Hedgecock and Russell (1975; Proc. Natl. Acad. Sci. USA 72: 4061) reported thermotaxis of C. elegans that can be observed in two ways: 1) when placed on a linear thermal gradient, the animals migrate toward a temperature at which they were cultivated; 2) when placed on a radial thermal gradient, the animals move along their adaptive growth temperature, thereby making isothermal circular tracks. By screening F2 progeny of EMS-treated N2 for animals which migrated abnormally in a linear thermal gradient, Hedgecock and Russell (1975) isolated 6 ttx (thermotaxis-defective) mutants. We are interested in identifying and analyzing genes involved in thermotaxis of C. elegans, and have begun isolating more ttx mutants with the modification of the procedure developed by Hedgecock and Russell (1975). We noticed earlier that it was difficult to assay reproducibly the thermotaxis of N2 strain. We thought that normal chemotactic behaviors of the N2 animals may cause poor or vague thermotaxis, and mutagenized CB1377(daf-6(e1377)) animals, which are che-, tive due to structural defect in amphidial sheath cells (Albert et al., 1981; J. Comp. Neurol. 198: 435), but show normal thermotaxis reproducibly. F2 progenies of 170 EMS-treated CB1377 animals were grown at either 15 C or 20 C, and were screened for animals which, after placed on a linear thermal gradient for 1hr, migrated from their growth temperature toward higher (from 15 C to 20 C, or from 20 C to 25 C) or lower (from 20 C to 15 C) temperature. Animals which migrated abnormally were picked among F2 progenies of 46 mutagenized parents, and thermotaxis of the isolated clones was assayed for several generations. We so far isolated 15 independent ttx mutants. Those include thermophilic (2 lines), cryophilic (7 lines), and atactic (6 lines) phenotypes, based on thermotactic behavior in the linear gradient. They can be further divided into several classes, based on whether or not the mutants are capable of making isothermal tracks on a radial thermal gradient. We are in process of mapping these ttx mutations. At present, we only know that ks5, showing cryophilic phenotype on two different thermotaxis assays, is probably linked to LGX. One of our ttx mutations, ks14, exhibits an interesting phenotype: it is atactic on the linear gradient, but has a leaky cryophilic phenotype on the radial gradient. In our new assay procedure using thin NGM square plate (14 x 10cm) with the linear gradient, individual ks14 animals grown at 20 C are capable of making isothermal tracks in a range of 17-22 C, whereas parental CB1377 animals grown at 20 C make sharp isothermal tracks around 20 C.