Counting Money: Rehousing the Greek and Roman Numismatic Collection in the Auckland War Memorial Museum Tāmaki Paenga Hira

Abstract

The numismatic collection in the Auckland War Memorial Museum Tāmaki Paenga Hira, Auckland, New Zealand, is an extensive although little-known collection. As part of a wider rehousing project for the coins, a trial was conducted to rehouse the Greek and Roman coins to establish the best storage and analysis practices. Records were also created or updated during this process to make them visible online and enable their use in future research. Best practice standards include handling and measurement practice, retaining all existing information about the collection, and, where possible, reuniting coins with their acquisition information. The project’s end goal is to create populated records in the museum database (Vernon) for the entire numismatic collection, either through updating existing records, creating records using available acquisition information, or creating records with temporary numbers where that information has been lost. This report gives a brief history of the collection and the rehousing process and suggests future research avenues.

1. Introduction

Auckland War Memorial Museum Tāmaki Paenga Hira, Auckland, New Zealand, houses approximately 30,000 coins in its numismatic collection. These coins originate from contexts worldwide and cover periods dating from ancient Greece and Rome to the modern day. While a rough count exists, according to an inventory project in 1996, around two-thirds of the collection is unregistered. Between November 2023 and February 2024, the Greek and Roman coins in the numismatic collection were catalogued and re-packed to create standards representing the best practices for cataloguing the remaining collection. While the numismatic collection includes coins, banknotes, tokens, stamps, and medals, the Greek and Roman collections were of particular focus as a starting point for future collection management, as it has already been identified by Christopher Ehrhardt [1,2,3,4,5,6,7,8,9,10,11,12,13,14]. Bringing the collection up to a best practice for conservation and collection management has made the coin collection available for future research and reduced risk to the museum in the form of unregistered coins. Another impetus for this work is to make the collection available online so that it may be incorporated into other corpora and contribute to wider numismatic discourse. This paper explains the process of re-cataloguing the Greek and Roman coins housed at the Auckland Museum.

Coins were some of the earliest objects obtained by the Auckland Museum when it first opened in 1852. Records of early numismatic acquisitions did not often record where they came from, thus losing their provenance information. Donors were often recorded, but as acquisition numbers were not transcribed onto the coins, the connection between the donor and the coin itself was also lost. A representative example of this from the Annual Report of the Auckland Institute and Museum, 1947-48, is “old Roman Coin from Mr Keith Gross in 1946” (Auckland Museum Annual Report [15]:96). While, as mentioned, most coins lack catalogue entries, others were catalogued previously, either through prior projects such as the current one on other parts of the collection or through bulk additions to the museum database. However, many of the extant records are sparse in detail, and updating these records was also an objective of the project.

The Auckland Museum acquisition registers and records indicate a general trend in the growth of the numismatic collections. There is a combination of larger donations, mainly from high-profile collectors, including early colonists in the history of Auckland, like J.C. Entrican and R.C. Barstow, and smaller donations by individuals, including many from former soldiers returning from overseas service, which was a common trend for the museum post-World War I and II (e.g., [16]). However, reconstructing these collections in some cases is difficult, as coins were separated into type and period, and therefore, there is little consistency or few assumptions that can be made about coins coming from a particular collection. In addition to a lack of information when they were first acquired, the collection has been moved several times and managed by many curators and record-keeping systems, causing various errors such as duplications and misattributions. Throughout the project, where possible, these issues were resolved.

2. The Greek and Roman Numismatic Collection

In total, 292 coins of Ancient Greek origin and 1150 of Roman origin were identified. The Greek and Roman collections range from Roman Britain to Greek India, chronologically from the 5th century B.C. to the 5th century A.D., not including modern forgeries and imitations. The Greek coins were provenance from the Classical and Hellenistic periods and Ptolemaic, Parthian, Bactrian, and Indo-Greek issues. There is a collection of Nabataean coins, all but two of which come from the wider ethnographic collection of J. Pearson. Only 11 coins are from the Roman Republic, with the remainder being Imperial, going up to the reign of Valentinian III in 455 A.D. A significant part of the Roman collection is a hoard of 296 Constantinian bronze coins, of which no provenance or acquisition is available. It is designated as a hoard because it was found in a single container and exhibits patina and wear consistent enough to indicate it was a single hoard [17].

Some significant donors of ancient coins to the Auckland Museum are R.C. Barstow, J.C. Entrican, C.D. Stanier, H. Green, and C.E. Tisdall. While we cannot precisely attribute coins to these collections for the most part, it is estimated that around 700 coins of the collection come from these early donors. Barstow was an early magistrate for Auckland during the late 19th century and one of the foundational members of the Auckland Institute. His donation is recorded in the annual report for 1878/79 [18] and includes a handwritten register of his collection and other lists and ephemera. The register lists a broad collection of much more than the “Roman Coins” listed in the annual report entry; however, no coins have yet been positively identified as being from Barstow’s collection. Cases like this demonstrate the difficulty of connecting coins to their acquisition information.

