66 ELIOT R. CLARK 



scribed for vessels elsewhere. In order to observe the fate of 

 such a capillary I completely isolated a small section of a capil- 

 lary by cutting through its connections. Much to my surprise, 

 this isolated capillary was found, two day.s later, to have formed 

 an anastomosis with one of the circulating capillaries (figs. 12). 

 Thus a capillary which has become isolated, does not lose its 

 blood-vascular-endothelial properties, and may be reincor- 

 porated in the vascular system. It is perhaps worth noting 

 that this vessel was in a region, near the margin of the fin, where 

 active new formation of vessels was taking place. 



What, then, are the factors which lead to the regression of 

 capillaries? In keeping a record of the developing vessels, a 

 record was also kept of the presence or absence of circulation in 

 each capillary, and of the approximate rate of the circulation. 

 On looking over the previous records of the circulatory con- 

 ditions of any vessel which has undergone retrogression it is 

 found that this process is preceded by a period in which the 

 circulation has ceased. This period is usually preceded in turn, 

 by a period in which the circulation has diminished in quantity. 

 This latter, of course, applies only to the vessels in which a cir- 

 culation has been established, for, as has been said, some ves- 

 sels are withdrawn before any circulation has been established 

 in them. Thus many definite records have been obtained in 

 which the retrogression of a capillary has been associated with 

 the stoppage or absence of circulation. The conclusion, there- 

 fore, seems justified that a vessel which is connected with the 

 rest of the circulating system of vessels, retrogresses and dis- 

 appears if the circulation within it ceases for a sufficient length 

 of time. 



The finding that the diminution in size of lumen which pre- 

 cedes the retraction of capillaries involves the property, on the 

 part of the endothelium, of reacting to the amount of blood flow- 

 ing through the vessel, agrees, in part, with Thoma's first histo- 

 mechanical law. According to this law, however, it is the rate 

 ('Geschwindigkeit') of blood flow which is the determining fac- 

 tor ('93, p. 37 ff). In these studies it appears that, in capillaries, 

 at least, it is not the rate but the amount of blood flow which is 



