Science provides us with one of the most powerful ways of understanding our world. The scientific method begins with observation of a pattern and then the creation of one or more hypotheses to explain the pattern. The next step is the collection of either experimental data or observations to try to disprove the hypotheses.

If one hypothesis is not disproved, we accept it as tentatively true. A careful scientist will test the hypothesis with different experiments or observation. If the hypothesis still stands, she feels more confident of the hypothesis.

It's easy to create hypotheses but it takes time and effort to test them. An untested hypothesis has no scientific validity. Nevertheless, a number of untested hypotheses to explain aspects of nature are treated as being true.

A recently published paper in the journal "Behavioural Ecology" tested an old explanation of why many butterflies and moths have circular patterns on their wings that look like the eyes of a vertebrate. More than 150 years ago, the suggestion was made that these eyespots mimic the eyes of the predators of the enemies of the butterflies and hence offer the butterflies some protection.

That explanation is quite plausible and in fact is a commonly cited example in animal behavior texts. But until now, it had never been tested. The researchers from Cambridge University crafted artificial moths out of paper and drew different kinds of markings on the models (bars, squares, eyespots, no markings). A small mealworm was attached to each moth model to tempt Blue Tits, Great Tits and House Sparrows.

The researchers found that artificial moths with eyespots were attacked as frequently as moths with bars or squares. The scientists found that more boldly marked moths, regardless of the shape of the markings, were avoided to a greater degree than lightly marked moths.

Thus, the old explanation has been tested and found wanting. Predators avoid moths and butterflies with eyespots because they are conspicuous, not because they mimic a predator's eyes.

Let's continue this thread with some of my own research. During the past decade, the concept of short-stopping has been applied to irruptive birds like northern finches and Red-breasted Nuthatches. The claim is that birds like Evening Grosbeaks and Purple Finches do not migrate as far south as they did 30 years ago because of the great increase in bird feeding during those 30 years.

Thus, the hypothesis goes, Purple Finches used to need to migrate to North Carolina to find food in the winter but now can take advantage of the many sunflower seeds that New Englanders put out for the birds.

Like the eyespot hypothesis above, this short-stopping hypothesis seems like a reasonable one but was offered without a scientific effort to disprove it.

In 1999, I published a paper that tested this hypothesis. Specifically, I examined the irruptions of Common Redpolls, Purple Finches, Evening Grosbeaks, Pine Siskins, White-winged Crossbills and Pine Grosbeaks from eastern North America.

I also examined Downy Woodpeckers, two species that do not migrate, as controls for my methodology.

I divided the period from 1970 to 1990 into five-year intervals. Using Christmas Bird Count data, I found that irruptions of Evening Grosbeaks and Purple Finches extended further south in the earlier time periods. However, if irruptive birds are staying further north in more recent years, one would expect a corresponding increase in these irruptive birds in northern areas.

I did not find such a pattern of increase and therefore my analysis failed to support the short-stopping hypothesis.

Herb Wilson teaches ornithology and other biology courses at Colby College. He welcomes reader comments and questions at whwilson@colby.edu. Previous columns and other information on Maine birding can be found at his blog: http://www.mainebirds.blogspot.com.