Dr Christopher Ehrhardt was a professor of classics and an honorary curator at the Otago Museum. Over his lifetime, he aimed to find and record any public collections of ancient Greek and Roman coinage in New Zealand museums and institutions. Ehrhardt travelled to Auckland in 1982 and took three groups of coins back to Otago to catalogue. From 1982 to 1991, he returned to Auckland and wrote a series of catalogues that identified and measured all the Greek and Roman coins from the Auckland Museum. Coins were first separated by region and period, then further by their minting authority, and ordered chronologically when possible. Ehrhardt mentions in his work the lack of information about the coins’ acquisitions and that he could no longer find some of the coins listed [17]. His work has formed the focus for a large section of the initial project, as his identification was retained by the museum when the collection was moved in 1997.

The numismatic collection was moved into its current storage in 1996 and 1997. During this process, the collection was indexed and counted, with the contents of each box listed and a broad count of the collection’s scale, both registered and unregistered. This process aided in discovering at least 50 other ancient coins that Ehrhardt did not count. While records of various acquisitions exist, the fundamental problem is identifying which coins belong to that acquisition. Barstow’s catalogue is vague and sometimes matches no known coins in the collection, so it is not easy to understand, without other references, which coins came from his donation. For example, Ehrhardt notes that he believes most Parthian coins are from the Entrican or the Tisdall collection [9]. Entrican donated his “extensive” collection of ancient and modern coins and numismatic texts in 1947 [19], several hundred of which have already been registered. Still, no catalogue lists the extent of the collection. Ehrhardt labels five specific Parthian coins as being from the collection of C.E. Tisdall [9]. C.D. Stanier’s donation of 75 “Roman Coins” is noted in the annual report for 1876–1877 [20], but as above, no further information has been found to indicate which coins these could be. There is a similar lack of existing catalogues or information to reunite “53 Roman Coins” from H. Green, listed in the annual report for 1896–1897 [21].

3. Rehousing the Collection

The methods used to process the Greek and Roman coins were designed to conform to best practices for object handling and to retain as much information as possible that may have been associated with each coin, whether that be bag labels, box labels, or other material related to the collection. For the collection to be easily accessible, the Vernon (Auckland Museum’s internal database) records created were as detailed as possible, particularly given the aim for the collection to be available to the public through Auckland Museum’s Collections Online. For each coin, any prior information was retained, such as box or drawer number and position within the container. Any information on a coin’s previous bag or slip was also kept, as well as other material stored with the coins, such as letters or envelopes.

Each coin was measured with plastic callipers, giving two measurements for diameter taken at a 90-degree angle and a measurement for maximum thickness and weight. A brief and broad description of its metal was given based on visual analysis. Around 80% of the coins catalogued were of various base metals, largely bronze and copper, with the other 20% being silver. Categories were added for period, cultural origin, and place of manufacture to correspond to Vernon fields. Coins were also checked to ensure that the references to Ehrhardt’s catalogues on their slips were accurate. Some were discovered to be misattributed, and the proper attribution was re-ascribed.

There is a variety of available studies in the literature which outline standards of collection care for numismatic collections. For example, The International Committee of Money and Banking Museums [22] provides a valuable resource for collection management and care. Numismatics studies were also useful but focused on identification standards rather than collection care or cataloguing standards. Furthermore, broader manuals and studies of proper museum collection care practices and collection management by Powell [23], Szczepanowska [24], and Decker [25] were useful for creating an overview of industry best practices for handling collections. This includes outlining issues relevant to working with coins, such as avoiding specific corrosion and wear processes. These texts are designed as guides to institutions and contain a good academic basis for understanding the collection care standards already in place at Auckland Museum that were utilised during the project.

The best practice for handling metal coins involves wearing gloves, particularly nitrile, as they are the museum conservation standard [23]. This reduces the risk of creating new possibilities for corrosion, as the acidic residues left by bare hands can lead to tarnishing on more sensitive metals like silver and copper alloys [22] which, while immediately obvious, is difficult to remove [24]. Some examples in the collection show tarnish that could have been caused or exacerbated by previous improper handling. One example is the top half of this Greek silver (Figure 1). Additionally, Tyvek-covered and cushioned tables were used as workspaces for further protection from wear or contact with acidic materials. The coins catalogued throughout this project showed a variety of wear, patina, and corrosion. While corrosion cannot be halted without treatment [24], it can be slowed down by storage in a protective environment until the appropriate treatment is applied.

