Although vascular occlusion has long been noted in peri-burn tissue, the literature is inconsistent regarding the nature of the occlusion, with articles in the 1940s claiming that erythrocytes were the culprit and in the 1980s–1990s that microthrombi were responsible. To better define the nature of vessel occlusion, we studied two porcine burn models, a hot comb horizontal injury model and a vertical injury progression model. In both cases, tissue from the first two days after burn were stained with hemotoxylin and eosin, or probed for platelets or for fibrinogen/fibrin. Erythrocytes, identified as nonstained, clumped, anuclear, 5 μm cells, occluded most blood vessels (BVs) in both burn models. In contrast, platelet or fibrinogen/fibrin antibodies stained BV occlusions minimally at early time points, and only up to 16% of deep dermal BVs at 48 hours in the hot comb model and up to 7% at 24 hours in the vertical injury progression model. Treatment of animals with a fibronectin-derived peptide (P12), which limits burn injury progression and can dilate peripheral microvasculature, reduced erythrocyte occlusion by at least 50%, speeded healing and reduced scarring. Early erythrocyte aggregation, rather than thrombosis, explains the ineffectiveness of anticoagulants to prevent burn injury progression.