102 Plant Genetics 



there is no segregation of determiners. This situation is 

 not merely shown by cytology, but is also in accord 

 with most of our breeding results. 



In the case of somatic segregation, therefore, there 

 seems to be only one possible conclusion, and that is that 

 sometimes reduction division occurs in the somatic 

 tissue, at least so far as certain chromosomes are con- 

 cerned. Such a conclusion is not unreasonable. Crosses 

 of certain factors might cause a very unstable condition in 

 certain chromosomes, so that they might not conform 

 to the ordinary mechanism of cell division. Certain 

 chromosomes, or certain determiners on the chromo- 

 somes, instead of dividing normally might pass un- 

 divided to one pole of the spindle at cell division and 

 be entirely lacking at the other pole. 



The only other alternative would be to discard the 

 old ideas of the mechanism of cell division and inherit- 

 ance, and this we are hardly ready to do. In any 

 event the phenomenon of somatic segregation opens 

 an important field for some critical cytological research. 



It will be realized that somatic segregation is a 

 much more fundamental situation to explain than 

 maternal inheritance. There is no reason to suppose, 

 however, that it contradicts Mendelian inheritance 

 seriously; perhaps it enlarges the scope of Mendelism. 

 Mendel's law is violated only in the fact that segregation 

 occurs at an unusual place; and that is no reason for 

 discarding Mendelism. It complicates the situation 

 somewhat because the ordinary reduction division 

 furnishes such a complete mechanism for segregation; 

 but MENDEL formulated his law before the reduction 

 division was discovered in plants. Somatic segrega- 



