As noted in chapter 1, many of the early descriptions of fluid-preserved specimens proclaimed them to be perfectly preserved, but often the writer was describing the preserved animal without having seen a living specimen of the species, or was making a subjective comparison to other preservation methods available. Fixation and preservation may affect specimens through dehydration, rehydration, shrinkage, swelling, alteration of colors, changes in skin texture, or extraction of chemical components. Despite the fact that some of these changes have been observed since the very beginning of fluid preservation, few have been investigated in detail. Most fixation- and preservation-induced changes are ignored by the workers who use specimens for scientific work, other than noting colors in preservative compared to colors in life. Nevertheless, numerous chemical mixtures have been proposed to reduce or eliminate fixative- and preservative-induced changes, but their efficacy is mostly anecdotal. There are few controlled studies investigating changes to specimens reported in the literature, with the notable exceptions of Lai (1963) and Lee (1982), and several papers addressing biomass changes in invertebrates and fish. A summary of reported changes in invertebrate specimens is provided in tables 13 and 14, and for vertebrates in tables 15 and 16.
In general, specimens tend to swell in fixatives and then shrink when transferred to preservatives. In formaldehyde-based fixatives, most swelling (including weight gain) occurs within the first few hours or days of exposure (depending on the fixative and the type of specimen). Transferring the specimen to an alcohol preservative usually induces shrinkage and weight loss, with most of the change occurring within the first few days of exposure (depending on the fixative and the type of specimen). These changes are due to the fact that formaldehyde-based fixatives are mostly water (the standard 1:9 dilution of formaldehyde is about 95 percent water), which causes specimens to hydrate, while most alcohol-based preservatives are around 50 percent to 30 percent water, which causes specimens to dehydrate. The use of more complex fixatives and preservatives tends to produce more complicated effects. In addition, some workers have recorded handling-induced changes, usually in the form of dehydration or interference with the penetration of fixatives or preservatives into tissues.
For invertebrates, the most commonly reported changes include specimen swelling, shrinkage, weight loss, color changes, embrittlement, and loss of structural integrity (some specimens become flaccid). For example, the pattern of swelling in fixatives followed by shrinkage in preservatives was reported in a number of taxonomic groups by Donald and Paterson (1977) and Mills et al. (1982). Changes in the structural integrity of specimens have been reported in microplankton (Stoecker et al. 1994), actinopods (Beers 1976a), crayfish (DiStefano et al. 1994), and octopuses (Voight 2001). The loss of minerals or organic components of specimens have been reported in crustaceans (Hopkins 1968) and isopods (Hendrickz et al. 2003). Variable preservation was reported in ciliates by Modigh and Castaldo (2005). See table 13 for more details gleaned from published reports.
For vertebrates, most of the work on fixative- and preservative-induced changes has been done with fish, particularly investigations of shrinkage to compare measurements of live or fresh specimens with those of preserved specimens. Some authors (e.g., Shetter 1936; Sigler 1949) have calculated conversion factors for comparing fresh and preserved specimens. Lai (1963) and Lee (1982) studied the effects of formaldehyde fixation and alcohol preservation on morphometric characters in fish and anura, respectively.
The loss of mineral content in fishes was reported by Gibbs et al. (1974) and damage to otoliths by McMahon and Tash (1979). Loss of muscle definition (Winokur and Hillyard 1992) and changes in skin pustularity (Nelson 1971) have been reported in anurans. Stuart (1995) investigated the darkening of color patterns with increasing time in formaldehyde in lizards. Scott and Aquino-Shuster (1989) reported changes induced by freezing frog and snake specimens prior to preservation. See table 15 for more information and further examples of induced changes in vertebrate specimens.
Changes in Body Dimensions and Biomass
The effects of fixatives and preservatives on specimens that has drawn the most attention of researchers are dimensional and other biomass changes, as summarized in tables 13 and 14 for invertebrates, and tables 15 and 16 for vertebrates. Nevertheless, little research has been done on the specific effects of fixatives and preservatives on the particular tissues involved. We do know, for example, that elastin (a fibrillar protein in connective tissue and blood vessels) swells with exposure to formaldehyde and alcohol, but does not go into solution (Mukherjee and Hoffman 1971), but such studies are rare.
