Genetic mapping provides a means of understanding the molecular basis of inherited diseases whose biochemistry is unknown. Adequate pedigrees, informative genetic markers, and accurate identification of the disease phenotype are necessary. For dominant inheritance, mapping studies can be done in a single large pedigree; the larger the number of affected individuals sampled the better the estimate of recombination between the gene causing the disease and one or more nearby genetic markers. For recessive inheritance, nuclear families with more than one affected sibling provide the best information. The development of many polymorphic DNA markers on the human genome has contributed to the success of mapping unknown genes because, as the genome is now densely covered with markers, the probability is good that at least one marker will be linked to the disease locus in a family that is segregating a disease allele. Most genetic markers now in use depend upon restriction fragment length polymorphisms (RFLPs), which are either the result of single-base-pair substitution or the presence of a variable number of tandemly repeated oligonucleotide units at a locus (VNTRs). RFLPs can be recognized by digesting DNA with restriction enzymes and separating the fragments by size on an electrophoretic gel. VNTRs can vary widely among individuals, and they provide more linkage information than single-site polymorphic markers because family members are more likely to be heterozygous. Genetic maps of each chromosome, constructed from linkage data relating marker loci to one another in normal reference families, permit rational choices of markers for disease-mapping studies.(ABSTRACT TRUNCATED AT 250 WORDS)