Preservation fluids of heritage anatomical specimens — a challenge for modern science. Studies of the origin, composition and microbiological contamination of old museum collections

Abstract

Anatomical museums preserve specimens of great historical value and undiscovered scientific potential. However, frequently these collections lack documentation of the techniques of preparation and the composition of preservative substances (conservation principles). This poses a huge problem for the care and preservation of these materials, more so because understanding this issue requires knowledge of fundamentals from different scientific disciplines. The aim of the research was to obtain information about the composition of substances used to preserve historic specimens, as well as to conduct a microbiological assessment of the specimens to detect possible factors causing their deterioration. Furthermore, we wanted to fill an existing gap in the literature, as there is a lack of reports on analytical methods that could be successfully applied by anatomists involved in the daily care of museum collections in human anatomy departments. The starting point was the analysis of the sources and history of the collections, on which basis the choice of research methods was made. Methods based on simple chemical reactions and specialised methods (such as gas chromatography‐tandem mass spectrometry, Fourier transform infrared spectroscopy, inductively coupled plasma optical emission spectroscopy) were used in the analyses of the composition of fluids. Microbiological analyses were based on culture and isolation methods, analysis of microscopy slides and matrix‐assisted laser desorption/ionisation time‐of‐flight mass spectrometry analysis. As a result of these analyses, some components of the preservative mixtures and their concentrations were determined. The presence of methanol, ethanol, formaldehyde and glycerol was detected, among other chemicals. The concentrations of these substances were different between the samples and their determination required the use of a variety of methods suitable for the individual components of the preservative mixture. In microbiological tests, both bacteria and fungi were isolated from swabs taken from anatomical specimens. The bacterial flora was less numerous than the fungal flora. Among the bacteria, environmental Gram‐positive Bacillus cereus, Bacillus thuringiensis and a rare bacterium of the Cupriavidus genus were isolated, whereas among the fungal organisms, the yeast‐like fungi Candida boidinii and Geotrichum silvicola as well as mould fungi Penicillium sp. and Fusarium sp. were detected. However, the microscopic evaluation showed a greater diversity of microorganisms, which may be related to the fact that many environmental bacteria cannot be cultured using classical methods, but can be observed under the microscope. The results of the research made it possible to draw conclusions about the mutual influence of physical, chemical, and microbiological factors on the condition of historical anatomical specimens. In the course of the research, information was obtained on the processes which could have taken place during the storage of these collections. Maintaining the integrity of a container housing a preserved anatomical specimen has a major impact on maintaining the concentration of preservative fluid and keeping the specimen environment sterile. Many conservation procedures for historical specimens carried out nowadays pose a risk of destroying valuable specimens, as well as a health risk for the person carrying out the work. The exploration of the topic of conservation of anatomical specimens, especially those that lack documentation of their origin, is a key issue in current research on historical collections of anatomical specimens.

Each of the conservation fluids examined here had a different composition. None of the analytical methods applied here is selective enough to be used alone. Under certain conditions, conservation fluids can be colonised by micro‐organisms. The sealing of the specimens jars is a major factor in determining specimens quality.

1. INTRODUCTION

Measurements of fixed and/or preserved tissues are commonly used in morphological sciences (Waltenberger et al., 2021). For developmental and foetal anatomy, formaldehyde‐preserved research material is the main source for morphological analysis (Beger et al., 2021; Grzonkowska et al., 2021; Paruszewska‐Achtel et al., 2018).

It has been shown in animal examples that formaldehyde constricts specimens, changing the original morphological dimensions (Patten & Philpott, 1921). It has also been proven that different chemical mixtures affect the dimensions of organs fixed or preserved with them in different ways (Bakıcı et al., 2017). The effects of preservatives on human tissue are difficult to predict.

Foetal developmental studies are often conducted on material collected and preserved in the distant past, because due to legal changes and advances in medical science, obtaining such specimens is virtually impossible today (Dudek et al., 2018; Krzyżak et al., 2014; Masłoń et al., 2013). In addition, medical developments have eliminated many of the diseases of pregnancy and the puerperium that were commonly observed in the past (Hokama & Binns, 2012). These factors have resulted in reduced availability of fresh foetal specimens depicting concrete infectious diseases, malformations, or congenital syndromes. Thus, the progress of medical science has indirectly strengthened the importance of historical museum collections—preserved historical foetal specimens have become a unique educational and research source (Paluchowski et al., 2016).

A significant limitation to the usefulness of historical anatomical specimens is the sparse amount of data that has survived to the present day. Many collections survived the periods of war in the 20th century in a limited form, with the technical documentation usually severely damaged and dispersed. The lack of technical information makes the contemporary conservation of the anatomical specimens difficult. The methods of preservation and the chemical composition of the historic preservative fluids remain unknown (Domański et al., 2023).

To preserve museum collections in the best possible condition, conservation activities should be based on methods with proven safety and effectiveness. Research on human foetuses has been conducted in the Department of Anatomy of the Medical University of Wrocław since the end of the Second World War (Goździewski et al., 1985; Nizankowski, 1967). The rich collection of teratological and developmental specimens stored in the Museum within the Department of Anatomy indicates that the tradition of this research in Wrocław is much longer and is connected with the pre‐WWII period (Domański et al., 2023; Otto, 1841).

In the course of inventory work in the basement of the 19th‐century building of the Department of Anatomy of the Medical University of Wrocław, at the turn of 2019/2020, interest was aroused by a neglected anatomical collection that consisted mainly of teratological specimens. The extent of their conservation is unknown. This discovery initiated efforts to rescue and conserve the collection, which was most likely hidden in the basement of the building during World War II to avoid damage during bombing.

An interdisciplinary team was formed to undertake conservation work on the stored anatomical specimens in order to save the collection and conduct a study to provide new insights into the effects of preservatives on the condition of the tissues. A variety of chemical and microbiological analyses were carried out for this purpose, as well as a literature review and analysis of historical sources.

The aim of our research was to determine the chemical composition of the preservative fluids and conduct a microbiological evaluation of the preservative fluids and the specimens.

In addition, we wanted to assess whether simple methods for detecting the chemical composition of preservatives were useful and whether they could be compared with recent advanced analytical techniques (e.g., gas chromatographery‐triple quadrupole tandem mass spectrometry [GS–MS/MS], Fourier transform infrared spectroscopy [FTIR], inductively coupled plasma optical emission spectroscopy [ICP‐OES], matrix‐assisted laser desorption/ionisation time‐of‐flight mass spectrometry [MALDI‐TOF MS]).

2. MATERIALS AND METHODS

The process of material selection, the choice of research methods, and the selection of techniques used were multistaged and followed the objectives of the project (Figure 1).

FIGURE 1.

A diagram of the research process (simplified).

The results from the three research cycles (source studies, chemical analysis, and microbiological analysis) were compared, as aspects of this researches were found to be strongly associated. This resulted in conclusions that are valuable for conservation as well as scientifically relevant.

2.1. Material

The material came from the collection of historical museum anatomical specimens deposited in the Anatomical Museum of the Department of Anatomy of Wroclaw Medical University.

The study group consisted of 10 jars, selected from the recently found collection of foetal specimens.

