DISCHARGE OF MITOCHONDRIA. 313 



of the spermatid (Figs. 11-19), reaching its maximum size at the 

 stage of Figs. 13, 14. In the mature and nearly mature sper- 

 matozoa it makes a slender rod, joining nucleus and flagellum, 

 and then is seen to have decreased in volume (Figs. 21-24). 

 The flagellum connected with it is a delicate, flattened thread, 

 evidently without spiral membrane or cytoplasmic sheath; in the 

 figure- only its proximal portion is shown. In some cases there 

 appeared to be a spiral skeleton around the nucleus, such as 

 Koltzoff (1908) has recently described in other species: but 

 examination proved that in Peripatns this is occasioned simply 

 by chance wrapping t a tla^dluin around tin- nucleus. 



In the mature e^i; and in deavag . - m> structures were 



tonml in any way resembling the mitochondria of the sperm cells. 



.\b\i-- [QOS, and later paper-) dra\\s the < < mclu-ii >n that 

 mitochondria are important hereditable elements, directing 

 Cytoplasmic acii\iiies as the chnmios, , nu .. direct tho-e of the 



inn leu-, self-perpetuating bodies differentiating during ontogeny 



in I -t of the tibrillar M ruct tires of the b< -d\ . \\"\\ \\\\\ entering 



into the rapidly i^rowini; literature no\\ , \ve will be content with 

 the Matemeiu that a considerable number of investigator- cor- 

 roborate the-e \ie\\s, and that the\' have been especially elabor- 

 ated b\ ( iiiJio-Tos and < .ranat.; 'o x . This hypothe-is more 

 than anv other has dire< ted attention to these bodies. V 

 they are far less conservative and regular than tin- chromosomes 

 in number, form and beha\ior and there is e\ idence that occa- 

 sionally some of them are eliminated during spermio-rnesis. 

 Thus l-'aure-l-'remiet has distinguished tour t \-pe- o| them: 

 i Those that undepco changes of position without profound 

 morphological chanu s. as in mammalian spermiogenesis. 

 Those that at the same time undergo -real structural changes, 

 a- in insect sprrmatids. 131 Those of which oiil\ a part change 

 into the i hondriosome or Nebeiikern of the spcnuatid, while 

 others degenerate, a- in spermatogenesis of certain gastropods. 

 And 4) those that transform wholly or partly into deutoplasmic 

 bodies, some cas,s of oogenesis. The fourth of these classes 

 cannot be said to be definitely established, but there can be little 

 doubt about the evidence for the third. Thus, besides the 

 in -astro pod spermatogenesis studied by Faure- 



