M, muscular wall with ends of glands (de) in it; Dv, decidua vera divided into compact (Comp.) and spongy (Amp.) layers; lc, leucocytes in the lymphatics. D c, decidua reflexa or capsularis. In this, up to drep, gland spaces are seen; above ii it consists mainly of fibrin. The ovum b b is resting on a spur of decidua. In the space between it and the reflexa are chorionic villi and processes of decidua; and into this space a vessel opens at c. Opposite h, a villus is becoming rooted.

the mucosa has the appearance which we have come to associate with the early months of pregnancy. There is a compact and spongy layer, the former showing the drawn out necks of the glands, the latter their convoluted bases. The new point brought out by this section is that these changes are present as early as the first week of pregnancy. It shows how rapidly the growth of the mucosa takes place that we should find it hypertrophied to this extent within ten days of a menstrual period. The ovum rests within the reflexa on a fold of mucosa 'like a ship resting on a ridge of rock.' There is no epithelium on the top of the ridge though there is on its lower slope. Dilated gland spaces with cylindrical epithelium are seen in this fold, and into the mouths of one of these glands a villus has extended. At one other point there is an appearance as if villi had grown into the glands, but as this is seen nowhere else, its occurrence is accidental. The villi are usually applied to the surface of the mucosa and united to it by a layer of fibrin. The reflexa is thick below, but it thins towards the upper pole of the ovum. Glands are present, provided with epithelium at its lower part, and dilated capillaries, but as it thins its structure becomes indistinct. The chorion is closely blended with it, and while cells are seen here and there of decidual character, it is chiefly composed of fibrin. Within the reflexa are seen villi cut in various directions. They have a single layer of epithelium. In their stroma, vessels may be seen beginning to form. A young villus appears as a wart of epithelium-' a heap of syncytium,' within which connective tissue and a capillary appear later.

We also note heaps of decidual cells forming between the villi, which are the beginnings of the trabeculæ of decidual cells which are found in the fully-formed placenta. The ends of the villi stimulate the decidua to proliferate. Thus arise the heaps of cells which cause a general opening out and loosening of the superficial layer of the decidua, and at the same time a dilatation and bursting of the capillaries.' According to Leopold, 'the ends of the villi are continually shooting out tendrils winding in all directions. From this period they are already washed with maternal blood from the open capillaries, and open up still more these vessels, so that the blood now flows unhindered between the villi up to the chorion. We may satisfy ourselves of this relationship best when we follow under the microscope different sections, from the capsularis of the one side across the serotina to the capsularis of the other. First we observe between the villi a structureless, partly granular, partly crumbling mass, which contains quite distinct fine round pale red structures, viz., red and white blood corpuscles. Although those elements are not present in quantity we can make no mistake in regarding them as lymph corpuscles, and the more that they are found especially at the border of the decidua. And just here we see at different points

quite plainly, although the preparation was not injected, how the surface capillaries open with their mouths into the spaces between the villi. It is thus demonstrated with certainty that the intervillous spaces are supplied with maternal blood from the earliest time of development. The italics are Leopold's.

The second preparation of a fourteen days' ovum embedded in the mucosa is not of the same value from the absence of exact clinical data. It was obtained from a suicide, and on inquiry it was ascertained that one period had been passed, but no further information was gained.

The chorionic vesicle is 14 mm. broad, 12 mm. long, and 5 mm. high, while that of the previous case is only 37 mm. broad and 4 mm. high. In the mucosa we note that the gland spaces are filled with irregular masses of swollen epithelium, and that the capillaries are greatly dilated both in the serotina and lower and middle portions of the reflexa, so dilated that 'it raises the question whether such an unusual dilatation of them may not be pathological.' At one point they appear almost as a lake. These spaces are lined by a distinct endothelium.

The reflexa here forms a complete envelope; its upper pole, which has no epithelium, being 25 to 5 mm. thick. It contains decidual cells similar to those found under the ovum. On the inner surface of the serotina no surface epithelium is seen, the blood bathing the surface of the decidua. Its tissue rises in processes, to the top of which a villus may be seen attached. The epithelial investment of the chorion is a well-preserved cubical epithelium in a single layer, with a nucleus at the bottom of the cell. Below this is a narrow layer of elongated cells with long spindle-like nuclei resembling somewhat the stroma, but more compressed and darker. The spaces between the villi are filled with blood, and their connection with the blood vessels can be distinctly seen. Near the mouths of these vessels there are hæmorrhages into the mucosa, such as are seen in normal menstruation.

The questions which we look to such preparations as these to settle are, how is the early ovum attached, and how is it fed? A great deal has been written on these subjects, and there has been much reasoning, chiefly deductions from what is known as to the condition found at a later period of development of the human ovum, and by analogy from these processes as they have been traced out in other mammals. We cannot build too much upon a single preparation, and there is always the difficulty which Leopold hints at, that the appearance found in it may be pathological. This preparation is, however, of the greatest value as giving us a basis to start from. Although we have but this one preparation, yet it is the only one, and it is priceless.