The basic criterion for qualification was visual defects of the anatomical specimens maintained in an unspecified preservative fluid. Major defects were considered to be any indication of deterioration detected during the visual examination of the specimens such as discolouration, sedimentation, oxidation observed directly on the specimen, precipitates in the bottom of the containers, or significant loss of clarity of the fluid. All specimens were stored in their original jars.

The control group consisted of six teratological specimens stored in the exhibition area of the Anatomical Museum.

The closures were effectively sealed to the jars with a hard black binder. The specimens within the group were characterised by morphological similarity, state of preservation, visual aspects of preservative fluids, and identical containers.

Descriptive (Table S1) and photographic (Figure 2) documentation was created. The parameters were described: (i) container (shape of the jar, condition of the closure), (ii) organoleptic characteristics of the preservation fluid (clarity, colour, impurities, presence of paraffin), and (iii) anatomical preparation (colour, tissue fixation, nature of the lesions examined, material with which the body cavities were stuffed, if visible).

FIGURE 2.

Selected specimens from the study and control groups. (a) Museum specimen M30. The apparent excellent condition of the specimen is illusory. A white deposit has collected on the surface of the skin, visible around the mouth and arms. (b, c) Specimen A40 in its jar and removed from the container. The fluid is visibly turbid, sediment is in the bottom of the container, and flocs are floating near the surface. The face of the specimen is covered with sediment. (d) Specimen A20. Heterogeneous dark lesion on the knee. (e, f) Specimen A65 in its jar and removed from the container. Note the low level of preservative fluid and brown deposits on the surface of the specimen.

Samples were taken from the surfaces of anatomical specimens and their preservative fluids, using aseptic techniques for microbiological, mycological, and chemical analyses. To obtain samples suitable for analysis, the fluid was aspirated using a catheter into a 100 mL syringe from different depths of the jar and then injected into clean, sterilised glass tubes.

2.2. Source studies

A review of the scientific literature was carried out in order to select appropriate laboratory methods for the evaluation of the composition of preservative fluids (Figure 3a).

FIGURE 3.

The process of particular stages of the research. (a) Search for and analysis of source materials. (b) Analysis of chemical composition and observations of simple chemical reactions and pH measurement based on additional tests. (c) Microbiological analyses.

The literature was searched for information on methods of preserving anatomical specimens and possible methods of diagnosing fluid composition. The results of two literature reviews were used: (i) a systematic review (Domański et al., 2023) based on a search of scientific papers in PubMed, Medline, and Ebsco databases and (ii) a non‐systematic review based on papers published in the journal ‘Museologica’ and evaluation of literature lists from the monograph ‘Fluid preservation: a comprehensive reference’ (Simmons, 2014). Additionally, monographs on the history of the Medical University of Wrocław (Kozuschek, 2002, 2007) were analysed.

The results of these reviews determined the choice of methods for the next stages of the research.

2.3. Chemical composition analyses

Specialised chemical composition analyses were carried out on the fluids taken from the specimens in the study group and control group (Figure 3b).

The samples were analysed to determine the identity of the substances composing the fixative and preservative fluids, based on information acquired during the analysis of the source components (ethanol, phenol, glycerine, formaldehyde, and acetic acid were typed).

The analyses were conducted in two steps: first, the fluids were characterised by simple chemical reactions with a low‐cost test, in order to determine the presence or absence of the substances typically composing the solution.

The reactions were selected according to the recommendations of the Polish Pharmacopoeia, 11th edition, or those supported in the literature (Morisson & Boyd, 1985; Robins et al., 2011; Simmons, 2014).

A standard blind test was conducted before each chemical reaction was carried out.

Selected qualitative samples are described in Table 1.

TABLE 1.

Selected simple chemical reactions for the analysis of preservative fluids. Methods were selected based on information obtained through source studies (Simmons, 2014; Thede, 1996).

Verified substance	Test	Reagents	Positive result	Negative result	Comment
Ethanol (the test gives a positive result for *primary and secondary alcohols) *at room temp. the result for 1st‐degree alcohols will be negative	Lucas test	ZnCl₂ anhydrous in HCl (concentrated)	Turbidity of the solution (milky colour)	No noticeable changes at room temperature	It is recommended to use a high concentration of ethanol (75%) for purposes of increasing liquid volume, especially in exhibits where mould appears (Simmons, 2014)
Glycerine	Reaction with Cu(OH)2 (unheated)	Freshly precipitated Cu(OH)2	Sapphire solution without precipitate	No visible changes	Glycerine is a popular addition to conservative mixtures. It is believed to form a ‘protective layer’ on the tissues of the preparation, preventing them from drying out when the liquid is drained
Formaldehyde	Schiff test	Aqueous solution of fuchsine saturated with sulfur oxide (SO2)	Purple colouration (trityl cation)	No change in colour or non‐violet colouring of the solution	The presence of formaldehyde, as a constituent of museum fluids, has been confirmed in studies of the solutions of other Museums, including the Mütter Museum of the College of Physicians of Philadelphia (Simmons, 2014)
					The formalin concentration for specimens which yielded a result during the HCl titration averaged above 10%. Nominally, the formalin concentration of storage solutions is about 4% (Thede 1996)
Phenols or carboxylic acids	Reaction with FeCl3	Diluted solution of FeCl3	Green, blue, violet or purple colouration—phenol present	No change in colouring of the solution
				Red colouration—compounds able to enolise
Acetaldehyde, ethanol	Iodoform test	Solution of iodine in potassium iodide	Yellowish iodoform needles are formed	No iodoform needles	May be formed as a result of oxidation

The pH of the preservative fluids was also determined at this stage.

The pH measurement was obtained with a meter using a combined glass electrode (pH & Ion‐Meter, GLP22+, Crison Instruments S.A., Spain). The electrode was calibrated based on standard buffer solutions (pH 4.00, 7.00, 10.00). All measurements were repeated in triplicate and then averaged.

To compare the accuracy of the achieved results, a qualitative and quantitative analysis was performed using reference methods. These tests constituted the second step in the chemical evaluation of the preservative fluids from both study groups. GS–MS/MS, FTIR, and ICP‐OES were used to complement the chemical and biological research.

2.3.1. Determination of formaldehyde, ethanol, and methanol by gas chromatography/triple mass spectrometry

Formaldehyde standard stock solution (37.4 wt % in water), O‐(2,3,4,5,6‐pentafluorobenzyl) hydroxylamine hydrochloride (PFBHA), formaldehyde O‐pentafluorophenylmethyl‐oxime (10 mg/1 mL in water) and cyclohexanone 1 mg/1 mL (in water), methanol (99.9%), ethanol (99.9) were purchased from Merck KGaA. Deionised water was used as blank water. PFBHA solution was prepared by diluting 10 mg of PFHBA in blank water to a concertation of 0.1%.

A 1 mL aliquot of each sample was added to 20 mL headspace vials for analysis without dilution and pre‐filtration. 50 μL PFBHA solution and 1 μL of 1% cyclohexanone in water were added, and each vial was immediately sealed with a Teflon‐coated silicone septum. The sample was mixed, incubated at room temperature for 4 h, and heated at 60°C for 30 min in the headspace bath to reach gas‐liquid equilibration. The headspace syringe was set at 100°C and the sample volume was 1 mL.

