The hypocretin (orexin) system is involved in sleep/wake regulation, and antagonists of both hypocretin receptor type 1 (HCRTR1) and/or HCRTR2 are considered to be potential hypnotic medications. It is currently unclear whether blockade of either or both receptors is more effective for promoting sleep with minimal side effects. Accordingly, we compared the properties of selective HCRTR1 (SB-408124 and SB-334867) and HCRTR2 (EMPA) antagonists with that of the dual HCRTR1/R2 antagonist almorexant in the rat. All 4 antagonists bound to their respective receptors with high affinity and selectivity in vitro. Since in vivo pharmacokinetic experiments revealed poor brain penetration for SB-408124, SB-334867 was selected for subsequent in vivo studies. When injected in the mid-active phase, SB-334867 produced small increases in rapid-eye-movement (REM) and non-REM (NR) sleep. EMPA produced a significant increase in NR only at the highest dose studied. In contrast, almorexant decreased NR latency and increased both NR and REM proportionally throughout the subsequent 6 h without rebound wakefulness. The increased NR was due to a greater number of NR bouts; NR bout duration was unchanged. At the highest dose tested (100 mg/kg), almorexant fragmented sleep architecture by increasing the number of waking and REM bouts. No evidence of cataplexy was observed. HCRTR1 occupancy by almorexant declined 4-6 h post-administration while HCRTR2 occupancy was still elevated after 12 h, revealing a complex relationship between occupancy of HCRT receptors and sleep promotion. We conclude that dual HCRTR1/R2 blockade is more effective in promoting sleep than blockade of either HCRTR alone. In contrast to GABA receptor agonists which induce sleep by generalized inhibition, HCRTR antagonists seem to facilitate sleep by reducing waking "drive".