658 Journal of Agricultural Research voi. xii. No. i© 



somes during reduction division, but was due to a lack of cytoplasm. 

 The abortion of the egg sacs, he again concluded, was not due to irregular 

 divisions, as all of the divisions following reduction were normal., but was 

 due to poor nutrition. 



Nakao (29) , working on the cytology of certain grain hybrids in which 

 very striking irregularities in reduction division took place, followed by 

 complete abortion of the microspore after liberation, concluded that 

 abortion in this case was due to an insufficiency of cytoplasm previous 

 to reduction division which resulted in abnormally early division and con- 

 sequent irregularities. He did not consider why these irregularities 

 should cause abortion of the microspores. 



In view of the conditions existing within an anther at the time of 

 degeneration of microspores and because of certain genetic results which 

 can only be explained on the basis of selective elimination of certain 

 gametic combinations, it is difficult to agree with the conclusions of 

 Tischler and Rosenberg that degeneration has nothing to do either with 

 the irregularities or normal repulsion which occur at reduction division 

 in hybrids. 



Although there are striking differences in the regularity with which 

 reduction takes place in sterile or partially sterile hybrids, there are 

 certain conditions, in both those which proceed normally and those 

 which show irregularities, which are alike and must be taken into account 

 in the consideration of the causes of pollen abortion. These conditions 

 are as follows : 



(i) At reduction division there is a sorting out of the parent chromo- 

 somes, resulting in new combinations in the daughter cells, the mnuber 

 of which depends upon the degree of difference between the two parents. 

 If division proceeds normally, there is an equal number of chromosomes 

 in each daughter nucleus. If it takes place irregularly, imequal numbers 

 are found in the resulting daughter cells. In either. case the combina- 

 tions are new and may or may not contain all of the properties necessary 

 for perfect metabolism of the cell. 



(2) If divisions take place regularly, there is an equal quantitative 

 and, as far as can be determined, also an equal qualitative division of 

 cytoplasm between the quadrants of a tetrad. If divisions have pro- 

 ceeded irregularly, the cytoplasm is divided between the members of the 

 tetrad in proportion to the amoimt of chromatin which they contain. 

 In either case at the time of liberation from the tetrad, or if liberation 

 does not take place, as in Drosera spp., at the period previous to enlarge- 

 ment, all of the microspores appear normal — that is, they contain an 

 organized nucleus and are filled with cytoplasm. In Minnesota 3 

 there is at this time as great uniformity in size and cytoplasmic content 

 of the individual microspores as is fotmd in entirely fertile plants of 

 F. virginiana. 