The literature on changes in dimensions and biomass during fixation and preservation has been reviewed by a few authors (e.g., Ellis 1987). These changes seem to be due primarily to dehydration as water in the tissues is replaced by alcohol. All preserving alcohols cause tissue shrinkage, but the amount of shrinkage is variable, with less shrinkage reported to occur in methyl alcohol and more shrinkage reported to occur in isopropyl alcohol compared to ethyl alcohol (Ciferri 1971). Preservation techniques can diminish or exaggerate some of the biomass changes—for example, preservation directly in alcohol (without a fixative) may result in shrinkage of cellular contents (syneresis) thereby causing distortion because alcohol penetrates tissues rapidly and has low specific gravity (Moore 1999). In general, the effects of biomass changes are much more pronounced in invertebrates than in vertebrates.
Many researchers have tried to establish a percentage of expected change that can be used to make better comparisons between preserved and nonpreserved specimens. Unfortunately, a review of the results of various studies shows too wide a range in variation to be useful, probably because of the variety of factors causing dimensional changes. For example, for anurans, Lee (1982) found dimensional changes of as much as two orders of magnitude in postpreservation in one toad species, with snout-vent shrinkage of 6.19 percent, while Deichmann et al. (2009) found snout-urostyle shrinkage of 0.31–5.6 percent in fourteen species. Fish studies have reported shrinkage of anywhere from 2–12 percent and more in body length (see table 15). Klauber (1943) and Reed (2001) reported shrinkage in length of snakes but no significant changes in body mass. Thibault-Botha and Bowen (2004) reviewed a large amount of the literature on body mass changes and concluded that gelatinous zooplankton are much more prone to shrinkage than other zooplankton due to their high water content and lack of strong structural features or hard coverings, and that body mass loss is generally higher in species with a higher surface to volume ratio. Some studies showed increases in length or weight following initial shrinkage of tissues. Radtke (1989) found significant shrinkage occurred at death in larval cod, but no shrinkage occurred when live larvae were placed in 95 percent ethanol, leading him to conclude that the shrinkages reported in the literature “were probably due to the handling associated with the preservation of specimens in these studies” (Radke 1989, 1893) rather than the preservatives. Theilacker (1980) found that handling of anchovy specimens may cause significant shrinkage prior to fixation or preservation. As summarized by one researcher who reviewed the literature, “The effects of preservatives on fish morphometrics are difficult to predict because of variance related to type of preservative, duration of preservative, origin of species (marine or freshwater fish), species, life stage, and others” (Sagnes 1997, 910).
Changes have also been noted in skin features due to differential preservation such as dermal pustularity of microhylids (Nelson 1971) and changes in the patterns of snakes (Smith 1955) and lizards (Smith 1975).
Changes in Color
Color changes may result from chemical alterations of the tissues, physical alterations of the tissues, or both, and may involve the loss, acquisition, or alteration of color. Color is conferred chemically by pigments, and structurally conferred by interference, light scattering, or refraction. Fluid fixatives and preservatives may both extract pigments and cause structural changes. For example, lipochromes (responsible for yellows, oranges, and reds) are alcohol-soluble carotenoids (Pettingill 1970). Fry (1985) described how the loss of color (yellow) due to submersion in alcohol in some bird specimens led to an erroneous description of a taxon as a new subspecies; Fry confirmed the color loss by subjecting dry specimens to submersion in alcohol, finding that green feather colors became blue-green, pure yellow became ivory, scarlet became pale buff, bright pink became buffy white, carmine became gingery brown, yellows and reds were suppressed, and that alcohol preservative had no effect on blue, orange-buff, or black. Another example of a color change that affected the usefulness of the specimen was in a note by Bliss (1872) concerning the unexpected appearance of a diagnostic vermillion spot on the abdomen of an alcohol-preserved cyprinoid fish that was not present when the fish was alive. Dissection demonstrated that the color was a true pigmentary color, and was probably visible only when the fish was in reproductive mode or preserved in alcohol.
