Neo-natal rats emit ultrasonic vocalizations (USVs) when isolated from their mothers and littermates. Clinically effective anxiolytics reliably reduce USVs, making this behavior a useful animal model of the anxiolytic potential of novel pharmacological approaches to the treatment of anxiety. Here, we assess the hypothesis that USV duration (total time spent vocalizing) is a more sensitive measure of anxiolytic and antidepressant efficacy than USV number by testing established and putative anxiolytics in this model. Negative geotaxis and righting reflex latency were measured to assess sedating properties. The benzodiazepines, CDP (1-10 mg/kg) and diazepam (0.3-3 mg/kg), the 5HT(1A) partial agonist, buspirone (0.3-3 mg/kg), and the mGluR5 antagonist, MTEP (1-30 mg/kg), reduced USV duration at lower doses and to a greater magnitude than USV number. The benzodiazepines, unlike buspirone and MTEP, produced measurable sedation, but it was dissociable from reductions in USV duration. The SSRI antidepressants, fluoxetine (1-30 mg/kg) and citalopram (0.3-30 mg/kg), reduced USV duration more than number with no measurable effect on sedation. The tricyclic antidepressants, imipramine (1-10 mg/kg) and amitriptyline (1-30 mg/kg), had no effect dissociable from sedation. These data support USV duration as a more sensitive and useful measure than USV number in the isolated rat pup model.