This paper reports an analysis of trisomy 18 in Sweden between 1978 and 1987 and a comparison with data from the Rh?ne-Alpes/Auvergne registry in France. A strong maternal age dependency is seen, which parallels that observed for Down syndrome. The patterns of registered malformations in the two data sets are quite similar. A progressive increase of the prevalence at birth of trisomy 18 is seen in Sweden during the ten years but not in France. This increase seems to result, at least in part, from an increased identification of the trisomy among malformed infants. A parallel decrease in the reporting of infants with the "cardinal" trisomy 18 malformations is indicated.