With 1 figure in the text)Heterosexual courtship trials were staged between dusky salamanders of two species (Desmognuthus fuscus and D. sanfeefluh) from two allopatric populations in the southern Appalachian Mountains of eastern North America. Two experiments were conducted in order to test for effects of experimental design on estimates of the strength of sexual isolation between these species. In Experiment I, frequencies ofinterspecific mating were determined when: (I) a male was presented with a conspecific and a heterotypic female successively; and (2) both types of females were presented simultaneously to a male. The frequency of interspecific mating by male D. fusci was greatest when females were presented successively. The frequency of interspecific mating by male D. santeetluh was not influenced by method of female presentation. An index of sexual isolation of +0.29 was obtained between D. fuscus and D. sunfeefluh when females were presented successively, and of +0.63 when females were presented simultaneously. In Experiment 2, interspecific mating frequencies were determined in courtship arenas of two different sizes. No significant effect of arena size on frequency of interspecific mating was found for D. ficus. In D. santeetlah, a trend towards greater mating frequency in small arenas was almost significant. These results suggest that interspecific mating frequencies (and thus estimates of the strength of sexual isolation between populations or species) may be sensitive to the experimental design employed. The validity and generality of this conclusion for other animals is an urgent, empirical issue.