STORAGE FACTORS INFLUENCING ETHANOL CONCENTRATION OF FLUID-PRESERVED INSECTS
Abstract
The scientific value of insect specimens stored in ethanol depends upon their adequate preservation, which is most directly impacted by the preservative solution’s concentration. Determining what storage factors influence ethanol concentration, including the size and type of container and closure, as well as the ratio of specimen to ethanol volume, can inform collections staff on how to manage ethanol-preserved insect specimens ideally. We hypothesize that ethanol solution concentrations would 1) decrease with increasing insect volume to ethanol volume ratios, 2) decrease with time, 3) vary by cap liner material, and 4) decrease more slowly in smaller vials. To test these hypotheses, we measured ethanol solution concentrations in 1,376 vials containing insects collected over an 8-year period and subsequently stored undisturbed for 4–11 yr. The ratio of specimen to ethanol volume was most influential, followed by time in storage, as increasing insect volumes and storage times resulted in lower ethanol concentrations. Cap liner material and vial size did not significantly affect concentration, but human error in ethanol solution mixing and overpacking of specimens in vials created poor preservation conditions. These results can inform collections management methods for ethanol-preserved insects, which are instrumental for keeping specimens scientifically useful in perpetuity.
Introduction
Insects are the most diverse taxon on Earth, and insect specimens in natural history collections are powerful sources of data that can be used in many different fields of science, including conservation, systematics, evolution, ecology, infectious diseases, and climate change research (Suarez and Tsutsui 2004, Kharouba et al. 2018). Although the focus of most entomology collections is pinned specimens, many collections also have holdings of specimens preserved in fluid (Anderegg et al. 2025). Although fluid preservation is less common in entomology than for other invertebrate taxa (e.g., myriapods, arachnids, crustaceans), it is the best preservation method for many important groups of specimens, including ecologically and taxonomically important immature life stages (Yang 2001, Meier and Lim 2009), small specimens too fragile to pin (Schauff 2001), and aquatic species (Hilsenhoff 2001, Baird et al. 2011). Fluids are also used to store large quantities of specimens collected with bulk-catch methods such as Malaise traps, pitfall traps, and kick nets (Montgomery et al. 2021). However, the best practices for the care of fluid-preserved specimens are still being established (Simmons 2014, Neumann et al. 2022). This is especially true for insects, which present a unique challenge because of their small size and large number in storage, relative to vertebrate taxa. Insect specimens preserved in fluid frequently do not receive the same detailed care as their pinned counterparts, and collections staff must rely on untested management and preservation techniques passed by word of mouth because of a lack of references and resources. Establishing the best practices in terms of methodology and quality of care for insect specimens stored in fluid is critical to ensuring that they are accessible and useful for scientific research.
Although fluid preservation is suitable for many insects and becoming increasingly relevant, managing fluid-preserved specimens involves many logistical and methodological challenges. Ethanol (ethyl alcohol) is the most common fluid preservative for zoology specimens, including insects (Simmons 2014, Anderegg et al. 2025), and that is the focus of the research presented in this article. A solution of 70% ethanol and 30% water is widely recommended, but a range of 70–95% ethanol has been used in invertebrate collections (Levi 1966, Simmons 2014). A complication with ethanol–water mixtures is that as the solution experiences evaporation and volume decreases, the concentration of ethanol decreases as well (Waller and Strang 1996; Notton 2010; Simmons 2014, 2019), because ethanol has a higher vapor pressure than water and evaporates out of solution more quickly (Simmons 1995, Waller and Strang 1996). As ethanol concentrations decrease because of evaporation, the changes in osmotic pressure cause specimens to absorb water and experience deterioration at a cellular level (Notton 2010, Simmons 2014). Additionally, ethanol solutions do not suitably inhibit bacterial and fungal growth at concentrations below 50% (Waller and Strang 1996). Low ethanol levels may also expose specimens to desiccation and mechanical damage when the weight of specimens is no longer properly supported (Simmons 2014). However, if containers are properly sealed and collections staff regularly monitor for and remediate fluid loss, ethanol can preserve insect specimens for many decades (Neumann et al. 2022).
To counteract the effects of evaporation and replace lost ethanol, collections personnel often add new ethanol on top of the existing solution, which is commonly referred to as “topping off” (Notton 2010, Simmons 2014). The concentration of the added ethanol is usually either the same as the target storage strength (e.g., 70%), or of undiluted stock strength (95.6% ethanol produced from distillation alone; higher percentages require chemical dehydration) if the collections personnel assume that evaporation has reduced the ethanol concentration low enough that a higher percentage is required to bring the solution back up to storage strength (Anderegg et al. 2025). However, these practices may result in concentrations much lower or higher than intended and can harm specimens and threaten their utility for future scientific study (Notton 2010). Calculating the exact amount and concentration of ethanol needed to top off a specific container and restore the ideal storage state requires measuring the current volume and concentration of ethanol (Notton 2010), which is prohibitive for most collections because of the costly technology and time involved.
