188 PROCEEDINGS OF THE ACADEMY OF [Fcb., 



from objects where the phenomena are simpler, where the chromosomes 

 show a definite long axis in early stages, where the mode of formation 

 of the bivalent chromosomes has been worked out, and where forms like 

 rings and crosses do not occur. We must seek to explain the more 

 complex from the more simple, not force an interpretation from the 

 more complicated upon the more simple. The strongest argument for 

 postreduction is that of McClung and Gross, and yet they are reasoning 

 from the basis of perplexing rings and crosses. That such forms can 

 be explained in quite a different manner, and their first division be 

 regarded reductional instead of the second, I have shown for Peripatus, 

 where the series of changes of the linin elements as well as of the chro- 

 matic are clearer than in any object yet seen by me. 



To the idea of postreduction we can apply the criticism " not proven." 

 No one can say that it does not occur, yet I do not hesitate to state as 

 my opinion, coming from observations of some years upon a number 

 of different animal forms, that it will be proved not to occur. And this 

 is said with no intention of any disparity of the work of those who take 

 the contrary stand, for they have accumulated very important and 

 hard-won facts; it is only one of their interpretations that is being 

 criticised. Prereduction is based upon a simpler reasoning and to 

 some extent upon more patent phenomena. 



So we reach the conclusion that maturation phenomena are all of 

 the pseudomitotic tj^e of Korschelt, and only of the prereductional 

 kind. There is a mass of evidence for the view that in all cases the 

 first maturation is the reductional one. Korschelt (1895) has described 

 this for Ophryotrocha, Henking (1890) for Pyrrhocoris, Paulmier (1899) 

 for Anasa, King (1901) for Bufo, Nichols (1902) for Oniscus, Lerat 

 (1902) for copepods, Schockaert (1902) for Thysanozoon, Schreiner 

 (1904) and Marechal (1904) for fishes, McGill (1904) for Anax, Bouin 

 and Collin (1901) for myriapods, and I for Hemiptera of different fami- 

 lies (1898, 1899, 1901a and b), for Peripatus (1900), for salamanders 

 (1903, 1904), and in the present paper for a grasshopper and a spider. 

 And it will be noted that it is the most recent work which supports this 

 view. Quite as conclusive evidence comes from an examination of 

 the heterochromosomes, as we shall see later. Most of the recent work 

 upon the botanical side corroborates this point of view, as that of 

 Gregoire (1904), Rosenberg (1904), Strasburger (1904), Berghs (1904) 

 and Farmer and Moore (1903). 



From what I consider to be the strongest evidence available at the 

 present time we find the following series of phenomena during the 

 spermatogenesis of animals. There are a number of successive genera- 



