it may be said the intervillal space (placental cavity) is continuous with the space intervening between chorion and inside of decidua reflexa all round the ovum.

In any case of pregnancy, supposing there was a placental cavity formed from the maternal vascular system, as Hunter maintains, and that this outer limit, supposing it also to exist, were lacerated, the blood must necessarily be poured out between ovum and decidua; and then, supposing the decidua lacerated, it would be difficult to understand how the patient should not succumb to hæmorrhage. But if there was no blood, but simply a transparent fluid, then the loss of a small portion daily is but of moderate moment; and, indeed, this would serve to explain the discharge of fluid which sometimes takes place during pregnancy, where the amnial membrane has not been ruptured, the occurrence of which fluid has been hitherto difficult to explain.

In relation to this outermost membrane of a sinus-system there is an assertion of some anatomists, as Van der Kolk, &c., which requires notice, namely, that the lobules are separated by septa from one another, and they instance in proof that injections thrown into the cavity of the placenta do not run everywhere, but are restricted to certain portions. Now, although I must, as before mentioned, refuse evidence of this kind as a proof, still I may answer that the like result has not attended my experiments; I have always invariably found that the injected fluid permeates all the spaces between the villi equally in all directions. This experiment I have tried very many times, always with the same effect; there was no evidence of any limit in any way.

From the consideration of these facts I feel warranted in asserting that from the beginning to the end of pregnancy there is no evidence of the sinus-system being in a transition state. It must either exist ab initio or not at all.

The discussion of the latter alternative will now occupy our attention.

III. That from dissections early or late in pregnancy there is no evidence of the existence of a sinus-system. Although

the placenta is generally said to be distinctly formed at about the end of the second month of pregnancy, yet in reality there is no definite period at which it can be said to be formed, because if we carefully examine the ovum at a much earlier period we notice the villi at a particular part are larger than the rest, so much so that even a month after conception it may be known at what spot the future placenta will appear. From that time till the completion of full term careful observation has failed to show me any essential difference in its structure, excepting in the addition of the decidual processes. It will also be noticed that the total amount of original villi included in this change is not so great as one might suppose from our inspection of a fully grown placenta. The area of the placenta seems, from numerous observations, to be determined by the extent of the allantois, although I have found one ovum where the vessels of the allantois were only to be noticed on that part of the ovum which had no villi.

It is a matter yet of uncertain explanation why the allantoic aspect of the ovum should nearly always be applied to the surface of the uterus on which it rests; but though it is not a part of the present paper to inquire into this subject, it may be remarked that possibly abnormalities in the adaptation of the allantois to the portions of chorion well covered by villi, and in the apposition of that facet of the ovum to the uterus, may be the source of some abortions.

Now, if we should find instances in which no blood is present amongst the villi, or, in other words, that the interior of the placenta does not furnish a trace of blood after the most careful examination, it appears to me that such evidence would be conclusive against the existence of the sinus-system; because, for the reasons I have already pointed out, one such fact is worth a hundred arguments to the contrary. For I may again repeat, when we consider the great number of chances there are against any cavity similar to the placenta being expelled from the uterus by compressive force, having one side porous, it would seem almost impossible for blood not to enter and not to be diffused by capillary action and

the compression of the uterus throughout the whole intervillal space. And, again, when we consider the tenderness of the walls of the dilated maternal vessels of the decidua and their liability to be ruptured under uterine force, under states of hyperæmia, shocks, &c., it is a wonder that the placenta is ever expelled without containing blood, and this would explain why it is so commonly the case that the intervillal space contains blood after expulsion.

I have already alluded to the comparative absence of blood in those placenta which are the most perfectly free from openings in their decidua. If in objection to this it is remarked that the blood might have been expressed during expulsion, then it ought to be borne in mind how very unlikely it is that that blood, once within, could be expelled so fully. But a much more complete answer to the objection would be given if it could be shown that when the placenta is examined in sitú no blood can be found in the intervillal space. If to this it were also objected that the blood had flowed from the sinus-sytem into the veins after death, the answer would seem sufficient that it would be next to impossible that blood should flow from these sinuses (if they exist), and yet remain in every other vein of the uterus, which is notably the case.

And even if we were to admit that no sinus-system existed, yet, considering the tender nature of the vascular walls surrounding the cavity, as I assert, it should not be surprising if blood were occasionally to appear in the intervillal space whilst still attached to the uterus, after violent concussions, in cases of hæmorrhagic tendency, or in the throes of death, or in the efforts of labour.

And this brings me to the consideration of an argument used by those who have supported the sinus-system, namely, that it is impossible to conceive the action of endosmosis to go on in the placenta at the extreme distance from the maternal surface, so that they argue the villi which were near the chorion would be useless. But the worthlessness of this argument is apparent at once when we consider that when the blood has once entered among the villi its position differs