Sst I, we detected a particular polymorphic DNA pattern in 17 of 28 patients (60.7%) with psoriatic arthropathy but in only 5 of 41 patients (12.2%) with psoriasis alone. Our findings suggest that genes in the immunoglobulin region confer susceptibility to the development of arthropathy in patients with psoriasis.Studies of families have shown that genetic factors play a major role in the development of psoriasis without arthropathy (PS) ( I ) and a less important role in the development of psoriatic arthritis (PSA) (2). Histocompatibility antigen studies have established the genetic basis of these entities and have shown an increased frequency of HLA-Cw6 in both PS and PSA (3). Armstrong and coworkers (4) and Murray et al (3) have reported the incidence of HLA-DR7, in linkage disequilibrium with Cw6, to be greater in PSA, although only Murray et a1 described a similar association for PS. In contrast, Gladman and colleagues ( 5 ) did not find PSA or PS to be associated with DR7. It has been suggested that HLA-Bw39 may be a marker for PSA (61, although Gladman et al (5). while confirming its increased incidence in PSA. also found an increased frequency of this antigen in patients with PS alone. The HLA system does not, therefore, appear to provide a reliable marker for distinguishing between PS and PSA.On human chromosome 14, there are immunoglobulin heavy chain (IgH) genes. These are of possible etiologic importance in PS and PSA, because approximately 66% of PSA patients and 33% of PS patients have higher levels of circulating IgA (7). However, studies of Gm allotypes, genetically determined markers on IgH, have failed to show an increased prevalence in patients with PSA (8). The availability of DNA probes for the IgH genetic region allowed us to reevaluate this question, using the powerful technique of restriction fragment length polymorphism (RFLP) analysis. The probe selected hybridizes to both Sp (the p heavy chain switch region) and Sa, (the a , heavy chain switch region) genes (9). We report an increased frequency of an RFLP band in PSA patients, as detected with this probe. This raised frequency is not observed in patients with PS alone.Patients and methods. Twenty-eight PSA patients (mean age t l SD 43.6 ? 16.3 years), 41 PS patients (42.5 k 16.4 years), and 132 healthy white control subjects were included in the study. The PSA patients were attending our rheumatology clinic. Clinical and radiologic findings showed that 15 patients had peripheral arthritis only, 2 had spondylarthropathy only, and 1 I had both peripheral arthritis and spondylarthropathy. Patients with any evidence of joint disease in the past or present were excluded from the PS