Each coin was removed from its previous housing and placed into an oil-free, chemically inert coin slip, made of Mylar on which an acid-free label was attached with the assigned acquisition number. Boxes and trays of acid-free material were specially made to fit the coin slips for this project (Figure 2). Each box contains four trays, which can be removed individually, with each containing a row of coin slips. Some coins were found to be too large for the width of the trays, and these were placed in larger inert coin slips and placed perpendicular to the tray sequence with the appropriate supports and separation to reduce movement.

Conservation standards recommend that metal coinage be stored in a chemically inert and acid-free environment to prevent corrosive vapours from coming into contact with artefacts and causing tarnishing [24]. Copper alloys oxidise in acidic environments and silver tarnishes in contact with sulfuric compounds, both of which can be created by certain woods, paints, fabrics, or adhesives [26]. This was an issue as much of the collection was stored in cardboard boxes, many of which contained the original wooden trays with velvet inlays that the coins were stored in previously. Both unsealed wood or wood treated with non-inert chemicals and velvet are susceptible to off-gassing over time, which will lead to damage to the coins that are either in direct contact or in the same microclimate as the gases [24]. Additionally, those coins from the Greek and Roman collections were placed in cardboard boxes, in either cellophane slips or small Ziplock bags. The cellophane had yellowed and, in places, split over time, indicating the need for a new storage system. The Ziplock bags also took up more space than necessary and allowed for too much movement of the objects within, possibly contributing to wear when moved. By removing the coins from their previous housing and placing them into individual, chemically inert oil-free slips, the goal was to halt any ongoing corrosion that may be occurring due to off-gassing as well as to reduce space and make each object more visible, reducing the need for direct handling.

4. Processing the Collection

Archival research was conducted to connect coins with acquisition information, with any findings recorded and added to Vernon. The primary resources for this research were the Auckland Museum Accession Registers, the annual records, Vernon records, and other collection files like the 1996 Numismatics Box Register. The project was not to analyse the significance of the coins in the collection but to create an easily accessible record that allows the numismatic collection to be available for future research. The assignment of acquisition numbers was, therefore, an important step. Where it was possible to relate coins to a particular acquisition, those numbers were assigned (e.g., 1918.27.## for the Pudsey Hoard). Following the registration process at Auckland Museum, new numbers beginning with 2024 were assigned for those objects without a known acquisition.

4.1. Examples of Coin Processing

The amount of information possible to ascribe to a coin varied, ranging from a full ascription to their acquisition to no information about how they were acquired. The best possible outcome is that acquisition data and any provenance information can be reconciled with coins, which is invaluable for future research. Here, four examples are presented that demonstrate the different levels of processing encountered.

4.1.1. Coin 1918.27.1

Coin 1918.27.1 (Figure 3) is one of four coins labelled by Ehrhardt as the “Pudsey Hoard” [13]. These coins were found in an envelope and letter, both degraded but still legible, which showed both the hoard’s provenance and their acquisition into the museum. Dated to 1918, the writing on both describes how the coins were excavated in a quarry near a Roman camp in Leeds and were acquired by Dr. C.E.A. Coldicutt, who then presented them to the museum in June of 1918.

The hoard contains four “radiates” of low-quality silver bullion from Roman Britain minted between c. 260 A.D. and 274 A.D. [13]. They depict two Roman emperors and one Caesar of the Gallic Empire, a splinter state created during a period of long-term political instability in the Roman Empire during the third century A.D. [27]. This coin was minted in the name of Gallienus, who ruled the Roman Empire between 253 and 268 A.D. [28]. The obverse features a bust of Gallienus with a radiate crown facing right, and the inscription IMP GALLIENVS AVG is partially visible. The reverse shows a stag walking right with its head facing left, with the inscription mostly illegible.

4.1.2. Coin 2024.x.313

Coin 2024.x.313 is a first-century B.C. Parthian coin which had been labelled as being from the Tisdall collection by Ehrhardt both in his catalogue of Greek coins at the Auckland Museum [9] and on the cellophane slip the coin was housed in (Figure 4). However, we have no further evidence to corroborate this, as Tisdall’s collection has no register. The annual record mention of this acquisition reads “67 Ancient Silver and Copper Persian Coins, Many of them of Greek Age” [29]. While the coins may be from Tisdall’s collection, without other evidence, Ehrhardt’s attribution is the only link these coins have to a possible acquisition.

The collection mainly comprises coins with no acquisition or provenance information. This creates issues in cataloguing them, as they cannot be assigned proper acquisition numbers and because it limits the amount of research that can be conducted. While studying the archaeological context of coinage has always been an essential part of numismatics, the field in its early form in the 19th century developed a focus on typology, dating, and artistic detail distinct from the developing discipline of archaeology [30]. Collectors of the 19th and 20th centuries, such as those who contributed to this collection, prioritised stylistic detail and historical context over the provenience of artefacts [31], exemplified by our current lack of information.

