Disodium octaborate tetrahydrate (DOT) dust, DOT aqueous solution, imidacloprid dust, and amorphous silica gel dust with synergized 1% pyrethrins were applied on wood surfaces to simulated attic modules. Modules (30 by 30 cm) with and without fiberglass insulation were exposed to dispersal flights of Cryptotermes brevis (Walker) in May and June of 1998 and 1999. Six months after flights, modules were disassembled and inspected for nuptial chamber location and contents. During both years, air and water control treatments contained 22.2 ± 9.94 (mean ± SD) nuptial chambers, 7.5 ± 5.7 live imagos, and 2.0 ± 1.4 chambers with brood. This survivorship indicated that the attic modules performed well as a colonizing platform for C. brevis. C. brevis dealates preferred constructing nuptial chambers in the crevices at the bases or tops of the modules instead of internal crevices. Modules treated in 1998 and 1999 with DOT or silica dusts contained no live termites, whereas zero of five modules treated with imidacloprid dust in 1998 and two of 20 modules treated with imidacloprid dust in 1999 contained single live incipient colonies. In 1998, 15% DOT solution, applied as a postconstruction treatment, yielded significantly fewer chambers and live termites than controls, but was not as effective as dusts in preventing successful colonization. In 1999, the DOT solution, applied as a construction-phase treatment, was equally as effective in preventing colonization as the dust treatments during that year. Results indicate that dust formulations of DOT, silica gel, and imidacloprid can be used to prevent drywood termite colonization in existing building voids and attics. Where the entire wood framing is exposed to treatment, such as during building construction, aqueous DOT solution can be equally effective as dusts in preventing colonization by C. brevis.