258 KANSAS UNIVERSITY SCIENCE BULLETIN. 



conception of the numerical and spatial relations of the chro- 

 mosomes may be observed than was possible in the earlier 

 stages. At this time it is very apparent from inspections of 

 polar views that the number of ordinary chromosomes is six- 

 teen. There exists also some variation in the sizes of the chro- 

 mosomes, and two of larger dimensions may always be recog- 

 nized on the peripherj'^ of the equatorial plate, and usually 

 together. The outlines of the chromosomes, in polar view, are 

 approximately circular and no rings are visible. A very com- 

 pact equatorial plate is formed with little space between the 

 chromosomes. No regular arrangement of the elements was 

 noticeable, except that the larger ones are on the periphery. 

 Occasionally the accessory chromosome, in addition to the six- 

 teen ordinary ones, may be seen in polar view (figs. 6, 7, 8, 9) . 



A lateral view of the first spermatocyte metaphase shows a 

 series of dumb-bell-shaped chromosomes with now and then a 

 ring-shaped element. There may also be observed lob ate chro- 

 mosomes like a letter E with the middle bar absent. In a large 

 majority of cases the accessory chromosome has already passed 

 to one pole of the spindle without division. The position de- 

 scribed in my former paper ('99, p. 190) is unusual, and there 

 can be no question that the accessory chromosome is not di- 

 vided in the first spermatocyte mitosis (figs. 10, 11). It is im- 

 possible to ascertain the character of the chromosome division 

 in this generation of cells, for the elements are almost all of 

 one shape, and offer no means for the determination of planes 

 or axes. In the case of the ring- and E -shaped chromosomes, 

 because of their position, I am inclined to believe that there 

 occurs a segregation division. I hope to present in another 

 paper the evidence for my belief that rings of this character 

 are divided transversely while those that lie in the plane of the 

 equatorial plate, such as I have described for other species of 

 the Orthoptera, are longitudinally cleft. The proof in support 

 of this view seems to me very conclusive, and offers the means 

 for reducing much of the discrepancy that exists in the conclu- 

 sions of different observers. Xinhidium, however, does not 

 afford material suitable for the determination of such points, 

 and I have made no effort to utilize it in this way. 



Telophases of the first spermatocyte show the very large 

 accessory chromosome in one daughter-cell and absent from the 

 other. There is evident at this period a marked separation of 



