A crucial role of the site of administration in the sensitivity of the alternative system using chick embryo for testing embryotoxicity was demonstrated by morphological evaluation of the effects of T-2 toxin and secalonic acid D, and by incorporation of [14C]sodium acetate radioactivity. Secalonic acid D, administered to 2-, 3-, and 4-day-old embryos in doses higher than 1 microgram produced mostly malformations of the face (bilateral cleft beak, microphthalmia) while the teratogenic effects of T-2 toxin were being limited to the embryonic trunk of 2-day-old embryos (rumplessness) after administering doses higher than 0.001 microgram. In case of subgerminal and intraamniotic injections, the doses of both mycotoxins needed for producing embryotoxic effects comparable to those obtained with the more commonly used yolk sac injections appeared to be lower by one and two orders of magnitude, respectively. The results stress the need of using the shortest transport channel of test substances from the site of application to the target tissues of the embryo, when the maximum sensitivity and reproducibility of the test system are to be expected.