A new method of measuring the range of the lateral deviation of a scoliosis is introduced. Here the Cobb angle is divided into two separate parts, consisting of the angles between each end vertebra and the horizontal plane. This angle is called the end vertebra angle. One hundred and twenty-one patients with adolescent idiopathic scoliosis, treated with Boston braces, were measured according to this technique and these angles were compared with the Cobb angles. All patients had S-shaped, right convex thoracic and left convex lumbar scolioses. Three end vertebra angles were measured and called A, B, and C. The middle end vertebra angle (B) was responsible for the improvement of the scoliosis in the brace, measured according to Cobb, and also the remaining improvement 2 years after weaning from the brace. The proximal and distal end vertebra angles (A and C), however, were unchanged or had increased at the time of the follow-up study 2 years after weaning from the brace treatment when compared with the status before the treatment. This could not be observed by using the Cobb method only. If end vertebra angles A and B are not equal, the thoracic curve is asymmetric. This asymmetry can be of two types depending on which of the two end vertebra angles is the greater one. If A is greater than B, the result of brace treatment was more successful than that of the symmetric curves. However, if end vertebra angle B was greater than A, the result of treatment was not good, and with asymmetry of a certain extent there was an increase of the curve despite brace treatment. In the lumbar curves the presence of asymmetry was of no importance for the result of brace treatment.