STUDIES ON CHROMOSOMES 361 



coarse, vague, lightly staining net-like structure, in which are 

 suspended the chromosome-nucleoli and the plasmasome, the 

 latter at its maximum size. In this stage the nuclei remain 

 throughout the greater part of the growth -period. 



h. Early prophases (figs. 105, 107). The staining capacity of 

 the chromatin increases, while more definite and apparently single 

 threads are evident. The sex-chromosomes are more elongate 

 and longitudinally split. The plasmasome now diminishes in 

 size and disappears. 



i. Middle prophases (figs. 108 to 114). The threads rapidly 

 condense, stain more deeply, and draw together to form tetrad- 

 rods, double crosses, double F's, or (rarely) double rings. The 

 sex-chromosomes are short rods, longitudinally split. 



j. Late prophases. In these all the chromosomes are converted 

 into compact, deeply staining dumb-bell shaped bodies, rarely 

 quadripartite in outline, which are ready to enter the spindle. 

 This stage is often found in the same cysts with the preceding, all 

 intermediate gradations being readily seen. 



k. The division-period, including the two spermatocyte-divi- 

 sions. 



1. Differentiation of the spermatids. Spermiogenesis in the 

 narrower sense. 



With various modifications the foregoing stages are found in 

 many Hemiptera, among the best of which for study of the early 

 stages are the pyrrhocorid species Largus cinctus and L. suc- 

 cinctus. Some doubt exists in regard to Stage a; and it is possible 

 that in some forms (of which Largus may be an example) the 

 spermatogonial chromosomes do not lose their identity at this 

 time but give rise directly to the massive bodies of Stage b. In 

 some cases (Largus, Pyrrhocoris, Alydus) the latter part of Stage 

 g is characterized by a second synizesis or contraction-figure, in 

 which the autosomes are again closely massed together. In such 

 cases the early prophases are much more difficult to analyze. 



I feel confident that the seriation of the stages is correctly deter- 

 mined indeed the only possible doubt concerns the earliest pre- 

 synaptic stages. The seriation is indicated by the general topog- 

 raphy of the testis, which consists of very definite lobes in which 



THE JOTJBNAL OF EXPERIMENTAL ZO6LOOT, VOL. 13, NO. 3 



