394 Edmund B. Wilson. 



of the idiochromosomes takes place the bivalent chromosome- 

 nucleolus again separates into its univalent constituents in the early 

 prophases of the first maturation-division. This process, which 

 at first greatly puzzled me, occurs at the time just preceding the 

 concentration of the larger chromosomes into their final condensed 

 form (corresponding to stage h in Lygaeus). In cysts of this period 

 every stage may be seen in the transformation of the single chro- 

 mosome-nucleolus into an asymmetrical tetrad, consisting of two 

 symmetrical dumb-bell shaped bodies of unequal size (Fig. 77), and 

 the separation of these unequal dyads to form the idiochromo- 

 somes of the first division (cf. Figs, jh, /, y). It is evident that 

 in these cases the final reunion or conjugation at the end of the 

 first division is not a primary synapsis, but a secondary process. 1 

 The facts observed in Brochymena make it very probable that 

 when only a single chromosome-nucleolus is present in Lygaeus, 

 Coenus and the other forms, it is there also a bivalent body as 

 Montgomery assumed; but the uniform separation of the idio- 

 chromosomes in the first division of all the eight species I have 

 examined is almost a demonstration that in all the forms a division 

 of the bivalent body must occur. On the other hand it seems 

 equally certain that in many of these forms the idiochromosomes 

 may fail to unite at the period of general synapsis, and may remain 

 separate through the whole growth-period; and in Brochymena 

 the same cysts that show the division of the single idiochromo- 

 some-tetrad may also contain nuclei in which the dumb-bell 

 shaped idiochromosomes are widely separated. In such cases it 

 seems probable that the conjugation of the idiochromosomes at 

 the close of the first spermatocyte-division must be regarded as a 

 true or primary synapsis that has been deferred to this late period. 



DISCUSSION OF RESULTS. 



The most essential fact brought out by a study of the idiochro- 

 mosomes is that in Lygaeus, Ccenus, Podisus, Euschistus, Brochy- 

 mena and Trichopepla a dimorphism of the spermatozoa exists, 

 there being two groups equal in number, both of which contain 



'The division of the bivalent chromosome-nucleolus is similar to the process described by Gross 

 ('04) in Syromastes; though it occurs at a much later period. For reasons that will appear in a 

 subsequent paper, I am, however, skeptical in regard to Gross's conclusion which is based on a study 

 not of the idiochromosomes but of paired microchromosomes similar to those of Anasa or Alydus. 



