This was an exploratory cross-sectional study comparing 45 children with ASD to 24 typically developing drug-naïve controls, group-matched on age, sex, and body mass index. Objective data was obtained using the following: an ambulatory circadian monitoring device; saliva samples to determine dim light melatonin onset (DLMO): and three parent-completed measures: the Child Behavior Checklist (CBCL); the Repetitive Behavior Scale-Revised (RBS-R); and the General Health Questionnaire (GHQ28). The CBCL and RBS-R scales showed the highest scores amongst poor sleepers with ASD. Sleep fragmentation was associated with somatic complaints and self-injury, leading to a higher impact on family life. Sleep onset difficulties were associated with withdrawal, anxiety, and depression. Those with phase advanced DLMO had lower scores for "somatic complaints"; "anxious/depressed" state; and "social problems", suggesting that this phenomenon has a protective role.