110 CHIM^MOSOMES IN THE SPERMATOGENESIS OF THE HE.MIPTERA HETEROI'TERA. 



NABIDJi. 



13. Nabis annulatus Reut. 



On my preiiarations there were no stages of the spei'niatogonia or earUer portion 

 of the growth jjeriod. 



First Maturatlou Mltods. — Very early prophases show G autosomes in the form of 

 long loops which are evidently to be considered tetrads with a very wide longitudinal 

 split. Besides these there is apposed to a plasmosome {PI, Plate X, Fig. 81) a still 

 larger body (!>/), safraninopliilous, of uneven contours, wliich the later history shows 

 to be a number of allosomes in close juxtaposition. Later the G autosomes shorten 

 and condense, and then each appears to consist of two parallel univalent elements 

 each longitudinally split, as illustrated by those marked m in Figs. 81-83 ; each of 

 these gradually condenses into a tetrad composed of four parallel rods, whereas in 

 most other Hemiptera the univalent elements come to lie end to end ; further, the 

 longitudinal split remains oj^en instead of closing temporarily. In these later pro- 

 phases the safraninopliilous body [Di, Fig. 81) separates into 4 allosomes, while the 

 plasmosome to which it is attached gradually dissolves (Figs. 82, 83). Two of these 

 compact allosomes are (piadripartite [DL '2), and each of these is therefore probably, 

 and the later history confirms this decision, a bivalent, longitudinally split chromo- 

 some ; these are the ones lettered Di. 2, di. 2 and Bi. 3, di. 3 in Figs. 82, 83 and 85. 

 Each is, that is to say, a bivalent diplosome with its components in close contact and 

 with these components of approximately equal volume. But the remaining pair of 

 allosomes consist of the largest and the smallest respectively, and are very unlike in 

 volume, while each is a dyad and not a teti-ad {Di. 1, di. 1, Figs. 82-85). These rela- 

 tions cannot be determined as long as these bodies ai'e in close contact, but very clearly 

 as soon as they become separate. These three pairs of diplosomes are readily distin- 

 guished fi'om the autosomes by their dense and rounded form and their strong affinity 

 for the safranine stain. There are accordingly three pairs of diplosomes in the sper- 

 matocyte, two of them tetrads, and one pair with widely separated components of 

 unequal volume. 



Pole views of the first maturation monaster show always 10 chromosomes (Fig. 

 86). Eight of these are clearly quadripartite, as can be readily determined when the pole 

 view is slightly oblique as that of the figure given, and these must correspond to the 

 8 tetrads of the prophases, namely, to the 6 bivalent autosomes, and to the 2 bivalent 

 diplosomes marked Di. 2, di. 2 and Di. 3, di. 3 ; which two, however, are these par- 

 ticular diplosomes, cann(jt be determined with certainty in the stage of the equatorial 

 plate. The two remaining elements are not tetrads Ijut dyads, they are of unequal 



