Recent investigations of the birth changes in the foetal circulation indicate that certain channels, namely, the ductus arteriosus, ductus venosus and extra-abdominal umbilical vessels, are quickly closed by the contraction of their muscular walls (Barclay, Franklin and Prichard, 1944). This immediate functional closure is followed by an anatomical obliteration which takes place during the first few weeks or months of life (Scammon and Norris, 1918; Kennedy and Clark, 1941). Little appears to be known regarding the functional response to birth changes of the intra-abdominal umbilical vein, but Barclay et al. (1944) note the following facts. First, that there is a marked difference between the structure of the intraand extra-abdominal parts of the umbilical vein: the former is thin-walled with only a small amount of smooth muscle, whereas the latter has a very thick muscular wall. Second, that the intra-abdominal umbilical vein is much reduced in calibre and contains little or no blood when examined some time after rupture of the cord. Sidbury (1923) first used the umbilical vein as a transfusion route (in infants suffering from icterus gravis neonatorum) by inserting a needle into the extra-abdominal umbilical vein just beyond the ligature on the cord. He found no clot in the vein and attributed this to the delay in clotting which is a characteristic of this condition. Diamond (1947) elaborated this method by severing the umbilical cord, some half-inch from the abdominal wall, and inserting a polythene catheter of 1 mm. internal diameter directly into the intra-abdominal umbilical vein. This technique is now the standard one for performing exchange transfusions in icterus gravis neonatorum and may be employed for as long as six days after birth (Diamond, 1954). When the cord is severed the vein occasionally bleeds violently, but 'if, as is usual, the vein does not bleed at all, it can be identified as a large patulous vessel which contrasts with the two smaller, tightly contracted arteries' (Mollison, Mourant and Race, 1948). The catheter meets no resistance until it impinges upon the far wall of the left branch of the portal vein and it may even be introduced into the ductus venosus and so into the inferior vena cava. Much information is available regarding the mode of anatomical obliteration of the intra-abdominal umbilical vein and the formation of the ligamentum teres hepatis (Robin, 1860; Wertheimer, 1886; Baumgarten, 1891; Butler, 1951). The account by Robin (1860) does not depict the relevant histological changes, so that interpretation of the obscure text becomes well-nigh impossible, and it is probable that sepsis was present in much of his young material. Wertheimer (1886) describes the formation of a central core of fibrous tissue completely filling the lumen of the vein. The obliterated lumen becomes surrounded by a musculo-elastic ring, while capillary formation in the central core of the fibrous tissue gives the spurious appearance of a tiny remnant of the original lumen of the vein. It is important to note that the youngest specimen described was from an infant aged 5 months. Baumgarten (1891) states that the lumen of the umbilical vein is not completely obliterated, and that a much reduced remnant, the Rest-Kanal, is to be found in most individuals. This opens into the left branch of the portal vein and extends for a varying distance along the ligamentum teres. Segall (1923) demonstrated this small vessel in 40% of injected adult human livers. In my opinion (Butler, 1951; 1952) the latter view is the correct one and is confirmed by the following observations, made on very young specimens. These observations also show why it is possible to use the intra-abdominal vein as a transfusion route in the early days of life.