Buspirone and other drugs that act as 5-HT1A agonists appear to be clinically effective anxiolytics in humans, yet their anticonflict effects, though robust in pigeons, are equivocal in rodents. In the present study we examined the effects of the benzodiazepine midazolam and a series of 5-HT1A agonists on punished responding of squirrel monkeys. Lever presses were reinforced according to a fixed-interval 3-min schedule; in addition, each thirtieth lever press was punished. Midazolam produced large increases in response rates, whereas none of the 5-HT1A compounds produced any increases in responding. Most of these drugs decreased response rates at the higher doses examined. Although the reasons for the discrepancy between species in the anticonflict effects of serotonergic anxiolytics cannot be specified, the different anatomical distribution of 5-HT1A binding sites across species may suggest a different functional role for this receptor.