It is rather difficult to draw conclusions from reported C-peptide values, as the methods for determination differ, and C-peptide may be measured in serum or in urine with the patient fasting or after stimulation. We have followed prospectively 49 children with IDDM with regular determinations of serum C-peptide fasting and after a standardized breakfast. A subgroup of seven patients have been studied more thoroughly with 24-hour-profile of serum C-peptide, C-peptide excretion in urine, and stimulation by i.v. glucose + i.v. arginine. Our results indicate that the stimulation of the beta cells usually reaches a maximum around a blood glucose level of 10-12 mmol/l leading to a curve linear relationship between serum C-peptide and blood glucose. Thus a simple quotient is not so useful but the degree of stimulation should be stated and actual blood glucose value noticed. Stimulation with a standardized breakfast gives roughly the same information as maximal stimulation with i.v. glucose + arginine, and little extra information is found by a 24-hour-profile. Urinary C-peptide may give valuable information if it is related to the actual degree of metabolic balance. It can be of special interest in patients with very low serum C-peptide levels.