PLACEMAN: THE SPERMATOGENESIS OF SCOLOPENDUA. 95 



together during the early prophase, as is usual in insects, hut are small in 

 proportion to the size of the nucleus and are evenly distributed through- 

 out the enclosing vesicle. This makes it possible to obtain an accurate 

 count without any chance of error. That this process of pseudo-reduc- 

 tion is universal, at least in the arthropods, would seem probable from 

 its occurrence in four of the classes of tins branch (Crustacea, Myriopoda, 

 Protracheata, and Insecta), and it may well be common to all sperm 

 cells. 



Perhaps the most interesting and valuable recent contributions to the 

 question of synapsis are those of Sutton (:02. :03). This is true, both 

 of the convincing manner in which he demonstrates the character of the 

 union, and of the significance ascribed by him to the process. In the 

 spermatogonia of Brachystola there are twenty-two chromosomes, ex- 

 clusive of the accessory chromosome. The ordinary chromosomes show 

 such constant relations in the matter of size that they may be readily 

 separated into eleven groups, the two forming each group being of 

 approximately equal size. Sutton shows that the two chromosomes 

 of any given size conjugate during synapsis by an end to end union and 

 during the division of the second spermatocyte are separated at this 

 point of union. 



The significance ascribed to this process is of the greatest interest. 

 Following the earlier suggestion of Montgomery (:Ol), he believes that 

 the two chromosomes uniting in synapsis represent similar characteristics, 

 one being derived from each parent. Then by the reduction division 

 it is brought about that no character possessed by the mother cell will 

 be either duplicated or lacking in any one spermatid or mature egg. 

 This serves as an explanation of the well-known fact that the nucleus 

 of either spermatozoon or egg possesses all the necessary characters for 

 the formation of a normal embryo (merogamy, artificial parthenogenesis). 

 However, the chromosomes going to one cell are not all necessarily either 

 maternal or paternal, and thus there arise chances of variation in the 

 offspring proportional to the number of chromosomes characteristic of 

 the cells of the organism. Thus the phenomena of the chromosomes is 

 brought into correlation with the known facts of heredity and variation. 



D. Formation of Tetrads. 



The literature upon tetrad formation and chromatin reduction has 

 been so often and so thoroughly discussed that it will not be necessary 

 for me to treat of it at great length. The observations seem in general to 

 warrant the conclusion that in vertebrates the maturation divisions involve 



