Anyone who has ever seen a horse caught in a barbed wire fence knows the terrible damage such an accident can create. Cuts requiring hundreds of stitches are not out of the question, and, in some cases, the animal dies.

Thus, streamside fences designed to protect fish habitat in northeastern Oregon are a mixed blessing.

Yes, cattle are prevented from congregating on streams, but in the process wild horses, deer and elk are killed or injured by the fences.

More than 600 miles of barbed-wire fencing has been added streamside in the past seven years, according to a story in The Oregonian.

It has successfully kept cattle at bay, but in addition to injuring animals it has made it easier for such plants as Russian knapweed and thistles — both damaging to rangeland — to move in.

At least some research suggests that, in fact, the fences may not be needed.

An Oregon State University study that is tracking the impact of wolves on cattle suggests that the fences are not needed and at some point should come down. That conclusion is disputed by some fish biologists, however.

Either way, the suggestion makes one thing clear. The science on rangeland management, like the science on so many other things, is evolving as research continues. Consider the use of antibiotics: 50 years ago they were given liberally; today, as it becomes clear that bugs are growing resistant to the drugs, they’re used far more carefully.

Knowing that, it may be time, perhaps not this year but soon, to reconsider the roles cattle and fences play in helping and hurting everything from streams to grasslands.

If further study makes clear that the fences do more harm than good, they should come down.

If it shows that more fencing, not less, is needed, that should be considered as well.

Either way, it should be current science, not old prejudices or studies that have since been proven wrong, that guides rangeland management.