The cephalometric points A and B after Downs are criticized and on the angle ANB their selective replacement by "standard-AS and BS points along with the angles ASNB and ASNBS is suggested. With 55 ideal cases of the permanent dentition it was proven by the t-test that the mean values of the latter angles vary only very little from the angle ANB. Steiner's "acceptable compromises" are compared with a new combination of "flowing norm values" using AS and BS. The correlation coefficients are considerably higher than Steiner's (r = 0.47--0.66). They are between r = 0.75 and 0.93. Also the mean deductive mistakes of the incisal parameters keep within relatively small limits. Steiner's expected relations of the ANB angle (reduction of same to one half of the original measure) seem doubtful. Probably it is mainly dependent upon the type of anomaly and the method of treatment, further upon the type of growth and age.