It’s thrilling to see an unusual bird. Who is not excited about the chance to see a snowy owl or an American oystercatcher in Maine? Yet there is the potential to experience wonder in watching our common birds.

My case in point today is the American crow. Here in Waterville, we are treated to an amazing spectacle late each afternoon. Hundreds of crows fly in from all directions to converge on a patch of forest behind the shopping center at Elm City Plaza in the fading daylight. The river of crows seems endless. No doubt, you have seen such behavior in your neck of the woods as well.

Why do the crows congregate to spend the night together? We don’t have a definitive answer, but some intriguing suggestions have been proposed.

Aside from humans with guns, the major threat to American crows is predation by great horned owls. Perhaps you have seen American crows mob a roosting great horned owl during the day. The crows are brutal, swooping down on the owl and pecking it. Usually the owl is forced to fly to find a more peaceful place to sleep.

Roosting might therefore offer protection from owl predation. It is awfully hard for an owl to approach when 1,000 or more crows are keeping a vigilant watch.

Another possible explanation comes from a somewhat controversial idea called the information-exchange hypothesis. The premise is that crows can learn about good food sources from each other. We have to be careful to avoid ascribing a human perspective to this information sharing. In the natural world, behavior that looks altruistic turns out to be selfish. In nature, individuals that watch out for themselves and their kin fare better than altruists. How does helping unrelated individuals survive increase the chance of getting your genes into future generations?

Honeybees, through their remarkable waggle dance, let other members of a hive know where good nectar sources are. But all individuals in the colony are related to each other, so sharing information indirectly benefits all colony members.

A roost of crows consists mostly of unrelated individuals. Crows are not so kind-hearted as to somehow share the location of a food bounty with unrelated individuals. However, it may be possible for crows to discover the whereabouts of good food. Crows may be able to assess the nutritional status of a crow as it comes back to the roost. Perhaps it is full of energy. If so, a crow in need of a good meal may follow the well-fed crow in the morning. This behavior can be seen as a type of parasitism. The definitive study to test for this type of information sharing in crows has yet to be done.

One other explanation for crow roosts is the patch-sitting hypothesis. This explanation entails roosting near a site where there is a reliable source of food. The food does not need to be the most nutritious, but will provide roosting crows with a breakfast to get them going in the morning in search of more substantial food and a snack in the evening before they go to sleep. Like the information-center hypothesis, this hypothesis needs more study as well.

Crow roosts are seasonal, occurring in the fall and winter. Some of these aggregations must be truly spectacular. I know of one report from Oklahoma of 2 million crows in a single roost! Roosts from 100 birds to tens of thousands are much more common.

We are seeing a trend of American crow roosts occurring in urban environments rather than in more rural or undeveloped habitats. Some have speculated the crows are taking advantage of the slightly warmer temperatures found in cities. Just a few degrees of warmth can make a big difference. Plus, the lights of a city make it easier to see great horned owls at night.

Herb Wilson teaches ornithology and other biology courses at Colby College. He welcomes reader comments and questions at whwilson@colby.edu