Materials and Methods Mice of the Prince Henry strain were used. The mice were inoculated at birth with a 1 (2 dilution of the stock solution of hepatoencephalomyelitis virus (the prototype of reovirus 3). The inoculum was given intraperitoneally. A 1 (2 dilution was chosen, since this proved the LD50 dosageat the eighth day of infection. Four mice were killed daily between the first and ninth days of infection. Small portions of liver were immersed in cold 1% osmium tetroxide in phosphate buffer at pH 7.4. The bile duct was then flooded with6. 25% lutaraldehyde and fixed in situ. With the aid of a dissecting microscope and while the tissues were still submerged in glutaraldehyde, the bile duct was carefuIly isolated and followed to its junction with the duodenum. The duodenumwas then opened, the site flooded with more glutaraldehyde, and the ampulla of Vater identified and carefully disscted away from the remainder of the duodenal tissues. In most casesthe bile duct and the ampulla of Vater were not severed from each other. The bile ducts and ampllue were fixed for 24 hr. in 6.25% glutaraldehyde in phos-phate buffer then washed for 24 hr. in buffer. Portions of all bile ducts and two of the four ampulae col n each successive day of the experiment were then postfixed for 60 mmi in 1% osmium tetroxide in phosphate buffer. ITe tisses were then dehydrated in graded solutions of ethanol and embedded in Araldite. Sections were cut on a Huxey microtome, sing glass knives, and picked up on copper grids.