STUDIES ON INSECT SPERMATOGENESIS. 349 



centriole takes us back again to the theories long ago proposed by 

 Van Beneden and Boveri, which contained a nucleus of real 

 truth that later workers have been inclined to overlook. 



This conclusion leads naturally to the consideration of a further 

 possibility, namely, that the division of the cytoplasmic structures 

 may be actually more or less meristic in character. It is impos- 

 sible in the present state of our knowledge to make any very 

 definite statement on this point, which has thus far received little 

 careful study. The results from the study of plant plastids indi- 

 cate something, however, of the possibilities for growth, multi- 

 plication, and distribution in cell division possessed by these 

 bodies. All that is needed to give such behavior the aspect of a 

 meristic division is a definite mechanism for the distribution of 

 the plastids in mitosis. That such an apparatus may exist is 

 demonstrated clearly by the surprisingly complicated mode of 

 division undergone by the Golgi bodies in Euschistus, in which a 

 partially meristic division almost certainly occurs. Such facts 

 as these coupled with the probable relation of many cytoplasmic 

 elements to a mechanism for orderly division indicate a possible 

 degree of cytoplasmic organization which has been little sus- 

 pected. Unfortunately our knowledge concerning it is as yet so 

 meager that we can scarcely even conjecture its nature and ex- 

 tent, but of its possibilities for a highly complex mode of division 

 there can, I believe, be no longer any doubt. 



II. The Origin of the Acrosome. 



No structure in the spermatid has been subject to more contra- 

 dictions of description and homology than the acrosome. 1 It has 

 in general been traced to a " sphere," the relations of which to 

 preceding cell structures have generally been merely inferred and 

 often quite mistaken. But aside from the purely descriptive 

 "sphere," the acrosome has been asserted to originate from the 

 mitochondria, spindle remnants, and combinations of mitochon- 

 dria and spindle remnants, from a centrosome, a centriole, an 



i The term " Akrosoma " was first applied by Lenhossek ('98) to the little 

 granule within the vesicle which is produced by the "sphere" (acroblast). 

 As a matter of fact the term has been applied so widely to the material as a 

 whole which forms the apical body that I see little reason for attempting to 

 restrict the word to Lenhossek's original meaning. 



