58 THE MATURATION OF THE EGG OF THE MOUSE. 



tion of the polar cell. Although the statement may be based upon the 

 sscond spindle, which Tafani mistook for the first, it nevertheless is 

 true of the first spindle. Sobotta (1895, P- 4 8 ) makes the unqualified 

 statement that the slender spindle (which he now calls the second) 

 turns from the paratangential position to the oblique and finally to the 

 radial just before the polar cell is cut off. He saw three cases of meta- 

 phase spindles, all oblique, but all of those in the telophase were radial. 

 Therefore, although he had not actually seen the process of abstriction, 

 he thought the first spindle was radial at the time the polar cell was 

 cut off. In a later paper (1899, p. 190) he describes the same process 

 for the first spindle and gives a figure (fig. 4) of the spindle during the 

 dyaster stage in what appears to be a radial position with one pole in 

 the polar-cell protrusion. The figure has a somewhat rigid diagrammatic 

 appearance and is not accompanied by any explanation to prove that the 

 spindle is radial with respect to the center of the egg as well as the center 

 of the section in which it lies. The relative shortness of the spindle 

 suggests the possibility that its axis is oblique to the plane of the section 

 and that consequently it may not be strictly radial in position. He 

 mentions having three other spindles in the stage of his fig. 4, but does 

 not state what their position is. In one of his recent papers Sobotta 

 (1907, p. 517) figures a dyaster stage of the first spindle (fig. 4) and 

 states that it is in an oblique position, having begun the rotation from 

 the tangential to the radial position. In a foot-note, however, he admits 

 that it really is never met with in a strictly radial position! He (1907, 

 p. 517) finds it difficult to decide whether the first spindle always rotates, 

 yet he argues that it must remain tangential in most cases (one polar 

 cell) because it is transformed in the monaster condition directly into 

 the monaster of the second spindle, which is likewise tangential. He 

 is not sure whether even in one-fifth of the cases (those in which it divides) 

 it may not be oblique when the polar cell is formed, but thinks it may be 

 assumed that as a rule it rotates, because the second spindle always 

 rotates, and because it (the first) takes up a position so near the surface 

 of the egg that no polar cell could be produced without its rotation. 

 Sobotta does not give any proof, except that contained in his first paper 

 (1895), tna t the second spindle is radial at the moment the polar cell 

 is abstricted. Moreover, he figures (1907, fig. 9) a dyaster of the second 

 spindle in a paratangential position and says (p. 525) that its not being 

 radial is purely accidental! Thus, except for his 1899 paper, which he 

 does not mention in this connection, there is no evidence that either 

 polar cell is cut off while the spindle is in a strictly radial position. Ger- 

 lach (1906, p. 10), while he does not take exception to the general conclu- 

 sion of Sobotta that there is a rotation of the spindle from a tangential 

 toward a radial direction, thinks that the strictly radial position is not 

 necessary to the formation of the polar cell. Neither Gerlach, Lams et 

 Doorme (p. 275), nor Kirkham (19076, p. 75) mention having seen any 



