The genetically diabetic db/db mouse exhibits symptoms that resemble human type 2 diabetes mellitus, demonstrates delayed wound healing, and has been used extensively as a model to study the role of therapeutic topical reagents in wound healing. The purpose of the authors' study was to validate an excisional wound model using a 6-mm biopsy punch to create four full-thickness dorsal wounds on a single db/db mouse. Factors considered in developing the db/db wound model include reproducibility of size and shape of wounds, the effect of semiocclusive dressings, comparison with littermate controls (db/-), clinical versus histologic evidence of wound closure, and cross-contamination of wounds with topically applied reagents. The size of wounds was larger, with less variation in the db/db mice (31.11 +/- 3.76 mm2) versus db/- mice (23.64 +/- 4.78 mm2). Wounds on db/db mice that were covered with a semiocclusive dressing healed significantly more slowly (mean, 27.75 days) than wounds not covered with the dressing (mean, 13 days; p < 0.001), suggesting the dressings may splint the wounds open. As expected, wounds healed more slowly on db/db mice than db/- mice (covered wounds, 27.75 days versus 11.86 days, p < 0.001; wounds not covered, 13 days versus 11.75 days, p = 0.39). Covered wounds, thought to be closed by clinical examination, were confirmed closed by histology only 62 percent of the time in the db/db and 100 percent of the time in the db/- mice. Topical application of blue histologic dye or soluble biotinylated laminin 5 to one of the four wounds did not spread locally and contaminate adjacent wounds. Multiple, uniform, 6-mm wounds in db/db mice heal in a relatively short time, decrease the number of animals needed for each study, and allow each animal to serve as its own control. The db/db diabetic mouse appears to be an excellent model of delayed wound healing, particularly for studying factors related to epithelial migration.