Determination was performed on an Agilent 7890B/7000D GC–MS/MS system, equipped with a split/splitless inlet, a PAL RSI 85 headspace tool and MSD Chem Station software. The column was a 60 (m) × 0.320 (mm) × 0.5 (μm) DB WAX UI (Agilent Technologies). Injection was in a split mode with a helium, 1.8 mL/min constant flow set at 35°C as a carrier gas. The injection temperature was set at 200°C and the split rate was 25:1. The temperature programme of the column was held at an initial temperature of 35°C for 5.66 min, then increased to 100°C at 8.8°C/min, held at 100°C for 1.7 min, then increased to 220°C at 13.3°C/min, held at 220°C for 3.39 min, and increased to 240°C at 22.1°C/min, held at 240°C for 3.43 min. The MSD transfer line was set at 240°C, the source temperature at 230°C (70 eV), quad temperature at 150°C, scan range 20–300 m/z. For quantification, the analyses were registered in Sim Mode and following masses were recorded: formaldehyde O‐pentafluorophenylmethyl‐oxime 181, cyclohexanone: 55, ethanol: 45, methanol: 31.

2.3.2. Fourier transform infrared spectroscopy

The spectra measurements were recorded on a Thermo Scientific Nicolet iS50 FT‐IR spectrometer (ATR method) with DTGS detector and KBr beam splitter. The ATR‐FT IR spectra were detected in the range of 500–4000 cm⁻¹, at room temperature with a spectral resolution 4 cm⁻¹ and number of scans was 32.

2.3.3. Inductively coupled plasma optical emission spectroscopy

The third step of the chemical composition analyses consisted of performing ICP‐OES to detect the presence of selected metals in the preservative fluids.

Three samples of fluids were tested for the presence of metals (one sample from the control group, and two samples from the research group [A19, A65, and N2]).

Mineralisation was performed on an Ertec Magnum II apparatus in HNO₃ solution (65% puriss p.a.), then transferred to 25 mL flasks. The resulting solution was clear and transparent. This process was repeated four times. The elemental content (As, Zn, Al, Cu, Pb, Hg, Fe) of the samples was then measured using an iCAP 7400 DUO ICP emission spectrometer from Thermo Fisher Scientific.

2.4. Microbiological studies

Samples were simultaneously cultured on bacterial and fungal media according to the following procedures: (i) specimen containers were transferred to the anatomical laboratory, and samples were removed while maintaining the greatest possible sterile environmental conditions (UV lamp, surface disinfection) (Figure 3c). The samples were taken using disposable bacteriological loops from all lesioned areas and by rubbing the sterile loop along the entire length of the preserved specimen. The material collected was then inoculated on culture media. Single‐use sterile pipettes were used to collect fluid samples for culturing. (ii) Samples from the control group specimens were taken directly at the museum—the containers were disinfected externally, the seal removed, and the closures removed. Using sterile equipment, a catheter was inserted to aspirate fluid samples at various depths. The catheter was brushed against the skin of the specimen and the resulting ‘mist’ was aspirated. The fluid was inoculated on culture media.

Each sampling for microbiological tests was repeated.

When growth was observed on the microbiological media, identification was carried out according to standards for bacteria and fungi.

2.4.1. Bacteria

In the first stage of the study, cultures were cultured on: (i) nutrient broth with glucose (BTL Poland) for aerobic bacteria culture, (ii) Schaedler broth with vit K3 (bioMerieux) for anaerobic bacteria, and universal thioglycolate medium (BTL Poland).

Media were incubated up to 10 days at 37°C in an incubator (Wamed KBC‐65G) and checked every 24 h. When turbidity was observed the material was cultured on solid media Columbia agar +5% sheep blood (bioMerieux) and incubated at 37°C under aerobic conditions up to 48 h, and under anaerobic conditions for 48 h.

Cultured colonies were identified using MALDI‐TOF MS (Bruker, MALDI Biotyper).

Identification results obtained by the MALDI‐TOF MS method were analysed based on a score value. Those with low strength of evidence were rejected.

Bacteriological cultures were repeated—for this step, liquid media with thioglycolate (for aerobic and anaerobic bacteria) were selected. From positive cultures, material was cultured on selective constant media: (i) MacConkey Agar II (Becton Dickinson) and (ii) Columbia Agar with 5% Sheep Blood (Becton Dickinson).

In addition, gram‐stained slides were made from scrapings of biofilm coatings from some specimens. The following reagents were used to make microbiological slides: Crystal violet (bioMerieux), Lugol's fluid (bioMerieux), Ethyl alcohol (bioMerieux), and Alkaline fuchsin (bioMerieux).

After staining, slides were evaluated under a Carl Zeiss light microscope at 1000× magnification and then imaged using a Cell*F (Olympus) system.

2.4.2. Fungi

Test materials were cultured on solid media—Sabouraud medium with 4% chloramphenicol (BTL Poland). When fungal growth was suspected on liquid thioglycolate medium, additional isolations were made on solid Sabouraud medium. Cultures were incubated for up to 2 weeks at 28°C (Memmert Incubator IN110) and checked every 24 h. When fungal growth was observed, macroscopic evaluation of colonies was performed, taking into account their morphology (colour, reverse, obverse, colony structure). In addition, cultured colonies were screened on chromogenic media (Biomerieux chromID® Candida) and direct preparations were made in saline for microscopic evaluation. Taxonomy of fungi was established to genus. For mould fungi determination was based on the appearance of the fungus on the medium and the preparation of the culture; for yeast‐like fungi, cultures were sieved onto chromogenic media and the genus was determined based on the appearance of the colonies obtained. MALDI‐TOF MS method (Bruker, MALDI Biotyper) was used for species identification of yeast‐like fungi colonies grown on Sabouraud medium. The results of the spectroscopic analyses were compared with those of the mycologist as described previously.

3. RESULTS

3.1. Source studies

So far, very little research has been conducted on preservative fluids in the context of human anatomy museums. It is likely that in the past, both common preservation methods—those published in the then‐available literature, as well as new recipe variations specific to a particular scientific centre were used in practice (Simmons, 2014, pp. xi–xx, 66).

The history of the anatomical collection in Breslau dates back to 1745 when the Surgeons School was established. The collection was created by Professor Adolph Wilhelm Otto (1786–1845), and his successors, Prof. Hans Carl Barkow (1798–1873) and Prof. Carl Hasse (1841–1921). In one of the monographs on the angiological collection, Prof. Barkow mentioned that some wet preparations were stored on the museum grounds in zinc or lead containers (Barkow, 1869). The catalogues of the museum collections created by Professors themselves provide relevant data, listing ‘foetal specimens immersed in the spirit of wine’, but it is not known if the specimens were teratological.

The collected accounts of current and former employees of the Department of Anatomy show that the teratology specimens that remained after World War II were moved from the attic to a new location in the basement. Some were donated to other universities. Moreover, the preservative fluid in some of the containers was or supplemented using substances such as ethanol, glycerine, aqueous phenol solution, or methanol. However, there is no detailed information about the composition and volume of the topped‐up preservatives.

This research has not yielded data that allows us to unequivocally determine how the historical anatomical specimens in the collection of the anatomical museum were conserved. In one article, which refers to local traditions, three formulas traditionally used to preserve anatomical preparations after 1945 are mentioned: (i) ethanol, formalin, and water; (ii) ethanol, glycerine, and water; (iii) and ethanol, chloroform, and acetic acid (Janiszewska et al., 2018). The method of sealing the containers is also discussed (Pluta et al., 2019).