It shows, for example, that there is no ground for the idea that the ovum becomes embedded in the raw surface, as some have

suggested. The desquamation of the mucous membrane at the menstrual period, and the fact that conception takes place shortly after it, gave colour to this hypothesis. The ovum is here, however, not embedded on a raw surface. It is rather perched on a fold of mucous membrane which has its surface epithelium still present on one side of it. The hypertrophy of the whole mucosa is also noteworthy. The rapidity with which that membrane can be regenerated after its destruction has been drawn attention to by Werth's investigations on uteri extirpated after curetting. Within five days after its removal by the curette he found the mucosa reproduced anew-with glands opening freely on the surface, and an unbroken covering of epithelium. In this case we have the additional stimulus of a growing ovum. But it is remarkable that the characteristic hypertrophy of the mucosa and its distinction into a compact and spongy layer, is present in about a week after impregnation. As we are speaking now of the hypertrophy of the glands, we note also that a week later we may have their hypertrophied epithelium breaking away and forming masses blocking their lumen.

Another view that has been suggested is that the villi become in the first instance rooted in the gland cavities. This view, which Sir William Turner combated, has recently been revived by Reichert from the study of his ovum, which until this preparation which Leopold has described took its place as the youngest ovum known. The appearance seen at two points of Leopold's section of a villus in the glands might at the first glance give colour to this hypothesis, but on looking all over the preparation it is evident that this is a mere accident, and that the attachment of the villus is to the surface of the decidua, and in the first instance by fibrinous adhesion, and in the later stage by the villus being embedded among the decidual cells. The attachment of the ovum, however, by the villi is at first loose, and as Leopold says it might easily be washed away were it not for the ring of the reflexa clasping it. The mode of closure of this ring in its upper pole is of interest, as it suggests a process similar to what Hubrecht describes in the hedgehog. In that animal a hæmorrhagic oedema occurs at the lips of the opening at the upper pole, and a blood extravasation takes place by which the ring is closed. This preparation then confirms the mode of attachment of the early ovum which has been generally received since the time of Hunter. He introduced the terms reflexa and serotina, through a mistake as to their mode of formation, the term reflexa implying that the ovum carried the hypertrophied mucosa before it when it entered the uterus, stripping a portion of the uterine wall bare, which was covered later by decidua hence the term serotina. For these terms Leopold substitutes decidua capsularis, and basalis, which are certainly more correct. Hunter's nomenclature, however, although it perpetuates a misconception, has passed into such general currency

that the changing of terms does not seem worth while, as the gain in scientific accuracy hardly compensates for the confusion caused by introducing new terms.

While in error as to the mode of formation of the decidua, Hunter's description of its disposition has been established by all that has been discovered since by microscopic investigation, for which, of course, he had not the methods which we have now. The new points brought out by this section of Leopold's are the mode of closure of the upper pole of the reflexa, and the mode of attachment of the ends of the villi.

While we are on firm ground in considering how the ovum gets its hold, we are plunged at once into a questio vexata when we consider how it gets its food. The gist of his conclusion is that at this early stage we already find what has been described in the fully-formed placenta at the fourth month, and that it arises in a simple way. We see here little processes of decidual cells growing out into the cavity between the chorion and decidua, which foreshadow the decidual trabeculæ of the fully-formed placenta. Further, we find in the space in which the villi lie, maternal blood, and the maternal vessels already opening into this space. His account of the commencement of the circulation which I quoted in full, is lucid, but not convincing. It is hard to see how a circulation can be established in this way. The difficulty is not as to the blood finding its way out of the vessels into a space, but how it finds its way from the spaces back into the vessels again.

The process described is analogous to that of blood extravasation, and the appearance of the blood as described in the first specimen is suggestive of this.

The channel must be built before the water finds its way along it, or the result is a marsh, not a stream. It seems to me that in the process of gradual dilatation of the capillaries without rupture, we must look for the commencement of maternal circulation in the placenta. It is quite possible that after the sinus has been formed, the villus may come to lie in it with only a single layer of epithelium. But the sinus must be formed first.

Is it not possible that in the dilated sinuses which are present in the second preparation, and which Leopold thinks pathological, we have the structures which are to become the sinuses of the placenta, and that at this stage the embryo is nourished by an exudation of lymph into the spaces in which the villi are now lying?

Besides these two sections of an early ovum, a series of beautiful sections are given of the placenta at the fifth and ninth months, as well as a drawing of the structure of the placenta made from an injected placenta by the removal of the smaller villi so as to make the stems of the villi and the septa of decidua stand out. results thus obtained are chiefly confirmatory of those reached before by Leopold and others. They show the maternal sinuses

The

G