Using Small Mammal Remains for Environmental Archaeology

Credit: Bresson Thomas

Archaeological remains of small mammals generally weighing under 1kg, or micromammals, are important as environmental indicators, partly because they tend to specialize in certain habitats and are sensitive to change. Many factors affect their ranges of distribution, including predators, food requirements, competition, fire, shifts in precipitation patterns, and shelter availability. Micromammals such as voles and mice also tend to live in dense populations and have evolved rapidly through high fecundity. Due to these diverse and interrelated factors, the interpretation of micromammal remains—bones and middens, mostly—requires a deep understanding of the rodents’ relationship with its environment. In other words, ecological information is imperative to accurate assessment of archaeological data on micromammals.

But sometimes micromammal remains have answered modern ecological questions. For example, packrat middens in arid North America offer relatively high temporal, spatial, and taxonomic resolution (i.e., small intervals with which to measure time, space, or species range), and contain what is possibly the “richest archive of dated, identified, and well-preserved plant and animal remains in the world” (Pearson & Betancourt 2002, p500). These packrat middens showed discrete temporal and spatial displacements of species, which helped solve the debate between Frederick Clements’ climax theory and Henry Gleason’s multiple-endpoints theory, a rigorously contested scientific argument of “superorganismal” versus individualistic ecological communities starting in the late 1920s. As Pearson and Betancourt point out, the rodent middens’ paleoenvironmental record of the last few thousand years in arid environments is valuable because it shows short-term climatic oscillations and long-term ecological changes.

Another type of midden, found in southern Africa and created by the rock hyrax Procavia capensis, also contains paleoenvironmental information, but through a different medium. Instead of macrobotanical remains that make up a packrat’s nest, the hyrax’s midden is a latrine; excrement and urine create “urino-fecal deposits” in a series of fine laminations sheltered in caves. The accumulation of dried urine seals microbotanical remains and isotope records within the midden, protecting them for as long as 35,000 years. These laminations, which form a sort of stratigraphy, contain stable carbon and nitrogen isotope variations that reflect the proportion of C3 and C4plants the hyraxes consumed. Since hyraxes eat as many as 79 species of grass, shrubs, and trees, but have particular preferences, it is possible to make connections between hyrax diet and the seasonality of southern Africa’s rainfall throughout the Holocene.

Dr. Chase sampling from a hyrax midden

How can the hyrax’s diet be deciphered from something like their urine? The information lies within the C3 and C4 plants mentioned above. C3 plants (generally trees and shrubs, in this case) fix carbon differently than C4 plants (which in this environment generally means grass): they produce a three- and four-carbon molecule, respectively, so the levels of carbon isotope-13 vary according to the fixation mechanism. The two types of carbon fixation also relate directly to the type of environments the plants are adapted to. C­4 fixation allows plants to photosynthesize without losing as much water as C3 fixation, so C4 plants can be more drought-resistant and survive higher temperatures than C3 plants, but C3 plants have more extensive root systems that access groundwater more easily. Thus, when rains become less regular, one can expect hyrax diet to change (both out of preference or out of necessity from altered resource availability), which then leaves a different isotopic range in urino-fecal deposits. Chase et al., who describe the great potential of hyrax middens for long-term variations in stable isotope analysis, point out that these excremental laminations can be high-resolution archives of environmental change in a region that “has hitherto relied on discontinuous and spatially disjunct geomorphological proxies” (2009, p706).

But these middens are collections of plant and animal material made by the rodents; the skeletons of the rodents themselves can also help our understanding of past and present environments. After all, rodents are generally ubiquitous and sensitive to changes in their local environment. For example, the English field mouse’s highest density is seen in scrubs and woodlands, whereas the field vole’s is in grasslands; finding large samples of one of these species reflects what sort of ecosystem may have been present in the past. Despite their ubiquity and density, however, micromammal remains are fragile, small, and easily altered (e.g. broken or moved).

