THE MECHANISM OF ASSORTMENT 79 



the second spermatocyte division it divides as do the others 

 at this time in the cell that contains it. If two are present 

 they do not behave as mates, but at the first spermatocyte 

 division may both go to the same pole (which may be 

 either pole in reference to the sex-chromosome), or they 

 may go to opposite poles. At the second spermatocyte 

 division each divides independently, and halves go to 

 opposite poles. These bodies then also move to either 

 pole without respect to other chromosomes — or at least 

 without respect to the sex-chromosomes ; but this behavior 

 can scarcely be used to advantage for the question of 

 assortment because these chromosomes have no mates (in 

 the cases so far described) and are so inconstant in their 

 occurrence that an appeal to their behavior as bearing 

 on the other chromosomes might not be conceded. If they 

 are pieces of other chromosomes (the bent ends, for exam- 

 ple) that have been broken off, we might expect them to 

 show some relation during synapsis to the original 

 chromosome from which they came, but as yet nothing 

 of the sort has been described. If they carry factors that 

 influence the characters of the individual, their presence, 

 especially when two occur, would give rise to unexpected 

 genetic results. 



The evidence furnished by cytolog}^ that has just been 

 given makes clear that whenever an opportunity has been 

 found to study the mode of assortment of the chromosomes 

 the result shows random distribution. If then the chromo- 

 somes carry the genes for the hereditary characters, we 

 should expect that the genes in different chromosome pairs 

 will ^* assort ^^ independently, and this, in fact, is what 

 Mendel's second law postulates. 



