Teratogenicity of vitamin A was firstly detected in experimental animals in 1953. Nearly 30 years later, teratogenicity of vitamin A analogue-isotretinoin was reported in humans. Isotretinoin induces serious birth defects of craniofacial and central nervous system, cardiovascular system and thymic malformations--in about 25% of babies exposed during the first trimester of their prenatal development. The biological interconversion of isotretinoin to vitamin A is known. That is why later epidemiological studies focused on high vitamin A intake in pregnant woman: Women who use daily vitamin A supplements during early pregnancy have approximately a two-fold increased risk of giving birth to a malformed baby. On the basis of these data, replacement of vitamin A has been recommended with its natural precursor B-carotene which is supposed to be more safe for pregnant woman due to its limited absorption from intestine. Aim of the present paper was to test a possible direct embryotoxic effect (i.e. lethality + teratogenicity) of B-carotene in chick embryos and to compare these results with known embryotoxicity of vitamin A in the same experimental model. Single subgerminal or intaamniotic injection of vitamin A or B-carotene within day 2-5 of incubation was used for estimation of the beginning of the embryotoxicity range determining the minimal embryotoxic doses. Vitamin A started to affect development between doses 0.3-0.3 microm [corrected] per embryo. Malformations of head, extremities and heart were detected similarly like in laboratory mammals and in man. B-carotene exhibited such an effect neither after injection of the highest tested doses-100 microm [corrected] per embryo. The results documented the strong difference in embryotoxicity between vitamin A and B-carotene. After theoretical extrapolation of the results achieved in the chick embryo, the minimal embryotoxic doses of vitamin A in mammals were estimated to be between 0.1-1 mg/kg of maternal weight and those of B-carotene to be more than 1000 mg/kg of maternal weight. Human epidemiological studies have proved teratogenicity of vitamin A after daily doses 25,000 i.u.-8.3 mg (0.13 mg/kg)- and reduction of its maximum intake has been recommended to 10,000 i.u. per day (0.05 mg/kg). The results about teratogenicity of vitamin A achieved in the chick embryo are in agreement with such a recommendation. Intake of vitamin A in the food is sufficient for pregnant woman in common Czech population. Therefore, an artificial supplementation of vitamin A brings risk of overdosage. If supplementation by vitamin A is unavoidable during pregnancy, B-carotene should be preferred.