Haemorrhage remains the leading cause of potentially survivable death on the battlefield, and whole blood transfusion close to the point of injury is now established as the resuscitation of choice. The doctrine that surrounds it, however, was written for the golden hour, a world in which a casualty reaches surgery within sixty minutes, and that world is disappearing. In large-scale combat operations against a peer adversary, evacuation may be delayed for many hours, and the medic who gives a single unit of whole blood and waits for a helicopter that does not come is left without a plan. This article develops that plan. Its original contribution is the Point-of-Injury Whole Blood protocol for Prolonged Field Care, a phased and titratable resuscitation algorithm built specifically for delayed evacuation, together with the underlying argument that the prolonged scenario requires four departures from golden-hour practice: a shift from a single fixed bolus to a titratable resuscitation governed by a sustained permissive perfusion endpoint; a serial walking-blood-bank cycle that turns a finite blood supply into a renewable one; an explicit time-indexed monitoring loop for the metabolic derangements, hypocalcaemia, hypothermia, and acidosis, that accumulate over hours of transfusion and that short-evacuation protocols can neglect; and the integration of these into one decision algorithm a far-forward provider can follow without laboratory or imaging support. The protocol is grounded in an evidence synthesis, presented as a narrative forest plot of mortality estimates from recent trials and cohorts, that supports the efficacy of whole blood and early transfusion while acknowledging the null findings that temper it. The protocol is presented as two figures and two tables: a clinical algorithm, a forest plot of the supporting evidence, a comparison of the golden-hour and prolonged-care paradigms, and a time-indexed schedule for the accumulating metabolic threats. The protocol is an evidence-informed design and not a validated standard; it generates specific, testable predictions that the article specifies for prospective field evaluation.