IN a previous communication we described an objective technique for observing and recording the unusual, stereotyped, motor and perceptual behaviors seen in autistic children.¹ The children were observed in a simplified, controlled environment, and behavior was recorded quantitatively on a 20pen, event recorder during six-hour observation sessions.

We found that the peaks of autistic activity occurred at random intervals without evidence of periodic or cyclic patterning of the behaviors. Although moment-to-moment variation was often considerable, the average amount of the behavior over prolonged periods of time was relatively consistent. Furthermore, there was no progressive increase or decrease in the amount of these behaviors during the sessions. Thus, the major finding of the previous study was the demonstrated persistence of autistic behaviors under conditions of prolonged observation in a controlled environment. These results are compatible with the notions that the autistic disturbances of motility are either serving as a source