The first horses in North America would not have been able to hold their own in the Triple Crown. At just about 5.6 kilograms the Sifrhippus sandrae hoofed onto the scene some 56 million years ago about the size of a small dog.

But then a funny thing happened. In the next 130,000 years during the Paleocene–Eocene Thermal Maximum, these small equines got even smaller, reaching the tiny size of 3.9 kilograms—some 30 percent lighter than their initial heft. Just 45,000 years later, however, the genus had bulked up to seven kilograms. And the horses were not the only ones. Many other mammals in the area followed the same pattern.

These animals' sizes likely resulted from relatively rapid climate change, suggest the authors of a new study published online Thursday in Science.

The study "highlights the importance of temperature on evolution—particularly mammal evolution," says Felisa Smith, a professor of biology at the University of New Mexico in Albuquerque, who wrote an essay on the findings in the same issue of Science. And it adds a new high-resolution tracking of body size and temperature during a crucial—and long puzzling—time in geologic history.

View a slide show of the visual history of ancient miniature horses

Looking the small horse in the mouth
The researchers did not have complete skeletons to measure for all of the animals, so to track the size of the horses over time they looked at their teeth—in particular, their molars. "It turns out that teeth are much better than femurs," Smith says. A leg bone "does tell you something about size, but teeth are much better." And as far as teeth go, she says, "the best thing to know is the area of the first molar."

The teeth came from a fossil-rich area called Cabin Fork in Wyoming and are part of a substantial collection at the University of Florida built in part by study co-author Jonathan Bloch, an associate curator of vertebrate paleontology there. From the collection, the research team could estimate the size of about 44 diminutive adult horses.

Some 40 percent of other mammals in the area seem to have experienced similar shrinking and subsequent growth, notes co-author Ross Secord, a vertebrate paleontologist at the University of Nebraska–Lincoln. They stuck with the small horses, however, because they had much more solid records from which to accurately date the samples.

The researchers used oxygen isotopes left by freshwater in the fossils to track mean annual temperature from when the animals had been alive. In particular, they sampled the isotopes from teeth of a large, water-dwelling mammal Coryphodon. With these isotope readings, "you get a little, tiny window as to what the temperature was at that time," Smith says.

This close reading has excited Smith and others who have been tracking animal size over the ages. "Although we knew that temperature might set a maximum for body size," Smith says, the new findings actually present a mechanism—and do so in a very detailed manner, showing "how animals responded to a particular temperature at a particular place at a particular time."

Backing Bergmann's rule
The concept that ambient average temperature likely influences body size is not new. Naturalists have long observed this trend geographically, but as Smith notes, Secord and his colleagues present a strong case for the correlation to occur over deep archeological time.

And the mechanisms behind this theory, known as Bergmann's rule, have been fiercely debated since the mid-19th century, when it was introduced.

One argument posits that temperature affects body size for the ease of keeping cool—or of staying warm. As the overall volume of an object increases, the relative amount of surface area decreases. This relationship is handy if you live in high latitudes and are a mammal that needs to retain as much warmth as possible. But if you live in the tropics and are trying to avoid overheating, it should be better to have a smaller body size, which would give relatively more surface area through which to shed heat.

But this direct temperature correlation might not be the only force at work in the case of the mini horses. Previous studies have suggested that temperature and, more specifically, atmospheric carbon dioxide levels influence body size more via an indirect impact on food availability and nutritional content.

But the sustained shrinkage of these horses over tens of thousands of years suggests a deeper genetic change that held fast over generations. "We can't say it didn't have an effect," Secord says of the nutritional changes. But, he notes, "we saw some fluctuations from wet to dry to wet to dry in these intervals, and the body sizes of these animals aren't changing" in parallel. Instead, the animals' sizes followed the single up and down of the average temperatures.

A smaller, hotter future?
The new findings hold implications for digging deeper into the past—as well as looking into our own warmer future. Smith suggests using the data to learn more about the other organisms in Sifrhippus's world to see if they were largely following the same pattern. "What about the predators?" she asks. "Were there some lineages that responded in another way? I think that would be phenomenally interesting."

Before we can understand what past climate change meant for more animals—"there needs to be a lot more work on modern animals," Secord says. Ancient animals, however, might give us an insight into how modern animals might fare with our predicted climate change.

Although the era Secord and his colleagues studied experienced a similar increase in temperatures (five degrees Celsius or more) as is predicted for us for the near future (four degrees C), he points out the ancient animals had tens of thousands of years to adapt to changing temperatures—rather than just centuries.

"The question is now, over the next century or two, are we going to see a shift in body size?" Secord asks. "Are they going to be able to adjust quickly enough?" He hopes that many species will be able to keep pace, especially those with shorter generations. Many bird species have already been getting smaller over the past few decades.

And if animals do undergo size changes with future climate change, as Secord points out, we are not going to be seeing smaller race horses—unless we breed them that way. "This is certainly something that is going to be restricted to wild animals," he says. "Anything that has a way of artificially regulating temperature or diet is going to take it out of the loop." That would certainly apply to jockeys and the rest of us humans, too.

View a slide show of the visual history of ancient miniature horses