A brief history of the North American Breeding Bird Survey (BBS) and a discussion of the technique are presented. The approximately 2000 random roadside routes conducted yearly during the breeding season throughout North America produce an enormous bank of data on distribution and abundance of breeding birds with great potential use. Data on about one million total birds of 500 species per year are on computer tape to facilitate accessibility and are available to any serious investigator. The BBS includes the advantages of wide geographic coverage, sampling of most habitat types, standardization of data collection, and a relatively simple format. The Survey is limited by placement of roads (e.g., marshes and rugged mountainous areas are not well sampled), traffic noise interference in some cases and preference of some bird species for roadside habitats. These and other problems and biases of the BBS are discussed. The uniformity of the technique allows for detecting changes in populations and for creation of maps of relative abundance. Examples of each are presented.