The feasibility of a microencapsulation process ultimately for cell transplantation was investigated by cncapsulating mammalian cells in hydroxyethyl methacrylate-methyl methacrylate capsules using an interfacial precipitation process. A simple colorimetric assay, based on cellular transformation of 3-(4,5-dimethylthiazol-2- yl)-2,5-diphenyl-tetrazolium (MTT), was adopted to assess metabolic activity and growth of encapsulated Chinese hamster ovary (CHO) and human hepatoma (HepG2) cells. Both CHO and HepG2 cells were shown to survive the encapsulation process despite exposure to shear forces and organic solvents/non-solvents and remain viable within the capsules. The encapsulation efficiency for robust CHO cells (65%) was higher than for the more sensitive HepG2 cells (20-50%). Both cell types were observed to attach to each other and grow within cellular aggregates. CHO cells in aggregates retained their ability to proliferate for an initial 1 week period after which no further growth was observed. HepG2 cells in aggregates, however, initially retained their metabolic activity and/or proliferated until ~1200 cells/capsule was reached, but eventually underwent a significant loss of viability. Encapsulation of a biological attachment substrate, Matrigel®, along with the HepG2 cells, however, resulted in rapid proliferation of cells (until ~3500 cells/capsule). The secretion of selected proteins, α₁-acid glycoprotein, α₁-antitrypsin, haptaglobin and fibrinogen, was demonstrated, indicating the retention of differentiated state by the microencapsulated cells. There was a large variation, however, in protein secretion among the individual capsules, with ~20% of capsules not secreting detectable amount of α₁-antitrypsin and ~70% of capsules not secreting fibrinogen, after 14 days of culture. This heterogeneity appeared to be due to a variation in cell number and in sieving effect of the capsule membrane. Implantation of Matrigel® encapsulated HepG2 cells into rats resulted in the delivery of α1-antitrypsin for only short duration (~1 week) with a gradual increase in human α1-antitrypsin-specific antibody levels in rat serum during the 4 week study period. Explanted capsules contained a significant population of necrotic cells, as assayed by toluidine blue staining of capsule cryostat sections. The relative contributions of cell death and antibody generation to the lack of α₁-antitrypsin delivery in vivo remains to be determined.