SUMMARYPigs were infected by the intranasal instillation of a large dose (ca. 107·0 ID 50) of a highly virulent strain of African swine fever virus (ASFV) and the progress of the infection was studied by the ‘routine titration approach’ (Mims, 1964) using pig bone marrow cultures.Virus growth was established within 16–24 hr. in the retropharyngeal but not in the alimentary or nasal mucosae or the tonsils. By 24–40 hr. the virus was consistently present in the retropharyngeal lymph nodes, almost invariably the medials; titres in these nodes exceeded those in the associated mucosa by 48–72 hr. Generalization, presumed to have occurred via the tracheal lymph ducts and the blood stream, was generally demonstrable after 72 hr., i.e. by the time of the onset of pyrexia or 24 hr. prior to this.On average 11% of the total infectivity in the blood was present in the plasma, with the rest assumed to be cell-associated. A mean of about 45% of the total infectivity was recovered in erythrocyte fractions in which the concentration of leucocytes had been materially reduced; fractions with increased leucocyte counts contained relatively little virus and it was concluded that the great majority of circulating virus was closely associated with the erythrocytes. Adsorption of ASFV to normal pig erythrocytes was demonstrated in vitro.The greatest concentrations of virus were recorded in the lymph nodes, especially those of the cephalic region, and in the spleen, where titres commonly attained 108·0 to 109·0 HAD 50/g. and exceeded those in the blood. They were, therefore, thought to be the source of much circulating virus, although there was some evidence that the liver, lungs and bone marrow may also have contributed, at least in some animals. There was no evidence that the mucosae of the alimentary and respiratory tracts or the kidney, myocardium and brain were a source of significant amounts of virus. The virus demonstrable in Peyer's patches did not exceed that in the intervening ileal mucosa.Although contact transmission of ASF does not normally occur during the first 12–24 hr. of fever, considerable amounts of virus were usually present in the nasal and intestinal mucosae at 72 hr. It was probable that this infectivity was due to the blood content and that excretion did not occur until the epithelium was breached.Three pigs, all of which had lesions of a portal cirrhosis, showed a delayed or restricted generalization of virus, in comparison with the other twenty-eight animals which behaved according to a regular pattern.We are grateful to Mrs M. O'Sullivan and Mrs E. Douglas for painstaking technical assistance and to Mr C. S. Rampton, A.I.M.L.T., for the preparation of the figures.