Canine abdominal aortas have been replaced with Dacron arterial prostheses to assess the effects of mesothelial cell seeding on graft prostacyclin and thromboxane A2 release. At both 2 weeks and 6 weeks after surgery, three seeded and two unseeded control grafts were examined for prostacyclin release. In addition, thromboxane release was assessed in one seeded and one unseeded graft. Sections of aorta and graft were removed and incubated in PBS containing either 10 microM calcium ionophore A23187 or 20 microM arachidonic acid. The incubation mixture was sub-sampled at 5 min intervals over a 20 min period to assess the progressive release of prostacyclin and thromboxane A2 using a radioimmunoassay for 6-keto-prostaglandin F1 alpha and thromboxane B2 respectively. In seeded grafts, 6-keto-prostaglandin F1 alpha release averaged 15 per cent compared with aorta at 2 weeks and 45 per cent compared with aorta at 6 weeks. By contrast, release from unseeded grafts was undetectable at 2 weeks; however, by 6 weeks there was some release amounting to 15 per cent compared with aorta. There was a statistically significant increase in the release of 6-keto-prostaglandin F1 alpha from mesothelial cell seeded grafts at 6 weeks compared with unseeded grafts (P less than 0.01). Thromboxane release from the graft sections was variable and unrelated to whether the grafts had been seeded or not. These preliminary results, showing that grafts seeded with autologous peritoneal mesothelial cells release more prostacyclin than unseeded grafts, further highlight the role of the mesothelial cell as an alternative to the endothelial cell for improving the patency of arterial Dacron prostheses in the early postoperative days.