Sir.—The article by Campos and MacGillivray1 in the August 1989 issue of AJDC suggested that gonadal steroids did not influence the somatic growth of the 18 children in the study. The graph that accompanies the article and the statement that the final adult height closely approximated or exceeded midparental height supports this suggestion. However, I suggest that more substantial evidence would have included a bone age at the beginning of sex steroidal therapy and a projected height based on the chronologic bone age before the onset of treatment. A demonstration that the final adult height closely approximated or exceeded the projected height would perhaps be a better demonstration of the hypothesis of their article.