8 days after the grafts had been applied considerable numbers of large basophilic cells, apparently identical with those that appear during immune responses to conventional antigens, appeared in the lymph draining from the homografts. No such cells appeared in the lymph draining from the autografts. At this time the homografts were already showing signs of rejection and were apparently dead well before the cellular response in the lymph reached a peak, about 350 hr (14–15 days) after the homografts had been applied. During the peak of the response up to 40% of the cells in the lymph were basophilic cells and in one experiment such cells were leaving the lymph node at a rate of 200 million per hr.

Peripheral lymph (afferent to the popliteal lymph node) draining from the sites of homografts of skin was collected from five sheep. This lymph contained few white cells (<1000 per mm³) and showed only an insignificant lymphoid cell reaction. Although the percentage of macrophage-like cells was increased significantly there were few signs of a lymphoid cell reaction; the lymph also contained much amorphous debris.

Experiments in which the basophilic cells from the efferent lymph were labeled in vitro with thymidine-³H and returned to the sheep by intravenous injections were carried out in six sheep. The presence of the labeled cells in the grafts, blood, and other tissue was detected by liquid scintillation counting of nucleic acid extracts of biopsy and postmortem material and by radioautography. 2–3 labeled cells out of every 1000 injected entered the homografts but hardly any entered the autografts. However, labeled basophilic cells that had originated in response to bacterial antigens entered the homografts with equal facility. It is thus hard to believe that the immunological specificity of a lymphoid cell endows it with a specific "homing" capability. Furthermore, in all the experiments the specific radioactivities of the nucleic acids extracted from the blood mononuclear cells were approximately of the same order as those of the nucleic acids extracted from the homografts. It was concluded that most of the mononuclear cells that infiltrate homografts represent a random selection from the mononuclear cell population of the blood.