12G CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETEROPTERA. 



Stages later than tliat of Fig. 210 Avere not found; but from the form and posi- 

 tion of the chromosomes there it is probable that the 5 autosomes divide equationally, 

 that the small diplosome {Di. 1) divides in the same way, but that the bivalent diplo- 

 some {Di. 2, di. 2) divides reductionally. 



Accordingly, there are two pairs of diplosomes ; in the maturation mitoses the 

 larger of them divide first equationally then reductionally, the smaller first reduc- 

 tionally then equationally, so that the phenomena of division are reversed in the two 

 pairs. 



Litcyaturc. — In my preceding account (1901c) the spermatogonia! number of 

 chromosomes was erroneously given as 1(5, since J had counted two of the larger con- 

 stricted ones as two each ; and the contrasted behavior of the two diplosome pairs was 

 overlooked because the second maturation mitosis was not studied. 



28. IcHNODEMUs FALicus Say. 



Spennatogoiiic Division. — On the clearest pole view (Plate XII, Fig. 211) 15 ele- 

 ments could be counted. There must, however, be 16 present at this stage as will be 

 shown by the later ones. Further, 4 must be diplosomes, of which the two marked 

 Di. 2, di. 2 must be the larger pair of diplosomes and Di. 1 be one component of a 

 smaller pair. The 12 largest bodies are certainly autosomes. 



Growth, Period. — Six bivalent autosomes are found in the form of V's or, as fre- 

 quently, parallel rods, that is, they may conjugate end to end or side to side; each 

 becomes longitudinally s])lit. Sharply distinguishable from these during the whole 

 growth period are 2 deep-staining, compact bodies, markedly different in volume, 

 attached to the nuclear wall [Di. 2, di. 2, Figs. 212-214). These are the larger pair 

 of diplosomes and represent the two similarly lettered ones in the spermatogonium 

 (Fig. 211). They are rarely in contact with each other so that it may be that they do 

 not conjugate. The larger of them {di. 2, Fig. 214) becomes longitudinally split, this 

 split continuing up to the following mitosis ; the smaller one is elongate, but only 

 in rare cases does it show signs of division {Di. 2, Fig. 213). Towards the close of the 

 growth j^eriod, which is not a rest stage, a large irregular [)lasmosome is developed 

 [PL, Fig. 214), to which one or the other of the large diplosomes is frequently attached. 



Fird Maturation Division. — In the early prophases reappear the pair of small 

 diplosomes [Di. 1, di. 1, Fig. 215) ; they are not connected and each is at first a small 

 bent rod with uneven contours and a longitudinal split. Each condenses and shortens, 

 the split still maintained {Di. 1, di. 1, Figs. 216-219), and they usually do not conju- 

 gate until the stage of the equatorial plate. The pair of larger diplosomes are recog- 

 nizable by their greater size {Di. 2, di. 2). Then there are in each nucleus 6 bivalent 



