Two of the major sites in the far south of India yielding a wide spectrum of early coins are the bed of the river Amaravati at Karur and the bed of the river Vaigai at Madurai. Karur and Madurai were both major commercial centres during the ancient period and both were also active in the field of Indo-Roman trade. Karur is now a small town in Central Tamilnadu, and it can be found on maps to the west of Trichy (Tiruchirapalli). A significant proportion of the Roman traders who reached Karur came by ship to the port of Muziris in modern Kerala (described by Pliny) and then went through the Palghat Pass and across the Kongu plain to Karur. The overland part of this route has been littered with coin hoards consisting mainly of early Imperial denarii and sometimes of aurei. Western copper coins do not feature in these hoards and are usually recovered from the river bed. From around the middle of the 4th century AD until the second half of the 5th century small, late Roman copper coins were themselves an important trade commodity, just as denarii had been during the earlier phase of Indo-Roman trade. These comments illustrate the background to Krishnamurthy' s study.
       Krishnamurthy has been studying the coins recovered from the river bed at Karur (and also at Madurai) for many years. His numerous papers have culminated in the publication of three books: Late Roman copper coins from South India: Karur and Madurai (1994), Sangam Age Tamil coins (1997) and the volume being reviewed here.
       Non-Roman ancient foreign coins from Karur in India complements his two previous books and presents coins in his own collection which originated from outside the Indian subcontinent. The book is arranged into nine chapters, some of which are subdivided. Also included are a glossary, bibliography and index. The chapters are organised around the places where the coins found at Karur had originally been minted. These are spread around the eastern Mediterranean plus the mainland to the east. They are, in order, Thrace, Thessaly, Crete, Rhodes, the Seleucids, Phoenicia, Askalon, the Priest-Kings of Judaea, the Roman governors of Judaea, Parthia, Edessa and Aksum. The kingdom of Aksum in modern Ethiopia lay along the Red Sea maritime trade route to India. Each chapter consists of a brief introduction to the region and its coinage, followed by a description of the coin, or coins, with details of size and weight, and accompanied by both photos and line drawings. A total of forty one coins is catalogued. Most are of small size and in worn condition, as is also the case with late Roman bronzes of similar provenance. In the majority of cases sufficient detail is retained to identify the issuing city or state, but a few attributions can be debated.
       This is an important study that opens up a new aspect to the investigation of ancient coin circulation and trade. Only a decade ago few scholars would have given serious consideration to any proposition that Hellenistic coins made their way to central Tamilnadu. Now this provenance is beyond doubt. I remember my own scepticism when hearing, on a visit to Nasik, that someone claimed to be finding Ancient Greek coins at Karur. So I went to have a look. This proved to be the first of several enjoyable meetings with Mr Krishnamurthy and sessions examining and discussing his coins. A visit to Madurai brought to light an Aksumite copper coin among a motley selection of generally worn, late Roman, Chola and Vijayanagar coppers from the river Vaigai. A visit to the groups of people digging the river bed at Karur yielded a Hellenistic copper coin from Cos (The coinage and history of Southern India; part 2). I found ample corroboration for Mr Krishnamurthy's assertion that Ancient Greek coins really did reach the far south of India.
       The interpretation of these coin finds provides the subject for the final chapter in the book: Observations. This is a subject that will no doubt continue to be debated for some time to come. Krishnamurthy favours the view that these coins reached southern India before the main phase of Indo-Roman trade, but discusses the alternative view that they travelled east during the Roman period (when many of the traders were Eastern Greeks, the Yavanas of Indian literature). These coin finds raise questions concerning trade between India and the West. They also raise questions concerning both the time span and the geography of circulation patterns for Hellenistic copper coinage. Krishnamurthy's book will be found relevant by those who are interested in Indian coinage, in Hellenistic coinage and also in ancient trade.

       Michael Mitchiner