Nearly half a billion dollars in resources are lost each time a drug candidate is withdrawn from the market by the Food and Drug Administration (FDA) for reasons of liver toxicity. The number of late-phase drug developmental failures due to liver toxicity could potentially be reduced through the use of hepatocyte-based systems capable of modeling the response of in vivo liver tissue to toxic insults. With this article, we report progress toward the goal of realizing an array of primary hepatocytes for use in high-throughput liver toxicity studies. Described herein is the development of a 64 (8 x 8) element array of microfluidic wells capable of supporting micropatterned primary rat hepatocytes in coculture with 3T3-J2 fibroblasts. Each of the wells within the array was continuously perfused with medium and oxygen in a nonaddressable format. The key features of the system design and fabrication are described, including the use of two microfluidic perfusion networks to provide the coculture with an independent and continuous supply of cell culture medium and oxygen. Also described are the fabrication techniques used to selectively pattern hepatocytes and 3T3-J2 fibroblasts within the wells of the array. The functional studies used to demonstrate the synthetic and metabolic capacity of the array are outlined in this article. These studies demonstrate that the hepatocytes contained within the array are capable of continuous, steady-state albumin synthesis (78.4 microg/day, sigma = 3.98 microg/day, N = 8) and urea production (109.8 microg/day, sigma = 11.9 microg/day, N = 8). In the final section of the article, these results are discussed as they relate to the final goal of this research effort, the development of an array of primary hepatocytes for use in physiologically relevant toxicology studies.