The local lymph node assay is a novel predictive test for the identification of contact allergens. The collaborative study reported here was performed to evaluate the reliability of the method when performed in independent laboratories. Eight chemicals were examined in each of 4 participating laboratories and results compared with predictions of skin-sensitizing activity made from concurrent Magnusson and Kligman guinea-pig maximization tests performed in a single laboratory. The local lymph node assay has as its theoretical basis the fact that contact allergens induce T-lymphocyte proliferative responses. In practice, predictions of contact-sensitizing potential are made following measurement of proliferation in lymph nodes draining the site of exposure to chemical, and derivation of a stimulation index using control values as the comparator. Although in the present study there was some variation between laboratories with respect to the absolute stimulation indices recorded, it was found that with all chemicals each laboratory made the same predictions of sensitizing activity. Six chemicals (2,4-dinitrochlorobenzene, formalin, eugenol, isoeugenol, p-phenylenediamine and potassium dichromate) yielded positive responses, and two (methyl salicylate and benzocaine) were negative, in each laboratory. Furthermore, with 7 of the 8 chemicals tested there was no significant difference between laboratories in terms of the characteristics of the dose-response relationships recorded. With the exception of one chemical (benzocaine), predictions made with the local lymph node assay were in accord with those derived from guinea-pig maximization tests. These inter-laboratory comparisons demonstrate that the local lymph node assay is a robust and reliable method for the identification of at least moderate and strong contact allergens.