Purple finches are often regular visitors to our feeders. The gorgeous males with their deep red breasts and heads can only be confused with male house finches. A male house finch has brown streaks on the flanks with less extensive red and more of a reddish-orange on its head.

Female purple finches are mostly brown and white. The breast has dark streaks. A bold white stripe lies just above the eye; this white supercilium is absent in female house finches.

Have you noticed that male purple finches are usually outnumbered by females at your feeder? Not so fast. First-year males are dead ringers for female purple finches. You really have to have them in hand to tell them apart by examining the wear of the primary coverts and the shape of the tips of the outer tail feathers. Some of those streaked purple finches at your feeder are first-year males.

Purple finches belong to the suite of irruptive finches popularly called northern finches. They breed across the northern tier of the United States from Maine to Minnesota and across the southern tier of the Canadian provinces. A breeding population also occurs west of the Cascades and Sierra Nevada from California to British Columbia.

Strong southerly irruptions occur every other year. These biennial irruptions are thought to be driven by variation in the production of cone crops of the conifers on which purple finches depend.

To gain some insight into these movements of purple finches, my wife and I analyzed all the banding data on purple finches from the Bird Banding Laboratory from 1921 until 2008. More than 745,000 finches were banded over this period and almost 20,000 of those banded birds were subsequently recaptured (or in a few cases found dead).

We were particularly interested in three questions: During irruptions, do birds from one area, like New England, move straight south, or do they spread across the continent? (Purple finches can be found throughout North America.) Do purple finches show fidelity to breeding sites? And do purple finches show fidelity to wintering sites?

The analysis of banding data presents many challenges. Banding effort is never constant. Most of the banding records come from 1960-85. Banding effort varies greatly among states and provinces. Nevertheless, some general patterns can be discerned.

Birds banded in Vermont, New Hampshire and Maine were re-encountered broadly but mostly in the eastern United States, curling eastward through the Gulf Coast states. A few reached Texas, and there were modest numbers in Michigan, Ontario, Wisconsin and Minnesota. A similar pattern emerged for birds banded in New York.

We also analyzed birds first banded in Pennsylvania, New Jersey and North Carolina. These areas are south of the breeding range and were banded in the winter. Re-encounters of these birds occurred mostly in New England and the eastern provinces. The re-encounters were consistent with data from birds originally banded in New England.

Moving to to the Midwest, birds originally banded in Michigan, Wisconsin and Minnesota also were re-encountered broadly but most irruptions were due south.

Only 275 purple finches banded in the Pacific states or British Columbia were re-encountered. But a consistent pattern was that those birds migrate west of the Cascades and Sierra Nevada.

We did find some evidence of breeding site fidelity across as many as five years. Some wintering site fidelity was evident as well over one to six years. Uneven banding efforts prevent us from knowing how prevalent such fidelity is.

Here’s the most impressive distance between captures. A purple finch banded in Maine in 1966 was caught two years later in Texas, a distance of 1,792 miles.

For a PDF copy of our paper, visit http://bit.ly/1BntjMb.

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

whwilson@colby.edu