The global mean temperature has risen 0.6°C in the last 100 years and is projected to increase between 1.4°C and 5.8°C over the next 100 years . This rate of increase (in temperature per century) is greater than any measured in the climate record of at least the last 100,000 years [2-4]. Several types of response to this rapid climatic change have been proposed for vertebrates, including range shifts to track resource changes, local adaptation to new resources, local extinctions, and developmental plasticity [5,6]. Although these changes are difficult to observe on a seasonal or annual time scale, empirical evidence from long-term field studies and historical records exists for range shifts and phenological shifts, especially earlier breeding events (see [4,5,7,8]). Clearly, long-term monitoring and historic data sets are critical to identification of meaningful biological responses to climatic change [9,10]. However, such information is scarce, with quality ecological records confined to only the last 100 years, and often far less . One unique way to address these problems is by using biologically relevant time series data provided by the fossil record. An ideal fossil record would span the last few hundred to few thousand years in order to detect changes over tens to hundreds of generations. Such a record would provide a temporal scale long enough to identify significant trends and enough resolution to reveal ecologically relevant population responses. Ultimately, this type of data will provide information fundamental for predictions of response to climatic change over the next century.
Amphibians are often considered sensitive indicators of environmental conditions [12-16]. Many amphibian species have a naturally complex life cycle that involves both aquatic and terrestrial environments, directly exposing individuals to changes in environmental temperature and moisture over a single lifetime. Amphibian skin and eggs are highly permeable, which makes them sensitive to small changes in chemicals in the environment . Furthermore, they are ectothermic, enabling differences in temperature to have direct effects on amphibian body temperature and metabolic processes [13,18]. Thus, projected global climatic change is expected to have significant effects on amphibian communities [18,19], but these effects have not yet been specified. Our study system provides the data necessary to initiate such predictions.
Lamar Cave is an exceptionally rich paleontological site located in Yellowstone National Park (YNP), Wyoming, USA . The fossils are primarily from a wood rat midden collection of raptor pellets and carnivore scats. The excavated fossil collection contains tens of thousands of bones encompassing mammals, amphibians, birds, reptiles, and fish, and spans the last 3300 years . Recent investigations of the small mammal record of Lamar Cave show morphological and genetic responses that correlate well with past climatic events in the area [22,23]. Lamar Cave also contains thousands of amphibian bones.
In particular, the most abundant amphibian in Lamar Cave, the tiger salamander (Ambystoma tigrinum) is an excellent species with which to study population level responses to climatic change. Many studies show that Ambystoma demonstrates developmental growth plasticity in response to environmental conditions (e.g. [24-27]): fossils record this as changes in mean body size. In addition, Ambystoma is able to exploit alternate life histories in response to different environmental conditions [26,28-31]. Ambystoma may undergo an aquatic larval stage and then metamorphose into terrestrial adults or alternately, be paedomorphic, i.e., achieve sexual maturity while remaining aquatic and retaining larval morphological features. Ambystoma tigrinum is unusual in that it retains the ability to metamorphose facultatively even after years in the aquatic form . The different life histories are detectable in diagnostic characters preserved in the fossil record [32,33] of populations represented in Lamar Cave. Thus, the cave records the potential effects of climatic change on several life history characteristics of A. tigrinum, and is documented here for the first time at this important temporal scale.
Paedomorphosis in modern A. tigrinum depends on various environmental stimuli [24,34]. Several biotic factors influence paedomorphosis and metamorphosis including thyroid hormone levels, larval density, and competition with fish [24,28,30,31,35,36]. Although important, these biotic factors are often overwhelmed by the influence of temperature and moisture. Populations in cold environments show high frequency of paedomorphosis  and are often large in size . Populations found in ponds of intermediate temperature often contain both paedomorphs and metamorphosing individuals. Likelihood of paedomorphosis is also closely linked to pond permanence, or perceived pond permanence [24,28,29,37,38]. Pond permanence is directly related to available moisture in the environment in the form of humidity, precipitation, or ground water.
Given the environmental forces that drive body size and metamorphic changes in A. tigrinum, we used a late-Holocene fossil record to track these traits through the last 3000 years. We analyzed trends within the context of known climatic change, and attempted to distinguish patterns of response that represent specific climatic periods through this time.