The genetic qualities of laboratory colonies of phlebotomine sand flies have not been compared with field specimens despite 1) probable genetic shifts due to the colonization process and 2) the problems associated with the extrapolation of experimental data derived from colonized organisms to field populations. The present study compared the genetic profiles of five laboratory colonies of geographic strains of the New World sand fly Lutzomyia longipalpis, and contrasted them with field populations. The profiles were based on the variability exhibited with polyacrylamide gels at 14 enzyme loci. A general pattern of a loss of infrequent alleles and decreased heterozygosity emerged as an apparent consequence of colonization. The average number of alleles per locus ranged from 1.2 to 1.6, and the average heterozygosity ranged from 4% to 11%. The field collection from Lapinha Caves (near Belo Horizonte, Brazil) averaged 2.1 alleles with a heterozygosity of 16%. In contrast, the LAPINHA laboratory colony established from that site 24 years earlier showed very low values (1.2 alleles/locus and 4% heterozygosity) and fixation for alleles not present or rare in the field collection from the same site. The genetic differences between the other Brazilian colonies and the Lapinha Caves field samples were due to presence of both different alleles and highly diverged allelic frequencies. Biological inferences based on colonized sand flies must be tempered by recognizing that the colony may represent a highly skewed genetic subsample of the L. longipalpis field genome.