In the diabetes-prone NOD mouse, there is a proven association between a systemic deficiency of NKT cells and the onset of type 1 diabetes. Numerous reports of similar defects within the NKT cell compartment of human type 1 diabetes patients suggested NKT cell levels might be a valuable predictor of susceptibility and could provide a target for therapeutic intervention. Two recent studies, however, found no association between type 1 diabetes and blood NKT cell levels in humans and consequently rejected a link between the onset of diabetes and NKT cell deficiency. This cast considerable doubts on the potential for NKT cell-based clinical applications and challenged the validity of the NOD mouse as a model of human type 1 diabetes. We now report that NKT cell levels in blood are a poor representation of those in other organs. Strikingly, systemic NKT cell deficiencies were identified in NOD mice with normal, or even raised, blood levels. This re-establishes the correlation between NKT cell deficiency and type 1 diabetes and raises important questions regarding the assaying of NKT cell levels in humans.