Observation of the dates when children with orofacial clefts were born, giv
es the impression that they were born in clusters, between which there mere
intervals of different duration. An analysis was therefore undertaken of t
he period between 1972 to 1997 when 1112 such children were born. For diagn
ostic checking of stationarity(1) of the time series the Box-Ljung test was
used. Since it was found that the time series was stationary, analysis of
autocorrelation(2) function was performed. The results confirmed the impres
sion that children with clefts are born in clusters, between which there ar
e gaps of different duration. This suggests that the birth of a child with
a cleft is not a random and independent event. (C) 2000 European Associatio
n for Cranio-Maxillofacial Surgery.