CIIUOMOSOMES IN THE SPERMATOGENESIS OF THE HEMII'TERA HETEROPTERA. 113 



the smaller ones [x) are bodies represented in neither the spermatogonic nor the sper- 

 matocytic mitoses. In the case figured (Fig. 9i)) the bivalent diplosome has each 

 component longitudinally split. 



First Matunilion Division. — In the early prophases (Figs. 100, 101) a bivalent 

 diplosome [Di, rf/) is frequentl}^ to be seen lying near the monosome {Mo), which might 

 indicate that previously it had been in contact with it, from which it would appear 

 possible that when the diplosomes are not discernible in the preceding rest period it 

 is because they may be closely applied against the monosome. The diplosomes seem 

 not to increase in size during the growth period. In these prophases the longitudinal 

 split of the monosome again appears. 



In the chromosomal plate (Figs. 102, 103) there are always present 1 bivalent 

 di[)losome [Di, di) that divides reductionally, and 1 monosome {Mo) that divides 

 through the plane of its longitudinal split. There may be either 5 bivalent autosomes 

 (Fig. 102, A, a-E, e) all of which divide reductionally ; or 4 bivalent autosomes {A, 

 a-C, c, E, e, Fig. 103) and 2 univalent ones {D, <!) ; in the latter case the 2 univalent 

 ones are regularly of the same form and volume, and therefore are evidently ones that 

 had either failed to conjugate or, more probably, ones that had precociously separated 

 from each other after conjugation, and which in this mitosis pass without division into 

 opposite daughter cells, i. c, divide reductionally as do the other autosomes. The 

 longitudinal split is well marked upon one or two of the larger autosomes. 



Second Maturation Division. — Here there are always 7 elements (Fig. 101, where 

 one of the autosomes has not yet taken its place in the equator of the spindle). The 

 smallest, the diplosome {Di), regularly divides, and so do the 5 autosomes, all of these 

 equationally. But the monosome {Mo) shows no sign of any division and passes bodily 

 over into one of the spermatids. The latter show correspondingly either 6 chromo- 

 somes (Fig. 105) or 7 (Fig. 106), the monosome being absent in the former case ; the 

 minute element in each spermatid is a diplosome. 



Literature. — My preceding accounts ( 1 90 1 r/, /;) were correct in the main, stated 

 the spermatogonia! number of chromosomes accurately, the variation in number in 

 the first maturation spindle, and the behavior of the monosome in the matui'ation 

 divisions. But what escaped me then was that the large allosome of the gi'owth 

 period is the monosome and not the bivalent diplosome. 



15. CoRizus ALTERNATUs Say. 



Spermatogonic Divisions. — There are 13 chromosomes (Plate X, Fig. 107). The 

 smallest elements, of slightl}' different volume, are the diplosomes {Di, di). Then 5 

 pairs of autosomes {A, a-E, c) ; of these the largest pair {A, a) is composed of 2 rela- 



A. P. S.— XXI. L. 23, 7, '06. 



