The differential diagnosis of Alzheimer's disease (AD) and frontotemporal dementia (FTD) remains a difficult clinical issue. A recent meta-analysis by Hutchinson and Mathias (2007) found that many cognitive tests do not adequately differentiate between these types of dementia. However, their study did not examine the ability of neurobehavioral scales to distinguish between AD and FTD. The data from 33 studies, published between January 1994 and December 2008, examining the neurobehavioral symptoms of persons with AD (N = 2,305) and FTD (N = 971) were therefore analyzed. Weighted Cohen's d effect sizes, percentage overlap statistics, confidence intervals, and fail-safe Ns were calculated for each scale. The most discriminating measures were the Schedules for Clinical Assessment in Neuropsychiatry and the Scale for Emotional Blunting. The Middelheim Frontality Score and the Frontal Behavior Inventory also had excellent discriminative ability, surpassing the cognitive tests examined previously. Numerous other scales additionally showed large and significant differences between the AD and FTD groups, highlighting the importance of assessing neurobehavioral symptoms in this context. While these latter measures may be useful in providing a clinical profile of patients with dementia, greater overlap in the scores of patients with AD and FTD limits their usefulness in the context of differential diagnosis.