Cardiff: Synapsis and reduction 297 



stage, of great chemotactic activity, which probably accounts for 

 the way it retains dyes. It is the critical stage in the history of 

 an organism. The heterotypic division immediately following it is, 

 then, not a true mitosis, but, as Farmer and Moore have suggested, 

 merely an intercalated phase in the ontogeny of an organism ; a 

 phase for the purpose of bringing about the distribution of the pa- 

 rental characters and causing the necessary variation in the prog- 

 eny of an organism. 



The bearing of this union of parental chromatin in the prere- 

 ductional stages upon the principles of Mendel was. first discussed 

 by Wilson ('02), Cannon ('02 and '03), and Sutton ('03). The 

 present work would indicate that the parental chromatins are 

 brought into much closer relationship than was at first supposed 

 by these workers. They are probably brought into such close 

 relationship that there is a more or less blending or complex in- 

 terchange of characters. DeVries ('03) suggests that in the case 

 of an intimate blending of the chromatins there is possible an in- 

 terchange of pangens which would, in many cases, approach Men- 

 del's ratios. From the cytologicai work that has been done in the 

 past six years it is evident that the explanation of Mendel's laws 

 lies in the structure of the germ-cells. While much has been ac- 

 complished along this line, and we seem nearer a solution of the 

 problem than ever before, it is equally evident that we are very 



far from a final explanation of heredity. 



From the results of recent investigators it seems possible that 

 there is not a strictly uniform behavior of the chromatin in syn- 

 apsis, and that there may be considerable variation in the distri- 

 bution of hereditary characters. 



I am indebted to Professor E. B. Wilson and Dr. C. C. Curtis 

 for kind criticism and advice during the progress of the work. 



Columbia University. 



Allen, C. E. 



LITERATURE 



'04. Chromosome reduction in Lilium cana dense. Bot. Gaz. 37 



464. 



'05a. Nuclear division in the pollen mother-cells of Lilium canadense 



Ann. Bot. 19 : 189. 