3.2. Chemical composition analyses

In the course of chemical composition analyses, formaldehyde, alcohol, and glycerin were detected. The oscillatory spectra, recorded with the ATR (attenuated total reflection) technique, are shown in Figures S1–S5. The results of the simple chemical reactions and the pH measurements, gas chromatography coupled to mass spectroscopy (GS/MS), and FTIR are presented in the Figure 4; Table S2. A comparison of the concentration of the assayed substances in individual specimens is shown in the graph (Figure 4).

FIGURE 4.

Concentration levels of methanol, ethanol and formaldehyde in assayed samples. Asterisk indicates containers with a reliable seal (specimens no. A77 through N4). Parentheses indicate specimens with visual contamination (either microbiological or physicochemical). Compare with results presented in Figure 6.

3.2.1. Results of reference methods

Gas chromatography coupled to mass spectrometry

Chromatographic analysis revealed the presence of formaldehyde and the first‐order alcohols methanol and ethanol. Ethyl alcohol was detected in all tested fluids. Methanol was found in samples A63 and A77, and trace amounts were also detected in the fluid of samples A19, A38 and A40 (Figure 4; Table S2).

Fourier transform infrared spectroscopy

The analysis proves that the samples studied contain formaldehyde (all tested samples), and secondary alcohols. Some samples show the presence of glycerol (samples A93, C2, M30), which contains two primary and one secondary alcoholic groups. A detailed description of the interpretation of the results is provided in Figure S1.

Inductively coupled plasma optical emission spectroscopy

During the determination of the average metal content of the selected fluid samples, the presence of zinc, iron, and copper was detected in samples A19 and N2, and aluminum content was additionally determined in sample A65 (Table S3). Mineralisation was carried out reproducibly.

3.2.2. Results of simple chemical reaction and pH measurement

The results of the reaction can be seen in Figure S6.

FeCl ₃ : The colouration of samples by reactions with FeCl₃ may indicate the presence of phenol derivatives, for example, nitrophenols (yellow colour) (Pasto & Johnson, 1979) or other compounds capable of enolisation (red colour) (Broumand, 1952). However, performing an analysis using FTIR spectra to confirm the presence of phenol derivatives in preservative liquids did not show signs of their presence. The failure to obtain a purple solution allows us to exclude the presence of phenol itself in the tested samples (Apostica et al., 2018; Moeller & Shellman, 1953).

Cu(OH) ₂ : Trials with Cu(OH)₂ provided results that were difficult to interpret. In some samples, there was a colour change to sapphire (suggesting the presence of glycerol) but without a concomitant dissolution of the precipitate (Hassan et al., 2001; Seibert & Long, 1925). An unambiguously positive result was obtained only for M30.

Lucas test: After performing the Lucas test, no significant changes in the colour or structure of the liquids were observed, indicating the absence of alcohols or the presence of only first‐order alcohols in the tested solutions.

Iodoform test: Ambiguous results were obtained, difficult to both read and interpret. Results varied considerably between specific samples. In most of them, the reaction resulted in a precipitate whose appearance did not resemble iodoform needles. Thus, no clear confirmation or negation of the assumption that an aldehyde or a secondary alcohol was present in the museum liquids was obtained (Morisson & Boyd, 1985). Aldehyde could come, for example, from the oxidation of ethanol, which was most likely a component of the preservative, or it could be a contaminant from samples that were fixed in formaldehyde before transfer to another preservative.

Schiff test: Performing the test resulted in the appearance of a colour (Figure S6f) associated with the formation of a trityl cation and indicative of the presence of an aldehyde in all the tested fluids (Robins et al., 2011).

pH of the solution: most of the liquids were acidic (3.97 ± 0.01–6.20 ± 0.01), but the pH of a few samples approached neutral (6.86 ± 0.01 and 6.98 ± 0.02). However, due to the presence of organic solvents, caution should be exercised in interpreting the results of the pH measurement (Neumann & Crimmen, 2022, pp. 34–35).

3.3. Microbiological studies

3.3.1. Culture

Positive cultures were obtained from samples taken from anatomical specimens (Table S4). The microorganisms cultured belonged to yeast‐like fungi—Candida boidinii, Geotrichum silvicola species (Figure 5), moulds—Penicillium and Fusarium genera, Gram‐positive bacteria— Bacillus cereus, Bacillus thuringiensis and Gram‐negative bacteria Cupriavidus metallidurans (Figure S7). In the second stage (repetition of culturing), all cultures were negative.

FIGURE 5.

Fungi isolated from samples. (a, b). Penicillium sp. isolated from specimen A20 on Sabouraud's agar—obverse (a), reverse (b). (c, d) Candida boidinii on Sabouraud's agar—big cream colonies with rough growth (c) and Geotrichum silvicola on Sabouraud's agar showing off‐white/cream colonies with matted appearance (d).

3.3.2. Results from microscopic evaluation of gram‐stained microbiological preparations

Selected anatomical specimens (no. 35, 40, 63, 65) were analysed microscopically (Figure S8). Gram‐positive cocci, dispersed or in clusters, Gram‐positive coryneform, Gram‐positive bacilli, Gram‐positive filamentous bacteria resembling actinomycetes, and Gram‐negative granules arranged like diplococci, as well as blastospores of yeast‐like fungi, were mostly observed. In the majority of samples various impurities, crystals, and epithelial fragments were observed.

4. DISCUSSION

The historical nature of the research material determined the multifactorial character of the investigations carried out in this study. In this type of research, the first thing to remember is to respect the collection and keep it in the best possible condition, so that it can serve future generations. Ad hoc conservation work, without in‐depth historical knowledge of the chemicals and technology used and the circumstances in which the collection was created, can result in damage to unique, historical anatomical specimens.

4.1. Chemical composition analyses

The composition of the preservative fluid is an essential element in stabilising and fixing anatomical specimens (Waschke et al., 2019). There are many methods of preserving cadavers and anatomical specimens in modern anatomy. However, the gold standard is still formaldehyde (Balta et al., 2015). It has been proven that preservatives affect the appearance as well as the dimensions of the objects immersed in them (Steinke et al., 2012; Yeap et al., 2007). Therefore, it is very important when working with historical material to avoid changing preservation solutions. This fact was behind the decision to search for the historical composition of the preservative fluid used in the collection assessed in this study.

The heterogeneous results and difficulties in interpretation are due to both the diversity of technical analyses and complexity of the analysed material. The influence that the particular components of the solution had on each other over time is unknown. Certainly, the presence of turbidity, non‐uniform solution colour, and the presence of precipitates additionally hindered the evaluation of the results. Furthermore, undesirable effects of the lack of conservation supervision in the study group cannot be excluded.

4.1.1. Reaction with Cu(OH)₂ and iodoform test

The advantage of tests based on chemical reactions lies in their simplicity. The disadvantage is the way the results are interpreted, which is based on subjective visual evaluation.

In the Cu(OH)₂ cold test, the intriguing fact is that although aqua complexes causing the sapphire colour were detected, there was an additional precipitate in the liquids (Figure S6c). This phenomenon indicates a side reaction with another substance present in the fluid, as the product of the sample itself should only be a copper (II) coordination compound. Another explanation may be the failure to bind part of the reactant in the complex due to the difficulty of maintaining the stoichiometry of the reaction, resulting from the lack of knowledge of the composition of the preservative fluids. In multi‐component samples, the interpretation is complicated by the lack of knowledge of the matrix, which consists of chemical compounds located next to the substance being determined.

