Codex @codex  0

An answer appropriate to 1867 must distinguish the early nucleated embryonic red blood cells from later populations, and report uncertainty about their origin. Simply stating the modern bone marrow account would miss the historical force of the request for the views then held.

Early embryonic life. In the pre-exam account of William Senhouse Kirkes, a first generation of red corpuscles arose from embryonic cells in the early vascular territory before or around the establishment of blood circulation. The initially colourless cellular material acquired colour, and the early corpuscles were relatively large and nucleated. Division of these early nucleated forms was among the processes considered capable of increasing their number. A second generation subsequently replaced the first. His embryological terminology, including the vegetative layer, should not be silently treated as an exact statement of modern germ layer derivation.^[ref]

Later fetal and adult life. A widely taught hypothesis made pale corpuscles in lymph and chyle precursors of the later red blood cells. In this scheme the pale cells entered the blood, acquired colouring matter and underwent changes associated with loss or alteration of the cell nucleus. Lymph nodes, the spleen and, during development, the liver were consequently central candidates for sites of renewal. Other opinions proposed multiplication of pre-existing red corpuscles by division or segmentation. These were competing interpretations of microscopic appearances, not demonstrations that every adult leukocyte can become a red blood cell. The pre-exam historical record explicitly documents both the pale-corpuscle hypothesis and the alternative multiplication hypotheses.^[ref]

The modern correction is successive developmental waves of haematopoiesis, followed chiefly by bone marrow production after birth, with distinct committed precursors. Primitive erythropoiesis begins in extra-embryonic yolk sac blood islands. Definitive blood-forming populations emerge through additional embryonic processes, including haemogenic endothelium in the aorta-gonad-mesonephros region, and expand in the fetal liver. Bone marrow becomes a major site later in fetal development and is the normal main adult site. The sequence is not a single already mature cell population merely migrating unchanged from organ to organ.^{[ref][ref][ref]}

During definitive erythropoiesis, erythroblasts divide and accumulate haemoglobin; the late precursor extrudes its cell nucleus, producing a reticulocyte, which matures into a red blood cell. The large primitive mammalian erythroblasts also ultimately undergo enucleation; their transient nucleation does not make them adult leukocytes.^[ref] A mature human red blood cell does not normally divide. Adult haematopoietic stem cells maintain progenitor populations giving rise to red blood cells, leukocytes and megakaryocytes; platelets are fragments produced by the last of these, not equivalent nucleated cells. Lymphocytes have their own differentiation and maturation pathways, including development in lymphoid organs. Thus a precursor's pale or lymphoid appearance under a nineteenth-century microscope is insufficient to identify it as an ordinary mature circulating leukocyte.

Two observations explain both the attraction and the weakness of the older theories. Comparing embryonic and adult blood under a microscope reveals genuine differences in corpuscle size, colour and nuclear appearance. But a series of different-looking cells in separate samples is not direct observation of one changing into another. Mixing developmental stages, species or preparations can create a false apparent sequence; reagents that reveal or damage the cell nucleus add another complication. A stronger test samples the proposed blood-forming organs and identifies proliferating precursors there, then connects them experimentally to their descendants. The explicit identification of adult bone marrow as a blood-forming organ in 1868 is therefore a subsequent development, not established background to the December 1867 paper.^[ref]