Green-to-blue color changes in many fluid-preserved vertebrates (particularly amphibians and reptiles) occur when xanthophores (responsible for yellows) are leached out by preservatives and the remaining iridophores are altered by dehydration, which affects the interference of light. In at least one instance, an artist, apparently working with a freshly preserved specimen, captured the green-to-blue color change in mid-process in a snake specimen (Simmons and Snider 2012).
Prum et al. (1994), investigating how colors are produced in bird facial skin, noted that in one genus of birds, a bright green structural color produced by the reflection of light on ordered collagen fibers was better preserved in a 2.5 percent gluteraldehyde fixative than in standard 10 percent formalin. In specimens fixed in formaldehyde and preserved 70 percent alcohol, the bright green changed to violet or blue, due to shrinkage of the collagen fiber arrays.
Color changes in specimens may also result from reactions with copper (in the form of copper wire or a copper container), oxidation of metal containers or lids, or metal needles, pins, or probes. Leonhard Stejnener’s instructions for field preservation (of amphibians and reptiles) recommended the use of a color standard due to the loss of colors in preservative solutions (Stejnener 1891), a suggestion which has been ignored by most fieldworkers since (Simmons 2002).
Specimens are sometimes darkened by exposure to formaldehyde, a condition commonly referred to as formaldehyde brown or formaldehyde gray, which may be caused by several factors including the interaction of acidic formaldehyde with metal trays or tags, or the formation of what is commonly referred to as formalin pigment which may form as the formaldehyde reacts with the hematin in hemoglobin that escapes from the red blood corpuscles at cell death. Studies have also shown that specimens become darker the longer they are left in formaldehyde (Stuart 1995; see also the section, Unwanted Effects of Formaldehyde in chapter 2).
As a consequence of the color changes during preservation, there have been many attempts to concoct a preservative mixture that will retain colors. Probably the best known of these is the Kaiserling’s method (see table 9), originally proposed for anatomical specimens. It is important to note that there are many variations of the recipes for Kaiserling’s solutions in the literature; Kaiserling himself published several variations (e.g., Kaiserling 1896, 1897, and 1900), as did other workers (e.g., Craig 1914; Edwards and Edwards 1959).
Various published methods proposed to preserve color in fishes were reviewed by Borodin (1930), who found all of them to be unsatisfactory, at the same time complained about the damage that formaldehyde does to scales and bones (most likely due to the use of inappropriately buffered or unbuffered formaldehyde). Based on the ingredients then found in British-made IMS that were used for a few jars of fishes from the Red Sea that had retained much of their color, Borodin recommended a formula consisting of thirty parts alcohol, two parts formaldehyde, one and one-half parts wood tar, and sixty-six and one-half parts water saturated with common salt for the preservation of color in fishes. Gerrick (1968) used fifteen commercial antioxidants dissolved in either 10 percent formalin or 40 percent isopropyl to preserve colors in four species of fish for a two-year period. The best preservation of color was 1 percent erythorbic acid in 10 percent formalin. Ionol CP-40 preserved reds, but the other antioxidants failed. He found that “isopropyl alcohol, while an excellent solvent for antioxidants, was ineffective as a color-preserving fluid because animal pigments were highly soluble in it” (Gerrick 1968, 240).
One of the more creative attempts to preserve color was based on experiments conducted at the Colombo Museum (Sri Lanka), using coconut oil and carbolic acid with glycerin (Haly 1892). More recently, an antioxidant (BHT, or butylated hydroxytoluene, sold under the trade name Ionol or Ionol-40), was advocated to preserve colors (Smith 1995; Waller and Eschmeyer 1965; White and Peters 1969). Windsor (1971) recommended the use of a 50 percent solution of ammonium sulfate to preserve color in frogs, although he noted that the preservative caused excess dehydration of the specimens. Hildebrand (1968) recommended two procedures for color preservation. The first was a variation of a method published by Sheim (1951), which called for fixing specimens in a solution of 50 mL of formaldehyde buffered with 5.9 g dibasic sodium phosphate and 4.7 g monobasic sodium phosphate with sufficient tap water to make 1000 mL for about a week; then preserving the specimens in a fresh solution of the same formula plus 5 g sodium hydrosulfite that was aged for a week and filtered (the author cautions that it is not suitable for fatty tissues and should not be used in metal containers). Hildebrand’s second recommendation was a variation of the Owen and Steedman (1956) method in which specimens are fixed for a minimal amount of time in formalin or alcohol to which 10 g of sodium acetate per liter has been added, then washed twelve to twenty-four hours in running water, and stored in either a 1 percent mixture of ethylene glycol monophenyl ether in water or a 0.2 percent solution of para-hydroxybenzoic acid in water. Zenke (1991) describes a method using ethylene glycol to preserve the silvery luster in some fish species; a review of several recipes for color preservation can be found in Kessler (1989). Similarly, botanical researchers have proposed a number of chemical preservatives or additives, most targeted toward preserving particular colors in plants (table 1; see discussion under Botanical Use of Fluid Preservation in chapter 3).
