Although the bronze material of the 2 monkeys has been dated to the 16^th/17^th century by a metallurgist at Oxford University (Dept of Materials), inexplicably, the 2 TL tests on the pottery core indicated a later date than the actual dating by metallurgical analysis. TL tests on the core indicated that the material of the sample was last fired less than 150 years ago. In view of the discrepancy of the tests on the metal itself and the internal pottery core it might, therefore, be useful to give a definition of TL testing and give information and opinions as to its reliability.

THERMOLUMINESCENT dating is an absolute method most widely used for ceramics and the residual traces of core material from the casting of bronze objects. No database is required for its use, but the method requires the drilling of a small sample from the object to be dated. A highly simplified description of the theory is outlined below. Impurities within clay deposits such as quartz and feldspar, absorb radioactivity from thorium, uranium and potassium causing the release of electrons which are then trapped within lattice defects. When a ceramic vessel is fired these electrons are freed and the build-up of radiation is cancelled. After firing the inclusions begin to absorb radiation again at a constant level, therefore the build-up of radioactivity will be proportional to the time elapsed since firing. When the sample is heated, the trapped electrons are freed and emit measurable light. A calculation involving the total TL emission and the annual radiation dosage are used to produce a date. Many variables have to be taken into account including rates of rainfall in certain areas and whether the original firing temperature was high enough to cancel the natural radiation. If a ceramic vessel is thought to be several thousand years old and a sample results in a TL date of a few hundred years then it is possible that the object is a forgery or possibly has been re-fired. When this method of dating was discovered it was thought to be a wonder method but now there are some reservations as to the reliability of the dates produced. From Apollo March 1990 Simple science in the service of authentication and attribution by Anna Bennett.

1 .On Renaissance bronzes the Victoria & Albert Museum Sculpture Department point out that TL testing has received some criticism recently and is usually considered in conjunction with some other information.

2. In respect of one monkey, the Research Laboratory for Archaeology and the History of Art, 6 Keble Road, Oxford, state that using standard methods and techniques it is estimated the material of the sample was last fired less than 150 years ago but it is only fair to point out their standard report includes a disclaimer of liability for financial loss incurred through an erroneous report being given. Sample 581n41.

3. Dr Mark L Evans, Assistant Keeper of the National Museum and Gallery, Cardiff, suggests that a THERMOLUMINESCENT test on the core of one of the monkeys can be tried to date the metal but the results are not invariably conclusive.

4. The State Museum, Berlin, (Prof Dr Josef Riederer) stated that the determination of the age of bronze sculptures by a thermoluminescence analysis of the clay core is not always very reliable and with bronze sculpture analysis of the metals usually promises more success

5  A conservator at the J Paul Getty Museum suggests that TL is just one technique that is used in the study and authenticity of bronzes and advises not to place too much on the results of just one type of testing. Attention could also be paid to a study of the surface with the naked eye and under low magnification, the facture, the metal and the core.

6. The British Museum (Department of Scientific Research) points out in the case of metal objects TL dating can only be applied to any residual amounts of ceramic core or mould which were heated sufficiently, at least 500 degrees C at the time the object was manufactured. The TL test measures the quantity, or dose, of ionising radiation the material has received since it was last heated. If the rate at which the dose has been acquired can be estimated then the ratio of the two, dose/dose rate gives an estimate of the age since the last heating. Whilst the dose can generally be measured precisely there will usually be some uncertainties in the estimate of the dose rate particularly for core materials so the age estimate will be subject to large margins of error.

7. An expert at the Ashmolean Museum, Oxford, has said that it is worth emphasising that TL tests are not infallible indicators of age as an object which has been subjected to high temperatures more recently such as a house fire or otherwise is likely to produce a recent reading.

8. Four Romanesque bronze Apostles, thought to be 12^th Century Lorraine, in the Ashmolean, Oxford, were considered to be of later date as a result of TL testing. Despite doubts cast by thermoluminescence dating on Moses and a Prophet, an origin in the period circa 1200 can be maintained. Research still in progress Pantheon 1986 Bruckmanns.

9. Artefacts discovered in Glozel, Auverne region of France, have been dated by TL testing as 700 BC to AD 100 but also have been dated to AD 1200 and AD 1350. The Oxford Research Laboratory delicately expressed for the first time the question: Is it not possible that in this particular instance, there is something mysteriously wrong with thermoluminescence? From fake, the Art of Deception, edited by Mark Jones, British Museum Publications.

10. Mrs D Stoneham (Research Laboratory of Archaeology, Keble Road, Oxford) was unable to convince a jury that her TL test was reliable in respect of Whieldon pottery and saltglaze wares. Guy Davies accused of making fakes of saltglaze and Whieldon pottery was acquitted. Antiques Trade Gazette report 2.4.94. Anne Crane.

11. Dr Jonathan Coleman, a specialist in nuclear radiation, physics and applied science, gave evidence that question aspects of TL methodology in particular how the possible changes in the quartz content could affect the results. See previous report Antiques Trade Gazette 2.4.94. Anne Crane.

12. TV programme, Horizon Out of Asia, BBC2, 25^th September 1997, 9.25pm questioned the reliability of TL testing. There are more reliable tests being devised.

13. Certain types of pottery give inconclusive results. To give an example TL test on slipware owl dated as circa 1700 Christie's London 24.2.97. was inconclusive but owl accepted by the experts as genuine.

14. Scientific techniques such as thermoluminescence dating have so far proved ineffectual in distinguishing Renaissance period stoneware from modern products. From Fake, the Art of Deception, British Museum Publications.

15. Sotheby's New York, Thursday, 30.1.95., lot 56, head of an angel, the model for the Habbakuk and the Angel group, terracotta 9 ¼" high covered in dark brown oil, by Bernini, dated by TL testing as 18^th Century but accepted by the trade as 17^th Century and genuine, realised $330,000 (£221,475) (Bernini 17^th Century Rome).

16. A prominent sculpture dealer (name to be withheld) believes TL testing to be unreliable. She had 4 differing dates on a terracotta from 4 different reputable laboratories. A well known historian of sculpture (name to be withheld)  believes that TL testing is a useful guide but accepts that occasional anomalies do occur.

Previous Page

Contents

Next Page