The association of duodenal ulcer with blood group 0 (Aird et al., 1954;Roberts, 1957) and with salivary ABH nonsecretion (Clarke et al., 1956;Doll et al., 1961) is firmly established. Thus the family studies of Doll and Kellock (1951), which showed evidence of a hereditary factor in duodenal ulcer, have been substantiated by identification of two of the responsible genetic loci-namely, those for the genes determining ABO blood group and salivary ABH secretor status. Though the underlying mechanisms whereby these genes influence liability to duodenal ulcer remain somewhat obscure, yet the present paper throws some light on one facet of the problem.Sievers (1959), using the " diagnex " blue test as a screening procedure for "achlorhydria " in several hundred patients, found evidence of free acid in 86.1% of group 0 patients but in only 73.0% of group A patients. Similarly, using the serum-pepsinogen level as an index of gastric secretion, Sievers found a significantly higher mean level in group 0 than in group A patients. These findings suggest the possibility that the possession of group 0 phenotype leads to the development of a greater gastric-acid-pepsin secretory-cell mass than that in the other ABO blood-group phenotypes. Further support for this possibility exists in the work of Purohit and Shukla (1960), who related the mean acid output in response to an alcohol test meal to the ABO blood groups, and in the clinical evidence, wherein there is not only decreasing gastric acidity but also a gradual switch from preponderance, relative to the control population, of group 0 to that of group A, in the following series of conditions-duodenal ulcer, gastric ulcer, gastric cancer, and pernicious anaemia (K0ster et al., 1955).While the present study was in progress Niederman et al. (1962), in a survey of 1,000 university intake students, stated that they found no significant differences between the mean serum-pepsinogen levels of the various ABO blood group and salivary ABH secretor phenotypic classes. Nevertheless, the mean level was higher in group 0 salivary ABH non-secretors than in group A salivary ABH secretors (P<0.2). For the following reasons it is felt that the negative results of Niederman et al. (1962) (1958) found correlation between the means of repeated basal estimates of acid and pepsin secretion and the serum-pepsinogen level. It thus seems likely that the latter affords an index of the gastric peptic-cell mass (Hirschowitz, 1955). Because under most conditions the output of acid and that of pepsin are roughly parallel (Spiro et al., 1956), it may be presumed with a fair degree of confidence that the serum-pepsinogen level reflects the size of the gastric secretory-cell mass (in this article the term " gastric secretory-cell mass " refers to the parietal and peptic cells but excludes the mucus-secreting cell mass).
Material and MethodsLarge numbers of healthy individuals aged 18 to 35 inclusive from the population were screened in order that the serumpepsinogen level of the 100 subjects could b...