1 1 2 CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETERORTERA. 



the other Hemiptera, bivalent in the spermatocytes, and so are probably the 2 quadri- 

 partite diplosomes ; the large and small diplosomes are undoubtedly univalent. 

 Therefore we can postulate for the spermatogonium with a high degree of certainty : 

 12 autosomes, and 6 diplosomes, the components of only one of these diplosome pairs 

 being very unequal in volume. . 



Liferahire. — My preceding account (1901a), which did not extend beyond the 

 first maturation mitosis, was entirely correct except for the conclusion that the sper- 

 matocyte had four bivalent diplosomes. My preparations of Coriscus fenis, another 

 member of the same family, had faded to such a degree that I could not test the cor- 

 rectness of my account of it (1901/^). 



COREIDJE. 



14. Harmostes reflexulus Say. 



Spermatogonic Divisions. — There are 13 chromosomes. One unpaired element 

 (Plate X, Figs. 94, 95, Mo) is the monosome, and it is not the largest. The 2 

 smallest are the diplosomes {I)i, di) and are not quite equal in volume. The remain- 

 ing 10 are autosomes and are seen to compose 5 readily recognizable pairs {A, a-E, e); 

 what is to be noted in them is that the two components of each pair seem to be of 

 slightly different form and volume, as is seen most clearly in the case of the pair A, a ; 

 and perhaps in each pair the larger element ma}' be the maternal one and the smaller 

 the paternal. The components of the 2 or 3 largest j^airs are regularly transversely 

 constricted. 



Growth Period. — The 10 autosomes conjugate to form 5 bivalent ones. The 

 monosome {Mo, Figs. 96-99) remains safraninophilous during this whole pei-iod. In 

 the synapsis (Fig. 96) it becomes elongated and concomitantly more or less bent, 

 thereby showing a great variety of forms ; frequently it is attenuated at the ends and 

 thicker at the middle. In the early "postsynapsis (Fig. 97) it becomes longitudinally 

 split so that the halves sometimes widely diverge from each other and at the same 

 time it becomes less dense and more or less granular, though to much less extent than 

 the autosomes (Fig. 98). In the rest stage, which is complete (Fig. 99), this split 

 becomes more or less closed ; and then the monosome {Mo) has usually a rod shape, 

 shorter than in the synapsis stage, with its arms parallel ; throughout the growth 

 period it lies against the nuclear membrane. I could not distinguish the diplosomes 

 in the earlier part of the gi-owth period before the plasmosome arises. In the rest 

 stage the latter (PI, Fig. 99) is a large body near the center of the nucleus. Quite 

 generally there are attached to its surface about 3 or 4 small safraninophilous bodies ; 

 the 2 larger that may or may not be in contact I take t(j be the diplosomes {Di, di) ; 



