Golden retriever muscular dystrophy (GRMD) is an excellent model for the study of the efficacy of gene therapy in dystrophin deficient myopathies for there are many similarities between affected dogs and Duchenne muscular dystrophy (DMD) in boys. GRMD is not caused by deletion mutation but results from a point mutation in the consensus splice acceptor in intron 6 of the canine dystrophin gene. As a result exon 7 is skipped during processing of the GRMD dystrophin messenger RNA. We have developed a rapid test which makes direct use of exon 7 specific genomic PCR products. We have undertaken preliminary experiments on gene therapy using the mini-gene and the full length gene alone and in combination with lipofectin and/or the bacterial beta-galactosidase reporter gene Lac Z. Following direct injection of the Lac Z plasmid, either alone or with lipofectin, about 50% of the sites showed expression when biopsied some 14 days later. The beta-galactosidase activity was present in muscle and granulation tissue but was never abundant. Pups injected intraperitoneally with Lac Z were found to have positive material in their mesenteric lymph nodes, liver and spleen. Those injected with Lac Z and lipofectin also had positive material in the diaphragm, intercostal muscles and abdominal muscles, but again only a small amount of positive material was present at any of the sites. In animals directly injected into the muscle with the dystrophin mini-gene, half had positive staining for dystrophin in biopsies taken 14 days later. Of the 6 sites in the muscles of animals given the mini-gene and lipofectin only one had fibres positive for dystrophin when examined 14 days later. Six pups were injected directly with full-length gene construct and when biopsies were taken 10 days later two of the animals had strongly stained peripheries to a small number of fibres.