In the case of this clear protuberance and of the similar protuberances which follow it, the protoplasm is not at first quite free from food-yolk, but only becomes so on being separated from the yolk-containing part of the ovum. We must therefore suppose that the production of the clear segments is in part at least due to the yolk-spherules becoming used up to form protoplasm. Such a formation of protoplasm from yolk-spherules has been clearly shewn to occur in other types by Bobretzky and Fol.
The protuberance soon becomes separated off from the larger part of the egg as a small segment composed of clear protoplasm. From the larger segment filled with food-yolk, a second small clear segment is next budded off, and simultaneously ([fig. 42], 2) the original small segment divides into two. Thus there are formed four segments, one large and three small; the large segment as before being filled with food-yolk. The continuation of a similar process of budding off and segmentation eventually results in the formation of a considerable number of small and of one large segment ([fig. 42], 3). Between this large and the small segments is a segmentation cavity.
Eventually the large yolk segment, which has hitherto merely budded off a series of small segments free from yolk, itself divides into two similar parts. This process is then repeated ([fig. 42], 4) and there is at last formed a number of yolk segments filled with yolk spheres, which occupy the place of the original large yolk segment. Between these yolk segments and the small segments is placed the segmentation cavity.
The segmentation of the ovum of Euaxes[43] resembles that of Unio in the budding off of clear segments from those filled with yolk, but presents many interesting individualities.
A very peculiar modification of the ordinary Gasteropod segmentation is that described by Bobretzky for Nassa mutabilis[44].
The ovum contains a large amount of food-yolk, and the protoplasm is aggregated at the formative pole, adjoining which are placed the polar bodies. An equatorial and a vertical furrow ([fig. 43] A), the former near the upper pole, appear simultaneously, and divide the ovum into three segments, two small, each with a protoplasmic pole, and one large entirely formed of yolk material. One of the two small segments next completely fuses with the large segment ([fig. 43] B), and after the fusion is complete, a triple segmentation of the large segment takes place as at the first division, and at the same time the single small segment divides into two. In this way four partially protoplasmic segments and one yolk segment are formed ([fig. 43] C). One of the small segments again fuses with the large segment, so that the number of segments becomes again reduced to four, three small and one large. The protoplasmic ends of these segments are turned towards each other, and where they meet four very small cells become budded off, one from each segment ([fig. 43] D). Four small cells are again budded off twice in succession, while the original small cells remain passive, so that there come to be twelve small and four large cells. In later stages the four first-formed small cells give rise to still smaller cells and then the next-formed do the same. The large cells continue also to give rise to small ones, and finally, by a continuous process of division, and fresh budding of small cells from large cells, a cap of small cells becomes formed covering the four large cells which have in the meantime pressed themselves together ([fig. 43] E). A segmentation cavity of not inconsiderable dimensions becomes established between this cap of small cells and the large cells.
Fig. 43. Segmentation of Nassa mutabilis (from Bobretzky). A. Upper half divided into two segments. B. One of these has fused with the large lower segment. C. Four small and one large segment, one of the former fusing with the large segment. D. Each of the four segments has given rise to a small segment. E. Small segments have increased to thirty-six.
Many eggs, such as those of the Myriapods[45], present an irregular segmentation; but the segmentation is hardly unequal in the sense in which I have been using the term. Such cases should perhaps be placed in the first rather than in the present category.