The incidence of hypsa over a 40-year period in 667 pts was noted to rise to a peak in 1958 and then to fall so that relatively few cases appeared in the early 1970's, with hypsa seen only rarely in the past 10 years. This diminishing incidence was confirmed in a major children's hospital. The incidence curve of rubeola is seen to follow a similar course; the possibility is discussed that subclinical infectious disease, and rubeola in particular as a prototype, may have played a more prominent role in the etiology of hypsa than has been considered in the past.