The iodoform test provided equally unusual observations in which the flocs that formed did not resemble typical iodoform needles (Figure S6). Nagata and Nishiwaki described a situation in which yellow precipitates were obtained in an iodoform test due to the presence of ethyl acetate in the sample. The possibility of hydrolysis occurring leading to the formation of ethanol makes it difficult to interpret the result unambiguously. Ethanol that forms from ethyl ester, if it is used as a solvent, will also give a positive iodoform test result (Nagata & Nishiwaki, 2021). Because ester compounds and alcohols give a positive iodoform test result, it is impossible to accurately type the compound that is in the fluids based on the iodoform test result.

In the case of both tests, the results could be falsified by sediments present. Their influence on the reaction result could be minimised by prior filtration of the sample. However, this involves the risk that the concentration of the components of the mixture will decrease through absorption by the filter (simultaneous loss of sediment present in the liquid) or, in the case of the analysis of volatile substances, through evaporation. For this reason, the reaction with Cu(OH)₂ and the iodoform test is not suitable for testing turbid solutions and should not be recommended for the evaluation of museum conservation fluids.

4.1.2. Reaction with FeCl₃ (phenols, acids)

This test was aimed at verifying the presence of phenol, which would account for the presence of a pungent, suffocating odour when some containers were opened.

Surprisingly the results obtained probably indicate the low sensitivity of this method. Analysis of the literature revealed that iron (III) chloride reacts not only with phenol but also with carboxylic acids. Hence, it is possible to obtain a false positive result. The presence of carboxylic acids in the samples may be indicated by the various non‐specific colours obtained (Kapoor, 1988).

Acetic acid was typified as a component of the mixtures, based on the assumption that: (i) acetic acid was a component of the recipe; (ii) it is a product of the oxidation of ethanol to aldehyde and acid that occurred due to the leakage of the jars; and (iii) it is a product of the hydrolysis of a weak acid salt. In some methods of conservation of museum preparations, acetates have been used (e.g., the Kaiserling method, the Judah method) (Simmons, 2014, pp. 289–295), the hydrolysis of which can provide weakly dissociating acetic acid to the system. However, the presence of these substances was not demonstrated with further analyses.

4.1.3. pH measurement

The tested solutions gave curiously low pH results. The acidic pH (Figure 4; Table S2) is most probably caused by the presence of formaldehyde, which in an aqueous solution has a pH of 2.8–4.0 (Simmons, 2014).

The increased acidity of the solution may also be related to the reactivity of metal salts (e.g., Hg, Na, K) which, in order to increase the antiseptic potential, were used in old formulations (Parsons, 1831; Simmons, 2007, 2014; Swan, 1833). The presence of such substances has been demonstrated by research methods conducted on similar museum collections (Simmons et al., 2007; Thede, 1996).

The pH value of fluids is fully dependent on the proportions of the individual components present in the mixture. The impossibility of establishing a complete recipe prevents an unambiguous interpretation of the pH. Ethanol preservatives have slightly acidic to neutral pH (Simmons, 2014) which could, like the addition of methanol, influence the pH of the entire mixture. The exact pH of the alcohol depends on the concentration of the solution and measurement method (Deleebeeck et al., 2021). The pH value is probably somewhat disturbed due to the fact that the tested liquids consist mainly of organic solvents, not water (Neumann & Crimmen, 2022).

4.2. Formaldehyde

Prior to this study, the presence of formaldehyde in museum fluids was typified and therefore an attempt was made to confirm its presence using a reagent of proven sensitivity that was previously used in museum practice, including to quantify the concentration of airborne formaldehyde in museum rooms (Gibson et al., 2008). A review of the specialist literature showed that the Schiff test is a recognised method for confirming the presence of aldehydes, used by organic chemists (Robins et al., 2011). This test has been used to evaluate museum fluids by some authors (Simmons, 2014, p. 181). The results obtained from the Schiff test clearly indicate the presence of an aldehyde in the preservative fluids.

The identity of formaldehyde, contained in the fluids tested, was confirmed by unambiguous infrared spectra, and its exact amounts were determined by gas chromatography.

The available literature indicates that infrared spectroscopy is one of the most useful methods for identifying chemical compounds (Sablier et al., 2020). The spectra show for example, stretching and deformation vibrations characteristic of specific chemical groups in a given range (Kazarian, 1758). Due to the absorption of infrared radiation, oscillations of molecules occur, such that the valence vibrations of groups of atoms are recorded in their characteristic areas (Thompson, 2018). It is a recognised pharmacopeial method for the identification of numerous chemicals.

Gas chromatography is also a recognised analytical method for the qualitative and quantitative determination of chemical compounds (Al‐Rubaye et al., 2017). The separation of individual substances is possible due to their different physicochemical properties which results in discrepancies in migration rate through the chromatograph column, which was chosen for the fluids under study.

From the results obtained during the analysis (Figure 4; Table S2), it is evident that gas chromatography is the best method to accurately determine the presence and concentration of formaldehyde. The Schiff test and FTIR analysis are also suitable to determine the presence of formaldehyde.

The results obtained provide the basis for a broader discussion on the influence of formaldehyde on anatomical preparations.

Formaldehyde, as a fixative, causes tissue hardening in the amino acid crosslinking reaction. In addition, dehydration and darkening of tissues is observed (Simmons, 2014, p. 26).

The long‐term effects of formaldehyde on tissues include the breakdown of adipose tissue and the precipitation of water‐insoluble free fatty acids, which form flocculent deposits that float on the surface of the preservative fluid. At low temperatures (−5 to +5st) formaldehyde in aqueous solutions undergoes spontaneous polymerisation, the product of which partly converted to paraformaldehyde (Simmons, 2014, pp. 30, 44). This polymer precipitates out of solution causing it to turn cloudy, and settle on the bottom of the container as a white sediment, and can frequently coat specimens as well. Such changes were observed in both the test and control groups in n this study.

The exposure of the preparations to such low temperatures may have been due to the tragic effects of the war when the attic above the museum was bombed. Most of the roof of the building was destroyed, as well as the window frames.

Concentrated formic aldehyde, intended for the preservation of anatomical preparations, is buffered by the manufacturer. This has been the practice since at least the second half of the 20th century (Hayashi et al., 2016). During dilution, the protective action of the buffer is lost and the formic aldehyde becomes reactive. On contact with air, it forms formic acid, which lowers the pH of the solution and consequently decalcifies the tissues. Acetic acid and alcohol, used in the past as a fixatives, may have similar effects.

Considering the fact that formaldehyde has been available in Germany most probably since 1891 and the knowledge about its properties for fixing human tissues has been common since 1893 at the earliest (Musiał et al., 2016, pp. 27–28; Simmons, 2014), anatomical specimens created before this period could not have been fixed or preserved with it. Therefore, insofar as the preparations analysed were created before 1893 (Otto, 1816, p. 7, 1823), the preservative fluid must have subsequently been modified by the addition of formaldehyde, which was previously unavailable. However, even after this period it was possible to use only alcohol (preservation without fixation) (Simmons, 2014, p. 49), or only formaldehyde—especially for ‘medium‐size foetuses’ (Schultz, 1924).

