaeological context determine what can be concluded from a plant specimen. Direct radiocarbon dates obtained from seeds are stronger evidence for the age of those remains than dates drawn only from nearby pottery. A seed in a medieval rubbish pit tells us that the plant was present in that context; it does not establish the moment of its first introduction to the region. This distinction is especially important for plants that have fragile remains or were used in small quantities. The earliest excavated trace of a crop is the first evidence archaeologists have found so far; it may postdate its first appearance in a region.
Genetic evidence can answer questions that excavated grains and written records cannot address on their own. Comparing the DNA of modern crops with their wild relatives helps trace domestication, genetic mixing and the ancestry of cultivated varieties. Ancient DNA successfully extracted from preserved archaeological specimens can sometimes connect these genetic patterns to particular periods. Yet the genome of a modern apple cannot by itself date a specific caravan journey. Researchers have to compare DNA with archaeological deposits, environmental conditions and written records, each of which provides a different kind of evidence.
Black pepper and the Indian Ocean trade
Black pepper grows on the vine Piper nigrum, a plant suited to the tropical climate of southwestern India. Dried peppercorns were easy to package in substantial quantities, and their sharp flavour made them distinctive among imported seasonings. The Periplus of the Erythraean Sea, sections 53–56, a Greek-language merchant’s guide usually dated to the middle of the first century CE, names Muziris and Nelcynda among the important ports of the Malabar region. It reports that large ships came for pepper and other products, and identifies the nearby district of Cottonara as a major pepper-producing area. The account identifies goods and ports, although the origin of each individual pepper shipment remains unknown.
Roman-controlled Egypt linked the Indian Ocean shipping lanes to markets around the Mediterranean. Merchants could send goods landed at Berenike or Myos Hormos over the Eastern Desert by pack animal to the Nile, from which they continued toward Alexandria. The monsoon system made long crossings of the Arabian Sea practicable in particular seasons, though sailors and merchants did not need to ‘discover’ those winds in the Roman period. The Muziris papyrus, a second-century CE commercial document preserved in Vienna, offers a rare view of financing and customs connected with an Indian Ocean cargo. It records financing and customs for one cargo, rather than a complete inventory of plant products circulating in the empire.
The pepper jar found at Berenike provides material evidence that can be read alongside written accounts of the trade. The archaeologist Steven Sidebotham describes the excavation in Berenike and the Ancient Maritime Spice Route, chapter 12. Archaeological work at the port documented peppercorns in multiple contexts, some charred, while the large cache was recovered in an Indian-made storage jar in a courtyard associated with the Serapis temple. The archaeological context does not establish that all 7.55 kilograms were charred, nor that the jar was an ordinary household store. Its position raises questions about ritual use or storage, but the precise purpose is uncertain. The cache nevertheless shows that sizeable quantities of pepper reached the Egyptian coast.
Roman authors also recorded how the imported spice was valued and used. Pliny the Elder, Natural History 12.28–29 records prices of four denarii for black pepper, seven for white pepper and fifteen for long pepper, all per Roman pound. He expresses surprise that people would pay so much for pungency alone. His figures describe a particular Roman market and period; they cannot be converted into a universal price for the entire trade network. Pepper was used in cooking and medicine, but the familiar assertion that spices were primarily bought to disguise spoiled meat is not supported by this evidence. Its buyers valued pepper for different culinary, medicinal and social reasons, depending on when and where it was available.
Frankincense and saffron followed very different paths
Frankincense and myrrh travelled widely even though their source trees grew in particular climates and regions. Frankincense comes from Boswellia trees, including B. sacra of southern Arabia; myrrh is extracted from various members of Commiphora. Harvesters make incisions in the bark and collect the resin after it hardens. Dried resin can be stored and transported without the moisture and temperature controls needed for living plants. Both resins were used in ritual, medicine and perfume, and travelled by ship or caravan.
Merchants’ accounts and excavated sites help trace where incense was produced and shipped. According to the Periplus, section 32, royal officials supervised the frankincense gathered at a harbour in southern Arabia. In Oman, the frankincense-producing landscape around Wadi Dawkah and the associated sites of Shisr, Khor Rori and Al-Baleed preserve traces of production and trade, recognised in the UNESCO Land of Frankincense listing. Overland traffic also connected southern Arabia with markets farther north, including the Nabataean sphere. No single caravan route or empire controlled all incense movements throughout antiquity. Maritime routes changed the balance of that trade and sometimes allowed merchants to avoid intermediaries on older overland routes.
