Warning: This Car May Cause Natural Selection

On roadkill and natural selection

Posted Mar 14, 2014

I can just see it now. A Ford Motor Company press release that reads “All Ford cars will now have the following printed on their dashboards: Warning: This Car May Cause Natural Selection.” Or at least that’s what Ford would do if they hired a few evolutionary biologists as consultants. Until that happens, here is the scoop.

In the United States, eighty million birds die each year as a result of road kill. Kill that many creatures, and you create new natural selection pressure. Or so evolutionary biologists like Charles and Mary Brown would argue. And they should know, as they study colonial cliff swallows (Petrochelidon pyrrhonota), which often form colonies near highways, making them an ideal species to examine whether natural selection has favored certain traits in response to mortality caused by road kill.

Each year since 1982, the Browns have studied many cliff swallow groups in Nebraska. As they travelled between colonies over the years, they also collected data on road kills involving swallows. What they’ve found is a significant decrease in the number of swallow road kills over the decades.  Why? Has selection favored certain traits that reduce road kill in swallows? Or is there, perhaps, a simpler explanation? The Browns wanted to know and they began a systematic analysis of possible explanations. 

They first checked the population size of the cliff swallow colonies they were studying. After all, if population size was decreasing, then a decrease in road kill would be expected, but this need not have anything to do with natural selection related to roadkill; in fact, the data show that cliff swallow population size has increased since the Brown’s studies began. Next they rechecked the routes they had taken each year, and, for the most part they were the same year after year: the drop in roadkill wasn’t due to their sampling different routes. And traffic on these routes increased during the course of their study, so the decrease in roadkill wasn’t just the result of fewer cars on the road for swallows to crash into.

The Browns also examined the possibility that roadkills decreased because of an increase in scavengers who removed the dead swallows before Brown and Brown could find them. The data don’t support this hypothesis because, at least for avian scavengers, scavenger population sizes stayed constant over the course of the Browns’ study. As each of these alternatives were ruled out, the possibility that the decrease in roadkill was the result of natural selection favoring some swallow trait that reduced mortality increased. But what trait?

The first clue the Browns had for answering this question came from the fact that the wing length of birds that died in road kills was significantly longer than the wing length in populations they censused. A more detailed analysis found that the average wing length of swallows in the populations at large had decreased over the course of their three-decade study, while the average wing length of swallows that died in road kills had increased over the same period. Wing length matters because long wing length reduces the “vertical take off” ability in birds: that is, the ability to get into the air quickly. Since swallows often sit on the road (eating food), reduced vertical take off will lead to increased collisions with cars.

Natural selection favors shorter wings then because it allows swallows the ability to survive with higher probability in the face of potential road kill.

Still, I’m not holding my breath for that call from Ford Motors, asking “About that consultant job…”

Brown, C.R. and M. Brown. 2013. Where has all the road kill gone? Current Biology 23, R223-R224.