Additional difficulties can arise when preparing specific storage strength solutions (e.g., 70%) from a 95–100% stock solution. This requires the addition of water in a specific volumetric ratio (Notton 2010), and ideally the added water is deionized to avoid impurities that could affect solution pH (Kotrba and Schilling 2017, Neumann et al. 2022). After water has been added, the ethanol solution should ideally be left to mix through diffusion for at least 24 hours in order for the ethanol and water to homogenize completely, and to allow the heat of mixing to subside (Simmons 2014). Other actions that can speed this process include stirring, shaking, decanting, and remixing, although 24-hour diffusion is recommended (Simmons 2014, Neumann et al. 2022). For example, a carboy first filled with 95–100% ethanol, then water, will be layered with a stronger concentration of ethanol on the bottom than on the top of the container. Allowing the solution to thoroughly mix for at least 24 hours reduces layering and ensures that the solution is a uniform concentration throughout. One common error is to remove liquid from the solution before homogenization, thus inadvertently using a solution either significantly higher or lower than the assumed concentration, depending on how the ethanol was extracted (e.g., siphoned from bottom or poured from the top). Therefore, ethanol concentrations may also be influenced by solution preparation and dispensing methods, before the solution is added to specimen containers.
Understanding how ethanol concentrations are impacted by different specimen storage factors can inform managers on what aspects of their ethanol-preserved collections they should consider when deciding how and when to top off containers, as well as prevent evaporation altogether, which can save valuable time and effort and contribute to an effective management strategy for fluid collections. Ethanol concentration may vary with a variety of storage factors, including the ratio of specimens to ethanol (Cato 1990, Palmer 1996, Pickering 1997), the length of time in storage, specimen taxa, and the size and material of the container and closure (Simmons 2014, Cruz-Rodríguez et al. 2021). Ethanol concentrations can become diluted as water is released from the specimens themselves (Taylor 1981), but the size of this effect may depend on the taxa and life stage of the specimens, specific fixation or collection methods, and the proportion of specimen volume to ethanol volume. Although previous studies have recommended ideal specimen-to-ethanol volume ratios [33% specimens to 66% ethanol (Zweifel 1966, Whitman et al. 2019); 30% specimens to 70% ethanol (Simmons 2014)], this relationship has never been directly measured. Evaporation also decreases the concentration of ethanol in solutions, but rates of concentration decrease are not well documented, as previous collections surveys have only captured concentration data at a specific moment in time (Cato 1990, Notton 2010, Palmer 1996, Pickering 1997). The permeability and diffusion of water and ethanol vapors through the sides and lid of a storage container may also influence evaporation rates of ethanol, and therefore the concentration within (van Dam 2000). Although the storage value of different container materials has been widely discussed (Levi 1966, Suzumoto 1992, Clark 1993, Warén et al. 2010, Cruz-Rodríguez et al. 2021), the exact measurement of how different container and closure materials affect ethanol concentration in collections surveys can provide meaningful recommendations that are directly applicable to fluid collections management practices.
Determining how these storage factors influence the preservation environment of specimens stored in fluid necessitates a thorough collections survey and evaluation. Previous surveys evaluating ethanol concentrations in natural history collections have been conducted for collections of mammals (Cato 1990, Palmer 1996), noninsect invertebrates (Schiller et al. 2014), reptiles and amphibians (Waller and Simmons 2003), and vertebrates and noninsect invertebrates (Pickering 1997; J. Pickering, Peabody Museum of Archaeology and Ethnology, pers. comm.), but never in detail for fluid preserved insects. Fluid preservation in insect collections (and other arthropod/invertebrate taxa) differs from that of larger taxa (vertebrates) because insects are smaller and usually stored in smaller containers (with the exception of bulk lots). This means that entomology fluid collections are composed of a greater quantity of small containers, making it more difficult and time consuming to assess and remediate ethanol concentrations and specimen preservation states.
In this research, we surveyed a collection of insects. These were mostly beetles sorted from pitfall trap samples that were collected over an 8-year period, preserved in 70% ethanol, and remained undisturbed in storage. The objectives were to determine how preservation state, specifically ethanol concentration, was influenced by different storage factors, including the ratio of specimen to ethanol volume, the number of years since collection and preservation, and different cap liner materials and vial sizes. This study addressed five specific questions:
How does the volume of insects in a container affect ethanol concentration? We hypothesized that ethanol concentration would decrease as the volume of insects increased.
