144 CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETEROPTERA. 



from the smallest pair of autosomes. In tlie next change woald appear a growing 

 disparity in size, which, if our last assumption be correct, would be due not to one 

 becoming smaller and to the other becoming larger, but ratlier 'to one retaining its 

 original volume and to the other becoming much larger. This second step would 

 then be one of differentiation of the two, a becoming-different, probably implying also 

 difference of metabolic activites. Tliis would account for the lessening affinit}^ of the 

 two as exhibited by the protraction of the time of conjugation. Then would be 

 attained the stage of the second type of diplosomes, no longer united but separate in 

 tlio first maturation spindle. And the last step would be that, instead of a reduction 

 division of them in tliis spindle, tliere would take place there an equational division 

 of each. 



In this interpretation, which serves at least to unify the diverse phenomena and 

 is in accord with them, we learn that the two kinds of diplosomes are not really radi- 

 cally different structures, but are rather extremes of a series of modifications. 



We may now pass to the question of the genesis of the monosomes. In most 

 cases these are larger than the diplosomes, sometimes the largest of all the chromo- 

 somes, more rarely are they very minute, as in Calocoris and Lyyus. Usually the 

 monosome remains dense and compact during the growth period, but in (Edancala it 

 becomes reticular and is then practically indistinguishable from the autosomes ; in 

 Hannostes it becomes reticular to a mucli less degree. A monosome like that of 

 (Edancala is clearly a less modified chromosome than are tlie monosomes of the other 

 Hemiptera. Then monosomes may divide in the first maturation mitosis but not in 

 the second (Hi/i/ufrcchus, Lhnnotreckus, Pyrrhocoris, all the Coreidae except Syromastex, 

 Gidancala, and the larger monosome of Calocorh) ; my recent observations show that 

 it is always an equation division, along the line in which the monosome splits in the 

 growth period. But in Archimerus and Banasa, according to Wilson, the monosome 

 does not divide in the first maturation mitosis but does in the second ; I find the 

 smaller monosome of Galocoris behaves in tlie same way, and that in Lyyus the minute 

 monosome does not divide in either mitosis. Thus with regard to the sequence of 

 division, three kinds of monosomes occur in the Hemiptera, of which the kind that 

 divides reductionally in the first maturation mitosis must be considered the least 

 modified because the one that behaves most like the autosomes. 



In an earlier paper (1901?;) I discussed the question of the genesis of the mono- 

 somes ; showed that a monosome might be produced by the hybridization of species 

 with different chromosomal numbers, but concluded tliis to be improbable; and 

 inclined to the view that monosomes arose by some abnormality in mitosis, as by 

 failure of two spermatogonia! chromosomes to separate, which led to my assumption 



