ORTHOPTERAN SPERMATOGENESIS 705 



cannot be laid upon the necessity for a thorough understanding 

 of the early formative periods in the history of the first spermato- 

 cyte chromosomes." In my own work, and in that of my stu- 

 dents, this stage has been carefully studied. It is nothing less 

 than amusing therefore to read Brunelli's statement that I have 

 not appreciated the real nature of the tetrads because of failure 

 to study their formation in the prophase. On the contrary, it 

 was just through such a study that I was saved from the errors 

 of de Sinety, Davis, JNlontgomery, Buchner and Granata in in- 

 terpreting the ring chromosomes, as I hope to show. 



The rings first appear clearly in the late prophase, following 

 a period in which the chromatin is much diffused and almost 

 unstainable. At this time the general forms of the chromosomes 

 may be recognized in the hazy shapes within the nucleus, and 

 among them are irregularly outlined rings. Very shortly after 

 this, sometimes even in other regions of the same cyst, a central 

 thread of darkly staining chromatin, split along its length, appears 

 in each of the diffuse masses. Immediately one recognizes in 

 the outlines thus established the various familiar rods, crosses, 

 V's, rings, etc. Early in this period of their development the 

 four chromatids of each chromosome lie approximately parallel 

 and the rings are therefore much flattened, but with the rapid 

 concentration of the hazy masses around their axial darkly 

 staining threads, separation occurs and the rings spring open 

 as if they were composed of elastic rods released from constraint. 

 Such rings are shown in figures 99 to 105 and 114 to 123. 



Misinterpretation begins here. As may be seen by inspection 

 of figures 100, 115, 118 and 119, the rings are not plain bands but 

 on one side there is an enlargement, and not infrequently there 

 may be a corresponding one opposite this (fig. 122). When such 

 structures are viewed laterally it is seen that each elevation from 

 the ring is continuous with the half of the ring upon the same side 

 as that on which it lies. When viewed en face (fig. 127) the ring 

 appears as a cross, in which case the upper and lower arms are 

 represented by the elevations which were seen in lateral view. 

 From this it is clear that each lateral half of these elevations is 



