The association of Parkinson's disease (PD) with an impaired sense of smell was first reported about thirty years ago. Since then, it has become quite firmly established that olfactory dysfunction is one of the first and most prevalent clinical manifestations of this disorder. Recent data from an ongoing prospective study indicate that otherwise unexplained hyposmia in first degree relatives of patients with sporadic PD is associated with an increased risk of developing clinical PD of at least 13%. In particular, a combination of impaired olfactory function and reduced striatal [123I]beta-CIT binding on a baseline SPECT scan appears to be a strong predictor of a subsequent diagnosis of PD. Pathological studies support these observations by demonstrating that the anterior olfactory structures may be one of the induction sites of PD pathology. Considering that there is a doubling rather than a loss of dopaminergic neurons in the olfactory bulb in PD patients, the pathophysiology of olfactory dysfunction in PD is far from being elucidated. Studying prodromal manifestations of PD, such as olfactory dysfunction, and their underlying pathophysiology may greatly contribute to the development of treatment strategies that focus on preclinical detection and slowing down disease progression.