Branching at, or near, the tips occurs and branches from adjacent columns make contact, forming loops.Loop formation is followed by the appearance of a lumen within the loop after which blood flow commences. A newly formed capil lary loop may generate further sprouts of endothelial cells and, eventually, fresh capillary loops so thatseveral tiers of loops may form resulting in a network of capillaries.Deposition of basement membrane on the abluminal side of the endothelial cells follows the estabiishment of blood flow in the capillary loop. It is clear from the described sequence of events thatthere is in the development of new capillary sprouts a way in which angiogenesis may further contribute to metastatic development.Thatis the broaching of an intact basement membrane, so important in transition from carcinoma in situ to frank carcinoma, need not be a consequence of tumour invasionperse.Rather the movement of new vessels through this structure, under the influence of an angiogenic stimulus, using many of the same enzymes as those used by invading tumour cells, could lead just as readily to membrane dissolution and the possibility for neoplastic dissemination as could movement in the opposi tion direction by tumour cells. With the availability of cultured capillary endothelial cells and the realisation that the stages of endothelial cell proliferation and migration could be monitored invi tro, there was fresh impetus in to the search for angiogenic factors.By the early s several polypeptide factors from a variety of normal, as well as tumour, tissues were shown to be capable of stimulating endothelial cell proliferation.Such factors in clude: heparinbinding endothelial cell growth factors, as exemplified by acidic and basic fibroblast growth factors, the polypeptide angiogenin, which was first isolated from conditioned medium of the HT hu man colon carcinoma line and has been shown to have sequence homology with pancreatic ribonucleases, tumour necrosis factor a another cytokine released by macrophages, plateletderived endothelial cell growth factor a protein of relative molecular mass of about which seems to be the sole endothelial cell growth factor in human platelets, heparin which, Yohimbine hydrochloride though in itself does not seem to be angiogenic per se, is capable of enhancing tumourinduced angiogenesis and a variety of miscellaneous factors such as prostaglandins E, and E. These authors pointed out thatthe various angiogenic factors either did or did not have an effect on the proliferation and motility of capillary endothelial cells invi tro. Those which stimu lated motility or mitosis of such cells invitro, they pro posed, had the vascular endothelial cell as their primary target in viuo, whereas factors which had no effect on these cells invi tro probably acted via some indirect pathway in vivo. This indirect pathway they felt could be via an activation and stimulation of infiltrating host leukocytes, could be via release of endothelial mitogens from storage in the extracellular matrix or could be by the release of intracellular Hematoxylin stores of endothelial growth factors. This concept of the possible duality of action of angiogenic factors, which need not be exclusive in particu lar tumours, raises an important point.