DEVELOPMENT PREVIOUS TO THE SWIMMING STAGE 21 



licle and by mutual pressure, since in some species akin to the oyster (Anodonta, Unio, 

 Cyclas) they are attached to the wall of the follicle by their micropylar ends, which may- 

 be short or long according to position or pressure. 



Brooks states that "three planes of cleavage run in towards the centre of the egg from 

 three equidistant points on the periphery." It is intelligible how cleavages that should 

 be successive may, through accumulation of yolk, be so modified as to occur close to- 

 gether; but when we study the statement in connection with the figures, the three nicks 

 that appear on the surface must resolve themselves into but two planes of cleavage in 

 the sense in which these terms are commonly used, viz., one horizontal, the other 

 vertical, and the horizontal one appears first (figs. 10, 11). The normal direction for 

 the first three planes of cleavage in well known eggs of other animals seems to be 1st, 

 meridional; 2nd, meridional, at right angles to the first; 3rd, equatorial, at right angles 

 to the other two; which would make it appear doubtful if the horizontal constriction 

 here is a true cleavage and at the same time look probable that there is only a single 

 true cleavage, viz., the one falling through from the polar bodies. 



Two alternatives are presented, depending upon whether the equatorial constric- 

 tion represents a true fission into upper and lower halves, or whether it results as a com- 

 bination of gravitation with the first vertical fission. If the first segmentation plane 

 is equatorial (horizontal), then the oyster will fall in with the diagram given by Kors- 

 chelt and Heider for the class, the first segmentation spindle will be in the principal 

 (vertical) axis, and one of the daughter nuclei must be hidden in the mass of deutoplasm, 

 which will represent a true blastomere (macromere). Moreover, the first spindle as 

 figured by Nelson and by Brooks (1905) will be not the first but the second segmentation 

 spindle. The diagram referred to does not show a polar body which should be ex- 

 pected at the apex of the micromere. A figure by Horst (his fig. 3; my plate V, fig. 17) 

 does show a polar body, but it is on the side, at the plane of cleavage, where it does not 

 agree with his fig. 2. 



If on the other hand the first segmentation plane is meridional (vertical) and fission 

 only partial, not dividing the deutoplasm, the first segmentation spindle will be trans- 

 verse; each daughter nucleus goes into one of the blastomeres, the deutomere is without 

 nucleus and is not a true blastomere, and the first spindle figured by Nelson and by 

 Brooks is really the first segmentation spindle. This view would bring the oyster 

 oosperm into unison with the great number of others, in that the first two cleavage planes 

 fall at the polar body and are constant for both holoblastic and meroblastic eggs, i.e. for 

 those that have comparatively little yolk deposited in the protoplasm, so that the whole 

 egg can divide into equal halves, as well as for those that have such a large deposit of 

 yolk that the active protoplasm and first attempts at segmentation are confined to a 

 small part of the surface. The oyster oosperm would stand between the two extremes, 

 nearer the first, and present many irregularities. The first fission is only partial and 

 the two blastomeres are not quite equal. The larger retains connection with the yolk 

 and seems to absorb some of it, while at the resting period it becomes completely con- 

 fluent with it, forming a macromere (Nelson) with a single nucleus in the formative end 

 (Brooks, Nelson). Horst's fig. 3 turned over and rotated until the polar body comes up- 

 ward, will agree with Brooks' fig. 14. The free blastomere (micromere) then looks 

 like a bud. But the second fission (Brooks' fig. 16) has every appearance of being as 

 normal as the first, dividing both of the original blastomeres vertically at right angles 

 to their first separation — one of the new ones again retaining connection with the yolk. 

 The egg, the oosperm, and the early stages of the first cleavage are radially sym- 

 metrical about the chief (vertical) axis, but the later stages show a bilateral symmetry. 

 The early appearance of this feature is doubtless due to the deutomere, which always 

 retains more complete connection with one of the blastomeres, so as to appear shifted 

 slightly to one side from the axis. What we might call the plane of symmetry would 

 fall through both blastomeres, as well as the deutomere, i.e. where the second cleavage 

 arises. But after the second cleavage is completed the only plane of strictly bilateral 

 symmetry would fall through the deutomere, the blastomere more immediately above 

 it, and the middle one of the other three, while two other blastomeres of different origins 

 lie one on each side. The plane of symmetry is now a different plane from the original 

 one, and in like manner it would shift with each succeeding cleavage. It can hardly be 

 claimed that the bilateral symmetry of this period is strictly comparable with that of 

 the adult, since it is too variable and temporary, radial symmetry predominating again 

 during later periods (morula, blastula, gastrula). 



All the foregoing features of the segmentation of the oyster's oosperm can be in a 

 measure understood by considering that the smallest and simplest eggs possess an 

 heredity of structure and of activity, that they are subject to the physical and chemical 

 laws of matter as well as the biological, and that the oyster's egg is further complicated 

 by the deposition of a large amount'of food yolk for an egg of its size. This diminishes 

 3 



