Objective The aim of this study was to determine if multiple sclerosis (MS) shows association with variants of single nucleotide polymorphisms (SNP) within chromosome 19p, where previous studies resulted in conflicting observations. Subjects and methods The transmissions of 569 SNP variants and 608 haplotypes from unaffected parents to their affected children were tested in 257 Caucasian families by using the pedigree disequilibrium test (PDT), the TRANSMIT version 2.5 program and the family- and haplotype-based association tests (FBAT and HBAT). The distribution of linkage disequilibrium (LD) among SNPs in 19p was assessed by ldmax and correlated with the location of MS-associated haplotypes. Results Individual SNP alleles did not show association after correction for multiple testing in PDT. Several marker haplotypes within potential candidate genes of intracellular enzymes, transmembrane proteins and receptors and signaling and adhesion molecules appeared to be weakly associated with the disease in both TRANSMIT and HBAT. However, none of the associations was strong enough to survive correction for multiple testing. Conclusions The present study is in the line of previous studies with negative conclusions concerning the role of the 19p region in MS. Multiple Sclerosis 2008; 14: 433—439. http://msj.sagepub.com