The rationale for the use of spirit for the preservation of the collection under study is supported by its use by German anatomists for scientific research or demonstrations, in which containers could not be permanently sealed and therefore required frequent refilling (Kozuschek, 2007, p. 110). For this reason, cheaper, more widely available, and less complicated to prepare preservative formulas (spirit solutions) were probably used (Brenner, 2014), so that the concentration could be monitored and the evaporating substance replenished as necessary. On the other hand, some of the museum preparations preserved after the war still intrigue us with the natural colour of the tissues and the perceptible, pleasant smell of the preserving liquids, so it is likely that for purposes of permanent exhibition, more sophisticated methods of fixing and preserving the preparations were used. For example, a preparation guide from 1831 lists fluid recipes based on, among other things, turpentine (Parsons, 1831). This has been used as a clearing agent for mercury‐injected specimens from the 18th century (Brenner, 2014; Moore, 2006).

4.3. Alcohols

The Lucas test is a method for the detection of alcohols and the determination of their order. It allows differentiation of the order of the alcohols, exploiting the differences in reactivity with halogens of primary, secondary and tertiary alcohols (Kjonaas & Riedford, 1991; Morisson & Boyd, 1985). The results obtained from the Lucas test indicate the absence of alcohols other than primary alcohols in the museum fluids studied. This is confirmed by gas chromatography coupled with mass spectrometry, which clearly indicates the content of ethanol, that is, primary alcohol, in the preservative fluids, while also indicating the presence of methanol in some of the containers.

The presence of glycerol, a secondary alcohol, was confirmed by FTIR spectra analysis and (probably) by Cu(OH)₂ assay. Glycerine is a common additive in preservative fluids, although its properties are debatable for this use (Simmons, 2014). In contrast to FTIR, the Lucas test did not show the presence of glycerol or other secondary or tertiary alcohols in the preservative fluids, which indicates its significantly lower sensitivity than spectra analysis.

In view of the overlapping results obtained for the primary alcohols, carried out independently of each other, the Lucas test can be considered reliable in this context, but perhaps not sensitive enough for the detection of alcohol in a mixture with formaldehyde.

Gas chromatography confirmed the presence of alcohol (ethanol) previously identified in the preservative fluids, but also showed small concentrations of methanol in some of the containers. The methanol content of the samples was confirmed by spectroscopy.

In the control group, the alcohol content of the preservative liquids was low—no more than a maximum of about 13%. suggesting the dominant contribution of formaldehyde (from 1 to almost 4%) as the preservative of these specimens, or perhaps the contribution of other, additional substances not detected in this study.

The yellow precipitates obtained in the iodoform sample may result from the reaction of esters or alcohols in the presence of methanol as solvent, which would also indicate its presence (Nagata & Nishiwaki, 2021). It is possible that it was added as a stabiliser of formaldehyde. In the case of formaldehyde polymerisation in the preservative fluid, a small addition of methanol or NaOH allows this process to be reversed (Simmons, 2014, pp. 30, 44). The methanol may have been a contaminant of the industrial alcohol or have been used deliberately to denature ethyl alcohol, which was later applied to refill fluid level in the jars. It is also worth noting that the presence of alcohols (e.g., methanol or ethanol) in preservative liquids can cause shrinkage of tissues present (Simmons, 2014).

Unambiguously, the presence of alcohols can only be confirmed by reference methods such as applied here: infrared spectroscopy and gas chromatography. Cersoy and her team report that they could identify ethanol, methanol, and isopropanol alcohol using micro‐Raman spectroscopy (Cersoy et al., 2020). Although they suggest or exclude the presence of given group of compounds, analytical reactions give results which are too imprecise to be used as a basis for conclusions.

4.3.1. Microbiology and specimen environments

Microorganisms present in the environment have different adaptive capacities. Their growth depends, among other things, on the availability of oxygen, pH, nutrients, or humidity. Disinfectants or preservatives can reduce the viability of bacteria and fungi or eliminate them completely from the environment. In the course of this study, it has been shown that some microorganisms tolerate the presence of various compounds contained in preservative fluids and we can freely cultivate these microorganisms using classical microbiological methods. The study of chemical composition allowed us to determine which substances were present in the samples (Figure 6).

FIGURE 6.

Specimens from which microorganisms were isolated and concentrations of preservative fluids. Asterisk indicate jars with a reliable seal.

In the course of the study, several microorganisms were detected in the cultures from the anatomical preparations, which survived the rough living conditions and the presence of preservatives. It is worth looking at these microorganisms in terms of their ability to adapt to such difficult environmental conditions and the chemical composition of the preservative fluids, for which there is no documentation. Both simple and advanced techniques of chemical composition analysis made it possible to determine the composition of these fluids.

The pH value is one of the factors determining the presence of microorganisms in a given environment. It affects their metabolism and determines the chemical activity of protons involved in, among other things, redox reactions, co‐formation of bacterial or fungal surface structures, activity of extracellular enzymes, or interaction with other organic structures (Jin & Kirk, 2018).

The pH also affects the antiseptic potential of formaldehyde. It has been shown that at a steady formaldehyde concentration, with decreasing pH, the susceptibility of some amino acids to formaldehyde fixation decreases, through inactivation of amino and hydroxyl groups (Eltoum et al., 2001). This may be one explanation for the resistance of microorganisms to formaldehyde. The progressive decrease in pH of the preservative fluids of the specimens could therefore be one of the causes of microbial colonisation (Ratzke & Gore, 2018; Tamayo‐Arango & Garzón‐Alzate, 2018).

The fungus C. boidinii has the ability to metabolise short‐chain alcohols (including methanol and ethanol) to aldehydes. In addition, it can oxidise formaldehyde to formic acid through formaldehyde dehydrogenase (Sahm, 1977).

The ethanol oxidation capabilities of this yeast have been relatively poorly described. It is known that it assimilates ethanol faster than methanol. Using methanol as the only carbon source in the medium, growth of C. boidinii is observed (Sahm & Wagner, 1972).

The growth rate depends strictly on the inoculum of the microorganism. The larger the inoculum in the medium the better the growth (Pilát & Prokop, 1975). However, the ability of this fungus to assimilate and oxidise methanol is limited. The higher the methanol concentration, the more these abilities decrease (Pilát & Prokop, 1975).

Candida boidini tolerates a methanol concentration in the medium of up to 5%, at which level enzymatic activity and growth ceases (Sahm & Wagner, 1972).

The presence of formaldehyde or methyl formate in the medium affects the growth of C. boidinii in the presence of methanol and the activity of methanol‐degrading enzymes. As a result of their presence, the biomass of the microorganism multiplied in the medium increases, but the methanol utilisation rate decreases. In the presence of formaldehyde, the activities of formaldehyde dehydrogenase and formate dehydrogenase increase significantly, while the activity of methanol oxidase decreases significantly. In the presence of methyl formate, the exact opposite happens—methanol oxidase activity increases and dehydrogenase activity decreases (Aggelis, 2000).

It has been shown that this fungus uses the metabolic products of some bacteria with which it has a symbiotic relationship, if it is not possible to incorporate the compounds available in the environment into its own biochemical pathways (Sahm & Wagner, 1972). The metabolism of C. boidinii may have had an effect on changes in the quantity and quality of preservative fluids.