How does ethanol concentration change over time? We hypothesized that ethanol concentration would decline over time.
How does cap liner material affect ethanol concentration? We hypothesized that ethanol concentration would vary by cap liner material.
How does vial size affect ethanol concentration over time? We hypothesized that larger vials would have larger ethanol concentration reductions over time than smaller vials.
What storage factors are the most important determinants of ethanol concentrations in preserved insect samples? We hypothesized that the ratio of insect volume to ethanol volume and time since initial preservation would be most influential on current ethanol concentrations.
Methods
Study System
Insect samples, predominantly beetles but also solifuges and lepidopterans, were preserved from research collections acquired by a member of the research team (C.M.M.) from general arthropod surveys conducted during 2010–2012 (McCain et al. 2018, McCain 2021), and several carrion beetle surveys conducted primarily by students in the summers of 2013, 2017, and 2018 from the Front Range of the Rocky Mountains. All pitfall trap samples used propylene glycol as an initial preservative and euthanizing agent, and arthropods from the samples were subsequently cleaned and transferred to vials with 70% ethanol, then stored in a university laboratory. Analyzed insect samples were undisturbed during storage, and no additional ethanol was added to vials at any point. Insect vials were stored in archival trays on open shelving in an active research lab using both natural light and electric lighting, average temperatures of 68–72°F were generally maintained throughout the year, and relative humidity was typically low (average 25% because of regional climate and building age).
Collection Storage State
Samples were stored in three different combinations of container and cap liner materials (Fig. 1):
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
wide-mouth high-density polyethylene (HDPE) bottles with foam-lined HDPE caps;
borosilicate glass vials and black phenolic plastic screw caps with a polyethylene lining;
borosilicate glass vials capped with black phenolic plastic screw caps lined with a vinyl disk (2013: 4-dram vials only).
Borosilicate glass vials included eight sizes (1/2 dram, 1 dram, 2 drams, 4 drams, 5 drams, 6 drams, 8 drams, 9.5 drams, and 11 drams). Each sample was given a “fullness class” rating based on the visual appearance of the proportion of insects to the volume of the container (Fig. 2A; six classes: <25%, 25%, 50%, 75%, >75% to <100%, and 100% full). This visual estimation was compared with a measurement of the ratio of specimen to ethanol volume (see the following).
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
Ethanol Concentration Measurement
We hypothesized that ethanol concentration would have an inverse relationship with the volume of insects stored in each vial. A higher ratio of insects to ethanol may decrease this concentration as body water is released from the specimens, diluting the ethanol. To determine the proportion of insect to ethanol volume, we removed the preservative solution and insects from the vials, measured the overall volume including the insects and ethanol (milliliters), then carefully separated the insects out from the ethanol using fine strainers and cheesecloth, and measured the volume of the ethanol alone (milliliters). The ethanol concentration (%) of each sample was then measured to the nearest 0.01% using an Anton Paar Snap 40 series digital density/specific gravity meter. The meter was flushed three times with deionized water and completely emptied between measurements to ensure accurate readings. For 0.5-dram vials (1.85 ml), which did not have enough ethanol to provide a reading (the meter requires 3 ml of fluid; Schiller et al. 2014), ethanol from two vials were combined for a concentration measurement (Cushing and Slowik 2007). Vials were only combined if they stored the same taxa and life stages (adult, larvae, or both) at the same fullness class and were from the same year of collection. The percentage volume of insects for each vial was calculated as: ([total volume − ethanol volume]/total volume) * 100. After analyses, specimens from vials with a fullness class higher than 50% were moved to a larger vial, and vials refilled with fresh 70% ethanol.
Analysis
We ran separate analyses to measure the effects of storage factors (insect volume, time, cap liner materials, and vial size) on ethanol concentration for each question, and we also ran a multivariate model to test the significance of all the storage factors simultaneously. All analyses were performed using JMP Pro 16 (SAS Institute 2022). The five questions and their analyses were as follows:
Question 1. How does the volume of insects in a container affect ethanol concentration?
We hypothesized that ethanol concentration would decrease as the volume of insects increased. For these analyses, measurements were split into two groups: samples collected and preserved in 2018, and samples from all other years (2010–2013, 2017). Visual inspection of the samples showed that many of the vials in 2018 were overpacked and exhibited large variation in insect volumes; thus, the decision was made to analyze those specimens separately. To test our hypothesis that ethanol concentration would decrease as the volume of insects increased, we used an ordinary least squares linear regression. For all analyses, percentage volume of insects was used instead of fullness class as a predictor variable because of its greater accuracy.