Stahl, in his description and analysis of the recovery and interpretation of small bone assemblages, states that data can be skewed based on recovery technique, partly due to the small size of some bones. Species are often difficult to identify with a partial skeleton because zoological characteristics, such as soft tissue anatomy, can be irrelevant to archaeologists, and in many cases specimens are left “indeterminate.” One of the most essential factors to take into account when attempting to explore specimens’ significance is their manner of accumulation and modification. Stahl warns, “The effects of postmortem processes are crucial for archaeological interpretation, as they can further distort, subtract from, or completely obliterate the interpretive value of microvertebrate bone assemblages” (1996, 56). These are risks that other specimens, such as large mammal remains, also face, but the size and number of small mammal samples affects the potential severity of misinterpretations.

Interpretations can be useful to compare archaeological taxa (i.e. taxonomic groups, such as species or family) to contemporary taxa. For example, a current class of mouse might be analogous to one in the past and thereby show certain preferences in habitat according to ecological conditions; Stahl recommends using “suites” of taxa where possible, to improve correlations between multidimensional variables.

kangaroo ratA recent application of this sort of method is described by Lyman and Cannon, who analogized a modern suite of rodents in Utah with a large sample of Late Holocene remains collected from Camels Back Cave, and used archaeological data to create a pre-Columbian baseline for the environment in Dugway Valley, Utah. Working with thousands of partially digested small mammal remains in carnivore scat and egested raptor pellets, Lyman and Cannon have identified at least 16 species with typical low desert characteristics, but with differences in microhabitat preference. Lyman and Cannon used precipitation data and surveyed the contemporary small mammal population between 1954 and 1998, and in doing so were able to postulate relationships between rainfall, vegetation, and rodent population growth. For example, knowing that Heteromyids—members of a family that includes the little pocket mouse and kangaroo rat—rely almost exclusively on seeds of native grasses, they could predict that high rainfall would lead to more grass seeds, and therefore higher numbers of Heteromyidae. However, the expansion of invasive, nonnative grasses such as Bromus tectorum caused significant regional vegetation change, so that periods of high rain meant denser ground cover by B. tectorum. Heteromyids characteristically avoid thick vegetation, preferring open habitats, but Murids—members of the Muridae family, such as deer mouse and the desert woodrat—generally inhabit thick, exotic annual grass cover. The arrival of B. tectorum, therefore, in part catalyzed a shift from more Heteromyids to more Murids over time.

It is important to emphasize the “in part” of the previous sentence, because small mammals are very vulnerable to environmental changes; higher predation, less shelter, altered microclimate can all have varying degrees of effect on a rodent population, which means that a discerning understanding of the intricacies in an ecological web is very necessary for accurate interpretation of archaeological data. Returning to the Dugway Valley example, it is important to note that some areas of Dugway are military testing zones, and that other changes, such as fluctuations in numbers of owls or foxes, might also be responsible for any alterations in small mammal diversity or density. Environmental archaeology, especially when using micromammal samples, requires strong supplemental knowledge of ecosystems and communities. Given many of the limits of finding and categorizing small mammal bones and the dangers in relying on assumptions, it is best to use a holistic, multi-disciplinary approach in studying environmental archaeology; relying only on micromammal remains will not provide a full understanding of an area’s past ecology.


Further Reading:

Chase, B. M., Thomas, D. S. G., Meadows, M. E., Scott, L., Marais, E., Sealy, J., & Reimer, P. J. (September 08, 2009). A record of rapid Holocene climate change preserved in hyrax middens from southwestern Africa. Geology, 37, 8, 703-706.

Luff, Rosemary-Margaret. 1984. Animal remains in archaeology. Aylesbury, Bucks, UK: Shire Publications.

Lyman, R. Lee, and Kenneth P. Cannon. 2004. Zooarchaeology and conservation biology. Salt Lake City: University of Utah Press.

Pearson, S., & Betancourt, J. (January 01, 2002). Understanding arid environments using fossil rodent middens. Journal of Arid Environments, 50, 3, 499-511.

Rackham, D. James. 1994. Animal bones. Berkeley: University of California Press.

Stahl, P. W. (January 01, 1996). The Recovery and Interpretation of Microvertebrate Bone Assemblages from Archaeological Contexts. Journal of Archaeological Method and Theory, 3, 1, 31-74.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s