ORTHOPTERAN SPERMATOGENESIS 669 



may be grouped into those having free ends and into those with 

 the ends joined. In most species the majority of the chromo- 

 somes belong to the first of these types, but in some cases more 

 of them are joined terminally. With but few exceptions the 

 members of any complex will be found included in these two cate- 

 gories of the first group. As a result of these flexures the chromo- 

 somes have the form of V's, U's, incomplete rings and complete 

 rings. There is no essential difference between the V-shaped 

 chromosome and the annular form except the extent to which 

 the bending has taken place, or, if there was a previous parasyn- 

 apsis, the degree of separation of the ends. Movement of the 

 chromatids in a plane normal to the plane of the longitudinal 

 cleft results in the formatio^j of various cruciform tetrads. These 

 changes are likely to occur in the smaller members of the com- 

 plex and result in the gradual separation of daughter chromatids 

 and the approximation of homologous derivatives. If carried 

 to the extreme the final result is the production of the same 

 I-shaped chromosome as that present at the beginning of the 

 process but with a transposition of the two axes, the shorter, 

 transverse one coinciding with the original, longitudinal cleft 

 of the two homologous chromosomes. 



Commonly there are flexures of the longer arms of the cross 

 producing a variety of forms depending in part upon the point of 

 view. When placed obliquely the cross may appear X- or K- 

 shaped. Divergence of the four chromatids, except at the ends, 

 constituting the third class of movements, produces the double 

 V-shaped element which appears under a variety of forms when 

 viewed from different angles. One of the included spaces repre- 

 sents the longitudinal cleft of the homologous chromosomes, now 

 bent at an acute angle to each other; the remaining one, lying in 

 a plane at right angles to the first, is the space between the ho- 

 mologous conjugants. It seems impossible to distinguish these 

 apart in most cases. Such a separation of the chromatids is only 

 temporary and a double-V chromosome is never seen in the meta- 

 phase. All the tetrad forms are most clearly differentiated in 

 the late prophase, just before going over into the homogeneous 

 condition of the metaphase, for at this time there is less of irregu- 