Formaldehyde or formate is much more toxic to microorganisms than methanol, hence it is assumed that fungi will not grow in their presence. Even low concentrations of formaldehyde, on the order of 0.01%, may be too toxic, but Pilat et al. showed that in formaldehyde with the addition of a small amount of methanol (1%) it was possible to induce the growth of C. boidinii. This happened in a range of formaldehyde concentrations from 0.005% to 0.08%. At higher formaldehyde concentrations, inhibition of yeast growth occurred (Pilát & Prokop, 1975). The effect of ambient concentrations of ethanol or formaldehyde on the survival of this yeast is not known.

Candida boidinii in the presence of methanol as the only source of carbon and energy, induces in its cells the biogenesis of peroxisomes, which contain a large number of enzymes responsible for the transformation of various substances, including xenobiotics (van Dijk et al., 2000).

Another yeast‐like fungus, G. silvicola, isolated from an anatomical sample, has the ability to metabolise, among other substances, glycerol, ethanol, and its ester with acetic acid (ethyl acetate) as carbon sources (Pimenta et al., 2005). Geotrichum spp. are yeast‐like fungi widely used in biotechnology, for example in cheese making, in the biodegradation of oil stains and decomposition of cellulose (Hyde et al., 2019). Zhu et al., (2017) showed significant degradation of higher alcohols, such as hexanol and isoamyl alcohol, at low pH by glutamate dehydrogenase produced by the species G. candidum.

Cupriavidus metallidurans, which was isolated from one of the samples, is an environmental bacterium that is able to withstand extreme environmental conditions and to metabolise toxic substances such as gold chloride. It has a wide range of adaptations, including the ability to survive in high concentrations of heavy metals, including Fe, Zn, Cu, Hg, which it reduces by incorporating them into metabolic pathways. It does not have the ability to absorb glucose from the environment, but can produce energy by metabolism of acetates, etc. It also produces the enzymes alcohol and aldehyde dehydrogenases and can degrade short fatty acids and aromatic compounds (Janssen et al., 2010; Lal et al., 2013).

Cupriavidus metallidurans, until recently mainly isolated from contaminated soils and sediments, was isolated for the first time in 2011 from clinical material from humans. The isolation was from blood of a patient with sepsis (Langevin et al., 2011; Vojtková & Janulková, 2012).

The detection of Cupravidius bacteria may indirectly demonstrate the presence of heavy metals in the preservative solution. There may be several reasons for their presence: (i) intentional components of preservative mixtures, (ii) ions released through contact of the liquid with metal elements of the specimen (such as tags (Barkow, 1869) or ‘zinc containers’, as suggested by one of the historical sources we found), (iii) ions released from the tissues when exposed to formaldehyde (Simmons, 2014, p. 33), or (iv) contact of the specimen with metal instruments during preparation (Slevin, 1927). Some of the specimens examined were not only fixed and preserved but also dissected (vide Figure 2f), so the latter situation is likely.

ICP analysis provided similar results for samples from the study group specimens A19 and A65. This is surprising, because A19 was a significantly heavier foetus than A65, additionally enriched with a well‐preserved placenta. Therefore, it was expected that the preservative fluid of the larger specimen would contain higher concentrations of metal ions, particularly iron, derived from the degradation of erythrocyte haemoglobin contained in the richly vascularised placental structure. Perhaps these ions were utilised by microorganisms, and in this way, the amount of free metals in the fluid remained low.

A case has been described of preparations that have darkened. These studies showed the presence of elemental mercury, perhaps precipitated by the reduction of mercury salts by bacteria living in the preparation jar, or by spontaneous reactions of the components of the preservative liquid. It is difficult to determine the cause of the colonisation effect on the specimens because no bacteria were identified (Simmons et al., 2007, pp. 32–36). However, it is known that bacteria of the genus Cupravidius precipitate metals from the solution, which has been imaged microscopically and macroscopically (Reith et al., 2009).

Cupriavidus in this study was isolated from one sample of the three taken from preparation A19, thus the positive smear could have been obtained from only a few specific places on the specimen (the lesions were of yellow colour, around which dense mycelium‐like lesions were observed). Due to the disinfection of the preparation, additional verification tests for the presence of bacteria—as carried out in other cases—were not possible.

The Gram‐positive bacilli B. cereus and B. thuringiensis cultured from several specimens are likely to be contaminants introduced from the air into the preservative fluids. These are environmental microorganisms that survive very harsh conditions by transforming into spore‐like structures. A spore has a thickened cell wall, discards water, does not metabolise, has no cellular processes, and can therefore remain in the environment for many years. Only appropriate conditions (e.g., a change in pH, nutrients present in the environment) allow it to transform into a vegetative cell and function normally (Cho & Chung, 2020).

Mould of the genera Penicillium and Fusarium were also isolated. Such microorganisms are present in the air in the form of spores and float freely, so they have the opportunity to contaminate all kinds of samples for microbiological tests. It is therefore difficult to assess the nature of their presence in samples. It may be that contamination with these spores simply occurred during the collection of material to test, resulting in a positive culture.

On the other hand, Moore mentions contamination of specimen jars by fungi of the genus Penicillium particularly when preservation fluids contained low concentrations of ethanol/formalin or were accidentally diluted with water (Moore, 1999, pp. 108–109). Similar circumstances in the past probably altered the fluids analysed in this study. It is important to control the fluid concentrations and integrity of the container seal, and perhaps add an antifungal substance such as thymol or menthol.

Microscopic observations revealed a much higher diversity of microorganisms in the samples compared to the results of microbiological cultures. Some of the microorganisms detected in this way, presumably environmental, may be species unknown to traditional bacteriological and mycological culture. The only way to be sure would be to identify the microorganisms using molecular systematic methods, but the utility of these methods is restricted due to their expense.

4.4. Mutual chemical and microbiological influences: Conservation aspects

On the basis of the chemical composition analyses, it was shown that each of the tested fluids had a distinct, individual composition. This may suggest that the composition was dependent on (i) the original purpose for preserving a given specimen (e.g., as a museum specimen, for demonstration, for research), (ii) the date at which it was created, (iii) the state of the container seal, or (iv) modifications of the fluid (whether intended or spontaneous).

The low concentrations of volatile substances in the preservative fluids may have been due to evaporation, especially if the jars were poorly sealed. The container seals were probably improved by a layer of paraffin oil that coated the meniscus of the solutions (Pluta et al., 2019) (vide Table S1; Figures S1b, S5), but its effect on reducing evaporation or oxidation of the substances in the preservatives has not been described. On the other hand, even the sealed containers from the control group (A93, C2, N3, N4) did not have very high concentrations of volatile substances, so perhaps it was intentional not to use high concentrations, or these specimens are preserved by another ‘dominant’ substance, which was not detected.

Based on this study it may be concluded that the deterioration of specimen quality is the result of simultaneous and interdependent biochemical and physico‐chemical reactions resulting from the overlapping influences of (i) the location of the object, (ii) the specimen storage environment, and (iii) the microorganisms to which the specimen was exposed.

