Making Tracks

Carpenter occasionally travels to consult map collections at the New York Public Library and the University of Connecticut. He has to be alert for dramatic, modern changes to the landscape: Many lakes that exist now, for instance, are the result of rivers dammed since 1946. At some point, Carpenter decided he wanted to show the direction of flow of every river and creek, with tiny blue arrows as the creeks leave the frame of the map. "Sometimes that's the hardest thing of all to track down, which way that stream was going," he says.

Carpenter is an amateur mapmaker, but he is not an amateur geographer. He spent his career as a regional planner (he hand-drew his maps there, too), so he's been thinking about the landscape, and the impact of people and development on the landscape, for 50 years. He started the research and prep maps for what has become the atlas 10 years before he retired. In the late 1990s, Carpenter sent a few of his maps to an old friend, a professor at Rutgers, who recommended that Carpenter get in touch with George Thompson, head of the Center for American Places, which sometimes teams up with Johns Hopkins to publish books about the American landscape. Thompson connected Carpenter to Johns Hopkins, which publishes the book in cooperation with his center.

Carpenter's passion for railroads, and for documenting them, goes back to college, and before. In some ways, the atlas is the work that, at 70, he's been preparing for his whole life.

One of the odder indexes in Carpenter's atlas is also one that reveals how much more than a set of maps it is. It is the "Index of Track Pans." In the 1940s, and even earlier, some rail routes were so intensely competitive that railroad companies couldn't afford to waste a minute. The problem: On long routes, steam locomotives needed to be resupplied with water.

So some rail lines were equipped with track pans. For hundreds of feet between the rails there was an open trough, perhaps 6 inches deep, filled with water. As a steam locomotive and tender roared over the pan at 60 or 70 miles an hour, a fireman could lower a scoop and refill the tender with water at full speed.

"These were concentrated on two lines competing for 16-hour travel time between New York and Chicago," says Carpenter. "You just couldn't stop and take on water." In the context of 1946, the track pans are as vivid a cultural benchmark as instant messaging is today. They are a reminder that we do not live in the first age of urgency or ingenuity.

Carpenter expected to turn in the finished maps for volume two (New York and New England) in January. He planned to pack them in a box, put them in the backseat of his car, and drive them down to Baltimore, to deliver them to Johns Hopkins University Press in person, as he did with the maps for the first volume.

Carpenter keeps his prep maps in blue three-ring binders on the shelves in his study. He appreciates not just his own skill but the remarkable fact that an academic press is publishing his work--and doing so in such fine fashion. He realizes that the maps in the blue binders could easily have stayed on the shelves of his study.

"A lot of people have something to say and never have the chance to say it," says Carpenter. "I'm lucky. And I like to express myself with maps."

Charles Fishman (cnfish@mindspring.com) is a Fast Company senior writer.