The present paper reports on results of ultrasonographic examination in the identification of severe congenital malformations in the period between 1975 and 1982. The incidence of severe congenital malformations in relation to the total number of births during this period was 159 out of 11,372 (1.4%). In 144 cases with severe malformations at least one antenatal ultrasonographic examination had been performed. According to their topographic location, 42% of these were head/neural tube defects, 38% trunk/organ defects, only 2% were severe defects of the extremities and 18% were rare fetal malformations. As a result of previous ultrasonographic examinations at specialists' practices 60% of the cases were referred to the authors' clinic for further clarification with a correct diagnosis or a suspected fetal malformation. Of all the sonographically demonstrable structural defects of the fetus, 81% of all severe fetal defects seen at the authors' clinic during the period in question were identified correctly. If the observation period is divided into the years 1975 to 1979 and 1980 to 1982, there is a striking rate of increase in the number of antenatal ultrasonographic diagnoses which were correct, from 71% in the first period to 86% between 1980 and 1982. Most of the false-negative ultrasonographic findings were congenital cardiac abnormalities, since up to that point no special fetal echocardiographic examinations had been performed. In the entire period covered by the investigation there was only one false-positive finding ("Potter's syndrome"). Forty-six per cent of the ultrasonographically demonstrated severe fetal malformations were diagnosed before the end of the 24th week of pregnancy, and 54% after the end of the 24th week of pregnancy. Only in 60 out of 141 cases (43%) with severe fetal malformations was the quantity of amniotic fluid found to be normal; 26% of the cases had hydramnios and 31% oligohydramnios. Pathologic movement behaviour had been registered ultrasonographically in 43% of the cases with severe fetal malformations; biometric dimensions of the biparietal cranial diameter and the transverse diameter of the thorax (greater than 10th percentile to 90th percentile, according to the percentile growth curves of Schmidt, 1982) corresponding to gestational age had only been measured in 30% and 50%, respectively, of the cases with fetal malformations. During the entire period covered by the investigation, from 1975 to 1982, only 16 children born at term (between the 38th and 42nd weeks) had severe malformations which had not already been diagnosed.(ABSTRACT TRUNCATED AT 400 WORDS)