The expanding use of genetically engineered mice has placed increasing demands upon research facilities to import mutant strains from noncommercial sources. Conventional quarantine strategies use relatively large amounts of space per animal and/or require specialized equipment (e.g., cubicles, isolators, and ventilated racks). We have retrospectively assessed a quarantine program instituted 4 years ago that used a small amount of space and minimized the need for special equipment. Shipments presumed, in light of health reports, to be free of agents excluded from our colonies were housed in static microisolator caging in a shared quarantine room. Rather than functioning as an "all-in-all-out" area, the room continually received new shipments, which were released intermittently as multi-shipment groups after testing was performed. Noninvasive testing of the imported mice was combined with nonsurvival sampling of sentinels that had been exposed to shipped animals via direct contact and/or exposure to soiled bedding. During the 4-year period examined, the vast majority of shipments presumed to be free of excluded agents showed no evidence of contamination when screened. When active infection was detected in the shared room, the procedures in place proved sufficient to prevent cross-contamination of other shipments. The use of sentinel animals to detect the shedding of infectious agents during quarantine was found to be an effective strategy that minimized the potential impact of invasive testing on quarantined animals. Although the program we describe is not appropriate for all situations, this type of approach may be considered by institutions wishing to explore alternatives to conventional quarantine strategies during periods of restricted space availability.