which would of itself produce not a specific embryo, but an amorphous monster unless directed by the influence of (in the rabbit) inanimate coats and the walls of the uterus.

I have come to the following conclusions upon certain disputed points.

(i) Van Beneden's description of the segmentation I consider to be inaccurate.

(ii) I find no trace of van Beneden's blastopore.

(iii) I find no trace of a "gastrulation."

(iv) There is no evidence in support of Robinson's1 speculations concerning the existence of a hypoblastic wall to the blastocyst surrounded subsequently by the epiblast.

(v) Rauber's layer fuses with the inner layer of epiblast as described by Balfour2 and Heape.3

(vi) This fusion has but slight morphological significance, since its existence and disappearance are caused mechanically by ontogenetic conditions.

(vii) The growth round of the hypoblast is apparent only, being due to the presence of a zone of specially active epiblast surrounding the embryonic disc, which zone is to be considered to be the equivalent of the träger in other rodents.

I have endeavoured to show how important is the albuminous layer, and how I believe that it is possible to account for many details of change up to the end of the sixth day by strictly epigenetic processes; and since these processes during this time are almost all directed by environment as between the embryo and the maternal influences rather than between cell and cell of the embryo itself, it follows that the palingenetic features of the development must be reduced to a minimum.

This paper is based upon the examination of upwards of 300 embryos between the 24th and 168th hours. The embryos have been examined fresh, and after treatment with various reagents -Perenyi, osmic acid, picric acid, Flemming, Hermann, chromic

1 'Quart. Journ. Micr. Sci.,' vol. xxxiii.

2 Comparative Embryology,' vol. ii.

3 Quart. Journ. Micr. Sci.,' vols. xxiii and xxvi.

acid, &c., and sections have been examined of all stages from the 30th hour onwards.

A large portion of the work for the present paper, and the three or four other papers I hope to publish shortly, was carried on while I held the post of Demonstrator of Zoology in the Owens College.

To the kind consideration of the late Professor Arthur Milnes Marshall I am indebted for many opportunities for work which I should otherwise have found difficult to obtain.

The remainder of the work has been done in the Morphological Laboratories of Cambridge through the permission of Mr. Sedgwick, to whom I wish to express my gratitude for the said permission and for his kindness in reading my papers and offering many valuable criticisms.

CHAPTER I.

SEGMENTATION OF THE OVUM.

Since the accuracy of van Beneden's account of the early stages of the development of the rabbit depends to a great extent upon the way in which the earliest cleavage planes succeed one another, I think it advisable to give in some detail the results of my own researches on this point. In fact the very first segmentation division, according to van Beneden, gives rise to an important question, which is as follows: is there any essential difference between the two first segments, and do the descendants of one segment give rise to the epiblast cells, and the descendants of the other to the hypoblast cells?

To this van Beneden replied that there is always a difference in size between the two first segments, and that the smaller of the two is the more granular, and that from that one are derived all the cells of the inner mass, and that the larger clearer segment gives rise to cells which gradually surround the descendants of the small segment and form the wall of the future "blastodermic vesicle."

From the descendants of the small segment van Beneden

derived the hypoblast and subsequently the mesoblast; from the descendants of the larger, the epiblast.

As far as the conversion of the whole of the inner mass into hypoblast and mesoblast is concerned, van Beneden has changed his opinion, admitting that Kölliker's (and others-Balfour, Heape) careful research in this matter is a truer account than his own.

Again, Heape, on the mole, has shown that a different interpretation may be placed upon the fact as seen in the rabbit, although he finds that there is a difference in the size of the two first segments.

The results of my researches show that there is, if not always at any rate very frequently, a difference in size between the first two segments; sometimes the difference is very marked, but more usually it is to be found only by careful measurement. In one rabbit I examined the following results were obtained.

Rabbit 48, was killed 24 hours after coitus. In the thinwalled part of the Fallopian tube of the right side I found four embryos at about 30 mm., 40 mm., 42 mm., 50 mm., from the funnel mouth of the Fallopian tube respectively.

No. 1, the highest up, had not as yet divided. The ovum was spherical with a distinct nucleus, the ovum not occupying the whole space within the zona radiata.

No. 2, the second on the right side, was divided into two segments, but this I did not measure.

No. 3, the third and lowest on the right side, was also in two segments. I could detect no differences at all in colour or texture, but on measuring the longest and shortest diameters of each the following results were obtained.

One of the segments measured as follows: longest diameter 1 mm., shortest diameter 076 mm. The other segment: longest diameter 096 mm., and shortest 068 mm.

The specimens were examined and measured in the fresh condition in a drop of aqueous humour of the rabbit. The measurements were taken with a micrometer eye-piece, No. 3, and Zeiss D objective, which gave a magnification of mm. for each division in the eye-piece.

In giving the measurements I have reduced these to actual measurements, giving them in decimal fractions of one milli

mètre.

No. 4 was also in two segments, but was not measured; each segment was composed of an inner denser and outer clearer part. In the Fallopian tube of the left side I found six ova, which were scattered along an area between two points 35 mm. and 65 mm. from the upper end of the Fallopian tube. The whole length of the Fallopian tube was about

100 mm.

Three of the ova were in the wide thin-walled upper part, and the other three were just within the lower thicker-walled portion of the Fallopian tube. No. 1, that is to say the one nearest to the funnel end, was unsegmented. It appeared, however, to be on the point of dividing (fig. 1). It did not occupy the whole of the cavity within the zona radiata. It was spherical, and two polar bodies lay between the ovum and the zona, the one larger than the other. The nucleus, however, though not very distinct, undoubtedly seemed to be double or dumb-bell shaped.

The diameter of the ovum was 11 mm.

No. 2 (v. fig. 3) was in two segments. The difference in size between these two was very marked. It was quite obvious without measurement. This specimen showed the greatest difference I have hitherto met with. There was no perceptible difference in density or in colour, either while fresh or after treatment with nitrate of silver followed by picro-carmine and aniline blue-black. The measurements of the two segments of this specimen were as follows:

Of the larger the longest diameter was 101 mm., the shortest diameter 088 mm. Of the smaller the longest diameter was 08 mm., and shortest was 064 mm. The nuclei were clear and round.

No. 3 was in two segments, but far more equal in size. Nuclei round and clear and distinct. Measurements were—

I could detect no difference except as regards size.

No. 4. I could make nothing of this one; it may have been an unfertilised ovum or pathological.

No. 5 was in two segments. Here again the only difference I could detect was in size, as follows:

Both

No. 6, the lowest down, that is to say the nearest the uterus on the left side, was still unsegmented, being spherical. polar bodies were visible between the ovum and the zona pellucida.

The ova from another rabbit, No. 51, were examined fresh in aqueous humour of the rabbit.

In the left Fallopian tube I found only one ovum, situated almost midway between the two ends. This specimen was not yet segmented, but the ovum had retracted from the zona pellucida, which seems to be a sign that it was a fertilised ovum. In the right oviduct I found four ova.

No. 1, the one nearest the funnel opening of the Fallopian tube, was in two segments. I could determine no difference in character, but on measurement a slight difference in size was made evident.

After being carefully examined and measured while fresh, this specimen was transferred to a 2 per cent. solution of osmic acid for thirty seconds, and subsequently stained with a mixture of aniline blue-black and picro-carmine, and mounted in glycerine jelly—a very pretty preparation. At no period could I detect any difference except in size between the two segments. In this specimen the two polar bodies were placed far apart from each other, a peculiarity I have not noticed in any other (v. fig. 4).

No. 2 was also in two segments, and was examined, measured,