CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETEROPTERA. 143 



account of the great mass of evidence uj^on this question, wiiich lias l^een fully dis- 

 cussed in earlier papers of mine, we shall assume it as proven that in the Hemiptera 

 the autosomes divide in this sequence. Therefore, the allosomes being modified 

 chromosomes, those allosomes that divide in the same way as the autosomes do would 

 be genetically closest to the autosomes. Such are the diplosomes of the C'oreidte 

 (except the smaller pair of Sijromastes), of the Reduviidpe and Tingis, Calocoris, the 

 smaller diplosomes of Nabi-'<, and one of the diplosorae pairs of Peliopelta and Ichno- 

 demus. These diplosomes correspond to the " M-chromosonies" of Wilson. They are 

 in most cases the smallest of all the chromosomes, sometimes very minute, and, except 

 in TiiKjis, are only very slightly different in size. Probably those of them that do not 

 remain dense but become reticular in the growth period, as is the case in Alt/dun, 

 Metapodms, (Miancala and Calocoris, are the least modified, because the most similar 

 in behavior to the autosomes. The other kind of diplosomes correspond to what 

 Wilson has called the "idiochromosomes," and he first distinguished between these 

 and the preceding kind. These usually do, sometimes do not, conjugate in the early 

 growth period, enter the chromosomal plate of the first maturation mitosis separately, 

 and divide there equationally, then in the second spermatocytes (usually but not 

 always after a conjugation in the center of the chromosomal plate) divide reduction- 

 ally ; they alwa3's remain more or less dense and compact during the growth period, 

 and are usually very difterent in volume, though Wilson has shown that in Nczara 

 they are nearl}' equal. Both kinds of diplosomes may occur in the same cell. 



We do not know intermediates between these two kinds of diplosomes, though 

 there can well be no doubt that the second is a further modification of the first ; 

 because sometimes in the first type the diplosomes may be unequal, and in the second 

 type sometimes almost equal in size, size difference cannot be taken as a criterion of 

 them, and for this reason it seemed to me inadvisable to consider them as quite dif- 

 ferent allosomes as Wilson has done. The most striking difference between the two 

 types is the discord with regard to the reduction division ; in the first tyjie it occurs 

 in the first matui-ation mitosis, in the second type in the succeeding mitosis. This 

 certainly stands in some relation with the time of conjugation of the elements of the 

 pair, which in the first type is always early in the growth period, while in the second 

 type it may occur then, but frequently does not take place until the stage of the 

 second spermatocyte or may not occur even at that stage. From the series of facts 

 now at hand, we niight conclude that the genesis of the diplosomes is as follows. 

 First a pair of autosomes V)ecame modified so as to retain their compact nature during 

 the growth period, still maintaining tlieir a])proximate equivalence in volume. Be- 

 cause such allosomes are usuall}' very small, we might conclude also that they arose 



