Beth Dumont, a fourth-year UW-Madison graduate student studying evolution and genetics, spends much of her time admiring the intricate internal architecture of mouse cells through a microscope.

But last week, Dumont found herself most mornings at 7 dealing with entire, whiskery mice, gingerly sticking her hand into each of 60 metal box traps set up on UW-Madison’s Biocore Prairie to see if a hungry rodent had taken her peanut butter or oatmeal bait overnight.

It was still science, just a little earthier and a bit more basic.

Usually, Dumont studies recombinant DNA, basically a process in which the shuffling of genetic material in chromosomes creates new genes but also can cause disease if it goes awry.

Mice are essential for her study. And if Dumont’s latest theory is correct, wild mice may become especially important, not only in her study but also in other areas of research.

The sequencing of human and mouse genomes has shown that 99 percent of the genes in mice have human counterparts and that only about 150 mouse genes don’t have similar versions in people. That makes mice a good choice for many lab experiments.

“What is going on in mice,” Dumont said, “is bound to inform us about what is happening in humans.”

Working in the lab of UW-Madison assistant professor Bret Payseur, who studies mouse genetics, Dumont noticed something interesting. Lab mice that are descended more closely from wild mice had higher rates of DNA recombination than classic lab mice, which have been bred for science for many generations.

That made her wonder: Would using truly wild mice for lab study give researchers more accurate information about nature’s workings than the traditional inbred lab mice raised in big mouse factories and shipped to labs by commercial vendors?

Clearly, Dumont said, lab mice are nothing at all like truly wild mice. Would those behavioral differences extend to the physiology of the mice?

“The behavioral differences are unbelievable,” Dumont said. “The lab mice are slow. And some will come right up to your hand and beg you to pick them up. ... The lab mice aren’t real, essentially.”

Payseur, who is as excited to see the results as Dumont, said the project is important, precisely because of the differences between wild mice and lab mice and the potentially different findings that might result from using one or the other.

“Lab mice are artificially selected to live in cages and harbor diseases,” Payseur said. “The question — given that humans don’t live in cages and aren’t artificially selected — is how relevant are the findings about things like genes that cause disease?”

All of that led Dumont and several helpers to try to catch mice at UW-Madison’s Biocore Prairie on the shores of Lake Mendota near the Eagle Heights vegetable garden. Eight years of studying genetics and evolution doesn’t really prepare you for capturing wild mice, as Dumont discovered on her first morning Tuesday when a cagey deer mouse slipped from her hand and made a clean getaway.

She hoped to collect at least 10 mice from which she can remove tissues that will allow her to study the rates of recombination and check the rates of wild mice against those in the classical mice back in her lab. By Friday, her expectations had been exceeded. Dumont captured 50 mice (and one very strong toad). She was able to keep eight mice and three meadow voles for her study.

Dumont suspects that she’ll find a higher rate of recombination in the wild rodents. If so, Dumont and Payseur said, it could lead to similar studies and, eventually, more research with wild mice and findings that are more reflective of what is truly happening in nature.

This is how science gets done. While it too often seems that science these days is a press conference and the announcement of a big discovery, it more often involves researchers such as Dumont rising at dawn, swishing through wet prairie grass and peering into a metal box to see if a mouse is peering back.