350 sleep EEG recordings in PKU children in the first year of life were analysed in order to characterize the abnormal features. Recordings were taken from serial studies carried out during the last 8 years on early and late treated children. Only quiet sleep was evaluated due to its longer duration after the second month of life and its important developmental characteristics. The criteria of abnormality were the following: lack of bioelectric features of sleep, abnormal " trac é alternant ", and other criteria of defined range, character and morphology. The number of abnormal records was referred to the three periods of the first year of life: first and second trimester, and the second semester of the first year of life, and also to treatment. EEGs before treatment were considered as records of untreated children. Table 1 compares the number of normal and abnormal records according to age and treatment. Table 2 illustrates the character and the range of abnormalities with reference to age and treatment. Tables 3, 4, 5, give the characteristics of morphological features of abnormal records. Our previous papers describe EEGs of PKU children of different age during waking and sleeping, while the present study is the evaluation of EEGs in children of the same age recorded in sleep only. The choice of these criteria is justified by the greatest variability of EEG records in relation to developmental factors and activating role of sleep. Among all EEGs, 60% were recorded during treatment. The biggest number of EEGs recorded in untreated children was obtained in the first three months of life and became gradually less. About 70% of the records were normal. No difference was noted between the number of normal and abnormal EEG records in different periods of the first year of life. The number of normal records was considerably higher in treated children than the number of abnormal records in untreated ones. This relationship was not noted in the first three months of life. In the first half year of life, generalized abnormalities as well as abnormal " trac é alternant " predominated and localized abnormalities were exceptional. On the other hand, in the second half year of life nearly half of the abnormalities were as follows: general paroxysmal abnormalities not observed in the first half year of life, considerably higher amount of abnormalities localized only in the temporal region, less generalized non-paroxysmal abnormalities, and lack of bioelectrical features of sleep.(ABSTRACT TRUNCATED AT 400 WORDS)