Question 2. How does ethanol concentration change over time?
We hypothesized that vials of insects would have lower ethanol concentrations the longer they had been stored because of ethanol evaporating out of solution with the passage of time. We used an ordinary least squares linear regression to test for a decline in concentration among samples from 2017 to 2010. Only vials containing less than 25% insects by volume were included to eliminate the effect of insect volume.
Question 3. How does cap liner material affect ethanol concentration?
Because different container closures are known to influence ethanol evaporation rates, we hypothesized that ethanol concentration would vary among the cap liner materials in this collection. We only included samples from 4-dram vials from 2013 in these analyses, as these samples included caps with both polyethylene and vinyl liners. Samples from large plastic jars with plastic caps were not included because of low sample sizes (n = 17). Means and standard deviations were calculated for each cap liner material. To detect if ethanol concentrations between vials with differed cap liner materials varied, we used a nonparametric Wilcoxon rank sum test to account for the nonnormal distribution of the data.
Question 4. How does vial size affect ethanol concentration over time?
We hypothesized that ethanol in larger vials would evaporate more quickly because of larger vial mouths, and potentially greater area for evaporation around the lid circumference. Samples were filtered to include 1-dram, 2-dram, 4-dram, and 6-dram vials with 25% or less insect volume, and to include only vials from 2010–2017. Unfortunately, the sample sizes of other container sizes (i.e., 0.5, 5, and 8–11 drams, and larger plastic jars) were too small for comparison. To test if the rates of decrease in concentration over time differed among vial sizes, we compared 95% confidence intervals of slope estimates from ordinary least squares linear regressions of ethanol concentration over time.
Question 5. What storage factors are the most important determinants of ethanol concentrations in preserved insect samples?
We hypothesized that the ratio of insect volume to ethanol volume and time since initial preservation would be the most influential factors on current ethanol concentrations. To account for multicollinearity and compare strength of predictors, we created a multivariate linear regression model for ethanol concentration, including the percentage volume of insects, time since preservation, cap liner material, and vial size. Multivariate data were filtered to have some consistency in sample size across parameters and included samples with ≤25% insect volume from 2010 to 2013 and 2017 collection years. Larger plastic jars were excluded because of their small sample size (n = 9).
Results
Overall, 1,376 vials of 12 different vial sizes were measured from collecting events in 2010–2013, 2017, and 2018. Vials mostly contained beetles, followed by lepidopterans, mixed bulk taxa, and a small number of solifuges.
Question 1. How does the volume of insects in a container affect ethanol concentration?
We hypothesized that ethanol concentration would decrease as the volume of insects increased. Vials exhibited a wide range of insect volumes (<0.10–70%, Fig. 2B) and ethanol concentrations (from 1.28% to 93.2%, Fig. 3). Although the initial target ethanol concentration for preservation was 70% for all vials in this collection, several samples were measured with >70% ethanol concentration, likely because of errors in ethanol mixing. Notably, the 2018 insect samples displayed the largest variation in ethanol concentration, with a few measuring over 75%, and substantial numbers with ethanol concentrations below 50% (Fig. 3), indicating an increased risk of decay. Vials from 2018 also were characterized by extreme overpacking of insects, with ∼1/3 of the vials exhibiting ≥75% and above fullness classes. In contrast, samples from other years had only a few vials with fullness classes over 75%.
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
The percentage volume of insects in each vial was roughly estimated by the observed fullness class (Fig. 2B), thus providing a helpful heuristic procedure for collections personnel. However, variation was high and increased with fullness class, as was overlap in classification assignment because of the visual appearance of the vials. This was because some vials contained a small number of insects with larger body sizes, thus leading to a “full” designation (because additional specimens could not be added) but with substantial ethanol between individuals. Thus, fullness classes systematically overestimated the actual percentage volume of insects, predicting that insect volume was higher than the exact measurements because of the vials’ visual appearance. The actual insect volume is about one third of the fullness class (e.g., 75% fullness ∼25% insect volume). For the analysis, the percentage volume of insects was used instead of fullness class as a predictor variable because of its greater accuracy.
Ethanol concentration decreased significantly as the percentage volume of insects in each vial increased (Fig. 4). This was observed in both the 2018 samples (Fig. 4A; r2 = 0.709, P < 0.0001, n = 368), which experienced significant overpacking and lower ethanol concentrations, as well as samples from all other years of collection (Fig. 4B; 2010–2013, 2017; r2 = 0.106, P < 0.0001, n = 1014). Samples from 2018 decreased by 0.78% for every year in storage, 10 times faster than the 0.08% decrease per year for samples from earlier years.
