STUDIES ON CHROMOSOMES 389 



we may here find a clue to the more extreme forms of diffusion 

 observed in the oogenesis of many animals and in the ordinary 

 somatic nuclei. '^ 



As regards the problem of synapsis and reduction, the existence 

 of the synizesis and diffuse stages renders the Hemiptera very 

 unfavorable objects as compared with Tomopteris or Batraco- 

 seps, and we are here thrown back upon analogies. Emphasis 

 may however be laid upon the essential similarity of the prophase- 

 figures in Tomopteris and these insects; and if my interpretation 

 of the diffuse stage be correct, it is probable that these figures 

 have essentially the same mode of origin from the diplotene- 

 threads. Following this analogy, I provisionally assume the 

 latter to follow an original side by side union, or parasynapsis — 

 not an end to end union or telosynapsis, as was assumed byPaul- 

 mier, Montgomery and (in Orthoptera) McClung, Sutton, and 

 more recently by Davis. In his latest paper ('11) Montgomery 

 rejects his former interpretation in favor of the one here adopted. 

 If the double cross-figures (or the tetrad-rods) arise in the 

 manner assumed, it is clear that their 'transverse' division is 

 the last remnant of the original longitudinal cleft of the diplo- 

 tene-thread; and it is certain, as Paulmier first showed, that this 

 'transverse' division corresponds to the planeof the first sperma- 

 tocyte-di vision. If we accept this, and if the original longitudi- 

 nal cleft of the diplotene corresponds to the plane of synapsis, 

 it follows that the first spermatocyte-divisioji is the 'reduction- 

 division,' as Paulmier and Montgomery concluded. I repeat, 

 however, that this conclusion is here adopted only in a tentative 

 way; since the case is by no means proved; and, as will appear, 

 my conception of the reduction-division differs materially from 

 the one more commonly held. 



As regards the sex-chromosomes, on the other hand, all is clear. 

 The observations here recorded remove every doubt, I think, in 



1^ A more or less wide divergence of the longitudinal halves of the diplotene- 

 threads appears to be the rule among many animals and plants. It has been espe- 

 cially emphasized by Gregoire ('04, '10) who has called attention to the striking 

 contrast in this respect between ^he bivalent chromosomes of the maturation- 

 period and the longitudinally split spireme-threads of the somatic divisions. See 

 also Strasburger, '09, pp. 98 to 100. 



