STUDIES ON INSECT SPERMATOGENESIS. 353 



orderly part of a logical scheme of sperm formation. The aero- 

 some of the animal sperm arises in connection with the Golgi 

 apparatus. When the differentiation of the acrosome has pro- 

 ceeded to a certain point, the remnant of the Golgi apparatus is 

 cast off and probably plays no essential role in the mature sperm. 

 The acrosome is, therefore, no fortuitous body contrived at the 

 last minute from spindle fibers or other chance inclusions, but is 

 to be considered in the light of a cell organ as definite in its origin 

 and role as are the nucleus, centrioles and mitochondria of the 

 spermatid. 



III. The Idiosome and the Structure of the Golgi Apparatus. 

 In conclusion I would like to refer briefly to some general 

 problems connected primarily with the structural aspects of the 

 Golgi apparatus in the male germ cells. In the first place it is 

 to be noted that the structures especially characteristic of the 

 spermatocytes of pulmonates and vertebrates and appearing in 

 the literature under the name of Golgi apparatus, idiosome, Ne- 

 benkern, sphere and other less familiar terms, are all one and the 

 same thing in a topographical sense. They represent a complex 

 of the Golgi apparatus plus a definite substance which forms 

 the idiosome of M'eves. Duesberg ('20) has pointed out, and 

 with reason, that the idiosome has nothing to do with the old 

 attraction sphere of Van Beneden, or, let me add, with the archo- 

 plasm which is supposed to form the spindle and astral rays of 

 the dividing cell. The idiosome of Meves was supposed to be a 

 mass of protoplasm surrounding the centrioles in the resting cell. 

 In other words, it derived its essential character from its relation 

 to the centrioles. Where then, it is fair to ask, is the idiosome 

 in the insect spermatocyte ? Obviously there is none. Fortu- 

 nately the spermatids furnish a clue which leads to a logical solu- 

 tion of the whole problem. In the vertebrate and pulmoiiate 

 spermatid there is seldom, if ever, any relation between the idio- 

 some and the centrioles, but there is a constant relation between 

 the idiosome and the Golgi apparatus, a relation which applies as 

 well to the condition seen in the insects. In the spermatocytes of 

 insects, however, the material of the spermatid "idiosome" is 

 scattered throughout the cytoplasm as a part and parcel of each 