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
Question 2. How does ethanol concentration change over time?
Excluding overpacked vials (>25% insect volume) and 2018 samples, ethanol concentration significantly decreased over time (Fig. 5; r2 = 0.035, P < 0.0001, n = 847). Ethanol concentration decreased by 0.24% per year in storage, but only a small fraction of variability was explained by time (see the following). Errors in ethanol preparation also likely contributed to concentrations greater than 70%.
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
Question 3. How does cap liner material affect ethanol concentration?
Comparing vials with polyethylene lined caps (n = 143) to vials with vinyl lined caps (n = 123) all from 4-dram vials preserved in 2013, there was no significant difference in ethanol concentration (Fig. 6; X2 [1, n = 266] = 0.89, P = 0.3443). Although polyethylene lined caps had a slightly lower mean and more variation in ethanol concentration (mean = 66.7; σ = 2.32) than vinyl lined caps (mean = 67.1, σ = 1.741), there was no significant difference.
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
Question 4. How does vial size affect ethanol concentration over time?
In our study, ethanol concentration did not vary by vial size. Decrease in ethanol concentration over time was similar for 1-, 2-, and 6-dram vials (r2 = 0.069, P = 0.0002, n = 194; r2 = 0.048, P = 0.0002, n = 280; r2 = 0.108, P = 0.0312, n = 43; respectively) but not for the 4-dram vial size (Fig. 7; r2 = 0.0007, P = 0.9891, n = 281). The 95% confidence limits for the slope estimates were largely overlapping among 1-, 2-, and 6-dram vials (−0.47 to −0.15; −0.50 to −0.15; −0.67 to −0.03; respectively). The 4-dram vials did not exhibit a trend in ethanol concentration over time. Several samples across all vial sizes had ethanol concentrations >70%, and when removed from analyses resulted in no significant differences in slope overlaps, but fits did improve for 1- and 2-dram vial sizes, but not 6-dram ones (r2 = 0.107, P < 0.0001, n = 187; r2 = 0.102, P < 0.0001, n = 270; r2 = 0.092, P = 0.051, n = 42; respectively).
Citation: Collection Forum 36, 1; 10.14351/0831-4985-36.1.23
Question 5. What storage factors are the most important determinants of ethanol concentrations in preserved insect samples?
In our study, the percentage volume of insects and time since collection had the most influence on ethanol concentration. The multivariate least squares regression model for variation in ethanol concentration included the percentage volume of insects, time since preservation, cap liner material, and vial size only included vials with ≤25% insect volume, preserved before 2018, and excluded plastic jars. The model was significant but only accounted for a small amount of the variation (r2 = 0.080, P < 0.0001, n = 838). Of the four variables, only the percentage insect volume (F ratio = 36.81; P < 0.0001) and time in storage (F ratio = 43.79; P < 0.0001) had a significant influence on ethanol concentration, whereas vial volume (F ratio = 2.15; P = 0.1428) and cap material (F ratio = 0.47; P = 0.4937) contributed negligibly. Overall, the percentage insect volume was the strongest predictor of ethanol concentration even at insect volumes at or under 25%, with nearly double the effect size as time in storage.
Discussion
The value of specimens stored in natural history collections comes from the ability to use them repeatedly for scientific research, as long as they remain adequately preserved over time. This survey of a collection of ethanol-preserved insects demonstrates that concentration (which was ideally and initially 70% ethanol in this collection) is influenced mostly by the ratio of specimen to ethanol volume (Fig. 4) and the amount of time in storage (Fig. 5). The size of vials (1–6 drams) and cap liner material (polyethylene or vinyl) were not significant predictors of ethanol concentration. Collection managers seeking to improve the storage state of ethanol-preserved insects in their collection should pay considerable attention to the volume of insects in each container, and how long the specimens have been preserved in the specific solution.
Question 1. How does the volume of insects in a container affect ethanol concentration?
In this study, the ratio of specimen to ethanol volume was the largest and most significant predictor of ethanol concentration, and this effect was especially dramatic and easily observed in samples from 2018 (Fig. 4A). Although the vials of specimens collected in 2018 had to be removed from most analyses because of their dramatic specimen overpacking, they offer insight into ethanol preservation methods. A majority of 2018 vials were not prepared in accordance with lab protocols for preservation and ethanol mixing. This was the only year of collection to have vials with ethanol concentrations lower than 50% (the threshold at which ethanol is no longer an effective biocide), despite being the most recent year of specimen collections. The 2018 samples exhibited a rate of ethanol decrease per year that was 10 times higher than samples from other years, indicating that overpacking vials results in a fast and drastic decrease in ethanol concentration. Many of the caps from 2018 samples were also loose (compared to other years), and although this was not tracked or quantified, this user error likely contributed to the pattern of lower ethanol concentrations for these samples, as loose container seals allow more ethanol to evaporate out of containers. Despite the eccentricities of the 2018 vials, ethanol concentrations declined with increased insect packing without their inclusion (Fig. 4B), and the ratio of specimen to ethanol volume was still the strongest variable in the multivariate model, even when samples were limited to ≤ 25% insect volume.
