In 1972, Valentine et al described, under the name of 'non-spherocytic haemolytic anaemia, high red cell ATP and ribose phosphate pyrophosphokinase (RPK; EC 2.7.6.1) deficiency', an obscure congenital haemolytic anaemia with the characteristic feature of red blood cell basophilic stippling. The activity of Embden-Meyerhof pathway and hexose monophosphate shunt were normal, and the concentrations of reduced glutathione and of ATP were raised 2 SD above the normal mean. The low values of RPK also encountered were considered to be an epiphenomenon rather than a causative defect. One year later, further studies performed in two new kindreds with the same haemolytic disorder associated with persistent basophilic stippling were described under the name of 'haemolytic disorders associated with increased ATP'. In 1974, two new and important observations contributed to the final identification of the disease: the patients' red blood cells (RBCs) contained large amounts of nucleotides (pyrimidine nucleotides), and in all cases they were deficient in a hitherto unrecognized enzyme called pyrimidine 5' nucleotidase (P5N). In conclusion, all these cases were formerly referred to as 'high ATP syndromes' because of the erroneous assumption that the large number of nucleotides within deficient RBCs were adenine phosphate rather than pyrimidine phosphate. Twenty-five years after its description, P5N deficiency has been reported in about 35 unrelated families from different parts of the world, and it has become one commonly identified cause of hereditary non-spherocytic haemolytic anaemia due to RBC enzymopathy. Genetic transmission is via the autosomic recessive mode, and only homozygous or compound heterozygous are clinically affected. Family members who are biochemically heterozygous are haematologically normal and difficult to detect. Unfortunately, the precise gene mutation or mutations causing the disease remain unknown.