Three liquid insecticide formulations were evaluated as barrier treatments against perimeter-invading ants at a multifamily housing complex in West Lafayette, IN. Several ant species were present at the study site, including (in order of abundance) pavement ant, Tetramorium caespitum (L.); honey ant, Prenolepis imparis (Say); odorous house ant, Tapinoma sessile (Say); thief ant, Solenopsis molesta (Say); acrobat ant, Crematogaster ashmeadi (Mayr); crazy ant, Paratrechina longicornis (Latrielle), field ants, Formica spp.; and carpenter ant Camponotus pennsylvanicus (DeGeer). Studies began in May 2001 and concluded 8 wk later in July. Individual replicate treatments were placed 0.61 in (2 feet) up and 0.92 m (3 feet) out from the ends of 46.1 by 10.1-m (151 by 33-foot) apartment buildings. Ant sampling was performed with 10 placements of moist cat food for 1 h within treatment zones, followed by capture and removal of recruited ants for later counting. All treatments led to substantial reductions in ant numbers relative to untreated controls. The most effective treatment was fipronil, where 2% of before-treatment ant numbers were present at 8 wk after treatment. Both imidacloprid and cyfluthrin barrier treatments had efficacy comparative with fipronil, but to 4 and 2 wk, respectively. Odorous house ants were not sampled before treatment. Comparisons of ant species composition between treatments and controls revealed an increase in odorous house ant frequencies at 1-8 wk after treatment in treated locations only. These results demonstrate efficacy for both nonrepellent and repellent liquid insecticides as perimeter treatments for pest ants. In addition, our findings with odorous house ant highlight an apparent invasive-like characteristic of this species that may contribute to its dramatic increase in structural infestation rates in many areas of the United States.