In a series of tissue cultures of fragments taken from different parts of young chick embryos, at various stages, it was found that differentiation frequently occurred, the characteristic cells of highly specialised tissues appearing in cultures in which they could not possibly have been present at explantation. The most striking differentiations were the development of red blood corpuscles, capillary vessels, nerve cells with axons and of rhythmically contracting cardiac muscle. This paper describes a simple method by which may be obtained regularly and quickly the development of very large numbers of red cells, and an account is given of the histology of the cultures. It is intended to be introductory to a physiological study, now in progress, of the conditions of hæmatopoiesis Literature. Erythropoiesis
in vitro
has been reported by several authors, but it has not been thoroughly investigated and, with the exceptions of the works of Slonimski (1930,
a
, 1931) and Shipley (1915-16), the earlier papers have concerned the somewhat sporadic appearance of small numbers of erythrocytes. In the earlier works, also, the explants were derived from hæmatopoietic organs, or from that part of the embryo in which blood would normally have developed at latest quite soon after the time at which the experiment was made. The present paper, on the other hand, is based principally upon cultures of fragments of the primitive streak—that is, upon explants of presumptively hæmatopoietic cells isolated before they had arrived at the normal hæmatopoietic region of the embryo. The literature may be very briefly summarised as follows : Shipley (1915-16) made plasma cultures from the
area opaca
of chick embryos at a time prior to the formation of the blood islands, and obtained the differentiation of erythrocytes from amœboid cells. Erythropoiesis is reported by N. G. and A. L. Ghlopin (1925) in cultures of Axolotl spleen; by Erdmann, Eisner, and Laser (1925-26), in cultures of embryonal rat spleen; by Freifeld and Ginsburg (1927) in cultures of rabbit adrenals; by de Haan (1928-29) in cultures of blood cells of the horse; by Timofejewsky and Benewolenskaja (1929) in cultures of blood from a case of acute myeloid leukaemia, and by Benewolenskaja (1930) in cultures of embryonal human liver. Slonimski (1930
a
, 1931), using
Rana fusca
and Axolotl embryos, excised the blood island zone at early stages, and kept it as a culture enclosed in a sheath of epiblastic epithelium. The little cyst became full of red blood in an abundant plasma, and there were vessels with endothelial walls.
The development
in vitro
of the blood of the early chick embryo