According to Mendel's second law, unlinked genomic loci segregate independently during meiosis, and this directly applies to non-homologous chromosomes. An exception to this rule is the phenomenon of quasi-linkage (QL) that describes the non-random segregation of non-homologous chromosomes. Molecular mechanisms of QL are not clear, and so far QL has been explained by the 'affinity hypothesis', that is the tendency of non-homologous chromosomes, through the mutual attraction of their centromeres, to migrate together. QL was initially observed in mice and plants but has not been investigated in humans. We were interested in determining if QL served as a significant confounding factor, that is quasi-linked loci falsely exhibiting co-segregation with disease, in genome scans for complex diseases. We investigated individual markers showing linkage to schizophrenia, asthma, multiple sclerosis, inflammatory bowel disease and type 1 diabetes. 'Disease linked' markers were tested for QL in a pairwise linkage analysis against all other markers exhibiting evidence for linkage in each specific disease. Marshfield genotyping dataset of eight CEPH families was used for this purpose. A wide range of QL maximal lod scores were observed. The highest lod score of 2.28 (theta F= 0.4; theta M= 0.32) was generated for two markers on chromosomes 15p13-q12 and 9q32-q34 which initially demonstrated evidence for linkage to schizophrenia (Kaufmann et al. 1998). A large number of QL lod scores varied between 1 and 2, and although did not reach the level of statistically significant linkage, are within the range of the overwhelming majority of "lukewarm" loci of complex diseases. The above data suggest that QL may be an important confounding factor in the genetics of complex traits, and if so should be taken into account when oligogenic and polygenic traits are investigated. Genome-wide QL maps of reference families would aid in accounting for such confounding effect.