This paper uses in-migration, out-migration, and net migration vectors to measure and portray the migration streams occurring between April 1, 1955, and April 1, 1960, for all persons five years old and over on April 1, 1960, among 38 contiguous state economic areas (SEA's) in Oklahoma, Colorado, Kansas, Missouri, Arkansas, Texas, and New Mexico. Both rectangular and polar coordinates were employed in computing three different types of vector representations. First, migration vectors for each SEA, expressed in total people-miles of in-movement, out-movement, and net movement represent the resultant direction of movement and the total people-miles of movement. Second, in- and out-migration vectors of the total people-miles of movement, with the widths of the vectors drawn proportional to the number of migrants, show the total size of each migration stream for each SEA, the resultant direction of movement, and the distance moved in total people-miles of movement. p ]Third, in- and out-migration vectors of the average distances moved to and from each SEA, with the widths of the vectors drawn proportional to the number of migrants, show the magnitude of movement, the average distance moved per migrant, and the resultant direction of movement. This study shows that migration vectors based upon polar coordinates are more accurate than the migration vectors based upon rectangular coordinates. Also, the in- and out-migration vectors are superior to net migration vectors in representing migratory streams among geographic areas.