a which is channeled in the walls of the gravid \\,m\l) does not consist of 

 veins or arteries enlarged for the purpose, but, on the contrary, that the proper 

 substances of the womb itself (muscles and connect he tissue i compose the actual 

 walls, which are lined by a tleliente uieinbnine answering to the lining mem- 

 brane of the vessels. The difference is very material, since the vessels are 

 enabled to expand ami contract themselves for increasing circulation, while in 

 the ca>'' of the sinuses, the walls of the womb would have to expand and contract 

 in order to HIKW- circulation in them, mid that this is the case will appear more 

 obvious as we proceed, the mechanics undoubtedly calling for such a condition. 



When a longitudinal section is made of a gravid womb at full term, this canal 

 ^v-teni is seen to traverse its substance in various directions from near the central 

 IM nt ions, but trending to the internal or mucous surface where they connect with 

 the placenta (Fig. 3 '2, 3). They never penetrate to the external surface to 

 form continuous tubes with the arteries or veins; but on the contrary, both 

 these systems of vessels connect with this system of canals by means of capil- 

 Itirit *, the arterial emptying and the venous discharging from these common 

 mains. Hence this blood is mixed as a matter of necessity. 



The next important point is the method this system of canals exhibits in con- 

 necting with the placenta! sinuses, which is done by projecting the lining mem- 

 brane in/') l/i'' t>>(lixtan<-e of the placenta (Fig. 4, c, c, c, c). This spreads out at once 



FIG. 4. VKIUII u. SI:CTHI\ OK I'I.ACF.NTA, showing arrangement of maternal ami f<rtul vessels. 

 A, n, chorion; l>, >>, ileciilua; 0, < . . . . , urillres of uterine sinuse.j. (After Dalton.) 



to form the large cavitary spaces in which the highly ramified placental tufta 

 are suspended and submerged in blood. 



The walls of these large cavities in which the lining membrane of 

 the uterine sinu>es is merged answers to the outer surface of these highly-rami- 

 fied tufts which tit into this as the fingers in a glove, the vascular loops being 

 contained inside of them (Fig. 5, a, a). It is easy to perceive that this out- 

 side wall and the wall of the capillary itself in// rr< n<- lx>tween the. maternal and 

 foetal blood, and that interchange would have to be effected through these two. 