4.1.3. Coin 2024.x.640

Coin 2024.x.640 was struck in A.D. 43/44 in Alexandria, under the emperor Claudius, and depicts his first wife Messalina on the reverse, indicated by the remaining section of the reverse inscription ΜΕΣΣAΛΙΝ (Figure 5). This coin is significant because there is evidence that it has been deliberately defaced, as its upper half is flattened with what appears to be a hammer. Ehrhardt notes that this may have been an ancient response to her death and damnatio memoriae or removal from the public record following accusations of treason and bigamy. While this coin is worthy of future research, the conclusions that can be made are limited to what can be observed on the coin itself because the object has no provenance or acquisition information.

4.1.4. Coin 2024.x.1312

Coin 2024.x.1312 is an example of a complete lack of information (Figure 6). In box Miscellaneous 6 were two trays of ancient coins, despite the 1996 box inventory list mentioning no ancient coins within this box. One tray contained largely Ptolemaic and Roman bronzes, and the other contained a collection of Hellenistic issues, all dull greyish metal encrusted with pale dirt. No associated information identifying these coins or acquisition records that align with the find exists. Coin 2024.x.1312 is one of the Hellenistic coins, depicting a common obverse image of Alexander the Great and his successors, the head of a young Hercules in a lionskin facing right. The reverse depicts a seated Zeus holding out an eagle to the left and holding a staff behind him to the right. Because of the quality of the metal and the crudeness of the depictions, there is a possibility that these are forgeries. However, because there is no information about them, no definitive conclusions can be drawn without further analysis.

5. Making the Most of What There Is

The lack of appropriate records and acquisition information is not unique to the numismatic collection at Auckland Museum. Cereti and Colliva [32] outline a similar project, where outside researchers helped three Iraqi museums measure, identify, and catalogue their numismatic collections and created a robust system for creating digital catalogues of coins. Valsi [33] is a study of a hoard of previously unpublished medieval Roman coinage found in the Capitoline Museum. The hoard has no definitive findspot or provenance, and the acquisition into the museum is only theorised based on 19th-century documents. However, it was still possible to carry out a hoard analysis. Another common type of literature is catalogues of legacy collections, like Ehrhardt’s catalogues, which formed the basis for much of this project. Jansari [34] created a catalogue of Roman coins from South India and Sri Lanka, which is part of a much more extensive collection of Indian coinage dating from British colonial occupation there. Williams [35] similarly worked with an ancient collection with limited definite provenance or acquisition data, as the objects were collected for the Florentine Archaeological Museum during the 16th and 17th centuries.

There is a debate about why coins are separated from other materials, often found in context with particular relation to those found in archaeological assemblages [30]. Coins do not exist in a vacuum but can provide a wealth of information about the economics of the ancient world, in addition to information about the formation of the archaeological record. However, this only applies to those coins whose archaeological contexts have been properly retained, which is often a minority of coins in legacy collections such as this one. Additionally, contemporary activities such as metal detecting have increased the rate of finds which, in many cases, are not recovered with proper excavation methods. For many coins in a museum context, as is demonstrated, any provenance information is often lacking, and when it does exist, it is often linked to a collector as opposed to a physical location. This further highlights the importance of adequately making all data about the coins available, a call made previously [36], and one that online resources have aided with [37].

In making the coins available for further study, remotely or in person, the Auckland Museum coins are now more accessible than they were previously. There are considerable opportunities for future research projects, both into the history of specific collections and hoard analysis. Additionally, archival coin images can, in the future, be added to more extensive databases like the Online Coins of the Roman Empire project, as they use coin images from institutions to visualise catalogue entries and expand available knowledge. This will enable the collection to be visible to numismatics scholars and contribute to ongoing research.

Further research and collection enhancements could involve 3D technologies such as photogrammetry to create accessible 3D models of the coins, which has significant potential to improve numismatic study. A model of Coin 2024.x.248, the largest coin catalogued, was created using photogrammetry to indicate the potential of this process (Figure 7). Three-dimensional models retain the details and physicality of a coin much better than traditional photos of the obverse and reverse and could allow for future work that would be able to recognise die links or create die-wear sequences automatically [38,39].

6. Conclusions

While lamenting the lack of data accompanying the numismatic collection is tempting, this project has further highlighted the need to work with what is available. Recording whatever information is available and appropriately housing the collection has opened the doors for further research and study of the Greek and Roman numismatic collection at Auckland Museum. Additionally, the project has created a standard of best practice for handling and cataloguing the numismatic collection at Auckland Museum and a process that can be followed in-house and adopted at other institutions. Further work is still required on the collection to rectify coins with original acquisitions as new information comes to light. While there will be different issues and considerations in future work focussing on the rest of the numismatic collection, following the workflow and best practices outlined by this project will create a world-class collection that can be shared with the public and contribute to broader research projects, in particular, by making this previously inaccessible collection available to the wider public and scholarly communities online, where it can be incorporated with larger corpora.