A polymerase chain reaction (PCR) technique was used to detect transgene presence after pronuclear microinjection of mouse zygotes cultured to various stages of development. The transgene was detected in 88% of 1-cell, 88% of 2-cell, 44% of 4- cell, 40% of morula, and 29% of blastocysts. By comparison, the integration frequency for transgenic mice made using the same DNA construct was 22%. After 5 days of in vitro culture, the injected construct was detected in 83% of arrested 1-cell, 85% of arrested 2-cell, and 85% of fragmented embryos. Only 28% of zygotes cultured after microinjection of DNA developed to the blastocyst stage compared to 74% of non-injected zygotes. When DNA buffer alone was injected, 63% of zygotes developed to the blastocyst stage. These data suggest that pronuclear microinjection of DNA is highly detrimental to subsequent embryonic development. Also, most injected DNA that is either unintegrated or that will not be integrated into the genome has been degraded by the blastocyst stage such that it can no longer be detected by PCR. The production of transgenic mice by cytoplasmic injection of DNA mixed with poly-L-lysine is also described. The effects of DNA concentration and stoichiometric ratio of positive charges provided by the polycation to negative charges provided by DNA on transgenic frequency and embryonic viability were studied. The highest transgenic frequency (13% of pups born were transgenic) was obtained when a polylysine/DNA complex having a stoichiometric charge ratio of one to one (equal positive charges as negative charges) at a DNA concentration of 50 ug/ml was used. The transgenic frequency by pronuclear injection of the same DNA construct was 22%. The percentage of zygotes, cultured in vitro, reaching the blastocyst stage which were injected cytoplasmicly was not different (p>0.05) than that of control zygotes that were not microinjected (65% versus 74%, respectively). The percentage of zygotes reaching the blastocyst stage after pronuclear microinjection with DNA at a concentration of 1.5 ug/ml was significantly lower (p<0.05) than control embryos (28% versus 74%, respectively). The overall transgenic pup production efficiency (percent of transgenic pups per embryos transferred) by cytoplasmic injection was 2.4% compared to 3.5% by pronuclear microinjection. / Ph. D.