Previous surveys of fluid concentrations in natural history collections also emphasized specimen-to-fluid ratios, although exact ratios were not quantified. Cato (1990) recorded the visual ratio of specimens to fluid in 400 jars of a mammal collection and concluded that a higher ratio indicated a lower ethanol concentration. Containers with <60% ethanol concentration had a relatively high number of specimens and a larger specimen-to-volume ratio (Cato 1990). Although they sought to measure the exact amount of ethanol (milliliters) in each jar for comparison, this metric was not consistently recorded by observers to use for analyses (Cato 1990). Palmer (1996) measured the ethanol concentrations of 272 jars in the Smithsonian National Museum of Natural History mammal collection, and also concluded that containers with a higher visual estimation of specimen-to-ethanol volume ratio had lower concentrations but did not quantify ratios. Pickering (1997) recorded a trend of decreasing concentration with increasing specimen-to-ethanol ratio in her evaluation of Oxford’s fluid collections, stating that jars with lower concentrations (<60%) generally had a higher ratio of specimens to ethanol, and that invertebrate jars with a higher number of specimens had lower concentrations as well. However, the exact ratio or volumes of specimens and ethanol were not recorded, and she only commented on the general trend and did not quantify it (Pickering 1997).
Several recommendations have been made on ideal specimen-to-fluid ratios, including 33% specimens to 66% fluid (Zweifel 1966), and 30% specimens to 70% fluid (Simmons 2014). Based on the results of this study, we recommend a ratio of 25% insect specimen volume to 75% ethanol for all container sizes. Collectors and museum staff may want to preserve the same taxon from a single collecting event together, which can lead to overpacking if specimens are numerous. However, this will lead to faster decreases in ethanol concentration and may expose specimens to substandard storage conditions. Ideally, larger containers should be selected for these larger lots of specimens for an ideal specimen–to–fluid-volume ratio. If larger containers are not available or collections facilities do not support the storage of larger containers, then the lot should be divided into multiple containers and clearly labeled as part of a series, on both physical labels and in collection management systems. Overfilling containers with specimens also may cost more over time, as containers will need to be topped off or have their fluid completely replaced more frequently if they are densely packed, using more fluid preservative, which can be costly, and a greater amount of staff time and resources.
Quantifying the exact ratio of specimens to ethanol in each container was critical for our study but is a time-consuming process that is likely unfeasible for most collections. As stated in the foregoing, previous studies used visual estimations of the ratio of specimens to fluid. For collections staff assessing storage states of many containers, this may be the only practicable and efficient method. Visual heuristics like the “fullness classes” used here of <25%, 25%, 50%, 75%, >75% to <100%, and 100%, although less accurate, helped identify overpacked vials (Fig. 2A). In our study, measured insect volumes were consistently about a third that of the estimated fullness class, a metric that could be applied to similar collections of insects (larger beetles in vials). At fullness classes >25%, variability increases, and the fullness classes are less predictive of actual insect volume. This is because even when no additional insects can be added to a container, there is still room for ethanol in between and around the insects. This variability means that collections staff profiling vials may inaccurately assess actual specimen to fluid ratios. But because accuracy between fullness classes and volume measurements was improved at the ≤25% levels, identifying those above that threshold and remediating to a larger vial size is an easy adjustment for efficiency.
Question 2. How does ethanol concentration change over time?
After specimen-to-ethanol volume ratio, time was the second most important and significant predictor of ethanol concentration (Figs. 4, 5). Vials that were packed ≤25% full lost about a quarter of a percent of ethanol concentration each year, or 1% loss every 4 years. Quantifying this effect is important for collections, because containers may stay unmonitored in storage for long periods. Previous studies surveyed containers that had been subject to various fluid management and topping off regimes over many years. These fluid management practices were not tracked or recorded in detail; thus, these studies were unable to assess how fluid concentrations changed over time in their collections (Cato 1990, Palmer 1996, Pickering 1997). Our study surveyed vials that had not been opened or disturbed since their initial preparation, making them an ideal system in which to study the change in ethanol concentration over time. Ideally, managers should track when vials were last topped off or refilled with fresh preservative so they have longitudinal data points for various containers, which can help determine which containers need more attention and need to be topped off more frequently based on different characteristics (size, closure type, location in collection, collection environmental conditions, etc.). Ideally, this includes measuring concentrations for random subsets of containers if collections are too large to assess every container (Notton 2010).
