CHROMOSOME COMPLEX OF SPERMATOCYTES. 323 



4. The synapsis of chromosomes in pairs or other combina- 

 tions does not occur in all species at the same time, nor do all 

 the elements of the cell enter into these relations synchronously. 

 There is accordingly no fixed synaptic phase. 



5. Heterotypical divisions are observed in spermatogonia, first 

 spermatocytes, and second spermatocytes, and in each case repre- 

 sent an ordinary longitudinal cleavage of the chromatin thread. 



6. In cases of unusual associations of chromosomes the acces- 

 sory is involved in the formation of the multiple chromosome. 



/. Under all conditions the accessory chromosome goes to but 

 one half of the spermatozoa. 



8. When pentivalent multiple chromosomes are formed whole 

 tetrads go into the second spermatocytes undivided. 



10. In a single mitosis some of the chromosomes may divide 

 qualitatively while the others divide quantitatively. This is due 

 to unusual associations of the elements, and occurs regularly in 

 the same way in all the cells of a generation. 



1 1. Unless chromosomes are associated into multiple elements 

 of higher valence than tetrads they divide longitudinally in the 

 first spermatocyte and transversely in the second. 



12. Tetrads divide longitudinally wherever found, either in 

 spermatogonia, first spermatocytes or second spermatocytes. 

 Multiple chromosomes of higher valence than tetrads divide 

 transversely. 



CONCLUSIONS. 



i. CJiroinosomes and Somatic Characters. 



Two facts of first importance stand out as a result of these 

 observations. The first is that a definite series of chromosomes 

 accompanies the exhibition of a group of somatic characters that 

 have been utilized by systematists for the inclusion of a number 

 of genera into a family. The second fact is that these genera are 

 characterized by a very precise architecture of their germ cells, 

 whereby the series of chromosomes presents a definite arrange- 

 ment and association which suffers modification in minor details 

 in the different species of the genus. At the bottom of all, of 

 course, lies the general fact that the chromosomes are morpho- 

 logical elements of the greatest constancy and importance. 