Saffron travelled as a dried spice, but its cultivation depended on moving living plants. It consists of the dried stigmas of the saffron crocus, Crocus sativus, which merchants could carry as a small but expensive good. Establishing a new saffron crop, however, requires living corms, the underground storage organs from which new plants grow. Cultivated saffron is effectively sterile because it is triploid, possessing three sets of chromosomes, and it is propagated vegetatively rather than through viable seed. The cultivated plant is closely related to wild Crocus cartwrightianus in Greece; research published in 2019 has identified a particularly close genetic connection with plants near Athens.
A 2022 review combining crocus genetics and Aegean art argues that saffron cultivation originated in ancient Greece, while acknowledging the problems involved in identifying a precise cultivated species in early images. The crop and its spice were known in several ancient societies, and their histories cannot be reduced to a single east-to-west introduction. Dried stigmas, wild crocus gathering and the movement of cultivated corms are three different phenomena. A profitable saffron harvest depended on knowing how to store and plant corms, manage moisture and propagate the crop, knowledge that could not be supplied simply by importing dried threads.
Apples, peaches and the orchards of Central Asia
Fruit trees also spread through Eurasian trade and farming networks. Fresh and dried fruit could be traded immediately, but a new orchard required suitable ground, water, sustained work and knowledge of how to reproduce good trees. The genetic history of the domestic apple illustrates both the reach and the limits of the Silk Road explanation. A 2017 study of 117 apple accessions traced the main ancestry of cultivated apples to Malus sieversii, especially populations in Kazakhstan, and identified substantial genetic contributions from the European crabapple, M. sylvestris. A 2020 genome study explored the same mixed ancestry and estimated that approximately 23 percent of the Gala apple genome examined had a hybrid origin.
The mixed ancestry of cultivated apples is consistent with movement across Eurasia and contact with wild relatives. These genetic results do not imply that every discarded pip could become a productive orchard variety. Apples grown from seeds differ genetically from their parent trees; people seeking to preserve a desirable fruit commonly propagate a selected tree by grafting, joining a shoot to a suitable rootstock. A seedling could introduce a new combination of traits, but sustained cultivation depended on growers recognising and maintaining useful trees. DNA can reveal ancestry and contact, but it cannot name the trader or identify the precise route and century when a variety first reached Europe.
Actual fruit stones and other excavated plant material offer evidence of what people consumed at particular settlements. At Tashbulak, a medieval site in the mountains of Uzbekistan, excavations recovered fruit stones, nutshells, legumes and cereals dating approximately to the ninth through eleventh centuries CE. The assemblage included evidence associated with peaches, apricots, grapes and apple or pear relatives, alongside locally available nuts and fruits. Because the settlement stood at an elevation unsuitable for growing some of these crops, researchers argue that fruit was brought from lower agricultural areas. The finds establish that supplies reached the mountain site; they do not demonstrate that its residents domesticated those fruits or planted every seed that survived.
These fruits have different histories of cultivation. Grapes have an old cultivation history in western Eurasia, peaches are associated more strongly with East Asia, and apricots were cultivated among different populations across Asia. The Tashbulak study reviews earlier finds from sites across Inner Asia, but also warns that some older identifications and dates lack the documentation required for certainty. The region combined established local agriculture with exchanges that brought in crops originally cultivated elsewhere.
The plant remains from Bukhara show what urban communities obtained and, in some cases, cultivated. A 2023 archaeobotanical study analysed plant remains from layers spanning late antiquity through medieval times. It identified cereals, cotton, flax, fruits, herbs and spices. The researchers found evidence of sumac, coriander, cumin and single specimens identified as black pepper and sesame, among other plants. A single sesame seed establishes that the plant was present, without revealing how much was traded or grown. More abundant crop remains and processing debris give stronger grounds for identifying regular local cultivation or processing.
Seeds, oil crops and the difficulty of establishing rice
Grains and oilseeds were simpler to carry over long distances than living fruit trees. Wheat, barley and millet might cross a region as rations, commercial cargo or seed for planting long before farming of that grain became widespread. Sesame could be transported as oil without the plant itself being cultivated at the destination. Its seeds can be pressed for oil and were valued as food; the Periplus, sections 14 and 41, refers to sesame oil in Indian Ocean commerce and production on the Indian coast. A reference to sesame oil does not prove that sesame plants were cultivated at the destination. Sesame oil, whole sesame seeds and seed kept viable for planting represent different kinds of trade, and only some of them imply that cultivation could spread.