Solvent Extraction in Fixatives and Preservatives
The preservative fluid around a specimen includes the components extracted from the specimen. Alcohol is a good lipid solvent, and has been used for years to extract lipids from tissues (Bloor 1943; Johnson 1971). A study by von Endt (1994) showed that in preservative fluids, fats are hydrolized and migrate within the specimens, and are ultimately removed by the alcohol. The amino acid profiles he found in preservative solutions indicate that the specimens undergo a general protein loss, as well as some structural protein loss (von Endt 1994).
Extracted lipids in preservatives tend to float to the top (lipids are insoluble in water) and oxidize into fatty acids, which can lower the pH of the fluid and cause tissue breakdown (Dingerkus 1982; Moore 2002a; Moore 2005b). Figure 4.1 shows lipids dissolved in an alcohol preservative; figure 4.2 shows lipids dissolved in a formaldehyde preservative (floating near the surface of the solution). Moore (2002a) reports on an instance of a stoat in a preservative so contaminated by emulsifying lipids that the pH of the alcohol preservative was 3.6, low enough to decalcify the animal’s skeleton.
Figure 4.1. Dissolved lipids in preservative with specimens.
Figure 4.2. Dissolved lipids floating at surface of formaldehyde preservative.
A relative estimate of the amount of lipids in the preservative can be made by pipetting a small amount of fluid into a Petri dish of distilled water on a black background and evaluating the turbidity—the lipids will appear as a white cloudiness around the pipette (Moore 2005b).
A study by Edwards et al. (2002) determined that formaldehyde fixation caused a decrease in stable isotopes in fish specimens.
Preservative fluids may become discolored due to solvent action, and this discoloration may affect the specimens. Figure 4.3 shows ink bleeding into an alcohol preservative (the label was added to the jar before the ink was sufficiently dry). Preserving fluids that have become discolored due to the presence of dissolved lipids may be cleaned by dropping charcoal particles in the fluid and shaking it, then filtering the charcoal particles out (Simon Moore, pers. comm.).
Figure 4.3. Ink bleeding from label in alcohol preservative.
A few studies have concentrated on specific changes that can affect the research use of collections. Hendrickx et al. (2003) determined that using a preservative fluid in a pitfall trap (4 percent formaldehyde with a small amount of detergent to decrease surface tension) could cause significant alterations in metal concentrations in woodlice. Hopkins (1968) found that some crustaceans underwent a carbon loss of 17–23 percent and nitrogen loss of 19.2–21.0 percent in fixatives and preservatives. Morris (1972) found that marine zooplankton lost up to a third of their total lipids, suffered degradation of polyunsaturated fatty acids in the lipids, and some hydrolysis of lipids when preserved in 10 percent formaldehyde or 100 percent methanol.
Much work remains to be done on the effects of fixatives and preservatives on specimens, particularly the changes that affect the comparison of live and preserved specimens for scientific studies. Based on the evidence available to date, it is unlikely that a simple correction factor can be calculated for most taxa, due to the many variables that contribute to changes induced by fixation and preservation, including how specimens are handled, how fixatives and preservatives are mixed, which buffers are used, and how specimens are stored. A further complicating factor is that for most preserved specimens now in collections, there are no reliable records of how specimens were handled, fixed, and preserved.