CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETEROPTEKA. 135 



by reason of its large size and deep stain, it does not remain completel}^ compact and 

 dense, but sometimes shows a loosening of its texture. Besides this there is a second 

 and much smaller monosome {Mo. 2, Figs. 276, 277), usually rod-shaped in the synapsis 

 and more spherical later, generally separated from the nuclear membrane ; it shows 

 no signs of a longitudinal split. Both of these monosomes increase considerably in 

 volume, then decrease again during the following prophases. Plasmosomes seem to 

 be absent, and there is no complete rest stage. 



First 3Iaturatio7i Division. — In the prophases (Fig. 278) the smaller monosome 

 (ilfo. 2) can be recognized by its unipartite aspect, the larger one {Mo. 1) by its form 

 of two more or less parallel rods. All the other elements are quadripartite autosomes 

 except the two smallest; one of the latter has the shape of two apposed spherules 

 (/>/. 1, Fig. 278), while the other {Di. S) eventually assumes this form but is the latest 

 of all the chromosomes to become dense in structure ; these two smallest elements are 

 probably bivalent diplosomes, because though they are not distinguishal>le during the 

 growth period they differ from the monosomes by much smaller volume and different 

 form ; and T judge that each is bivalent on account of its behavior in the two matu- 

 ration mitoses. 



In the spindle there are alwa3's 16 elements, all placed in one plane except one 

 {Mo. 2, Figs. 279-283) that lies invariably nearer one spindle pole than the other. 

 This is the only one that seems unipartite, and is the smallest of all ; it is undoubtedly 

 the smaller monosome, and has decreased in volume since the pro[)hases. Of the re- 

 maining elements one is the larger monosome and it can be recognized on side view 

 only, and then because its long axis lies in the plane of the equator {Mo. 1, Fig. 283). 

 Then there are 2 diplosomes {Di. 1, di. 2) which are very small and next larger than 

 the smaller monosome. The 12 remaining elements are 12 bivalent autosomes, each 

 quadripartite; one of them, that marked t in the Figs. 279-281, is unusually large, 

 and for this reason I had originally (1901/;) suppo.sed it to be quadrivalent ; l>ut since 

 there are 30 elements in the spermatogonium this one cannot be more than bivalent. 



The 12 bivalent autosomes divide transversely to their lengths, therefore probablj' 

 reductionally. The two diplosomes also divide, but in what way I have no means of 

 determining. The larger monosome divides and equationally. But the smaller mono- 

 some, which always lies a little out of the plane of the other elements, never divides 

 but passes wholly over into that spermatocyte of the second order to which it is near- 

 est. Half the second spermatocytes receive, accordingly, 16 chromosomes, and half 

 of them 15, the one that may be lacking being the smaller monosome. 



Second Maturation Division. — Pole views of the second spindle are shown in 

 Figs. 285, 286. One of them is a cell containing the smaller monosome {Mo. 2, Fig. 