Cotton could travel as seeds, processed fibre or finished cloth. The Periplus lists cotton cloth in commerce from western India; recovered cottonseed in Central Asian settlements may instead reflect local processing or cultivation. Cloth found in a distant port is not evidence that cotton was grown nearby. The water demands, labour requirements and growing season of cotton also mattered: an oasis could support cultivation only if farmers could allocate land and irrigation to it. The cotton recovered with grain and fruit in medieval Bukhara points toward varied agricultural production rather than a single-crop economy organized only for export.
Rice illustrates the long gap that could separate the arrival of imported food from its cultivation. Oryza sativa first developed as a domesticated crop in East Asia before spreading into South and Central Asia. The 2023 study of northern Iranian sites provides evidence of rice there by around 120 BCE, but the presence of rice grains alone does not show that irrigated paddies were widespread. A 2021 review in the journal Rice reconstructs two broad westward corridors, one involving southern maritime contacts and another crossing the inland exchange zones. The review finds rice in parts of West Asia and Europe by the end of the first millennium BCE, while widespread cultivation in much of West Asia developed considerably later.
People might buy and eat rice even in places where farmers had never planted it. Some rice grows on rain-fed or upland farms, so permanent flooding is not necessary for every variety. Nevertheless, growing rice reliably in arid regions often demanded irrigation, suitable soils, water rights and coordinated labour. It also had to compete with established grains such as wheat and barley. The first archaeological trace of rice cannot establish when it became an everyday staple. Its expansion involved changes in irrigation, labour, agricultural practices and diet over centuries rather than a rapid transformation caused by one imported shipment.
Moving a plant meant moving practical knowledge
When merchants transported dry pepper, aromatic resin or oil, the transaction could end with a buyer using the goods. Establishing a crop required growers to keep planting material alive and reproduce it successfully. Seeds had to remain viable; some fruit trees required grafting to reproduce desired varieties; cuttings could dry out; and corms needed careful storage. Ships had more cargo space than many caravans, but the surviving evidence does not establish that plants survived longer sea voyages more successfully. Ships could also be hot, salty and difficult environments for living cuttings, while short journeys between adjacent oases could allow plant material to be exchanged successfully. Living crops could be passed from one nearby community to another, rather than requiring a single shipment across the continent.
The separate histories of citrus species in the Mediterranean show that the first imported fruit could precede widespread cultivation by centuries. Dafna Langgut’s research on the citrus route draws on archaeological plant remains, written texts and visual representations to reconstruct its spread. The citron is attested in a Persian-period garden near Jerusalem in the fifth or fourth century BCE, while the earliest identified lemon remains in Rome date around the turn of the first century BCE and first century CE. These plants appear first in settings associated with elite gardens. Other citrus fruits became common much later. Citron, lemon and other citrus fruits reached Mediterranean societies at different times and followed different paths toward cultivation.
Ancient agricultural and medical writers recorded the methods and uses of some cultivated plants. In Enquiry into Plants, book 2, Theophrastus, writing during the fourth century BCE, explains planting conditions and the reproduction of trees and vines. His text does not document a specific Silk Road shipment, but it shows that successful cultivation required knowledge of soils, seasons and methods of reproduction. Dioscorides’ De Materia Medica, composed in the first century CE, records the properties and preparation of numerous plants and botanical substances. Their descriptions illuminate the practices and beliefs of their own periods, while any medicinal claim needs independent evidence before it can be treated as modern pharmacology.
Ancient texts tend to name rulers, merchants and ports while leaving out the gardeners and labourers who grafted trees, dug irrigation canals or planted unfamiliar seeds. Archaeobotanical evidence restores part of that missing history. It shows fruit reaching high mountain settlements, spices appearing in urban refuse and imported plants becoming components of local agricultural systems. Each new fruit or grain depended on local decisions about whether to buy, plant, propagate and keep growing it.
Plants associated with the Silk Roads reached different regions at different times. Wheat and millet crossed parts of Inner Asia before the historical Silk Roads; pepper reached Roman Egypt by sea from India; medieval Central Asians obtained fruit from lower valleys; and rice and citrus reshaped agriculture long afterward. The expensive spices noted by ancient writers were only part of this exchange. Farmers gradually adapted unfamiliar crops and selected varieties in fields and orchards, changing what people grew and ate.
Discussion
Comments