582 Transactions of the Society. 



membranes of embryos or newly born animals. Both of these 

 histologists described these cells as ultimately becoming connected 

 with or forming part of the circulation, but the former has specially 

 studied them in this relation and given to them the name they now 

 bear of " vasoformative cells." 



It is unnecessary for us at present to enter into his arguments 

 for considering these structures as connected with the development 

 of blood-vessels, as we are more concerned with certain remarks 

 which he makes at page 633 of his ' Traite d'Histologie,' where he 

 advances the hypothesis, only to reject it, it is true, that having 

 regard to the great changes continually taking place in the circula- 

 tion of the embryo, and after birth more especially in connection 

 with the obliteration of the branchial arches, the ductus arteriosus, 

 &c., it may be plainly argued that those cavities containing 

 blood-corpuscles are really portions of the circulation becoming 

 obliterated, and that the intercepted portions are really parts of 

 pre-existing capillaries. 



However startling such a hypothesis may appear at first sight, 

 we are surprised to find Professor Eanvier reject it as being 

 inapplicable, at all events, to most of the examples he has 

 studied. For our part, we subscribe fully to it, not as a hypo- 

 thesis, but as a fact, for there is certainly in this research no fact 

 easier of demonstration than it is. Our reasons are the followinof : 



O 



In the first place, there is complete identity between these vaso- 

 formative (so-called) cells containing blood-corpuscles, and the 

 intercepted portions of retrograding capillaries containing blood- 

 corpuscles, in animals where nutrition has been insufiicient. In 

 the second place, if one studies the omentum of newly-born kittens, 

 as recommended by Kanvier, nothing can be clearer than the fact 

 that retrogression and develoj)ment of blood-vessels are going on 

 side by side, and that the two processes are so distinct that there 

 is scarcely any possibility of confounding the one with the other. 

 We give a drawing of such an example, Fig. 13, where retrograding 

 and developing vessels are lying parallel and close to each other, 

 so that a glance ought to be sufiicient to distinguish between them. 

 In this camera lucida drawing a, a! are terminations of branches still 

 in connection with the original channels of the chculation, and 6, h' 

 are what are called vasoformative cells, containing blood-corpuscles, 

 but which in reality are portions of the capillary which originally 

 stretched from a to a', and are now identical with Figs. 15, 17, 

 and 19, from a rat which died of old age and inanition. In Fig. 13 

 we also see a new vessel c, from which three new branches d, d', d", 

 are being developed, in conformity with the process we described 

 in the early part of this paper. 



It is therefore evident that the physiological retrogression of 

 blood-vessels follows the same course, whether it be due to insuffi- 



