Some of the following studies have been based on objects in the Islamic ceramics collection of the Ashmolean. Where possible and enlightening, these objects are pointed out throughout this CD-ROM programme, with a short discussion of the methods of analysis used and why, and what the results were.

Allan et al (1973); Dayton and Bowles (1977)

These two articles both attempt to clarify elements in Abu’l-Qasim’s treatise on ceramics which are not fully understood, either because they are words that cannot be translated or ingredients that cannot be identified. The second article was written in response to the first, with some additional investigations; the first attempted to establish Abu’l-Qasim’s ingredients and to see how widespread they were in C12th and C13th Persia. This was done by analysing the glazes and bodies of 15 objects from the Ashmolean collection which were dated stylistically to c.1300 and thought to have been manufactured in or near Kashan.

The glazes were analysed for tin and lead oxide content (as described in Abu’l-Qasim) using X-ray fluorescence spectrometry, and two distinct groups were identified: the first group (7 objects, all opaque glazes) was found to have lead oxide content of 13-20% with concomitantly significant levels of tin oxide; the second group (8 objects, all transparent glazes except for one opaque-glazed anomaly with hign tin/low lead content) contained less than 0.5% lead oxide and also negligible quantities of tin. These results confirm the description in Abu’l-Qasim of the production of opaque glazes from tin and lead in conjunction.

The clay bodies were analysed by optical emission spectrometry, and samples were obtained by drilling with a fine tungsten carbide drill (c.1.5mm diameter). The silica content of all the bodies was found to be between 83-94%, with soda as the next most significant ingredient (2.25-5.5%) and all other ingredients present in trace quantities. This composition is very similar to ancient Egyptian faience and at the end of this article is an interesting discussion on the possible transmission of this technique from ancient Egypt to medieval Persia.

While the proportions of the major constituents were instructively identified, the actual ingredients used by Abu’l-Qasim and his family were not. Dayton and Bowles took a trip to Kashan and chatted to "some friendly locals" who took them on a tour of the mineral deposits and rocky outcrops of the area. It seems that they managed to identify the "Qamsari stone" mentioned by Abu’l-Qasim as a kaolinised granite that occurs 1 mile out of the village of Qamsar, and whose most significant constituents are silica and soda. They also believed that they found the source of the colourant "lajvard" in an outcrop of fibrous haematite along the river bed, also near the village.

This study shows how scientific analysis can be combined with more practical experiments to reach a fuller understanding of the conditions of the time. The very localised ingredients uncovered by Dayton and Bowles probably account for why Kashan was such a thriving production centre in comparison to other Persian sites, which did not have access to ingredients which produced such fine results.

There has been a collaborative programme of analytical investigations into early Islamic ceramics between the Royal Ontario Museum and the Research Laboratory for Archaeology and Art History, Oxford University, the results of whose efforts are published in a series of articles by Mason and colleagues, and brought together in Mason’s D.Phil. Dissertation. Many of the objects analysed are in the Ashmolean collection (list hopefully soon to be compiled).

In the first article (1991) Mason took a typical cross-section of "Samarran" pottery (see above on Samarra horizon) and subjected them to petrographic analysis, which was the method of analysis that in his experience most consistently produced relevant and meaningful results. He identified petrofabrics for Basra and Siraf by analysing pottery which had a definite provenance, ie. kiln-furniture and sherds which had been left in the kiln area for whatever reason. He then took a selection of pottery found at Siraf to identify which of the pots had actually been made in Siraf and which had been imported from Basra. The results showed that all the luxury glazed wares found at Siraf (transparent turquoise glazed wares and opaque-glazed wares with cobalt blue-painted or lustre decoration) were produced at Basra and exported to Siraf; occasionally these imports were supplemented by locally produced wares which tended to be more easily produced wares such as Splashed Lead glazed wares.

Mason’s D.Phil Thesis confirms the indication, seen in microcosm in the relationship between Basra and Siraf, of major production centres in the Islamic world with wide distributions. He applies the methods of petrographic analysis to ceramic samples from many different provenances around the Islamic world: one of his significant results has shown the definite dominance of Kashan as a production centre in C12th and C13th Iran. While abundant evidence has always pointed to this fact there has never been any definite proof that this was so: now there is. Mason also shows that Kashan was dominant as a production centre for other, coarser wares which dominate ceramic assemblages at such sites as Gurgan and Ghubeyra.

One other significant result of his investigations is that the developments of stonepaste bodies and tin-opacified glazes occur side by side in Iraq in the C9th, achieved through a number of experimental phases, and these exact techniques move in the C10th from Basra to Fustat. Coupled with similarities in decorative schemes, this helps to establish a case for the movement of potters from Iraq to Egypt where the ascendancy of the Fatimids (969-1171)was attracting craftsmen from all over the Islamic world. With the fall of the Fatimids at the end of the C12th, Mason has shown that the stonepaste/tin-glaze techniques move along with the craftsmen from Egypt to Syria and Iran. This was thought to be the case because of the movement of the highly specialised lustre-industry from Egypt to Iran: now it seems that there is proof.

Carr (1990); Middleton et al (1992).

As mentioned above this technique was first used to investigate mummies in museum collections which were too fragile to be unwrapped or taken out of their cartonnages. The Middleton (et al) article discusses such investigations by xeroradiography at the British Museum: results revealed features such as amulets included within the wrappings of the mummy, the configuration of the arms folded across the chest, evidence for broken bones, and even tricks of the embalmer’s trade: an examination of the mummy of Rameses II showed that a small animal bone had been used to support the nose to preserve its distinctive beak-like profile. More recently, xeroradiography has been used in the examination of Lindow Man, and allowed details of his ritual killing to be constructed.

Carr has applied xeroradiography to studying ceramics from southern Ohio, and he advocates this method as a helpful tool for archaeologists in identifying sherds to their vessels of origin and evaluating vessel function and formation. He has sorted more than 3000 sherds from 85 different provenances by this method, and can identify analystically individual vessels: for example, out of a selection of 352 sherds, he was able to identify 68 individual pots. The method is less reliable in identifying appendages (rims, handles, feet) when they are made in a different clay from the body. It also becomes less feasible when the sample may contain many thousands of sherds. However, Carr emphasises the value of this method in allowing a shift in the fundamental unit of analysis from sherd to vessel, thus making available a large enough area of vessel to estimate vessel shape/volume/function/decoration. He claims that museum curators can even use this method to check if past reconstructions of vessels have been done accurately or not!

N.B. an investigation into petrographic analyses of Timurid ceramics based on Tamerlane’s Tableware will stand on its own as case-study in main programme.