Pesticides on Museum Objects

Part of the Cook-Voyage project at the Pitt Rivers is to look at the question of pesticide residues on objects from the collections.  Chemicals have always been applied to objects in museum collections to protect them from pests, mainly insects, which can cause a lot of damage. 

As the Forster Collection has been in Oxford since 1776, and the Banks collection was at Christchurch even before that, the Cook-voyage collections potentially span virtually the whole history of pesticide use in museums.  Even though we know that chemicals such as arsenic, mercury and DDT, were routinely applied to museum collections, finding evidence of what was actually done is difficult.  

Label from a poncho from the Tahitian Mourner's costume, giving information on the condition of the feather cape.

A label on the Tahitian Mourner's costume, written just before the Ashmolean's ethnographic holdings were transferred to the Pitt Rivers in 1886, records that the cloak was 'now in a very bad condition' and an entry in the Accession Registers goes on to say that this was 'through the moth, and (it) has been removed to the Museum in the Parks (Oxford University Natural History Museum) and placed under the charge of Professor Moseley to be fumigated.'  We don't know, however, what fumigation involved. 

Sometimes, the Pitt Rivers annual reports give us more of a clue.  In 1946-47, the Curator, T.K. Penniman, wrote that 'we have tried the Clymax spray on all of our material, and have found it thoroughly effective in destroying every sort of pest that attacks every sort of material that human beings use.'  Powerful stuff indeed!  There is also a mention of a fumigation chamber being built so that methyl bromide gas could be used to kill pest infestations on museum objects, but it is not clear whether this was ever actually used.  

Jeremy Uden and Kloe Rumsey, conservation intern, taking samples of surface deposits from a Tongan mat.

  

The analysis of pesticide residues is being done for us by Andrew Charlton, an analytical chemist at the Wildlife Incident Unit, part of the Food and Environment Research Agency.  He usually works on investigating pesticide uptake by wildlife, especially honey bees.  The samples he works on are usually destroyed during the analysis - not something that is practical for museum objects.  We are using cotton wool swabs dampened with distilled water to take samples of the surface dirt and residues from objects - we know that the object is not damaged in the process, which is of paramount importance, but we have to accept that we may not be able to get large amounts of any residues remaining on the surface onto the swabs. 

Once Andrew receives the swabs, any residues on them are extracted with an organic solvent for analysis using gas chromatography - mass spectrometry (GC-MS), a highly sensitive technique which can detect parts per billion (ppb) levels of target pesticides.  Each target pesticide is detected in the mass spectrometer by the presence of characteristic molecular fragments providing a unique ‘fingerprint’.  To detect the presence of arsenic or mercury compounds, swabs are extracted with dilute acid for elemental analysis by inductively coupled plasma - mass spectrometry (ICP-MS).  Argon plasma at a very high temperature (approximately 6000 to 10000°C) is used to turn the sample solution into ionised atoms (ions). Each element produces ions with a characteristic mass and charge, which are detected by the mass spectrometer. ICP-MS can detect residues of arsenic or mercury pesticides in the swabs at ppb levels.

Cotton wool swabs used to remove surface deposits from a fragment of feather from the Tahitian Mourner's costume cloak.

The results of the analysis will help us to understand how pesticides were used in the museum, and may help answer questions we are starting to ask about whether the pesticides have contributed to damage, such as staining and deterioration in certain areas of some of the objects.  The results will also help us to make sure that the precautions we routinely take when we handle objects, such as wearing gloves and lab coats, are effective.

'A glaring ornament'

Forster No.133 Mus. No. 1886.1.1269

This breast ornament was collected on Tahuata in the Marquesas in April 1774.  Forster, in 'A Voyage Round the World' describes this 'glaring ornament', saying that 'their leaders wore a kind of gorget around the neck, or rather on the breast; it consisted of small portions of a light wood, like cork, glued together with gum, in a semicircular form; a quantity of scarlet beans (abrus precatorius, LInn.) are glued all round it with the same gum, forming a great number of rows, of the length of two or three inches.'

