Arterial oxygen saturation is now one of the important values in paediatries. It is not always possible to perform repeated estimations in infants and children after puncture of an artery. An apparatus that renders the determination of oxygen saturation of capillary blood possible is tested. The blood specimen is gained by prick incisions. These methods stand the statistical tests under the conditions applicable in paediatrie cases. The values are also sufficiently accurate in neonates and cyanosed children (e.g. congenital heart diseases) when certain conditions are taken into consideration. If the circulation at the site of the prick incision is poor and if the child screams vigorously, the value of the capillary oxygen saturation thus gained is not reliable. A comparison between the oxygen saturation values gained by the capillary and hemoreflectometer methods is also statistically valid, if 1 day or 2 lie between the tests. The apparatus is especially suitable in paediatrie cardiology and intensive care.

After establishing the methodical and technical prerequisites, the microoxymeter OSM 1 (Radiometer, Copenhagen) was used to determine the capfllary O2-saturation in routine clinical practice.1. Studies on 10 normal neonates in the first 4 days of life showed an initial hypoxia of an average O2-saturation of 86.3~ which then rose to 93.1~ on the 4th day of life. This observation has to be taken into consideration when judging the O~-saturation values of neonates.2. Studies in the effectiveness of the three different ways of O2-administration showed that the lowest rise of capillary O~-saturation oceured when 08 was administered by means of a plastic funnel placed in front of the mouth and nose of the infant; in the 08 tent (above all by administering large quantities of 02) and by the use of tightly closing plexiglass hoods the effect on the capillary O2-saturation was found to be favourable.3. Even children with cyanotic congenital heart diseases demonstrated a rise of the capillary 02-saturation after the breathing of 08. A residual hypoxia remains, irrespective of the normal O2-saturation of 96~ . The use of an "oxygen test" for the differentiation of cardiac and pulmonary cyanosis cannot be sanctioned.4. The practical use of the apparatus and its importance in paediatric cardiology and intensive care is described and discussed with a few examples.