CHROMOSOME COMPLEX OF SPERMATOCVTES. 333 



prophase, and their position in the equatorial plate, together with 

 their relation to the mantle fibers in the metaphase and ana- 

 phase. I cannot stop here to enter into a criticism of the 

 entirely erroneous interpretations of these rings by which de 

 Sinety proves a double longitudinal division and Montgomery 

 a prereduction. 



The probability of the occurrence of equational and reduction 

 divisions in one mitosis is one against which I have previously 

 argued strongly, because whenever reported the cases have lacked 

 the support of thorough observation. The failure of the acces- 

 sory chromosome to divide in the first spermatocyte mitosis, 

 which is equivalent to a reduction, was the first authentic instance 

 of this complexity, and now we have the case of the multiple 

 chromosome in Mermiria, which furnishes further indisputable 

 evidence of the two types of division in one mitosis. In the face 

 of this evidence we must accept the fact and seek an explanation. 



This is a matter of secondary importance compared with the 

 question of the probable distribution of the chromosomes to the 

 four spermatozoa arising from each first spermatocyte. It would 

 appear at first glance that from the complexity introduced by the 

 unusual arrangement and distribution of the chromosomes in 

 Mcnniria there must be a great many possibilities for variation 

 in the chromosome complex, but if we accept the theory of the 

 continuity of the chromosomes and remember that always the 

 same complex appears in each first spermatocyte, then we must 

 conclude that the mechanism of separation, distribution and re- 

 combination is a precise one. On the other hand, it is evident 

 that the spermatozoa cannot all be alike, and we have to deal 

 with different categories. This fact has been recognized for some 

 time with regard to the accessory chromosome, but by the un- 

 usual combination of parts in the multiple chromosome (decad) 

 of Mcnniria there comes about an additional differentiation. 

 The precise nature of this I have not been able yet to determine, 

 on account of the absence of some prophase stages in my ma- 

 terial. This hiatus I hope to remedy by the acquisition of a 

 larger series of slides, but meanwhile we are sure of some facts 

 and these are sufficient for the present discussion. 



Since entire tetrads pass into the second spermatocyte there 



