ROBERTSON: THE CHROMOSOME COMPLEX. 287 



SYRBULA ADMIRABILIS. 



With the mistakes of Montgomery in mind I have been care- 

 ful in this work to use the testes of several individuals, about a 

 dozen or more, and to examine and draw from these as many 

 cells as time would permit, showing the full number of the chro- 

 mosomes. The sections were cut twelve micra thick, and so in 

 a great many instances I have been able to show cells that con- 

 tained in a single section the whole number of chromosomes. 

 In every case so far examined the numbers for the first sper- 

 matocjrte and the spermatogonia have been respectively twelve 

 and twenty-three. In addition to this the cells of Syrbula 

 fusca-vittata, another species of the same genus that was ex- 

 amined, showed the same numbers. In view of these facts 

 Syrbula acuticornis either must be radically different in germ- 

 cell structure from the other members of the genus or there 

 must be some mistake in the reported numbers. Until some 

 of the material is inspected I am inclined to believe the latter, 

 for Montgomery in his meager illustrations has not furnished 

 proof sufficient to establish the truth of his statements. 



The next point in which we shall have to disagree with 

 Montgomery is in regard to the formation and division of the 

 tetrads. The conditions as they occur in Syrbula admirabilis 

 have already been described. There remains to be considered 

 the differing conditions that he has described for acuticornis, 

 and concerning which I believe he was mistaken. It seems that 

 his first object here was primarily to find evidence to confirm 

 his old "pre-reduction" theory, and accordingly he has forced 

 the interpretation of every phenomenon, whenever possible, 

 to its support. Generally speaking, it matters little whether 

 reduction occurs in the first or the second maturation division, 

 for the result is the same either way. In fact, some species 

 of the Acrididse (McClung, '05) show that both reduction and 

 equation divisions among ordinary chromosomes may occur 

 at the same time. 



As before shown, all the tetrads in the late prophase and 

 in the metaphase may be reduced to a common type — that of 

 a rod with one longitudinal and one transverse division. The 

 latter marks the point at which the proximal ends of the uni- 

 valent spermatogonial chromosomes have come together to 

 form the bivalent tetrad. This rod has the ability, as before 

 explained, to change its axis from a longitudinal to a trans- 

 verse direction. The former it possesses in the beginning of 

 synapsis and the latter in the final stages of the first sperma- 