The gorget, known as a tahi poniu, is made up of seventeen sections of wood.  The red and black abrus seeds have been set in a thick layer of brittle brown resin, said to be gum from the breadfruit tree.

Microscope image of gum

As part of this project, samples of the gum were taken and will be analysed by scientists at the University of Bristol.

Sampling from 1886.1.1269

The gorget is missing quite a few seeds - even in 1884 when the collection was transferred from the Ashmolean Museum it was noted that 'many of the berries have come off this ornament and have been stuck from time to time with glue and many are wanting,particularly from the back edge.'  The back of the gorget has been extensively repaired, with new sections of plywood added to replace missing or damaged areas.  This was done before conservation records were kept.

The back of the gorget.  Replacement plywood rays can be seen near the top of the image.

Hunting for Crystals

A significant part of the project to investigate the Cook-Voyage collections at the Pitt Rivers is to attempt to identify some of the materials used to produce the objects.  Over the years, there have been many informed guesses about what materials are present, but no physical identification has been carried out on the majority of the collections.  Once materials have been identified, it is possible that they may have some characteristic that could help identify similar material in future.  One of these is the presence of crystals in plant cells.  These crystals, or phytoliths, are found in many plant species, and their presence could be one of the features that allows for a positive identification of a sample.  Phytoliths are commonly made from calcium oxalate, but other crystals have been found that contain silica.

This week I took samples of plant material from two objects to Dr. Debra Carr at Cranfield University.  Deb has worked extensively on methods for characterising plant fibres, particularly those used by the Maori (see for example her article 'Approaches for Conservators to the Identification of Plant Material used in Maori Artefacts'  in Studies in Conservation Vol.53 (2008) http://www.iiconservation.org/node/1180).  The samples I took to her were from Tahitian objects.  One was a piece of aerial root from a type of palm (Freycinetia sp.) that was used to stabilise the structure of the cloak from the Mourner's costume (1886.1.1637 .4).  The other was a piece of hibiscus (Thespesia populnea) fibre from a woven mat (1945.11.130).  Both samples had already been identified, and Deb was interested to see if they contained phytoliths.

The samples were first examined by X-ray diffraction (XRD).  This technique allows for the identification of materials with a regular crystalline structure by focusing a beam of X-rays onto them.  The X-rays are diffracted by the crystals and the diffraction patterns detected and analysed - these are characteristic for different crystals.  The small size and dispersed nature of many of the phytoliths in plant material means that it's hard to detect them using this technique, although it was possible to say that there were crystals in the samples.

The sample of Freycinetia in the XRD machine

Next, X-ray fluorescence (XRF) was used to examine the samples.  Again, the sample is bombarded with X-rays.  The elements within the sample release energy when they are excited by the X-rays, and the amount of energy released is characteristic of each element.  The XRF machine at Cranfield can build up an image of the sample, mapping out where the different elements of interest, such as calcium in this case, are located within the physical structure.  Again, there was some evidence, particularly in the sample of Freycinetia, that calcium was present just under the epidermis of the aerial root, where the presence of crystals might be expected.

 

Dr. David Lane carrying out XRF analysis

A map of the elemental composition of the hibiscus fibres obtained from the XRF

Finally, we looked at the samples under the electron microscope to see if we could visualise phytoliths within the plant structure.  The microscope uses a beam of electrons to illuminate the sample and to provide the image, and has a much greater resolution that a light microscope.  Although I was surprised at how much of the physical plant structure remained after nearly 250 years, we did not conclusively find crystals in either sample. 

Looking at the hibiscus sample under the SEM

It's possible that with more time we could find phytoliths within the plant material, and this would help us carry out identification in the future.  It's also possible, particularly in the case of the hibiscus mat fibre, that the way the fibre was processed before being woven, by being soaked in seawater for several days, could have removed any evidence of phytoliths.