Women with inherited mutations in the BRCA2 gene have a very high lifetime risk of developing breast cancer. Previously, we used gene targeting with embryonic stem cells to generate mice with a mutation that disrupts exons 10 and 11 of the Brca2 gene. Mice that are homozygous for this mutation exhibit an embryonic lethal phenotype. To overcome this difficulty we have generated mice with loxP sites flanking Brca2 exon 27. Prior studies have shown this C-terminal domain of Brca2 interacts with Rad51, and cells that lack Brca2 exon 27 are hypersensitive to gamma-radiation. Therefore, site-specific recombination of loxP sites and deletion of exon 27 in this floxed Brca2 allele by a Cre recombinase should disrupt basic functions of Brca2 in DNA repair. The mammary-gland-specific removal of Brca2 exon 27 by Cre-mediated recombination in vivo has been accomplished by crossing the homozygous floxed Brca2 mice with a mouse mammary tumor virus (MMTV)-Cre strain D transgenic mice. Analyses of ROSA26 LacZ Cre reporter mice confirm that this MMTV-Cre transgene is specifically activated at the onset of puberty in mammary epithelial cells. In parallel studies a germline deletion of exon 27 was created by transiently electroporating embryonic stem cells carrying the floxed Brca2 allele with a Cre-expression plasmid. Surprisingly, mice homozygous for the germline deletion of exon 27 appear to be completely viable at birth, but preliminary studies suggest impaired male fertility. Gross phenotypic abnormalities in mammary gland ductal morphogenesis have not been shown by mammary whole mount preparations in these animals at up to six months of age. These mice are being observed closely for neoplastic development in mammary glands as well as other tissues. Mammary-specific Brca2^Δ27 mice should be a valuable experimental model mimicking the breast tumor development of women who have inherited a BRCA2 defect and then acquire a secondary somatic BRCA2 mutation.