MUSEUM OF COMPARATIVE ZOOLOGY. 359 



polar rays much more than those of the equator. These oscillations have 

 the effect of giving the figure at the end of the systole the form of a 

 spheroidal body, with flattenings, or even funnel-shaped depressions, at 

 the poles. 



Plmse 4 is of short duration, and is characterized by the formation of 

 an equatorial plate, during which the filament elements become grouped 

 parallel to the axis which is perpendicular to the plane of division. The 

 thickness of the plate is from one fifth to one third as great as the whole 

 length of the cell. At first, however, it is only the filaments in the axial 

 portion of the figure which are parallel, or slightly converging toward the 

 poles ; the more peripheral are still tortuous. But this '' equatorial- 

 plate " stage, which recalls the " nuclear spindle " of other authors, differs 

 in several points from the typical spindle. The equatorial thickenings 

 known collectively as the " Kernplatte " are wanting here,* and the fila- 

 ments terminate at the equator in free ends. Whether or not this equato- 

 rial interruption in their course is induced by an immediately preceding 

 division, it is difficult to say. Certain preparations which show a continu- 

 ity still existing between some of the fibres belonging to the opposite halves 

 of the figure favor the view that they were recently continuous through- 

 out, and yet this appearance may have been brought about by a tempo- 

 rary apposition of filaments which are really distinct, and were actually 

 separated into two groups at an earlier stage. The occasional existence 

 of double stars during the preceding phase strengthens the latter hy- 

 pothesis ; for, if the substance of the daughter nuclei in these cases was 

 already segregated in such a manner that each star embraced all the 

 nuclear substance destined for the nucleus of its own pole, then a tem- 

 porary apposition of the filaments must necessarily be predicated. But 

 if the "double-star condition does not normally come in the series of 

 changes, it is possible that a separation of the nuclear halves may 

 have transpired before the equatorial-plate stage. In this connection 

 Flemming explains how the longitudinal fission of the filaments may 

 possibly be equivalent to such a separation, one half of each filament going 

 to one of the new nuclei, the other half to the other. This stage may 

 possibly be comparable with Mayzel's spindle with outlying nuclear plate 

 (see p. 348). Continuous observation of this " nuclear-cask " stage shows 

 that the cask becomes broader, and that its peripheral filaments become 

 alternately curved and straightened at the equatorial end. 



Phase 5. Separation of the halves of the nuclear figure. This is 

 also of short duration. Each half has the form of a broad fish-weir. f 



* See also Mayzel for Triton. t This is the basket stage of Eberth. 



