farr: cytokinesis of pollen-mother-cells 267 



varying courses and may anastomose. These he believes are 

 not the spindle fibers but rather that they have arisen after the 

 spindle, resulting from the homoeotypic karyokinesis, has dis- 

 appeared. In this case the cell-walls are formed in the cyto- 

 plasm without reference to the fibers. His figure 26 mdicates 

 the development of a cell-plate from the center of the tetranucleate 

 mother-cell outward in three (in all, six) directions toward the 

 mother-wall. This should be described then as neither a centri- 

 petal nor a centrifugal development, for the latter terms should 

 be used with respect to the individual spindles, and not to the 

 cell as a whole; it might better be referred to as progressive from 

 one side of the spindle to the other, such as Treub noted in his 

 study of the orchids. In view of the following study it is evident 

 that further investigation of cell-division in the mother-cells of 

 Anthoceros should be undertaken. 



The mosses have scarcely been studied at all in this respect. 

 Sachs on page 13 of ^his textbook (59) reports tetrahedral quadri- 

 partition in the spore-formation of Funaria. Hofmeister (32^) 

 shows that the mother-wall of Phascum swells to eight times its 

 original thickness, and the cell doubles its diameter during the 

 reduction divisions. 



In the ferns several cases of quadripartition have been reported. 

 Calkins (9) describes such in Pteris, where division occurs tetra- 

 hedrally and by cell-plates, but no figures are given of the process. 

 In Osmunda a similar instance is given by Smith (65). Here the 

 tetrahedral arrangement of the nuclei is said to be brought about 

 by a rotation of the spindles and chromosome aggregates during 

 the homoeotypic karyokinesis, the heterotypic spindle having 

 meanwhile broken down into a granular mass. In Botrychtum 

 Stevens (66) describes a similar dense equatorial plate of cyto- 

 plasm following the heterotypic spindle. In this species the 

 nuclei at the time of quadripartition may be arranged either in a 

 tetrad or a rhomb. The division is by cell-plates and the mother 

 cell-wall is not shown to thicken during the process. In Polypo- 

 dium, Russow (58) described successive bipartition, the mother- 

 cell-wall swelling in water. _ . 



Among the fern allies there is a considerable range of variation 

 in the cytokinesis incident to spore formation. For Eqmsetum 

 Russow (58) reported tetrahedral quadripartition, the division of 