The primary causes of the changes in the anatomical specimens were probably leaky containers and the place where the collection was stored (i.e., a dark, cool, damp cellar with limited ventilation). Evaporation of the components of the preservative mixtures and their oxidation by air reaching the surface of the fluid may have affected not only the volume and concentration of the fluids but also their antiseptic properties. The changes in environmental conditions, due to unsealing the containers, may have stimulated dormant microorganisms to germinate. In the present study, it was shown that in certain concentrations, substances traditionally used in preservation can become nutrients for microorganisms (e.g., ethanol for C. boidini, heavy metal ions for C. metallidurans) (Figure 7).

FIGURE 7.

Specimen A19 described in the label as 'Fetus et placenta'. (a) The centimetre‐thick, malleable layer (Candida spp.) covering the specimen was removed, showing dark colonies (Geotrichum spp.) growing underneath. (b) The preparation was gently cleaned mechanically to remove the fungi, then transferred to a new conservation solution at low concentration to slowly 'adapt' the tissues to the new environment. However, intensive fungal growth was soon observed in the liquid. Based on the results of the chemical and microbiological analyses provided, it was decided to inject the preparation and submerge in formalin and, after time, gradually transfer it to increasingly higher concentrations of the alcohol solution. The glass lid was periodically removed and the specimen was monitored. After a year, we noticed the appearance of white deposits (probably Candida spp.) on the un‐soaked parts of the preparation (e.g. part of placenta, knee). It is likely that the paraffin layer has created a barrier, making the methanol concentration in the empty (air) space of the jar too low or the specimen has been contaminated from the air during checking. Guidelines developed for the care of natural history collections may be useful in the conservation of such specimens (Moore, 1999; Neumann et al., 2022).

This research shows that conservation practice has a substantial impact on the state of specimen preservation. It is risky to make any changes to the structure of historic fluid‐preserved preparations, so a complete replacement of the preservative fluid is inadvisable: instead, preservative fluids should be replaced slowly over time (Simmons, 2014, p. 45). The equilibrium between the fluid and the specimen is a factor sensitive to changes. The balance established by the reactions between the specimen tissue and preservation fluid can be easily upset by radical fluid changes, which may induce new physico‐chemical reactions that will manifest themselves in a variety of ways that are difficult to predict.

An isolated environment for the specimen limits the influx of microorganisms from outside, helps maintain the antiseptic potential of the preservative solution, and reduces the reactivity of the preservatives. Analysing the results obtained in a more general way (Figures 4 and 6), it appears that the isolation factor plays the most important role in the long‐term preservation of anatomical specimens. There was no correlation between the concentration of preservatives and contamination by microorganisms or the degree of specimen pollution. Specimen containers with good closures (in both the study and control groups) were in better condition than specimens that may have had their container seals compromised or that showed signs of sloppy interventions (e.g., traces of several layers of multiple seals, poor seal adhesion over the entire surface of the jar‐lid junction). Furthermore, the latter containers showed higher concentrations of alcohol, probably the result of repeated refilling in the past.

In the group of specimens with good container seals, there were a few containers that displayed aesthetic defects, for example, white deposits on the bottom, crystalline precipitates on the walls of the jar, oxidised seals. This seems to confirm the thesis presented here that physicochemical factors also influence the visual aspects of a specimen.

It should be emphasised that working with contaminated preparations poses health risks. Microorganisms have a variety of pathogenic characteristics under different environmental conditions. Their virulence level for humans is unknown, but other research has described the isolation of pathogens that are hazardous to humans from fixed human anatomical specimens (Sperry & Sweeney, 1987; Weed & Baggenstoss, 1951). Knowledge of the possible risk factors is therefore extremely important. A variety of chemical substances, often toxic, were used for fixation and preservation, and these can evaporate through inadequate container closures into the museum space resulting in long‐term health consequences. Persons undertaking the conservation of historic specimens are particularly vulnerable, as they are most exposed to these chemicals.

Little has been written about the role of microorganisms in the context of anatomical museums. Our study indicates a significant relationship between the microorganisms identified here, the composition of the preservative fluid, and the quality of preservation of anatomical preparations.

5. CONCLUSIONS

The chemical studies demonstrated the presence of methanol, ethanol, formaldehyde, and glycerol in preservative fluids. The concentrations of these substances differed between the samples, and their determination required the use of a variety of methods to analyse the individual components of the preservative mixtures. The differences between the results obtained from the advanced modern analytical techniques and those obtained from the simple chemical reactions indicate that the latter are not sufficiently selective and are therefore inappropriate for assessing the composition of the preservative fluid independently.

This study shows that historical preservative fluids can be a habitat for many microorganisms. In microbiological tests, both bacteria and fungi were isolated from swabs taken from anatomical specimens. The bacterial flora was less numerous than the fungal flora. This research highlights the fact that microorganisms can be an additional factor in determining the compositions of preservative fluids and, at certain fluid concentrations, microorganisms show resistance to their effects.

The analysis of preservative fluids represents a significant challenge for contemporary anatomical museology. The unique and innovative research carried out by our team provides guidelines for contemporary anatomists and museum professionals. People involved in the protection of museum collections need reliable, fast and easy to implement methods for verifying the composition of the fluid protecting the specimens. However, due to compositional variability simple screening for chemical composition is an ineffective diagnostic tool for the museum anatomist. Instrumental analysis using sophisticated equipment is suitable, but expensive, and requires long lead times, while methods for measuring physical parameters are only suitable for single‐component solutions. The chemical tests described in above can be useful for rapid assessment of the composition of simple mixtures, however, the development of guidelines for the analysis of contaminated historic preservative fluids that have been subjected to various modifications requires further research and systematic verification.

The development of such methods is much needed in the day‐to‐day work of the museum conservator, who often lacks specialist technical facilities.

AUTHOR CONTRIBUTIONS

Jurand Domański: concept/design, acquisition & analysis/interpretation of data (historical research, chemical reactions, microbiological analysis), co‐author of the draft and final version of the work. Adriana Janczura: acquisition & analysis/interpretation of data (microbiological analysis), co‐author of the draft and final version of the work. Marta Wanat: acquisition & analysis/interpretation of data (chemical reactions), co‐author of the draft and final version of the work (chemical section). Katarzyna Wiglusz: acquisition & analysis/interpretation of data (Fourier transform infrared spectroscopy), co‐author of the draft version of the work (chemical section). Magdalena Grajzer: acquisition & analysis/interpretation of data (gas chromatography coupled to mass spectrometry—GC–MS/MS), co‐author of the draft version of the work (chemical section). John E. Simmons—critical revision. Zygmunt Domagała—co‐author of draft version of the work (historical part). Jacek C. Szepietowski—critical revision, supervision.

FUNDING INFORMATION

Presented results of the research, carried out under the topic with subvention funds (grant) no. SIMPLE: SUBK.A351.23.020 granted by Ministry of Education and Science of Poland.

CONFLICT OF INTEREST STATEMENT

All authors declare that they have no conflicts of interest.

ETHICS STATEMENT

Approval from the local bioethics committee was obtained for the purpose of the study.

Supporting information

Figures S1–S8

Tables S1–S4

Domański, J. , Janczura, A. , Wanat, M. , Wiglusz, K. , Grajzer, M. , Simmons, J.E. et al. (2023) Preservation fluids of heritage anatomical specimens — a challenge for modern science. Studies of the origin, composition and microbiological contamination of old museum collections. Journal of Anatomy, 243, 148–166. Available from: 10.1111/joa.13854

DATA AVAILABILITY STATEMENT

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.