ORTHOPTERAN SPERMATOGENESIS 703 



split and at the same time show the space between homologous 

 chromosomes. This produces a double V, the enclosed spaces 

 of which are in planes perpendicular to each other. In such a 

 structure it is not possible to distinguish these apart and it is 

 theoretically possible that, in the subsequent concentration, 

 either of these might represent the plane of division in the first 

 spermatocyte. Such forms do not appear in the metaphase and 

 they are comparatively rare in the prophase. In principle all 

 the tetrads would be of this type if there were parasynapsis and 

 equal separation of the chromatids. 



3. Cross-shaped tetrads. This simple form of chromosome has 

 been subjected to the most inexplicable misinterpretation possible, 

 ranging from a practical denial of its presence by Wilcox, Otte, 

 Vejdovsky and Montgomery to the strange accounts of its 

 formation by de Sinety and Brunelli. It is correctly represented 

 in prophase by Stevens, Sutton, Davis, Otte and Robertson, 

 and, in a modified form, due to the shortness of the chromatids, 

 by Jordan. Failure to appreciate the true composition of this 

 element can be due only to very imperfect or superficial obser- 

 vation. Nothing is clearer than the fact that the four arms, at 

 the middle, lie in the same plane and that each is split along its 

 length. Such appearances as are shown by de Sinety in figures 

 123 and 124, and by Brunelli ('11) in figures 12 and 13 never 

 occur, and Montgomery's supposed X-shaped chromosomes in 

 figure 31 is in reality a V with the synaptic ends extended. Both 

 de Sinety and Brunelli are inexcusable for describing the cross 

 as two superimposed chromosomes. Even the most casual in- 

 spection will show that the limbs lie in the same plane and have 

 a clear diamond-shaped opening where their clefts intersect. 

 Such figures are shown in the first spermatocyte of Brachystola 

 by Sutton ('02, figs. 6 and 7) ; by Stevens ('05 in Stenopelmatus, 

 figs. 56, 58, 59 and 64) ; by Robertson ('08 fig. 29) ; by Davis 

 ( '08, figs. 60, 166 to 168) ; by Nowlin ('08, figs. 3, 5 and 7, pi. 29, 

 and figs. 3, 4, 13, 14, pi. 32); and by Carothers ('13, fig. 28). 



4. Ring-shaped tetrads. Such tetrads I have always regarded 

 as of the greatest significance, and I am now even more firmly 

 of the opinion, if possible, that they present most clearly the real 



