Several investigators discussed the need to know more about conditions which facilitate normal human development, especially the need for a better understanding of the processes at work during the development towards greater independence in normal and pathogenic human relationships. The study reported in this paper aims to provide a description of the processes at work during the development of contact- and distance regulation for free-living chimpanzee mother-infant relationships. We believe that such an ethological study contributes to human studies by providing methods of observing and analysing behaviour, and by providing hypotheses that can be tested. Our study takes a systems approach to mother-infant contact- and distance regulation. We looked at the history of mother-infant relationship in terms of the effects of changes in the mother's and infant's behaviour on the mother-infant dyad as a self-regulating homeostatic system. A single-subject research design was used, because it has special advantages for the study of developmental processes, not shared by the usual experimental- or correlational designs. We found that the infants' progress towards greater independence proceeds discontinuously in 5 jumps over the first 24 months. With each jump the following drastic changes are found in the quality of the distance-regulation between mother and infant and/or the amount of time spent at a certain distance: 1. After month 2 maternal behaviour related to carrying and supporting the babies in the ventro-ventral position decreased sharply and the babies were forced to cling to the mother themselves. 2. In month 5-6 a period of mother-infant conflict was found in which aggressive maternal rebuffs were aimed at breaking nipple- and ventro-ventral contact. Before this age the mothers were primarily responsible for ventro-ventral contact, and after this age the infants were. A relation was found between the rebuffs and the onset of dorsal riding and eating solid food. These changes in the mother-infant interaction coincided with physical changes such as a change in the speed of growth and the eruption of teeth. The findings are placed in the wider framework of mammalian development. 3. After month 7 an explosion of the frequency of the infants' excursions was found and the infants now made short lasting excursions and remain within arm's reach from their mothers. It is argued that such excursion behavior expresses the infants' concern with the distance to their mothers at this age. 4. Around month 12 and month 18 periods of mother-infant conflict were found in which the aggressive maternal rebuffs were aimed at breaking body contact. Both periods of mother-infant conflict were associated with peaks in the infants' responsibility for body contact and with rises in the amount of time spent in this contact. Both periods of mother-infant conflict were followed by sharp drops in the amount of time spent in body contact, and after each drop time spent in body contact remained at a newly reached level. Furthermore, after both periods of mother-infant conflict all infants made use of space more distant from their mothers. We found that mothers do not promote the infants' independence as a whole in each confict period, but that they do so only for a particular aspect in training the infants how to adapt to other individuals and to the physical outer world when using the new ability. It is argued that mothers recognize the ages that their infants are ready to reorganize their behaviour, upon which they force them to do so. The periods of mother-infant conflict around month 12 and 18 are preceded by periods of regression: temporary shifts back to mainly staying in closer contact (= qualitative regressive shifts) and temporary increases in the amount of time spent in ventro-ventral contact (= quantitative regressive peaks). It is suggested that regressive behaviour is a common feature of normal development after a certain age. We called a period in which the succession of regression and/or conflict and/or jump towards greater independence was found a labile period, as opposed to stabile periods. In the general discussion the following topics are attended to: a) The possibility that "labile periods at specific ages" in the mother-infant relationship are a common feature in normal development. b) The possibility that changes in maternal behaviour in each labile period are responsible for the phenomenon of jumps in the growing independence, provided that the infants are not pushed beyond their maturational abilities. c) The possibility that regression, which precedes mother-infant conflict, is associated with maturational changes in the infant is discussed. Several authors associated regressive behaviour with spurts in development. d) The possibility that maturational changes are associated with developmental steps. Infants contribute to changes in the mother-infant systems as they change maturationally. The latter changes were established independently of their effect on the mother-infant relationship by PLOOIJ (1984). He looked for qualitative changes over development in behaviour and recognized 5 steps in the infant's first year of life. It is discussed that the onsets of labile periods in the mother-infant relationship are associated with the developmental steps. It seems that a developmental step triggers mother-infant conflict, and that mother's behaviour during this conflict is of vital importance in realizing the infant's developmental potential. PETERFREUND's (1971) thinking on the relation between maturation and learning (reprogramming) suggests mother's helpful contributions to what happens around the times of occurrence of developmental steps. We related our findings to BOWLBY's (1969, 1973, 1980) attachment theory. The possibility is discussed that attachment takes different forms over age, depending on the proximity involved and the skills available to the infant, and that it is active and present from birth onwards.