to spindle formation. This /.nnc hcc.imc uianiilar. and then the 

 g-ranules arranj^ed themselves in rows lo lorni tilii ilhe ( I'l--. i s ^/ ), 

 which, hcnvever, did not extend int.. the evtnplasni radially, but 

 lay -enerally parallel to one another, so that the >i)indle appeared 



was -enerally lornied considerably in advance oi the other (Fi<^. 

 13 /). The fibers did not meet at a common i^omt but over a 



mallyno multipolar sta<;-e m Osmunda, althou,<;h tripolar spmdies 

 {Vi<y. I ij^r) were occasionally found. During anai)ha.se secondary 

 fibers were put forth from the vicinity of the daughter nuclei and 

 these met in the equatorial region of the cell. 'I'he spindle of 

 the second mitosis was formed exactlv as in the hrst. After this 

 di\i>ion the lour granddaughter nuclei lav connected with one 

 another by six spindles (two primary and four secondary). Cell 

 plates were laid down in the equatorial regions ol these spindles 

 so that the protoplasm became divided simullaneousl}- and sym- 

 metrically into tetrahedral spores. 



The studies of Calkins ('97) and Stevens ('98^0 were chiefly 



nection with reduction phenomena Calkin-; believed that the 

 processes of sporo,i;cnesis tollowed the same ^.m^c as the matu- 

 ration o[ sexual cell, in animals, with a tiansxeisc dixisionto 

 ^ive a qualitative reduction in Weismann's sense. Stevens dis- 

 ^i,i;ree(l \\\{h Calkins in several particulars, holding that the 

 lediution was merely quantitative. Reduction phenomena in 



i)apei.s which dealt with ])!ants miicii more difficult to study than 



Mder the subject m Section \\ but mav state now that Calkins' 

 conclusions have not been sustained 



Strasburger (:oo) gives considerable attention to ^pmdle 

 formation in his well known review and criticpie of cytological 

 literature. He proposes the following classification of spindles 

 in higher plants which lack centrosomes. Those that pass 



