4O2 EDMUND B. WILSON. 



stages) is stated to be present in all the secondary spermatocytes 

 and also in all the spermatids, where its maximum size is attained. 

 Duesberg ('08) found it only after the first division and believed 

 "qu'il persiste pendant la mitose et se divise pour son propre 

 compte," though the division was not actually seen. Both 

 observers agreed that it degenerates in the spermatid (cf. also 

 Meves, '99). Regaud ('10) again describes it in the rat, but 

 finds that it fragments into small granules during the first division 

 and is afterwards reconstituted to form a single body which 

 passes to one pole in the second division, beyond which stage its 

 history was not traced. Such a process of fragmentation cer- 

 tainly does not take place in Pentatoma, nor can there be the 

 least doubt of its absence from the greater number of spermatids. 



I suspect that the body described by King ('07) as an "aero- 

 some" in the spermatogenesis of Bufo belongs in the same general 

 category, though it is described as having a very different history 

 from that seen in either the insects or the mammals. Lastly, 

 I may point out the probable identity of the chromatoid body 

 in Pentatoma with that described by Doncaster ('10) in the gall- 

 fly Neuroterus, which likewise passes to one pole in the spermato- 

 cyte-di vision. This observer tentatively suggests a possible 

 connection between this body and sex-determination; but the 

 facts seen in the Hemiptera evidently lend no support to this. 

 In none of these cases is the real nature of this body yet clear. 

 In Pentatoma it is obviously not a centrosome, centriole, acrosome 

 or extruded nucleolus. Since no definite idiozome is seen in 

 the spermatocytes I suspected for a time that it might be such 

 a body; but this too seems to be excluded by the conditions de- 

 scribed in the mammals, where an idiozome is also present. 



The nature of the chromatoid body thus remains problematical, 

 but the facts are worthy of serious attention for another reason. 

 Were the chromosomes very small, numerous, closely crowded, 

 or otherwise unfavorable for exact study, and could not the 

 entire history of the chromatoid body be so clearly traced, even 

 an experienced observer might fall into the most confusing 

 error concerning the relations of the chromosomes. It may seem 

 superfluous to urge the danger of confusing with chromosomes 

 other compact and deeply staining bodies that may lie near or 



