-l'I KMATOGENESIS OF Till-: CADDIS-FLY. 63 



lects more or less to the outside of the nucleus, forming a hollow 

 sphere, just inside the nuclear wall. The remains of the spindle 

 lie near the nucleus as an oval body in a clear zone. This body 

 apparently divide-; as Baumgartner (i) has described and a part 

 of it -eem- to pa-- around the nucleus. Soon after these stages 

 -howii in 1 ii;. ;>4 the nuclei become very sensitive to the fixing 



gents .iml .1- a <"n sequence practically all of them have col- 

 lap-ed. Tin- would probably be the stages when the chromatin 

 i- in the form of .1 tube with a very small lumen. The fully 

 developed -perm i- -hown in Fig. 36. 



I In detail- v;i\ en .ibovc arc in the main the same as h,i\ e 1 >een 

 dc-i ribed many time- for animal spermatogenesis. In the action 

 oi (.mi- p nt- of the ( rll mechanism there is always quite a little 

 variation and it i- to the-e parts that particular attention \\ ill be 

 p.iid. The t\\o -tun tures to be especially noted are the centro- 

 .iinl tin- "chrontatin nuclcolus." 



1 UK CENTROSOMK. 



h i- impo ibU- to locate the centrosome except \\hen it is 

 o. iiipxiii- a p.i-ititui at the end of the sj>iiulle. At any other 

 pl.iee it i- ini| >o--il ! to say wlii-ther a certain dark staining body 

 i- .i centrosome or not. HtKlies appear near the nucleus where 

 the ceiitro-. .me -hould be anil have all the characteristics of 

 Centrosomes, U-in^ darkly staining little granules surrounded b\- 

 .1 i lear space -ind trequeiitly with what ajjpear to be rays running 

 up to them, ilioii^h these latter are very faint and small, but the 

 miihiplii it\ of -urh bodies makes it impossible to trace the hi-tory 

 of the ( entrosome u ith any certainty for any distance or even to 

 be -tire tli.it -mil a centre is present at all times. At tin end- 

 ot the -pimlle. it- position is of course always easy to determine. 

 It u-u.ilh lir-t appears so as to be definitely recogni/ed at the 

 tetrad >ta^e a- a dark body on the plasma membrane. It ap- 

 peal - as .1 \ ei \ -mall blark -raiiiilc on which the fibres terminate 

 and lie- ju-i in-ide the pla-ma membrane and apparently at- 

 lai hed to it. hi- mo-t eon-|m-uous during the meta]iha-e \\ lien 

 the >harp ])ointed -pindle end- in this little grannie at the plasma 

 membrane. It ma\ be. of course, that here it i> not a bod\' at 

 all but onl> the i omnion point of att.ichment of the Iil)res of the 