Question 3. How does cap liner material affect ethanol concentration?
In this collection, cap liner material did not significantly affect ethanol concentration. Other data sets may have a stronger ability to assess these effects, as we could only assess the difference between two different liner types for black phenolic plastic screw caps. Studies that survey a wider variety of cap and lid types that vary in material, size, and closure type can more readily comment on the performance of different vial caps and jar lids. Indeed, Cruz-Rodríguez et al. (2021) found that container closures with a pressure and twist design were more effective at preventing evaporation than simple pressure caps without threading in their survey of 2,434 zoology specimen lots. In contrast, Schiller et al. (2014) did not find a difference in ethanol concentration between containers with screw top or ground glass closures. Although Schiller et al. (2014) observed a significant interaction between their “closure type” and “collection” variables (Arachnoidea, Crustacea, Evertebrata Varia, Mollusca, Myriapoda), this could be due to different management regimes and histories between their collections. The environmental conditions of a collection can also exacerbate the effects of container material, as the sides of the container, air in the headspace of the container, and the preservative itself expands and contracts with changing temperature and relative humidity (van Dam 2000, Simmons 2014). Additionally, how tightly the container lid is attached can also significantly affect concentration over time.
Question 4. How does vial size affect ethanol concentration over time?
We hypothesized that larger vials would have larger ethanol concentration reductions over time than smaller vials, but our results showed that the size of vials did not significantly affect ethanol concentration. Our study was only able to assess glass vial sizes of 1–6 drams, which are all smaller vials when compared with the variety of container sizes used in natural history collections. Additionally, the 4-dram vials were predominately utilized in 2013 and had high variability in ethanol concentrations, thus likely swamping the overall time signal. Despite the limitations, these results match that of Pickering (1997) and Schiller et al. (2014), who both concluded from their collections surveys that jar size does not affect alcohol concentration. Palmer (1996) reported that the largest containers had lower ethanol concentrations, but this was likely because larger jars were more packed with specimens and had a higher specimen-to-volume ratio. Cato (1990) concluded that jar size did not impact fluid concentration, but that the combination of jar size and closure type is likely more important. However, they only measured fluid concentration in jars larger than 450 ml due to the constraints of using a float hydrometer.
Question 5. What storage factors are the most important determinants of ethanol concentrations in preserved insect samples?
Both time in storage and the percentage volume of insects were the most important factors in determining ethanol concentrations in this study. However, although the multivariate least-squares regression model was statistically significant, it only accounted for a small amount of variation (r2 = 0.080) and was weak compared to the individual analyses. The model only included samples stored in glass vials with ≤25% insect volume and from 2010 to 2013 and 2017 collection years, because of large variations in 2018 samples and the dramatic effect of insect volume at percentages above 25%. This analysis was unable to include the effects of human error in ethanol mixing and other preparation steps that are likely to influence ethanol evaporation rates, including how tightly vial caps were applied. Although the model and overall study design were unable to account for every factor that could influence ethanol concentration, the percentage volume of insects and time in storage significantly influenced ethanol concentrations, whereas vial volume and cap liner material did not. Thus, collections staff should pay particular attention to these two storage factors when managing their ethanol-preserved specimens.
Recommendations for managers of ethanol-preserved insects
Collection staff can combat rates of ethanol evaporation and decreasing concentration with the proper instruction of personnel, assessment, remediation of containers with high insect to ethanol volume ratios, and the establishment of a fluid-collection monitoring schedule. Our recommendations for best practices for entomological ethanol collections include:
Avoid more than 25% insect volume in containers. If collections staff want to keep a large number or volume of insects together (because they are from the same collecting event and/or taxon), specimens can be moved to an appropriately sized container that will allow the specimen-to-ethanol ratio to be kept at or below 25% specimens by volume, or a fullness class of 25% or below in this study. If larger containers are not available, specimens can be divided up into different lots and labeled accordingly. However, measuring the exact volume of insects in a container to calculate the specimen-to-ethanol volume ratio may be prohibitively time consuming and subject specimens to unnecessary handling. Instead, staff can first fill containers 75% full with ethanol when preparing specimen lots, and then place specimens into the fluid until the container is full.
Track ethanol concentrations on a set schedule. Ethanol concentrations decrease over time as solutions experience evaporation, and determining the rate of concentration decrease in a specific collection can allow staff to establish a set schedule for assessing and remediating ethanol concentrations and levels in containers (Notton 2010). We recommend checking ethanol concentrations in containers with ideal specimen-to-ethanol volume ratios every 4 years, and containers with a higher specimen-to-ethanol ratio and/or inadequate containers and closures every year. Concentration assessments should be recorded for each container (measurement and date checked), as well as additional storage information that may impact concentration (container and closure type, specimen to ethanol volume, collection environment). Tracking the fluid volumes, concentrations, and storage states of individual containers is important in establishing exactly how containers lose ethanol over time and can help staff determine if evaporation rates vary between different containers in their collection. This can be accomplished by giving each container a specific numeric identifier, separate from catalog numbers if there are multiple catalog numbers in a single container. If the collection is too large to measure the concentration in every container, then a subset can be chosen and tracked as representatives of other containers that have the same characteristics (e.g., size, material, closure, taxa, storage location). Small vials that do not have enough ethanol to assess concentrations can be filled completely (to the shoulder of the vial) with a solution of known concentration, and then regularly monitored for visual decreases in fluid volume, which likely indicates decreasing ethanol concentration as well.
Combat user error in specimen preparation and ethanol mixing steps. Ethanol concentration is significantly influenced by the ratio of specimens to fluid in containers. Collections staff may instruct personnel who prepare ethanol-preserved samples and researchers who will be donating their material to the collection that containers should not be filled with insects above a certain amount (25% or less insects by volume recommended here). This practice can save time and money in the long run, as full vials will not have to be split into separate containers later, and ethanol concentrations will decrease more slowly, which saves money on expensive ethanol and eliminates the need to completely replace solutions with extremely low concentrations. When mixing ethanol of stock strength (95–100%) with water to create storage strength solutions (e.g., 70%), wait at least 24 hours before using to allow the fluids to fully mix (Simmons 2014). Check the concentration of the ethanol before using, either with a digital density meter or a hydrometer. Failing to wait for complete homogenization can result in solutions either below target concentration (potentially subjecting specimens to concentrations that do not provide an adequate preservation environment) or above (which can dehydrate specimens and cause them to become brittle). If ethanol is usually removed from a carboy via a spigot at the bottom, measure the ethanol concentrations at both the top and bottom of the container to determine if the solution has finished mixing.
Conclusions
Although fluids have been used to preserve natural history specimens for hundreds of years, much is still unknown about the best practices of fluid preservation and how fluid collections should be managed, and this is especially true for insects. As ethanol-preserved entomology specimens are becoming more frequently used because of recent scientific advances in DNA analysis, it is increasingly important for collections personnel to focus on the preservation and management of their ethanol-preserved insect specimens. This study can inform insect collectors and entomology collections personnel on how the preservation of their specimens is impacted by different storage conditions (e.g., the ratio of specimens to ethanol) and the passage of time, allowing them to update preservation environments and combat factors contributing to specimen decay. Implementing best practices for the preparation and management of ethanol-preserved insects can keep these specimens useful to science for long periods of time after collection, the ultimate goal of museum collections.
Container and lid materials used to store samples in this study. Borosilicate glass vials (left) and phenolic plastic screw caps with either vinyl (inset, left) or polyethylene lining (insert, right), and high-density polyethylene (HDPE) bottles with foam-lined HDPE caps (right).
Illustration of the visually assessed fullness classes with 6-dram vials of carrion beetles (top) and the fullness classes compared to the measured insect volume percentage, including vials from all years (bottom). Box and whisker plots are for median (midline in box) and quartiles with gray dots showing samples within each fullness class.
Fluid concentrations of samples for each year of collection and preservation. Gray line denotes initial and ideal 70% ethanol concentration.
Fluid concentration of storage ethanol by the percentage of insect volume for vials from (A) 2018, and (B) 2010–2013 and 2017. Red lines show significant linear regression line of fit. Gray lines denote initial and ideal 70% concentration.
Ethanol concentration with time. Ethanol concentration declined slightly with time since collection. Red line shows significant linear regression line. Gray line shows initial and ideal 70% concentration.
Ethanol concentration with cap liner material. Ethanol concentration did not differ between cap liner materials (polyethylene, vinyl). Box and whisker plots are for median (midline in box) and quartiles with gray dots showing individual measurements.
Percentage change in ethanol concentration over time by vial size using only those vials with ≤25% insect volume and from all years except 2018. (A) 1 dram, (B) 2 drams, (C) 4 drams, (D) 6 drams. Red line shows linear regression line (significant = solid; nonsignificant = dotted), and gray line indicates the ideal ethanol concentration (70%).
Contributor Notes
Associate Editor.—Mariana di Giacomo